Decoupling citeproc and highlighting-kate from pandoc

Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Pandoc currently includes built-in support for citation processing
(using citeproc-hs) and syntax highlighting (using highlighting-kate).
This is not ideal, since many users don't use either of these features,
and few, I would imagine, use both.  Both citeproc-hs and
highlighting-kate are big libraries, which bloat the pandoc binary.

So I am moving towards decoupling these components from core pandoc.
As noted earlier, I've added a --filter flag to pandoc, which makes it
easy to run JSON filters.  So, to process citations, you'll now do

    pandoc --filter pandoc-citeproc input.txt

and specify the bibliography and csl file in the document's metadata.
Similarly, to highlight source code using highlighting kate, you'll do

    pandoc --filter pandoc-highlight input.txt

(Highlighting style will be specified in the metadata rather than on
the command line.) Both pandoc-citeproc and pandoc-highlight will be
optional add-ons. The packages that provide them will also provide
Haskell libraries for processing citations and doing code highlighting
in pandoc documents, for those who use pandoc as a library.

I have already made the citations change; I have not yet made the
highlighting change, and could still back down if there is a lot of
protest.  The main drawback I can see is that performance will be a
*little* worse -- but the difference is pretty negligible.  The
command line will also be a bit longer for those who use highlighting.
But I think this change makes a lot of sense.  It will also make it
easier for users to use their own highlighting libraries; they just need
to write a simple filter.

Another nice feature I'm adding is inline references.  Instead of
putting your bibliographic references in an external file, you can
now include them directly in the document, in a YAML metadata block.
(This can occur at the end of the document if you like.)
This makes the document more self-contained.  The YAML format is
basically a YAML representation of the JSON format used by citeproc-js.
Here's an example:

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Some claim [fenner12a].

---
csl: tests/mhra.csl
references:
- id: fenner12a
  title: One-click science marketing
  author:
   family: Fenner
   given: [Martin]
  container-title: Nature Materials
  volume: 11
  url: 'http://dx.doi.org/10.1038/nmat3283'
  doi: '10.1038/nmat3283'
  issue: 4
  publisher: Nature Publishing Group
  page: 261-263
  type: article-journal
  issued:
    year: 2012
    month: 3
...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Of course, you can also store the YAML bibliography in a separate file,
and provide it to pandoc on the command line with the main file.

The pandoc-citations package will include an executable, biblio2yaml,
that translates from common bibliographic formats (e.g. bibtex) into
this YAML format.

I'm pretty excited about these changes, which I would like to include
in the 1.12 release.  The release has been held up waiting for Andrea
to release citeproc-hs.  I haven't heard from Andrea in a long time, so
I now have a contingency plan.  I will simply include the code from
citeproc-hs, including a couple of bug fixes, in the pandoc-citations
package.  When Andrea comes back, as I hope he does, I'll remove this
and depend once more on citeproc-hs.  But at least this will let us
go ahead with the release.

If you want to play with the new pandoc-citations stuff, it's on github:
http://github.com/jgm/pandoc-citations
It assumes you've installed a very recent dev version of pandoc and
pandoc-types.

John

Re: Decoupling citeproc and highlighting-kate from pandoc

[Joseph Reagle]

On 08/25/2013 03:15 PM, John MacFarlane wrote:
> and specify the bibliography and csl file in the document's metadata.

Does using the filter require this? Is it because one can't pass
arguments to the filter?

I often already specify CSL in the metadata using my wrapper/build
utilities so that's fine by me. For instance:

~~~~

Title: DRAFT: The Obligation to Know
Author: Joseph Reagle
Date: 20130601
md_opts: --toc --style-csl apa --british-punctuation

~~~~

However, I am concerned with having the bibfile in the metadata because
I generate those dynamically and temporarily to deal with the problem of
my larger bibtex file (~3MB) and pandoc/hs-citeproc's slow processing.
(I extract the keys from the markdown file then use that to generate a
specific and per-document bibtex file.)

Re: Decoupling citeproc and highlighting-kate from pandoc

[BP Jonsson]

[-- Attachment #1: Type: text/plain, Size: 2735 bytes --]

I've been meaning to start using the citation feature for quite some time
and this change might actually get me going. When working on other people's
documents as I do quite a bit I usually get a ready-formatted bibliography
section so then the need hardly arises.

As for using either syntax highlighting or citations but not both it kindof
applies in that I hardly would use both with the same document, but I might
well use them both at different times. Anyway I'm not opposed to the change.

Like Joseph I'm wondering whether it is or will be possible to pass
arguments to filters?

/bpj

Den måndagen den 26:e augusti 2013 skrev Joseph Reagle:

> On 08/25/2013 03:15 PM, John MacFarlane wrote:
> > and specify the bibliography and csl file in the document's metadata.
>
> Does using the filter require this? Is it because one can't pass
> arguments to the filter?
>
> I often already specify CSL in the metadata using my wrapper/build
> utilities so that's fine by me. For instance:
>
> ~~~~
>
> Title: DRAFT: The Obligation to Know
> Author: Joseph Reagle
> Date: 20130601
> md_opts: --toc --style-csl apa --british-punctuation
>
> ~~~~
>
> However, I am concerned with having the bibfile in the metadata because
> I generate those dynamically and temporarily to deal with the problem of
> my larger bibtex file (~3MB) and pandoc/hs-citeproc's slow processing.
> (I extract the keys from the markdown file then use that to generate a
> specific and per-document bibtex file.)
>
> --
> You received this message because you are subscribed to the Google Groups
> "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:;>.
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:;>
> .
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pandoc-discuss/521B730A.50901%40reagle.org
> .
> For more options, visit https://groups.google.com/groups/opt_out.
>


-- 
/BP

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAFC_yuR_8EKbhs8LwzwzNXANmS6%3Dxs7m3JtFR1vYXhmqbuwi3Q%40mail.gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3704 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 958 bytes --]



On Sunday, August 25, 2013 9:15:45 PM UTC+2, fiddlosopher wrote:
>
>
> The pandoc-citations package will include an executable, biblio2yaml, 
> that translates from common bibliographic formats (e.g. bibtex) into 
> this YAML format. 
>

"biblio2yaml" sounds exciting. Could it be used as a filter, too, 
converting external biblio databases at runtime, and replacing bibutils?

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/d479b8cf-95f2-4182-bc16-8dca53420eb8%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1373 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Aug 27 13 06:18 ]:
>    On Sunday, August 25, 2013 9:15:45 PM UTC+2, fiddlosopher wrote:
> 
>      The pandoc-citations package will include an executable,
>      biblio2yaml,
>      that translates from common bibliographic formats (e.g. bibtex) into
>      this YAML format.
> 
>    "biblio2yaml" sounds exciting. Could it be used as a filter, too,
>    converting external biblio databases at runtime, and replacing
>    bibutils?

Yes, I've set it up so it will work as a pipe -- I'm not sure if
that's what you meant.
Note that it uses the bibutils library behind the scenes.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ John MacFarlane [Aug 27 13 06:53 ]:
> +++ Nick Bart [Aug 27 13 06:18 ]:
> >    On Sunday, August 25, 2013 9:15:45 PM UTC+2, fiddlosopher wrote:
> > 
> >      The pandoc-citations package will include an executable,
> >      biblio2yaml,
> >      that translates from common bibliographic formats (e.g. bibtex) into
> >      this YAML format.
> > 
> >    "biblio2yaml" sounds exciting. Could it be used as a filter, too,
> >    converting external biblio databases at runtime, and replacing
> >    bibutils?
> 
> Yes, I've set it up so it will work as a pipe -- I'm not sure if
> that's what you meant.
> Note that it uses the bibutils library behind the scenes.

Note that you can also just include

    bibliography: mybib.bib

in the metadata, and it will grab the bibliography from a bibtex or
biblatex file.  (Other formats can also be used.)

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1680 bytes --]

That's what I wanted to know, thank you.

If bibutils is used, I'm curious how the conversion step from MODS to "YAML 
representation of CSL variables in the JSON format" is done? If the 
existing citeproc-hs routines are used, please note that these still 
contain a number of bugs, and I’d be very happy if these could be ironed 
out at this opportunity.

On Tuesday, 27 August 2013 16:53:49 UTC+3, fiddlosopher wrote:
>
> +++ Nick Bart [Aug 27 13 06:18 ]: 
> >    On Sunday, August 25, 2013 9:15:45 PM UTC+2, fiddlosopher wrote: 
> > 
> >      The pandoc-citations package will include an executable, 
> >      biblio2yaml, 
> >      that translates from common bibliographic formats (e.g. bibtex) 
> into 
> >      this YAML format. 
> > 
> >    "biblio2yaml" sounds exciting. Could it be used as a filter, too, 
> >    converting external biblio databases at runtime, and replacing 
> >    bibutils? 
>
> Yes, I've set it up so it will work as a pipe -- I'm not sure if 
> that's what you meant. 
> Note that it uses the bibutils library behind the scenes. 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/049d081f-6a1a-479b-9b44-c36c8967cf6c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2204 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Bruce D'Arcus]

[-- Attachment #1: Type: text/plain, Size: 4649 bytes --]

Sounds good.

One minor thing: we put together that CSL JSON input schema fairly 
informally, and I've long thought we ought to look again at a few details 
(like, say, how dates are represented).

Just to say it may not be 100% stable.

On Sunday, August 25, 2013 3:15:45 PM UTC-4, fiddlosopher wrote:
>
> Pandoc currently includes built-in support for citation processing 
> (using citeproc-hs) and syntax highlighting (using highlighting-kate). 
> This is not ideal, since many users don't use either of these features, 
> and few, I would imagine, use both.  Both citeproc-hs and 
> highlighting-kate are big libraries, which bloat the pandoc binary. 
>
> So I am moving towards decoupling these components from core pandoc. 
> As noted earlier, I've added a --filter flag to pandoc, which makes it 
> easy to run JSON filters.  So, to process citations, you'll now do 
>
>     pandoc --filter pandoc-citeproc input.txt 
>
> and specify the bibliography and csl file in the document's metadata. 
> Similarly, to highlight source code using highlighting kate, you'll do 
>
>     pandoc --filter pandoc-highlight input.txt 
>
> (Highlighting style will be specified in the metadata rather than on 
> the command line.) Both pandoc-citeproc and pandoc-highlight will be 
> optional add-ons. The packages that provide them will also provide 
> Haskell libraries for processing citations and doing code highlighting 
> in pandoc documents, for those who use pandoc as a library. 
>
> I have already made the citations change; I have not yet made the 
> highlighting change, and could still back down if there is a lot of 
> protest.  The main drawback I can see is that performance will be a 
> *little* worse -- but the difference is pretty negligible.  The 
> command line will also be a bit longer for those who use highlighting. 
> But I think this change makes a lot of sense.  It will also make it 
> easier for users to use their own highlighting libraries; they just need 
> to write a simple filter. 
>
> Another nice feature I'm adding is inline references.  Instead of 
> putting your bibliographic references in an external file, you can 
> now include them directly in the document, in a YAML metadata block. 
> (This can occur at the end of the document if you like.) 
> This makes the document more self-contained.  The YAML format is 
> basically a YAML representation of the JSON format used by citeproc-js. 
> Here's an example: 
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 
> Some claim [fenner12a]. 
>
> --- 
> csl: tests/mhra.csl 
> references: 
> - id: fenner12a 
>   title: One-click science marketing 
>   author: 
>    family: Fenner 
>    given: [Martin] 
>   container-title: Nature Materials 
>   volume: 11 
>   url: 'http://dx.doi.org/10.1038/nmat3283' 
>   doi: '10.1038/nmat3283' 
>   issue: 4 
>   publisher: Nature Publishing Group 
>   page: 261-263 
>   type: article-journal 
>   issued: 
>     year: 2012 
>     month: 3 
> ... 
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 
>
> Of course, you can also store the YAML bibliography in a separate file, 
> and provide it to pandoc on the command line with the main file. 
>
> The pandoc-citations package will include an executable, biblio2yaml, 
> that translates from common bibliographic formats (e.g. bibtex) into 
> this YAML format. 
>
> I'm pretty excited about these changes, which I would like to include 
> in the 1.12 release.  The release has been held up waiting for Andrea 
> to release citeproc-hs.  I haven't heard from Andrea in a long time, so 
> I now have a contingency plan.  I will simply include the code from 
> citeproc-hs, including a couple of bug fixes, in the pandoc-citations 
> package.  When Andrea comes back, as I hope he does, I'll remove this 
> and depend once more on citeproc-hs.  But at least this will let us 
> go ahead with the release. 
>
> If you want to play with the new pandoc-citations stuff, it's on github: 
> http://github.com/jgm/pandoc-citations 
> It assumes you've installed a very recent dev version of pandoc and 
> pandoc-types. 
>
> John 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/28ef090f-b33e-4fd1-91cb-5da1eadf386b%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5621 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Rintze Zelle]

[-- Attachment #1: Type: text/plain, Size: 1033 bytes --]

(I commented at 
http://blog.martinfenner.org/2013/08/08/csl-is-more-than-citation-styles/ about 
the current state of Citeproc JSON)

On Friday, August 30, 2013 11:41:13 AM UTC-4, Bruce D'Arcus wrote:
>
> Sounds good.
>
> One minor thing: we put together that CSL JSON input schema fairly 
> informally, and I've long thought we ought to look again at a few details 
> (like, say, how dates are represented).
>
> Just to say it may not be 100% stable.
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/aa3f83b9-9e91-4cc0-ae03-e872dda16d9d%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1574 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Joseph Reagle]

On 08/30/2013 11:41 AM, Bruce D'Arcus wrote:
> One minor thing: we put together that CSL JSON input schema fairly 
> informally, and I've long thought we ought to look again at a few details 
> (like, say, how dates are represented).

On this point, as I consider moving from mostly biblatex/latex to mostly
CSL/HTML I've been looking at some of the mappings. My data is kept in
mindmaps and can be output to biblatex, bibtex, or Wikipedia. If I add a
YAML output (to feed directly into pandoc so as to avoid bibutils, its
bugs, and too many conversions) it also might be worthwhile moving to
that schema as well. I've briefly looked more closely at which fields
are making it into pandoc and comparing the input and output of
`biblio2yaml`.

1. It appears the biblatex field 'original-date' is disappearing (even
though there is a CSL original-date).
2. In biblatex we have a 'series' field. It appears the equivalent in
CSL is 'collection-title'? This seems to go missing as well.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Jorge Peña]

[-- Attachment #1: Type: text/plain, Size: 2610 bytes --]

I'm not entirely sure I completely understand the proposal, but I think 
it's relevant to me and my use case, which might help one of you to come to 
a conclusion.

I am one such user. I don't use either the citation or the built-in 
highlighting features. What I do instead is that I created an AST 
transformer that collects CodeBlocks and runs the contents through Pygments 
(backed by some caching to speed things up a bit) and embeds the resulting 
html in a RawBlock; I'm not sure if there's a better way.

When you say that you'd add a filter for highlighting, for example, would 
that basically be an AST transformer like mine, except using 
highlighting-kate? And so the highlighting would be decoupled in that 
manner, where users can run the filter/transformer if they want to (rather 
than by default, hence no longer requiring highlighting-kate as a 
dependency)? If so, I take it my kind of setup will not be affected?

On Friday, August 30, 2013 12:36:05 PM UTC-7, Joseph wrote:
>
> On 08/30/2013 11:41 AM, Bruce D'Arcus wrote: 
> > One minor thing: we put together that CSL JSON input schema fairly 
> > informally, and I've long thought we ought to look again at a few 
> details 
> > (like, say, how dates are represented). 
>
> On this point, as I consider moving from mostly biblatex/latex to mostly 
> CSL/HTML I've been looking at some of the mappings. My data is kept in 
> mindmaps and can be output to biblatex, bibtex, or Wikipedia. If I add a 
> YAML output (to feed directly into pandoc so as to avoid bibutils, its 
> bugs, and too many conversions) it also might be worthwhile moving to 
> that schema as well. I've briefly looked more closely at which fields 
> are making it into pandoc and comparing the input and output of 
> `biblio2yaml`. 
>
> 1. It appears the biblatex field 'original-date' is disappearing (even 
> though there is a CSL original-date). 
> 2. In biblatex we have a 'series' field. It appears the equivalent in 
> CSL is 'collection-title'? This seems to go missing as well. 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/242ec07d-4cfa-4599-b324-b2ad4e367fa6%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3086 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1651 bytes --]



On Friday, 30 August 2013 22:36:05 UTC+3, Joseph wrote:

>
> 1. It appears the biblatex field 'original-date' is disappearing (even 
> though there is a CSL original-date). 
> 2. In biblatex we have a 'series' field. It appears the equivalent in 
> CSL is 'collection-title'? This seems to go missing as well. 
>

So it would seem that biblio2yaml indeed uses the existing buggy 
citeproc-hs routines for converting MODS to CSL. Both bugs you describe 
have been reported at http://code.google.com/p/citeproc-hs/issues a long 
time ago. 

Note that proper mapping of the biblatex series field (to CSL 
'collection-title', that's correct) depends on proper handling of the whole 
title/booktitle/maintitle/series complex. I proposed a solution of how to 
implement this in MODS, and both bibutils and citeproc-hs have begun 
implementing this, but there have been no attempts at fixing the remaining 
bugs for a long time.

Again, I'd be very happy if a fresh attempt could be made to convince the 
citeproc-hs and bibutils maintainers that these really need to be fixed.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/b00f2f74-ae3a-4a4b-9f3e-a6ce2a5e5e04%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2080 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Aug 31 13 07:19 ]:
>    On Friday, 30 August 2013 22:36:05 UTC+3, Joseph wrote:
> 
>      1. It appears the biblatex field 'original-date' is disappearing
>      (even
>      though there is a CSL original-date).
>      2. In biblatex we have a 'series' field. It appears the equivalent
>      in
>      CSL is 'collection-title'? This seems to go missing as well.
> 
>    So it would seem that biblio2yaml indeed uses the existing buggy
>    citeproc-hs routines for converting MODS to CSL. Both bugs you describe
>    have been reported at http://code.google.com/p/citeproc-hs/issues a
>    long time ago.
>    Note that proper mapping of the biblatex series field (to CSL
>    'collection-title', that's correct) depends on proper handling of the
>    whole title/booktitle/maintitle/series complex. I proposed a solution
>    of how to implement this in MODS, and both bibutils and citeproc-hs
>    have begun implementing this, but there have been no attempts at fixing
>    the remaining bugs for a long time.
>    Again, I'd be very happy if a fresh attempt could be made to convince
>    the citeproc-hs and bibutils maintainers that these really need to be
>    fixed.

I have not heard from Andrea Rossato in months, which is one reason
I've taken the citeproc-hs code as the basis for a new package,
pandoc-citeproc, which I can modify.  I hope that Andrea will come
back to the project, in which case I can depend on hs-bibutils again.
In any case, I can try to fix the bugs, but my time is limited.
Probably best to get the bibutils side working first.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2463 bytes --]



On Saturday, August 31, 2013 6:23:09 PM UTC+3, fiddlosopher wrote:
>
> I have not heard from Andrea Rossato in months, which is one reason 
> I've taken the citeproc-hs code as the basis for a new package, 
> pandoc-citeproc, which I can modify.  I hope that Andrea will come 
> back to the project, in which case I can depend on hs-bibutils again. 
> In any case, I can try to fix the bugs, but my time is limited. 
> Probably best to get the bibutils side working first. 
>

Well, the bibutils author has not reacted for a long time either. Maybe it 
would help if you, the pandoc author, underlined this message on bibutils 
issues on the bibutils forum.

To clarify what's still buggy: For mapping title/booktitle/maintitle/series 
to MODS, nested <relatedItem type="host"> elements are used. The bibutils 
author has opted for a rigid system that always has three levels (top-level 
element plus two nested <relatedItem type="host"> elements) for "book", 
"collection", etc. entry types and four levels for "inbook", 
"incollection", etc. entry types. Currently, this works ok for some types, 
but levels are missing, or biblatex fields end up on the wrong level for 
others. (See 
http://sourceforge.net/p/bibutils/discussion/general/thread/d5909c51/). 
IMHO, the fix would be easy, it's just that the schema that has been 
introduced has to be used consistently.

In addition, MODS offers nice <genre> elements to further specify levels. 
bibutils, however, only uses <genre authority="marcgt">book</genre> or <genre 
authority="marcgt">collection</genre> one the first, resp. second level. To 
avoid errors, it would probably be better to use <genre 
authority="marcgt">multivolume monograph</genre> and <genre 
authority="marcgt">series</genre> on the appropriate levels as well, but 
bibutils currently does not. There hasn't even been any reaction when I 
suggested this.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/b423b3ba-5a81-4d16-aa32-83131b6c441b%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3411 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 959 bytes --]

What I'm meaning to say is that since it's clear what is required, the 
citeproc-hs and biblio2yaml bugs could easily be fixed even before bibutils 
is ready. For some entry types, MODS generated by bibutils is ok, but 
citeproc-hs doesn't do the right thing, e.g., it does not set 
"collection-title" even if a MODS element corresponding to the biblatex 
series field is present.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/03d50535-8cba-4844-958d-2a8fc1ee77b1%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1218 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I'm open to fixing these bugs.  The code base is a bit forbidding, but
I'm starting to understand it better.

+++ Nick Bart [Aug 31 13 21:07 ]:
>    What I'm meaning to say is that since it's clear what is required, the
>    citeproc-hs and biblio2yaml bugs could easily be fixed even before
>    bibutils is ready. For some entry types, MODS generated by bibutils is
>    ok, but citeproc-hs doesn't do the right thing, e.g., it does not set
>    "collection-title" even if a MODS element corresponding to the biblatex
>    series field is present.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/03d50535-8cba-4844-
>    958d-2a8fc1ee77b1%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/03d50535-8cba-4844-958d-2a8fc1ee77b1%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Matthias Hüning]

[-- Attachment #1: Type: text/plain, Size: 720 bytes --]



Am Montag, 2. September 2013 17:57:11 UTC+2 schrieb fiddlosopher:
>
> I'm open to fixing these bugs.  
>


That would be great!!

Matthias 

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/10c1dd20-c44b-46b3-8fea-89406bbb941c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1143 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2697 bytes --]

On Monday, 2 September 2013 17:57:11 UTC+2, fiddlosopher wrote:
>
> I'm open to fixing these bugs.  The code base is a bit forbidding, but 
> I'm starting to understand it better. 
>

Brilliant. This is much appreciated.

Here's my summary of how title / (booktitle) / maintitle / series should be 
mapped to MODS and then to CSL variables:

book, collection, proceedings, etc.:

biblatex   MODS       CSL
---------  ---------  ------------------------------------
title      top level  if maintitle non-empty: volume-title
                      if maintitle empty: title
maintitle  level +1   if non-empty: title
series     level +2   collection-title


inbook etc.:

biblatex   MODS       CSL
---------  ---------  ------------------------------------
title      top level  title
booktitle  level +1   if maintitle non-empty: volume-title
                      if maintitle empty: container-title
maintitle  level +2   if non-empty: container-title
series     level +3   collection-title

Notes: 

- MODS "levels" refer to nested <relatedItem type="host"> elements

- "maintitle", if non-empty, will be mapped to the CSL variables "title" 
(for books etc.) or "container-title" (for inbook etc.) and will be 
included in the output by all current CSL styles. 

- volume-title (i.e., title of a single volume within a multi-volume 
monograph) has been proposed 
(https://github.com/ajlyon/zotero-bits/issues/54) but is not an official 
CSL variable yet. Still, I see no harm in starting to implement it now, 
even if currently only customized CSL styles will be able to use it. 
Standard CSL styles will not include this information in their output 
(yet), but the title of the whole multivolume monograph will be (plus 
volume number, if provided), so while not ideal, this is perfectly 
acceptable; much better in any case than to provide nothing but the the 
single-volume title.

As to other citeproc-hs bugs identified so far, I think most are described 
rather fully on https://github.com/jgm/pandoc/issues.

If you need any further information, or assistance that I might be able 
provide, please let me know. 

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/4d09f924-b9ca-4f29-8c64-de355407a90a%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3757 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Joseph Reagle]

On 09/03/2013 03:42 AM, Nick Bart wrote:
> Here's my summary of how title / (booktitle) / maintitle / series should be 
> mapped to MODS and then to CSL variables:

I've been trying to wrap my head around biblatex <-> CSL/yaml as I hope
to make the transition myself. Sections 2.1 and 2.2 of the biblatex
documentation have a really nice description of the entry types and
fields (i.e., for each type, what are the required and optional/expected
fields). Does anyone know if something similar exists for CSL? The CSL
JSON schema [2] is syntactical only. The best sense of this that I've
gotten is from a zotero doc.

[1]:
http://mirrors.rit.edu/CTAN/macros/latex/contrib/biblatex/doc/biblatex.pdf
[2]:
https://github.com/citation-style-language/schema/blob/master/csl-data.json
[3]: http://aurimasv.github.io/z2csl/typeMap.xml

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1665 bytes --]

http://citationstyles.org/downloads/specification.html#appendix-iv-variables 
contains a list of all CSL variables, each with a brief explanation.

On Tuesday, 3 September 2013 13:54:10 UTC+2, Joseph wrote:
>
> On 09/03/2013 03:42 AM, Nick Bart wrote: 
> > Here's my summary of how title / (booktitle) / maintitle / series should 
> be 
> > mapped to MODS and then to CSL variables: 
>
> I've been trying to wrap my head around biblatex <-> CSL/yaml as I hope 
> to make the transition myself. Sections 2.1 and 2.2 of the biblatex 
> documentation have a really nice description of the entry types and 
> fields (i.e., for each type, what are the required and optional/expected 
> fields). Does anyone know if something similar exists for CSL? The CSL 
> JSON schema [2] is syntactical only. The best sense of this that I've 
> gotten is from a zotero doc. 
>
> [1]: 
> http://mirrors.rit.edu/CTAN/macros/latex/contrib/biblatex/doc/biblatex.pdf 
> [2]: 
> https://github.com/citation-style-language/schema/blob/master/csl-data.json 
> [3]: http://aurimasv.github.io/z2csl/typeMap.xml 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/ff6335de-a8cd-44e9-88e0-d95167fc8c9e%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2414 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Bruce D'Arcus]

I don't have time to think again about the details, but note: that
proposal for a "volume-title" variable is for zotero; not CSL. The
variables in CSL tend to be a little more abstract.

Bruce

On Tue, Sep 3, 2013 at 3:42 AM, Nick Bart <nickbart1980-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> On Monday, 2 September 2013 17:57:11 UTC+2, fiddlosopher wrote:
>>
>> I'm open to fixing these bugs.  The code base is a bit forbidding, but
>> I'm starting to understand it better.
>
>
> Brilliant. This is much appreciated.
>
> Here's my summary of how title / (booktitle) / maintitle / series should be
> mapped to MODS and then to CSL variables:
>
> book, collection, proceedings, etc.:
>
> biblatex   MODS       CSL
> ---------  ---------  ------------------------------------
> title      top level  if maintitle non-empty: volume-title
>                       if maintitle empty: title
> maintitle  level +1   if non-empty: title
> series     level +2   collection-title
>
>
> inbook etc.:
>
> biblatex   MODS       CSL
> ---------  ---------  ------------------------------------
> title      top level  title
> booktitle  level +1   if maintitle non-empty: volume-title
>                       if maintitle empty: container-title
> maintitle  level +2   if non-empty: container-title
> series     level +3   collection-title
>
> Notes:
>
> - MODS "levels" refer to nested <relatedItem type="host"> elements
>
> - "maintitle", if non-empty, will be mapped to the CSL variables "title"
> (for books etc.) or "container-title" (for inbook etc.) and will be included
> in the output by all current CSL styles.
>
> - volume-title (i.e., title of a single volume within a multi-volume
> monograph) has been proposed
> (https://github.com/ajlyon/zotero-bits/issues/54) but is not an official CSL
> variable yet. Still, I see no harm in starting to implement it now, even if
> currently only customized CSL styles will be able to use it. Standard CSL
> styles will not include this information in their output (yet), but the
> title of the whole multivolume monograph will be (plus volume number, if
> provided), so while not ideal, this is perfectly acceptable; much better in
> any case than to provide nothing but the the single-volume title.
>
> As to other citeproc-hs bugs identified so far, I think most are described
> rather fully on https://github.com/jgm/pandoc/issues.
>
> If you need any further information, or assistance that I might be able
> provide, please let me know.
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "pandoc-discuss" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/pandoc-discuss/-SajbqoPX8k/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pandoc-discuss/4d09f924-b9ca-4f29-8c64-de355407a90a%40googlegroups.com.
>
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1468 bytes --]

On Tuesday, 3 September 2013 16:54:01 UTC+2, Bruce D'Arcus wrote:
>
> I don't have time to think again about the details, but note: that 
> proposal for a "volume-title" variable is for zotero; not CSL. The 
> variables in CSL tend to be a little more abstract. 
>
> Bruce 
>

The proposal in https://github.com/ajlyon/zotero-bits/issues/54 is both for 
a new Zotero field "Volume Title" *and* a new CSL variable "volume-title".

"Proposal: add a field "Volume Title" to Book and Book Section, map it to 
new CSL variable "volume-title""

I think such a variable is definitely necessary: CMoS requires the 
distinction between single volumes and multivolume monographs (and series, 
which is again something different). biblatex has it, MODS has it … and CSL 
should introduce it, too.

How exactly are decisions on introduction of new CSL variables being made?

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/9703dae4-9816-4508-a1cd-8272c0c6ea0c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1906 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Bruce D'Arcus]

Decisions: I'd call it a loose consensus process. So someone proposes
something (ideally on the github issue tracker, maybe with a note on
the ML), and if sufficient implementers support it, without
significant objection, then it gets adopted. Otherwise, it doesn't.

But admittedly, variables are an easier conversation than other types
of proposals.

On the substance, that may well be the solution, except that it needs
to fit logically into the current title, container-title,
collection-title logic, which is fairly abstract. As I said, I haven't
thought about the details in awhile.

On Tue, Sep 3, 2013 at 11:46 AM, Nick Bart <nickbart1980-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> On Tuesday, 3 September 2013 16:54:01 UTC+2, Bruce D'Arcus wrote:
>>
>> I don't have time to think again about the details, but note: that
>> proposal for a "volume-title" variable is for zotero; not CSL. The
>> variables in CSL tend to be a little more abstract.
>>
>> Bruce
>
>
> The proposal in https://github.com/ajlyon/zotero-bits/issues/54 is both for
> a new Zotero field "Volume Title" *and* a new CSL variable "volume-title".
>
> "Proposal: add a field "Volume Title" to Book and Book Section, map it to
> new CSL variable "volume-title""
>
> I think such a variable is definitely necessary: CMoS requires the
> distinction between single volumes and multivolume monographs (and series,
> which is again something different). biblatex has it, MODS has it … and CSL
> should introduce it, too.
>
> How exactly are decisions on introduction of new CSL variables being made?
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "pandoc-discuss" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/pandoc-discuss/-SajbqoPX8k/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pandoc-discuss/9703dae4-9816-4508-a1cd-8272c0c6ea0c%40googlegroups.com.
>
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAF-FPGOjxF%3D8vDBOtL3RY5dRO-zetra3EXvddNgKZMf8tiEB-A%40mail.gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3683 bytes --]

Thank you for the clarification. So far, I took the proposal in 
https://github.com/ajlyon/zotero-bits/issues/54 to be directed at both 
Zotero and CSL developers.

If I understand you correctly, however, proposals concerning CSL should 
rather be made on https://github.com/citation-style-language/styles/issues 
and the xbiblio-devel mailing list.

Introducing "volume-title" has indeed not been proposed there yet. Would it 
help if I did this, or would you want to do it yourself?

Nick

On Tuesday, 3 September 2013 17:56:28 UTC+2, Bruce D'Arcus wrote:
>
> Decisions: I'd call it a loose consensus process. So someone proposes 
> something (ideally on the github issue tracker, maybe with a note on 
> the ML), and if sufficient implementers support it, without 
> significant objection, then it gets adopted. Otherwise, it doesn't. 
>
> But admittedly, variables are an easier conversation than other types 
> of proposals. 
>
> On the substance, that may well be the solution, except that it needs 
> to fit logically into the current title, container-title, 
> collection-title logic, which is fairly abstract. As I said, I haven't 
> thought about the details in awhile. 
>
> On Tue, Sep 3, 2013 at 11:46 AM, Nick Bart <nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org<javascript:>> 
> wrote: 
> > On Tuesday, 3 September 2013 16:54:01 UTC+2, Bruce D'Arcus wrote: 
> >> 
> >> I don't have time to think again about the details, but note: that 
> >> proposal for a "volume-title" variable is for zotero; not CSL. The 
> >> variables in CSL tend to be a little more abstract. 
> >> 
> >> Bruce 
> > 
> > 
> > The proposal in https://github.com/ajlyon/zotero-bits/issues/54 is both 
> for 
> > a new Zotero field "Volume Title" *and* a new CSL variable 
> "volume-title". 
> > 
> > "Proposal: add a field "Volume Title" to Book and Book Section, map it 
> to 
> > new CSL variable "volume-title"" 
> > 
> > I think such a variable is definitely necessary: CMoS requires the 
> > distinction between single volumes and multivolume monographs (and 
> series, 
> > which is again something different). biblatex has it, MODS has it … and 
> CSL 
> > should introduce it, too. 
> > 
> > How exactly are decisions on introduction of new CSL variables being 
> made? 
> > 
> > -- 
> > You received this message because you are subscribed to a topic in the 
> > Google Groups "pandoc-discuss" group. 
> > To unsubscribe from this topic, visit 
> > https://groups.google.com/d/topic/pandoc-discuss/-SajbqoPX8k/unsubscribe. 
>
> > To unsubscribe from this group and all its topics, send an email to 
> > pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > To view this discussion on the web visit 
> > 
> https://groups.google.com/d/msgid/pandoc-discuss/9703dae4-9816-4508-a1cd-8272c0c6ea0c%40googlegroups.com. 
>
> > 
> > For more options, visit https://groups.google.com/groups/opt_out. 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/1ae7a028-9095-4365-a874-179cabf6b702%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4860 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Bruce D'Arcus]

[-- Attachment #1: Type: text/plain, Size: 4890 bytes --]

Sure; feel free to propose yourself.

CSL is closely connected to Zotero, but independent.
On Sep 3, 2013 1:51 PM, "Nick Bart" <nickbart1980-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:

> Thank you for the clarification. So far, I took the proposal in
> https://github.com/ajlyon/**zotero-bits/issues/54<https://github.com/ajlyon/zotero-bits/issues/54>to be directed at both Zotero and CSL developers.
>
> If I understand you correctly, however, proposals concerning CSL should
> rather be made on https://github.com/citation-style-language/styles/issuesand the xbiblio-devel mailing list.
>
> Introducing "volume-title" has indeed not been proposed there yet. Would
> it help if I did this, or would you want to do it yourself?
>
> Nick
>
> On Tuesday, 3 September 2013 17:56:28 UTC+2, Bruce D'Arcus wrote:
>>
>> Decisions: I'd call it a loose consensus process. So someone proposes
>> something (ideally on the github issue tracker, maybe with a note on
>> the ML), and if sufficient implementers support it, without
>> significant objection, then it gets adopted. Otherwise, it doesn't.
>>
>> But admittedly, variables are an easier conversation than other types
>> of proposals.
>>
>> On the substance, that may well be the solution, except that it needs
>> to fit logically into the current title, container-title,
>> collection-title logic, which is fairly abstract. As I said, I haven't
>> thought about the details in awhile.
>>
>> On Tue, Sep 3, 2013 at 11:46 AM, Nick Bart <nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>> > On Tuesday, 3 September 2013 16:54:01 UTC+2, Bruce D'Arcus wrote:
>> >>
>> >> I don't have time to think again about the details, but note: that
>> >> proposal for a "volume-title" variable is for zotero; not CSL. The
>> >> variables in CSL tend to be a little more abstract.
>> >>
>> >> Bruce
>> >
>> >
>> > The proposal in https://github.com/ajlyon/**zotero-bits/issues/54<https://github.com/ajlyon/zotero-bits/issues/54>is both for
>> > a new Zotero field "Volume Title" *and* a new CSL variable
>> "volume-title".
>> >
>> > "Proposal: add a field "Volume Title" to Book and Book Section, map it
>> to
>> > new CSL variable "volume-title""
>> >
>> > I think such a variable is definitely necessary: CMoS requires the
>> > distinction between single volumes and multivolume monographs (and
>> series,
>> > which is again something different). biblatex has it, MODS has it … and
>> CSL
>> > should introduce it, too.
>> >
>> > How exactly are decisions on introduction of new CSL variables being
>> made?
>> >
>> > --
>> > You received this message because you are subscribed to a topic in the
>> > Google Groups "pandoc-discuss" group.
>> > To unsubscribe from this topic, visit
>> > https://groups.google.com/d/**topic/pandoc-discuss/-**
>> SajbqoPX8k/unsubscribe<https://groups.google.com/d/topic/pandoc-discuss/-SajbqoPX8k/unsubscribe>.
>>
>> > To unsubscribe from this group and all its topics, send an email to
>> > pandoc-discus...@**googlegroups.com.
>> > To post to this group, send email to pandoc-...@googlegroups.**com.
>> > To view this discussion on the web visit
>> > https://groups.google.com/d/**msgid/pandoc-discuss/9703dae4-**
>> 9816-4508-a1cd-8272c0c6ea0c%**40googlegroups.com<https://groups.google.com/d/msgid/pandoc-discuss/9703dae4-9816-4508-a1cd-8272c0c6ea0c%40googlegroups.com>.
>>
>> >
>> > For more options, visit https://groups.google.com/**groups/opt_out<https://groups.google.com/groups/opt_out>.
>>
>>
>  --
> You received this message because you are subscribed to a topic in the
> Google Groups "pandoc-discuss" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/pandoc-discuss/-SajbqoPX8k/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pandoc-discuss/1ae7a028-9095-4365-a874-179cabf6b702%40googlegroups.com
> .
> For more options, visit https://groups.google.com/groups/opt_out.
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAF-FPGM19xs%2B9xipOMSa3wEXgumqBE3ro841cN%3DvBe-j6fz4Pg%40mail.gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 6597 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Rintze Zelle]

[-- Attachment #1: Type: text/plain, Size: 2120 bytes --]

On Tuesday, September 3, 2013 1:51:34 PM UTC-4, Nick Bart wrote:

> Thank you for the clarification. So far, I took the proposal in 
> https://github.com/ajlyon/zotero-bits/issues/54 to be directed at both 
> Zotero and CSL developers.
>
> If I understand you correctly, however, proposals concerning CSL should 
> rather be made on https://github.com/citation-style-language/styles/issuesand the xbiblio-devel mailing list.
>

The xbiblio-devel mailing list is the most central location to discuss CSL 
development. We also use the GitHub issue trackers of the CSL project, but 
not everybody follows those, so it's not the best place for any decision 
making. If any significant proposals develop on GitHub, we usually 
crosspost to xbiblio-devel to inform all stakeholders and to seek consensus 
approval.

The zotero-bits repository was started by several Zotero power users in 
anticipation of a long-promised redesign of Zotero's data model. We have 
been using it to keep track of changes to the Zotero data model that users 
have requested, as well as the changes that would be required in CSL to 
accomodate those. The data model redesign is still pretty far ahead on the 
horizon, and many of the proposals in the zotero-bits tracker still have to 
be vetted by the Zotero developers. When that time comes, any desired 
changes on the CSL side of things should be discussed on xbiblio-devel.

(the fact that most CSL contributors are Zotero users might explain some of 
the apparent chaos in this communication arrangement)

Rintze

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/e770d4d1-14ad-4f08-8c32-add1aef89279%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2781 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Joseph Reagle]

On 09/03/2013 11:56 AM, Bruce D'Arcus wrote:
> On the substance, that may well be the solution, except that it needs
> to fit logically into the current title, container-title,
> collection-title logic, which is fairly abstract. As I said, I haven't

Nick pointed to the CSL variables in an appendix of the spec, but this
is one of the things I haven't been able to figure out. What is the
logic between collection vs. container?


-- 
Regards,
Joseph Reagle                    http://reagle.org/joseph/
(Perhaps using speech recognition, sorry for any speakos.)

Re: Decoupling citeproc and highlighting-kate from pandoc

[Bruce D'Arcus]

Take a book chapter and an article: they share the characteristic that
they are discrete items within a larger, autonomous, item. So an
edited book is a container for the former, and a series is the
collection. For the latter, the journal is the container.

Now if you're paying attention, you're saying that's not so
consistent, and somewhat arbitrary. Why isn't a journal issue the
container, and the journal itself the collection?

And, apropos Nick's point here, what about things like multi-volumed works?

There's no good reason: I think I/we just designed the spec to cover
as much as we could as easily as possible. It seemed sensible at the
time, but am not so sure it does now.

Think I need to add this to the new ticket Nick added.

Bruce

On Tue, Sep 3, 2013 at 4:44 PM, Joseph Reagle <joseph.2011-T1oY19WcHSwdnm+yROfE0A@public.gmane.org> wrote:
> On 09/03/2013 11:56 AM, Bruce D'Arcus wrote:
>> On the substance, that may well be the solution, except that it needs
>> to fit logically into the current title, container-title,
>> collection-title logic, which is fairly abstract. As I said, I haven't
>
> Nick pointed to the CSL variables in an appendix of the spec, but this
> is one of the things I haven't been able to figure out. What is the
> logic between collection vs. container?
>
>
> --
> Regards,
> Joseph Reagle                    http://reagle.org/joseph/
> (Perhaps using speech recognition, sorry for any speakos.)

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2753 bytes --]

I think it's rather straightforward: From 
http://citationstyles.org/downloads/specification.html#appendix-iv-variables:

*collection-title*
title of the collection holding the item (e.g. the series title for a book)
*container-title*
title of the container holding the item (e.g. the book title for a book 
chapter, the journal title for a journal article)

"container-title" contains the title of the whole book or the journal in 
types such as book chapters, articles, etc. It corresponds to biblatex's 
"booktitle" or "journal".

According to the "volume-title" proposal, "container-title" would always 
hold the title of the whole book, even if it consists of multiple volumes 
(such as "Collected Works", "Handbook of Psychology"), and "volume-title" 
would hold the title of a single volume ("Tragedies", "Clinical 
Psychology", etc.).

"collection-title" on the other hand contains the title of a book series in 
the sense of CMoS 6e 14.128, e.g., "Johns Hopkins Jewish Studies", "Studies 
in Illinois Archaeology 6", which is not italicized, and may even "be 
omitted to save space".

1. Gershon David Hundert, *The Jews in a Polish Private Town: The Case of 
Opatów in the Eighteenth Century,* Johns Hopkins Jewish Studies (Baltimore: 
Johns Hopkins University Press, 1992).

Fowler, Melvin L. *The Cahokia Atlas: A Historical Atlas of Cahokia 
Archaeology.* Studies in Illinois Archaeology 6. Springfield: Illinois 
Historic Preservation Agency, 1989.

Grenier, Roger. *Les larmes d’Ulysse.* Collection l’un et l’autre. Paris: 
Gallimard, 1998.

Wauchope, Robert. *A Tentative Sequence of Pre-Classic Ceramics in Middle 
America.* Middle American Research Records, vol. 1, no. 14. New Orleans, 
LA: Tulane University, 1950.

Since I have not been able to spot any other use, I imagine 
"collection-title" might just as well be renamed to "series-title".

On Tuesday, September 3, 2013 10:44:53 PM UTC+2, Joseph wrote:
>
> Nick pointed to the CSL variables in an appendix of the spec, but this 
> is one of the things I haven't been able to figure out. What is the 
> logic between collection vs. container?
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/83ebfc01-27a8-4f45-a3f8-3ee4378e66e2%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3300 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Bruce D'Arcus]

[-- Attachment #1: Type: text/plain, Size: 3088 bytes --]

Yeah, I'm saying that a journal is (should be) really a collection, as is a
book series, a legal code, etc.

But let's move this to the CSL venues.
On Sep 4, 2013 5:02 AM, "Nick Bart" <nickbart1980-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:

> I think it's rather straightforward: From
> http://citationstyles.org/downloads/specification.html#appendix-iv-variables
> :
>
> *collection-title*
> title of the collection holding the item (e.g. the series title for a book)
> *container-title*
> title of the container holding the item (e.g. the book title for a book
> chapter, the journal title for a journal article)
>
> "container-title" contains the title of the whole book or the journal in
> types such as book chapters, articles, etc. It corresponds to biblatex's
> "booktitle" or "journal".
>
> According to the "volume-title" proposal, "container-title" would always
> hold the title of the whole book, even if it consists of multiple volumes
> (such as "Collected Works", "Handbook of Psychology"), and "volume-title"
> would hold the title of a single volume ("Tragedies", "Clinical
> Psychology", etc.).
>
> "collection-title" on the other hand contains the title of a book series
> in the sense of CMoS 6e 14.128, e.g., "Johns Hopkins Jewish Studies",
> "Studies in Illinois Archaeology 6", which is not italicized, and may even
> "be omitted to save space".
>
> 1. Gershon David Hundert, *The Jews in a Polish Private Town: The Case of
> Opatów in the Eighteenth Century,* Johns Hopkins Jewish Studies (Baltimore:
> Johns Hopkins University Press, 1992).
>
> Fowler, Melvin L. *The Cahokia Atlas: A Historical Atlas of Cahokia
> Archaeology.* Studies in Illinois Archaeology 6. Springfield: Illinois
> Historic Preservation Agency, 1989.
>
> Grenier, Roger. *Les larmes d’Ulysse.* Collection l’un et l’autre. Paris:
> Gallimard, 1998.
>
> Wauchope, Robert. *A Tentative Sequence of Pre-Classic Ceramics in Middle
> America.* Middle American Research Records, vol. 1, no. 14. New Orleans,
> LA: Tulane University, 1950.
>
> Since I have not been able to spot any other use, I imagine
> "collection-title" might just as well be renamed to "series-title".
>
> On Tuesday, September 3, 2013 10:44:53 PM UTC+2, Joseph wrote:
>>
>> Nick pointed to the CSL variables in an appendix of the spec, but this
>> is one of the things I haven't been able to figure out. What is the
>> logic between collection vs. container?
>>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAF-FPGOb%2Bh48xYH9S04gNi-hZ3m7Hph9gdWLzBFcZHc6-qoUbw%40mail.gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4161 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 918 bytes --]

Continued at the xbiblio-devel ML (see 
http://sourceforge.net/mailarchive/forum.php?forum_name=xbiblio-devel)

On Wednesday, 4 September 2013 13:00:47 UTC+2, Bruce D'Arcus wrote:
>
> Yeah, I'm saying that a journal is (should be) really a collection, as is 
> a book series, a legal code, etc.
>
> But let's move this to the CSL venues.
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/b1c9a0f2-11af-4e30-bb26-57cfb5503571%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1334 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1579 bytes --]

So, thu upshot for pandoc/citeproc-hs so far seems to be that the proposal 
for a new CSL variable, "volume-title" is on the table, but it's not so 
clear if and when a consensus might be achieved.

In the meantime, pandoc/citeproc-hs could fix the issues concerning the 
series field, and also

(a) implement "volume-title" as I suggested earlier in this thread, or

(b) map the content of the biblatex maintitle field, if non-empty, or else 
the content of the title field (for type book etc.) or the content of the 
booktitle field (for type inbook etc.) to the CSL variable 
"container-title"; if the maintitle field is non-empty, the content of the 
title field (for type book etc.) or the content of the booktitle field (for 
type inbook etc.) will have to be discarded.

(b) is tolerable. It would still output, e.g., "Collected Works, vol. 7", 
which is much better than just "Tragedies", or even "Tragedies, vol. 7" 
which would be just plain wrong.

Still, I'd be happier if it'd be (a) ...

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/6f1be56f-1995-4e54-84ed-201881300237%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1867 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Just don't expect anything too soon...

+++ Nick Bart [Sep 03 13 00:42 ]:
>    On Monday, 2 September 2013 17:57:11 UTC+2, fiddlosopher wrote:
> 
>      I'm open to fixing these bugs.  The code base is a bit forbidding,
>      but
>      I'm starting to understand it better.
> 
>    Brilliant. This is much appreciated.
>    Here's my summary of how title / (booktitle) / maintitle / series
>    should be mapped to MODS and then to CSL variables:
>    book, collection, proceedings, etc.:
>    biblatex   MODS       CSL
>    ---------  ---------  ------------------------------------
>    title      top level  if maintitle non-empty: volume-title
>                          if maintitle empty: title
>    maintitle  level +1   if non-empty: title
>    series     level +2   collection-title
>    inbook etc.:
>    biblatex   MODS       CSL
>    ---------  ---------  ------------------------------------
>    title      top level  title
>    booktitle  level +1   if maintitle non-empty: volume-title
>                          if maintitle empty: container-title
>    maintitle  level +2   if non-empty: container-title
>    series     level +3   collection-title
>    Notes:
>    - MODS "levels" refer to nested <relatedItem type="host"> elements
>    - "maintitle", if non-empty, will be mapped to the CSL variables
>    "title" (for books etc.) or "container-title" (for inbook etc.) and
>    will be included in the output by all current CSL styles.
>    - volume-title (i.e., title of a single volume within a multi-volume
>    monograph) has been proposed
>    (https://github.com/ajlyon/zotero-bits/issues/54) but is not an
>    official CSL variable yet. Still, I see no harm in starting to
>    implement it now, even if currently only customized CSL styles will be
>    able to use it. Standard CSL styles will not include this information
>    in their output (yet), but the title of the whole multivolume monograph
>    will be (plus volume number, if provided), so while not ideal, this is
>    perfectly acceptable; much better in any case than to provide nothing
>    but the the single-volume title.
>    As to other citeproc-hs bugs identified so far, I think most are
>    described rather fully on https://github.com/jgm/pandoc/issues.
>    If you need any further information, or assistance that I might be able
>    provide, please let me know.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/4d09f924-b9ca-4f29-
>    8c64-de355407a90a%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/4d09f924-b9ca-4f29-8c64-de355407a90a%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Just for kicks, I made a little bibtex -> pandoc yaml converter
that doesn't depend on bibutils.  It's bibtex.hs in the pandoc-citeproc
repository, and can be compiled with 'ghc --make bibtex' (you'll need
to 'cabal install bibtex split' and also pandoc-citeproc itself).

It's very limited, only supports bibtex fields so far, and probably
doesn't properly support even then.

It uses pandoc's LaTeX reader to parse bibtex fields, so that all
latex will be properly interpreted.  (Including math.)

Perhaps this is one way forward...

Re: Decoupling citeproc and highlighting-kate from pandoc

[Joseph Reagle]

On 09/07/2013 01:16 PM, John MacFarlane wrote:
> Just for kicks, I made a little bibtex -> pandoc yaml converter
> that doesn't depend on bibutils.  It's bibtex.hs in the pandoc-citeproc
> repository, and can be compiled with 'ghc --make bibtex' (you'll need
> to 'cabal install bibtex split' and also pandoc-citeproc itself).

I was curious how it'd handle fragment identifiers in URLs and it looks
to be escaping them with a backslash. That doesn't seem to cause any
problem in generated HTML, but neither does not having the slash...

~~~~

@online{Winer2007uvp,
   author = {Winer, Dave},
   custom2 = {01-intro.mm},
   day = {01},
   keyword = {comment},
   month = {01},
   organization = {Scripting News},
   title = {{The} Unedited Voice of a Person},
   url = {http://scripting.com/2007/01/01.html#theUneditedVoiceOfAPerson},
   urldate = {2013-02-15},
   year = {2007},
}

~~~~

~~~~

---
references:
- author:
  - family: Winer
    given:
    -
    - Dave
  id: Winer2007uvp
  issued:
    month: 01
    year: 2007
  title: The Unedited Voice of a Person
  type: 'no-type'
  url: 'http://scripting.com/2007/01/01.html\#theUneditedVoiceOfAPerson'
...

~~~~

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1671 bytes --]

Interesting. It chokes on uppercase chars in bibtex field names, though: 
"author={Doe, John}" works; "Author={Doe, John}" doesn't. Also, just 
calling it "bibtex" might not be ideal - what about bibtex2pandocyaml or so?

And of course pure bibtex doesn't cover a number of common types - @online, 
e.g., is converted to "type: 'no-type' " (see post by Joseph in this 
thread) … so I'm looking forward to biblatex support …

On Saturday, 7 September 2013 19:16:30 UTC+2, fiddlosopher wrote:
>
> Just for kicks, I made a little bibtex -> pandoc yaml converter 
> that doesn't depend on bibutils.  It's bibtex.hs in the pandoc-citeproc 
> repository, and can be compiled with 'ghc --make bibtex' (you'll need 
> to 'cabal install bibtex split' and also pandoc-citeproc itself). 
>
> It's very limited, only supports bibtex fields so far, and probably 
> doesn't properly support even then. 
>
> It uses pandoc's LaTeX reader to parse bibtex fields, so that all 
> latex will be properly interpreted.  (Including math.) 
>
> Perhaps this is one way forward... 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/fd109d63-2add-4956-b1f3-dbaf706634fb%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2083 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Sep 12 13 01:57 ]:
>    Interesting. It chokes on uppercase chars in bibtex field names,
>    though: "author={Doe, John}" works; "Author={Doe, John}" doesn't.

Fixed.

>    Also,
>    just calling it "bibtex" might not be ideal - what about
>    bibtex2pandocyaml or so?

Provisionally renamed to bibtex2yaml.  Of course this is just
experimental at this point.

>    And of course pure bibtex doesn't cover a number of common types -
>    @online, e.g., is converted to "type: 'no-type' " (see post by Joseph
>    in this thread) so I'm looking forward to biblatex support

Sure.  Should just be a matter of toil to add these.  In fact I may
try to make the code clear enough that a non-Haskell speaker could
do it.

John

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've renamed to bibtex2pandoc.hs and made several improvements,
including parsing of non-dropping-particles.

I've added all the biblatex entry types, and I've started adding support
for biblatex data fields.  (In the biblatex list of data fields,
2.2.2 of http://www.math.upenn.edu/tex_docs/latex/biblatex/biblatex.pdf,
I've gotten as far as 'annotation'.)

Nick, if you (or others) want to work on this, it should be pretty
self-explanatory how to add to the code, even if you don't know Haskell.
(Find the part of the code with all the arrows.  ==> means the field is
to be interpreted as latex; --> means it's a raw string; *=> is an
author list.)

Recompile with 'ghc --make bibtex2pandoc.hs'

John

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4747 bytes --]

Fantastic. I'm amazed by your energy. And, yes, I'd love to contribute. 
Time to learn a little Haskell after all, I guess.

Still, a few questions first:

1. Would it make sense to cleanly separate code for bibtex vs. code for 
biblatex? Handling of dates, and a number of other things, see below, 
differs a lot.

2. (How) can the bibtex and biblatex "crossref" mechanism for inheritance 
be implemented? (biblatex has introduced an alternative cross-referencing 
mechanism using the xref field, but I'm not sure how many people have 
actually started using this. I haven't seen any advantage in using this 
instead of crossref yet.)

3. Some fields will have to be concatenated,

    - title + subtitle + titleaddon (same for booktitle, maintitle), and 
journal/journaltitle + journalsubtitle. How will this be handled? 
(titleaddon etc. are supposed to be printed "in a different font", i.e., 
unitalicized or outside of speech marks, but this currently does not seem 
to be possible in CSL.)

    - note + addendum. biblatex has addendum at the end of an entry, and 
note somewhere in front of that, but in CSL there's only one suitable 
field: note. – Differs from biblatex "annotation" (and its alias "annote") 
which map to CSL "annote", and contain reader's notes.

4. Some fields have to be omitted if some certain other/newer field(s) are 
present: In biblatex, "date", if present, is used instead of "year", 
"month", and "day"; "location" instead of "address", etc. How will this be 
handled?

5. Proper mapping of some fields or entry types is conditional upon either 
the existence or the content of other fields. How can this be made to work?

    - for @book, @collection, @proceedings: title ==> title, unless 
maintitle exists, in which case maintitle ==> title (and, once consensus on 
new CSL variables is reached, title ==> volume-title; as discussed earlier 
in this thread),

    - for @inbook, @incollection, @proceedings: booktitle ==> 
container-title, unless maintitle exists, in which case maintitle ==> 
container-title (and booktitle ==> volume-title),

    - mapping of editor, editora, editorb, editorc depends on the content 
of editortype, editoratype, editorbtype, editorctype,

    - An entrysubtype of "magazine" should change the CSL type from 
"article-journal" to "article-magazine" (this is not from biblatex itself 
but from biblatex-chicago, but I think it'd be very useful to borrow 
certain bits from this package; an entrysubtype of "newspaper" could of 
course map to CSL "article-newspaper").

    - series maps to CSL collection-title, except for @article (and 
@periodical?), in which case it should appear after the journal title (the 
only solution in CSL is probably to concatenate the two fields).

6. "keys" are used liberally in biblatex fields, as shorthands, and for 
localization (see biblatex manual 4.9). These should be translated to 
natural language terms. Let's focus on English first - but what about 
localization …?

7. In addition to "name lists" (authors etc.), biblatex has "literal lists" 
with elements separated by "and", too: institution, location, organization, 
origlocation, origpublisher, publisher, list[a-f], address, school. Will 
routines used for names work unmodified?


On Friday, 13 September 2013 08:24:45 UTC+2, fiddlosopher wrote:
>
> I've renamed to bibtex2pandoc.hs and made several improvements, 
> including parsing of non-dropping-particles. 
>
> I've added all the biblatex entry types, and I've started adding support 
> for biblatex data fields.  (In the biblatex list of data fields, 
> 2.2.2 of http://www.math.upenn.edu/tex_docs/latex/biblatex/biblatex.pdf, 
> I've gotten as far as 'annotation'.) 
>
> Nick, if you (or others) want to work on this, it should be pretty 
> self-explanatory how to add to the code, even if you don't know Haskell. 
> (Find the part of the code with all the arrows.  ==> means the field is 
> to be interpreted as latex; --> means it's a raw string; *=> is an 
> author list.) 
>
> Recompile with 'ghc --make bibtex2pandoc.hs' 
>
> John 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/b01350cd-72e7-49e5-aa50-6c25382376f6%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5387 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

You raise some good points.  All of this is possible, but I think
I'd like to work a bit on the underlying framework to make it all easier
to implement.

+++ Nick Bart [Sep 13 13 07:26 ]:
>    Fantastic. I'm amazed by your energy. And, yes, I'd love to contribute.
>    Time to learn a little Haskell after all, I guess.
>    Still, a few questions first:
>    1. Would it make sense to cleanly separate code for bibtex vs. code for
>    biblatex? Handling of dates, and a number of other things, see below,
>    differs a lot.

Since so much is the same, the thing to do is probably to add an
ifBibTeX conditional.

>    2. (How) can the bibtex and biblatex "crossref" mechanism for
>    inheritance be implemented? (biblatex has introduced an alternative
>    cross-referencing mechanism using the xref field, but I'm not sure how
>    many people have actually started using this. I haven't seen any
>    advantage in using this instead of crossref yet.)

I don't know.  Can you give an example of the use of crossref?

>    3. Some fields will have to be concatenated,
>        - title + subtitle + titleaddon (same for booktitle, maintitle),
>    and journal/journaltitle + journalsubtitle. How will this be handled?
>    (titleaddon etc. are supposed to be printed "in a different font",
>    i.e., unitalicized or outside of speech marks, but this currently does
>    not seem to be possible in CSL.)
>        - note + addendum. biblatex has addendum at the end of an entry,
>    and note somewhere in front of that, but in CSL there's only one
>    suitable field: note. Differs from biblatex "annotation" (and its alias
>    "annote") which map to CSL "annote", and contain reader's notes.
>    4. Some fields have to be omitted if some certain other/newer field(s)
>    are present: In biblatex, "date", if present, is used instead of
>    "year", "month", and "day"; "location" instead of "address", etc. How
>    will this be handled?
>    5. Proper mapping of some fields or entry types is conditional upon
>    either the existence or the content of other fields. How can this be
>    made to work?
>        - for @book, @collection, @proceedings: title ==> title, unless
>    maintitle exists, in which case maintitle ==> title (and, once
>    consensus on new CSL variables is reached, title ==> volume-title; as
>    discussed earlier in this thread),
>        - for @inbook, @incollection, @proceedings: booktitle ==>
>    container-title, unless maintitle exists, in which case maintitle ==>
>    container-title (and booktitle ==> volume-title),
>        - mapping of editor, editora, editorb, editorc depends on the
>    content of editortype, editoratype, editorbtype, editorctype,
>        - An entrysubtype of "magazine" should change the CSL type from
>    "article-journal" to "article-magazine" (this is not from biblatex
>    itself but from biblatex-chicago, but I think it'd be very useful to
>    borrow certain bits from this package; an entrysubtype of "newspaper"
>    could of course map to CSL "article-newspaper").
>        - series maps to CSL collection-title, except for @article (and
>    @periodical?), in which case it should appear after the journal title
>    (the only solution in CSL is probably to concatenate the two fields).
>    6. "keys" are used liberally in biblatex fields, as shorthands, and for
>    localization (see biblatex manual 4.9). These should be translated to
>    natural language terms. Let's focus on English first - but what about
>    localization ?
>    7. In addition to "name lists" (authors etc.), biblatex has "literal
>    lists" with elements separated by "and", too: institution, location,
>    organization, origlocation, origpublisher, publisher, list[a-f],
>    address, school. Will routines used for names work unmodified?

No, because this routine includes code for dividing a single name into
given, last, and prefix.  But it will be easy to write something that
will work for these.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1859 bytes --]

Example:

@incollection{item1,
    Author = {Doe, John},
    Crossref = {item2},
    Pages = {12-34},
    Title = {The Incollection Title}}

@collection{item2,
    Date = {2012},
    Editor = {Editor, Ed},
    Location = {Location},
    Publisher = {Publisher},
    Title = {The Collection Title}}

item1 should inherit all missing fields from item2.

The bibtex program used to do this in a rather simple fashion, and required 
an additional "Booktitle = {The Collection Title}" in item2. 

biblatex in combination with the biber program provides a much better 
solution: it remaps field names of item2, e.g., title -> booktitle, 
according to a set of cross-referencing rules. See biblatex manual, §§ 
2.4.1 and 4.5.9, and Appendix B. 


On Friday, 13 September 2013 16:36:45 UTC+2, fiddlosopher wrote:

>
> >    2. (How) can the bibtex and biblatex "crossref" mechanism for 
> >    inheritance be implemented? (biblatex has introduced an alternative 
> >    cross-referencing mechanism using the xref field, but I'm not sure 
> how 
> >    many people have actually started using this. I haven't seen any 
> >    advantage in using this instead of crossref yet.) 
>
> I don't know.  Can you give an example of the use of crossref?
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/2513322d-2156-40f2-9546-64eb2cce12bf%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2491 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 23243 bytes --]

For what it's worth, this is my list of mappings from all biblatex fields 
(as listed in the biblatex manual, §§ 2.2.2–2.2.5) to matching CSL 
variables, with a few comments thrown in. I hope it's useful both for 
developing bibtex2pandoc and for testing existing conversion routines such 
as biblatex -> MODS -> CSL.

Feedback on content and formatting is welcome, and maybe some improved 
version could even be integrated into the general pandoc documentation.

2.2.2 Data Fields

| biblatex field  | biblatex field type        | CSL                      | 
comment                                         |
|:----------------|:---------------------------|:-------------------------|:------------------------------------------------|
| abstract        | field (literal)            | abstract                 | 
!                                               |
| addendum        | field (literal)            | note                     | 
concat w/ note                                  |
| afterword       | list (name)                | ---                      | 
no corresp. CSL var                             |
| annotation      | field (literal)            | annote                   | 
!                                               |
| annotator       | list (name)                | ---                      | 
no corresp. CSL var                             |
| author          | list (name)                | author                   | 
!                                               |
| authortype      | field (key)                | ---                      | 
no corresp. CSL var                             |
| bookauthor      | list (name)                | container-author         | 
!                                               |
| bookpagination  | field (key)                | ?                        | 
no corresp. CSL var (?), but check CSL locators |
| booksubtitle    | field (literal)            | (container-title)        | 
concat w/ booktitle                             |
| booktitle       | field (literal)            | container-title          | 
or volume-title if maintitle exists             |
| booktitleaddon  | field (literal)            | (container-title)        | 
concat w/ booktitle                             |
| chapter         | field (literal)            | chapter-number           | 
at least if numeric                             |
| commentator     | list (name)                | ---                      | 
no corresp. CSL var                             |
| date            | field (date)               | issued                   | 
!                                               |
| doi             | field (verbatim)           | DOI                      | 
!                                               |
| edition         | field (integer or literal) | ---                      | 
---                                             |
| editor          | list (name)                | editor, or other type    | 
depending on editortype: CSL has composer,      |
| editora         | list (name)                | editor, or other type    
|    director, editorial-director, illustrator,   |
| editorb         | list (name)                | editor, or other type    
|    interviewer, recipient                       |
| editorc         | list (name)                | editor, or other type    
|                                                 |
| editortype      | field (key)                | ---                      | 
determines editor's CSL type                    |
| editoratype     | field (key)                | ---                      | 
determines editora's CSL type                   |
| editorbtype     | field (key)                | ---                      | 
determines editorb's CSL type                   |
| editorctype     | field (key)                | ---                      | 
determines editorc's CSL type                   |
| eid             | field (literal)            | ---                      | 
---                                             |
| entrysubtype    | field (literal)            | ---                      | 
may modify entry type of @article               |
| eprint          | field (verbatim)           | PMID                     | 
only if eprinttype = {pubmed} ...               |
| eprintclass     | field (literal)            | ---                      | 
... other eprinttypes have no CSL counterpart   |
| eprinttype      | field (literal)            | ---                      
|                                                 |
| eventdate       | field (date)               | event-date               | 
!                                               |
| eventtitle      | field (literal)            | event                    | 
!                                               |
| file            | field (verbatim)           | ---                      | 
no corresp. CSL var                             |
| foreword        | list (name)                | ---                      | 
no corresp. CSL var                             |
| holder          | list (name)                | ---                      | 
no corresp. CSL var                             |
| howpublished    | field (literal)            | publisher                | 
h.+p. never co-occur; h.+orga. do *             |
| indextitle      | field (literal)            | ---                      | 
---                                             |
| institution     | list (literal)             | publisher                | 
i. + p. never co-occur *                        |
| introduction    | list (name)                | ---                      | 
no corresp. CSL var                             |
| isan            | field (literal)            | ---                      | 
no corresp. CSL var                             |
| isbn            | field (literal)            | ISBN                     | 
!                                               |
| ismn            | field (literal)            | ---                      | 
no corresp. CSL var                             |
| isrn            | field (literal)            | ---                      | 
no corresp. CSL var                             |
| issn            | field (literal)            | ISSN                     | 
!                                               |
| issue           | field (literal)            | issue                    | 
can issue + number co-occur?                    |
| issuesubtitle   | field (literal)            | ---                      | 
concat after issuetitle                         |
| issuetitle      | field (literal)            | ---                      | 
CSL var proposed: volume-title/issue-title      |
| iswc            | field (literal)            | ---                      | 
---                                             |
| journalsubtitle | field (literal)            | (container-title)        | 
concat after journaltitle                       |
| journaltitle    | field (literal)            | container-title          | 
!                                               |
| label           | field (literal)            | citation-label (?)       
|                (???)                            |
| language        | list (key)                 | ---                      | 
no corresp. CSL var (!!!)                       |
| library         | field (literal)            | call-number              | 
[at least close]                                |
| location        | list (literal)             | publisher-place          | 
!                                               |
| mainsubtitle    | field (literal)            | (title/container-title)  | 
concat w/ maintitle                             |
| maintitle       | field (literal)            | title/container-title    | 
!                                               |
| maintitleaddon  | field (literal)            | (title/container-title)  | 
concat w/ maintitle                             |
| month           | field (integer)            | ---                      | 
see date (?)                                    |
| nameaddon       | field (literal)            | ---                      | 
no corresp. CSL var                             |
| note            | field (literal)            | note                     | 
concat w/ addendum                              |
| number          | field (literal)            | issue                    | 
if in @article (?)                              |
|                 |                            | number                   | 
if in @report (?)                               |
|                 |                            | collection-number        | 
all other types, at least if series exists      |
| organization    | list (literal)             | ---                      | 
*                                               |
| origdate        | field (date)               | original-date            | 
!                                               |
| origlanguage    | field (key)                | ---                      | 
no corresp. CSL var                             |
| origlocation    | list (literal)             | original-publisher-place | 
!                                               |
| origpublisher   | list (literal)             | original-publisher       | 
!                                               |
| origtitle       | field (literal)            | original-title           | 
!                                               |
| pages           | field (range)              | page                     | 
!                                               |
| pagetotal       | field (literal)            | number-of-pages          | 
!                                               |
| pagination      | field (key)                | ?                        | 
no corresp. CSL var (?), but check locators     |
| part            | field (literal)            | ?                        | 
concat w/ volume (?)                            |
| publisher       | list (literal)             | publisher                | 
! *                                             |
| pubstate        | field (key)                | status                   | 
! (translate localization keys!)                |
| reprinttitle    | field (literal)            | ---                      | 
"BibTeX only"=obsolete?                         |
| series          | field (literal)            | collection-title         | 
if @article, concat after container-title       |
| shortauthor     | list (name)                | ---                      | 
no corresp. CSL var                             |
| shorteditor     | list (name)                | ---                      | 
no corresp. CSL var                             |
| shorthand       | field (literal)            | ---                      | 
no corresp. CSL var                             |
| shorthandintro  | field (literal)            | ---                      | 
no corresp. CSL var                             |
| shortjournal    | field (literal)            | container-title-short    | 
!                                               |
| shortseries     | field (literal)            | ---                      | 
no corresp. CSL var                             |
| shorttitle      | field (literal)            | title-short              | 
!                                               |
| subtitle        | field (literal)            | (title)                  | 
concat after title                              |
| title           | field (literal)            | title                    | 
!                                               |
| titleaddon      | field (literal)            | (title)                  | 
concat after title+subtitle (not optimal)       |
| translator      | list (name)                | translator               | 
!                                               |
| type            | field (key)                | genre                    | 
!                                               |
| url             | field (uri)                | URL                      | 
!                                               |
| urldate         | field (date)               | accessed                 | 
!                                               |
| venue           | field (literal)            | event-place              | 
!                                               |
| version         | field (literal)            | version                  | 
!                                               |
| volume          | field (literal)            | volume                   | 
!                                               |
| volumes         | field (literal)            | number-of-volumes        | 
!                                               |
| year            | field (literal)            | issued                   | 
! only use if date empty                        |


! means straightforward mapping is possible

* Some entry types may contain two biblatex fields, for both of which only 
mapping to CSL publisher seems to make sense:
  howpublished + organization co-occur in @misc;
  organization + publisher co-occur in @manual, @proceedings, 
@mvproceedings, @inproceedings.
  Maybe simply always concat institution + organization + howpublished + 
publisher to CSL publisher, no matter what the entry type is?

    
2.2.3 Special Fields

| biblatex field | biblatex field 
type                                        | CSL      | 
comment                              |
|:---------------|:-----------------------------------------------------------|:---------|:-------------------------------------|
| crossref       | field (entry 
key)                                          |          | field for 
cross-referencing entries  |
| entryset       | field (comma-separated 
values)                             |          
|                                      |
| execute        | field 
(code)                                               |          
|                                      |
| gender         | field (Pattern matching one of sf, sm, sn, pf, pm, pn, 
pp) | ---      | no corresp. CSL var                  |
| hyphenation    | field 
(identifier)                                         | language | 
!                                    |
| ids            | field (comma-separated list of entrykeys (biber 
only))     | ?        | YAML entries w/ >=1 id allowed?      |
| indexsorttitle | field 
(literal)                                            |          
|                                      |
| keywords       | field (comma-separated 
values)                             | keyword  | 
!                                    |
| options        | field (comma-separated \keyval 
options)                    |          | at least a subset of options 
could   |
|                
|                                                            |          | 
prob. be used, e.g., url=false to    |
|                
|                                                            |          | 
suppress the URL                     |
| presort        | field 
(string)                                             |          
|                                      |
| related        | field (comma-separated 
values)                             |          
|                                      |
| relatedoptions | field (comma-separated values (biber 
only))                |          |                                      |
| relatedtype    | field (identifier (biber 
only))                            |          
|                                      |
| relatedstring  | field (literal) (biber 
only)                               |          
|                                      |
| sortkey        | field (literal) (biber 
only)                               |          
|                                      |
| sortname       | list 
(name)                                                |          
|                                      |
| sortshorthand  | field (literal) (biber 
only)                               |          
|                                      |
| sorttitle      | field 
(literal)                                            |          
|                                      |
| sortyear       | field 
(literal)                                            |          
|                                      |
| xdata          | field (comma-separated list of entrykeys (biber 
only))     |          |                                      |
| xref           | field (entry 
key)                                          |          | field for 
cross-referencing entries  |

2.2.4 Custom Fields

| biblatex field | biblatex field type | CSL | comment |
|:---------------|:--------------------|:----|:--------|
| name[a-c]      | list (name)         |     |         |
| name[a-c]type  | field (key)         |     |         |
| list[a-f]      | list (literal)      |     |         |
| user[a-c]      | field (literal)     |     |         |
| verb[a-c]      | field (literal)     |     |         |

2.2.5 Field Aliases

| biblatex field | biblatex field type | CSL             | 
comment                |
|:---------------|:--------------------|:----------------|:-----------------------|
| address        | list (literal)      | publisher-place | alias for 
location     |
| annote         | field (literal)     | annote          | alias for 
annotation   |
| archiveprefix  | field (literal)     |                 
|                        |
| journal        | field (literal)     | container-title | alias for 
journaltitle |
| key            | field (literal)     |                 
|                        |
| pdf            | field (verbatim)    |                 
|                        |
| primaryclass   | field (literal)     |                 
|                        |
| school         | list (literal)      | publisher       | alias for 
publisher    |



On Friday, September 13, 2013 2:36:45 PM UTC, fiddlosopher wrote:
>
> You raise some good points.  All of this is possible, but I think 
> I'd like to work a bit on the underlying framework to make it all easier 
> to implement. 
>
> +++ Nick Bart [Sep 13 13 07:26 ]: 
> >    Fantastic. I'm amazed by your energy. And, yes, I'd love to 
> contribute. 
> >    Time to learn a little Haskell after all, I guess. 
> >    Still, a few questions first: 
> >    1. Would it make sense to cleanly separate code for bibtex vs. code 
> for 
> >    biblatex? Handling of dates, and a number of other things, see below, 
> >    differs a lot. 
>
> Since so much is the same, the thing to do is probably to add an 
> ifBibTeX conditional. 
>
> >    2. (How) can the bibtex and biblatex "crossref" mechanism for 
> >    inheritance be implemented? (biblatex has introduced an alternative 
> >    cross-referencing mechanism using the xref field, but I'm not sure 
> how 
> >    many people have actually started using this. I haven't seen any 
> >    advantage in using this instead of crossref yet.) 
>
> I don't know.  Can you give an example of the use of crossref? 
>
> >    3. Some fields will have to be concatenated, 
> >        - title + subtitle + titleaddon (same for booktitle, maintitle), 
> >    and journal/journaltitle + journalsubtitle. How will this be handled? 
> >    (titleaddon etc. are supposed to be printed "in a different font", 
> >    i.e., unitalicized or outside of speech marks, but this currently 
> does 
> >    not seem to be possible in CSL.) 
> >        - note + addendum. biblatex has addendum at the end of an entry, 
> >    and note somewhere in front of that, but in CSL there's only one 
> >    suitable field: note. Differs from biblatex "annotation" (and its 
> alias 
> >    "annote") which map to CSL "annote", and contain reader's notes. 
> >    4. Some fields have to be omitted if some certain other/newer 
> field(s) 
> >    are present: In biblatex, "date", if present, is used instead of 
> >    "year", "month", and "day"; "location" instead of "address", etc. How 
> >    will this be handled? 
> >    5. Proper mapping of some fields or entry types is conditional upon 
> >    either the existence or the content of other fields. How can this be 
> >    made to work? 
> >        - for @book, @collection, @proceedings: title ==> title, unless 
> >    maintitle exists, in which case maintitle ==> title (and, once 
> >    consensus on new CSL variables is reached, title ==> volume-title; as 
> >    discussed earlier in this thread), 
> >        - for @inbook, @incollection, @proceedings: booktitle ==> 
> >    container-title, unless maintitle exists, in which case maintitle ==> 
> >    container-title (and booktitle ==> volume-title), 
> >        - mapping of editor, editora, editorb, editorc depends on the 
> >    content of editortype, editoratype, editorbtype, editorctype, 
> >        - An entrysubtype of "magazine" should change the CSL type from 
> >    "article-journal" to "article-magazine" (this is not from biblatex 
> >    itself but from biblatex-chicago, but I think it'd be very useful to 
> >    borrow certain bits from this package; an entrysubtype of "newspaper" 
> >    could of course map to CSL "article-newspaper"). 
> >        - series maps to CSL collection-title, except for @article (and 
> >    @periodical?), in which case it should appear after the journal title 
> >    (the only solution in CSL is probably to concatenate the two fields). 
> >    6. "keys" are used liberally in biblatex fields, as shorthands, and 
> for 
> >    localization (see biblatex manual 4.9). These should be translated to 
> >    natural language terms. Let's focus on English first - but what about 
> >    localization ? 
> >    7. In addition to "name lists" (authors etc.), biblatex has "literal 
> >    lists" with elements separated by "and", too: institution, location, 
> >    organization, origlocation, origpublisher, publisher, list[a-f], 
> >    address, school. Will routines used for names work unmodified? 
>
> No, because this routine includes code for dividing a single name into 
> given, last, and prefix.  But it will be easy to write something that 
> will work for these. 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/73c9fc49-d61e-457e-9304-cb42ca1ccac7%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 67950 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Bruce D'Arcus]

I will just emphasize, I view the biblatex and csl models as both a
superset of bibtex. E.g. bibtex isn't a great target format.

On Tue, Sep 24, 2013 at 9:11 AM, Nick Bart <nickbart1980-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> For what it's worth, this is my list of mappings from all biblatex fields
> (as listed in the biblatex manual, §§ 2.2.2–2.2.5) to matching CSL
> variables, with a few comments thrown in. I hope it's useful both for
> developing bibtex2pandoc and for testing existing conversion routines such
> as biblatex -> MODS -> CSL.
>
> Feedback on content and formatting is welcome, and maybe some improved
> version could even be integrated into the general pandoc documentation.
>
> 2.2.2 Data Fields
>
> | biblatex field  | biblatex field type        | CSL                      |
> comment                                         |
> |:----------------|:---------------------------|:-------------------------|:------------------------------------------------|
> | abstract        | field (literal)            | abstract                 |
> !                                               |
> | addendum        | field (literal)            | note                     |
> concat w/ note                                  |
> | afterword       | list (name)                | ---                      |
> no corresp. CSL var                             |
> | annotation      | field (literal)            | annote                   |
> !                                               |
> | annotator       | list (name)                | ---                      |
> no corresp. CSL var                             |
> | author          | list (name)                | author                   |
> !                                               |
> | authortype      | field (key)                | ---                      |
> no corresp. CSL var                             |
> | bookauthor      | list (name)                | container-author         |
> !                                               |
> | bookpagination  | field (key)                | ?                        |
> no corresp. CSL var (?), but check CSL locators |
> | booksubtitle    | field (literal)            | (container-title)        |
> concat w/ booktitle                             |
> | booktitle       | field (literal)            | container-title          |
> or volume-title if maintitle exists             |
> | booktitleaddon  | field (literal)            | (container-title)        |
> concat w/ booktitle                             |
> | chapter         | field (literal)            | chapter-number           |
> at least if numeric                             |
> | commentator     | list (name)                | ---                      |
> no corresp. CSL var                             |
> | date            | field (date)               | issued                   |
> !                                               |
> | doi             | field (verbatim)           | DOI                      |
> !                                               |
> | edition         | field (integer or literal) | ---                      |
> ---                                             |
> | editor          | list (name)                | editor, or other type    |
> depending on editortype: CSL has composer,      |
> | editora         | list (name)                | editor, or other type    |
> director, editorial-director, illustrator,   |
> | editorb         | list (name)                | editor, or other type    |
> interviewer, recipient                       |
> | editorc         | list (name)                | editor, or other type    |
> |
> | editortype      | field (key)                | ---                      |
> determines editor's CSL type                    |
> | editoratype     | field (key)                | ---                      |
> determines editora's CSL type                   |
> | editorbtype     | field (key)                | ---                      |
> determines editorb's CSL type                   |
> | editorctype     | field (key)                | ---                      |
> determines editorc's CSL type                   |
> | eid             | field (literal)            | ---                      |
> ---                                             |
> | entrysubtype    | field (literal)            | ---                      |
> may modify entry type of @article               |
> | eprint          | field (verbatim)           | PMID                     |
> only if eprinttype = {pubmed} ...               |
> | eprintclass     | field (literal)            | ---                      |
> ... other eprinttypes have no CSL counterpart   |
> | eprinttype      | field (literal)            | ---                      |
> |
> | eventdate       | field (date)               | event-date               |
> !                                               |
> | eventtitle      | field (literal)            | event                    |
> !                                               |
> | file            | field (verbatim)           | ---                      |
> no corresp. CSL var                             |
> | foreword        | list (name)                | ---                      |
> no corresp. CSL var                             |
> | holder          | list (name)                | ---                      |
> no corresp. CSL var                             |
> | howpublished    | field (literal)            | publisher                |
> h.+p. never co-occur; h.+orga. do *             |
> | indextitle      | field (literal)            | ---                      |
> ---                                             |
> | institution     | list (literal)             | publisher                |
> i. + p. never co-occur *                        |
> | introduction    | list (name)                | ---                      |
> no corresp. CSL var                             |
> | isan            | field (literal)            | ---                      |
> no corresp. CSL var                             |
> | isbn            | field (literal)            | ISBN                     |
> !                                               |
> | ismn            | field (literal)            | ---                      |
> no corresp. CSL var                             |
> | isrn            | field (literal)            | ---                      |
> no corresp. CSL var                             |
> | issn            | field (literal)            | ISSN                     |
> !                                               |
> | issue           | field (literal)            | issue                    |
> can issue + number co-occur?                    |
> | issuesubtitle   | field (literal)            | ---                      |
> concat after issuetitle                         |
> | issuetitle      | field (literal)            | ---                      |
> CSL var proposed: volume-title/issue-title      |
> | iswc            | field (literal)            | ---                      |
> ---                                             |
> | journalsubtitle | field (literal)            | (container-title)        |
> concat after journaltitle                       |
> | journaltitle    | field (literal)            | container-title          |
> !                                               |
> | label           | field (literal)            | citation-label (?)       |
> (???)                            |
> | language        | list (key)                 | ---                      |
> no corresp. CSL var (!!!)                       |
> | library         | field (literal)            | call-number              |
> [at least close]                                |
> | location        | list (literal)             | publisher-place          |
> !                                               |
> | mainsubtitle    | field (literal)            | (title/container-title)  |
> concat w/ maintitle                             |
> | maintitle       | field (literal)            | title/container-title    |
> !                                               |
> | maintitleaddon  | field (literal)            | (title/container-title)  |
> concat w/ maintitle                             |
> | month           | field (integer)            | ---                      |
> see date (?)                                    |
> | nameaddon       | field (literal)            | ---                      |
> no corresp. CSL var                             |
> | note            | field (literal)            | note                     |
> concat w/ addendum                              |
> | number          | field (literal)            | issue                    |
> if in @article (?)                              |
> |                 |                            | number                   |
> if in @report (?)                               |
> |                 |                            | collection-number        |
> all other types, at least if series exists      |
> | organization    | list (literal)             | ---                      |
> *                                               |
> | origdate        | field (date)               | original-date            |
> !                                               |
> | origlanguage    | field (key)                | ---                      |
> no corresp. CSL var                             |
> | origlocation    | list (literal)             | original-publisher-place |
> !                                               |
> | origpublisher   | list (literal)             | original-publisher       |
> !                                               |
> | origtitle       | field (literal)            | original-title           |
> !                                               |
> | pages           | field (range)              | page                     |
> !                                               |
> | pagetotal       | field (literal)            | number-of-pages          |
> !                                               |
> | pagination      | field (key)                | ?                        |
> no corresp. CSL var (?), but check locators     |
> | part            | field (literal)            | ?                        |
> concat w/ volume (?)                            |
> | publisher       | list (literal)             | publisher                |
> ! *                                             |
> | pubstate        | field (key)                | status                   |
> ! (translate localization keys!)                |
> | reprinttitle    | field (literal)            | ---                      |
> "BibTeX only"=obsolete?                         |
> | series          | field (literal)            | collection-title         |
> if @article, concat after container-title       |
> | shortauthor     | list (name)                | ---                      |
> no corresp. CSL var                             |
> | shorteditor     | list (name)                | ---                      |
> no corresp. CSL var                             |
> | shorthand       | field (literal)            | ---                      |
> no corresp. CSL var                             |
> | shorthandintro  | field (literal)            | ---                      |
> no corresp. CSL var                             |
> | shortjournal    | field (literal)            | container-title-short    |
> !                                               |
> | shortseries     | field (literal)            | ---                      |
> no corresp. CSL var                             |
> | shorttitle      | field (literal)            | title-short              |
> !                                               |
> | subtitle        | field (literal)            | (title)                  |
> concat after title                              |
> | title           | field (literal)            | title                    |
> !                                               |
> | titleaddon      | field (literal)            | (title)                  |
> concat after title+subtitle (not optimal)       |
> | translator      | list (name)                | translator               |
> !                                               |
> | type            | field (key)                | genre                    |
> !                                               |
> | url             | field (uri)                | URL                      |
> !                                               |
> | urldate         | field (date)               | accessed                 |
> !                                               |
> | venue           | field (literal)            | event-place              |
> !                                               |
> | version         | field (literal)            | version                  |
> !                                               |
> | volume          | field (literal)            | volume                   |
> !                                               |
> | volumes         | field (literal)            | number-of-volumes        |
> !                                               |
> | year            | field (literal)            | issued                   |
> ! only use if date empty                        |
>
>
> ! means straightforward mapping is possible
>
> * Some entry types may contain two biblatex fields, for both of which only
> mapping to CSL publisher seems to make sense:
>   howpublished + organization co-occur in @misc;
>   organization + publisher co-occur in @manual, @proceedings,
> @mvproceedings, @inproceedings.
>   Maybe simply always concat institution + organization + howpublished +
> publisher to CSL publisher, no matter what the entry type is?
>
>
> 2.2.3 Special Fields
>
> | biblatex field | biblatex field type
> | CSL      | comment                              |
> |:---------------|:-----------------------------------------------------------|:---------|:-------------------------------------|
> | crossref       | field (entry key)
> |          | field for cross-referencing entries  |
> | entryset       | field (comma-separated values)
> |          |                                      |
> | execute        | field (code)
> |          |                                      |
> | gender         | field (Pattern matching one of sf, sm, sn, pf, pm, pn,
> pp) | ---      | no corresp. CSL var                  |
> | hyphenation    | field (identifier)
> | language | !                                    |
> | ids            | field (comma-separated list of entrykeys (biber only))
> | ?        | YAML entries w/ >=1 id allowed?      |
> | indexsorttitle | field (literal)
> |          |                                      |
> | keywords       | field (comma-separated values)
> | keyword  | !                                    |
> | options        | field (comma-separated \keyval options)
> |          | at least a subset of options could   |
> |                |
> |          | prob. be used, e.g., url=false to    |
> |                |
> |          | suppress the URL                     |
> | presort        | field (string)
> |          |                                      |
> | related        | field (comma-separated values)
> |          |                                      |
> | relatedoptions | field (comma-separated values (biber only))
> |          |                                      |
> | relatedtype    | field (identifier (biber only))
> |          |                                      |
> | relatedstring  | field (literal) (biber only)
> |          |                                      |
> | sortkey        | field (literal) (biber only)
> |          |                                      |
> | sortname       | list (name)
> |          |                                      |
> | sortshorthand  | field (literal) (biber only)
> |          |                                      |
> | sorttitle      | field (literal)
> |          |                                      |
> | sortyear       | field (literal)
> |          |                                      |
> | xdata          | field (comma-separated list of entrykeys (biber only))
> |          |                                      |
> | xref           | field (entry key)
> |          | field for cross-referencing entries  |
>
> 2.2.4 Custom Fields
>
> | biblatex field | biblatex field type | CSL | comment |
> |:---------------|:--------------------|:----|:--------|
> | name[a-c]      | list (name)         |     |         |
> | name[a-c]type  | field (key)         |     |         |
> | list[a-f]      | list (literal)      |     |         |
> | user[a-c]      | field (literal)     |     |         |
> | verb[a-c]      | field (literal)     |     |         |
>
> 2.2.5 Field Aliases
>
> | biblatex field | biblatex field type | CSL             | comment
> |
> |:---------------|:--------------------|:----------------|:-----------------------|
> | address        | list (literal)      | publisher-place | alias for
> location     |
> | annote         | field (literal)     | annote          | alias for
> annotation   |
> | archiveprefix  | field (literal)     |                 |
> |
> | journal        | field (literal)     | container-title | alias for
> journaltitle |
> | key            | field (literal)     |                 |
> |
> | pdf            | field (verbatim)    |                 |
> |
> | primaryclass   | field (literal)     |                 |
> |
> | school         | list (literal)      | publisher       | alias for
> publisher    |
>
>
>
> On Friday, September 13, 2013 2:36:45 PM UTC, fiddlosopher wrote:
>>
>> You raise some good points.  All of this is possible, but I think
>> I'd like to work a bit on the underlying framework to make it all easier
>> to implement.
>>
>> +++ Nick Bart [Sep 13 13 07:26 ]:
>> >    Fantastic. I'm amazed by your energy. And, yes, I'd love to
>> > contribute.
>> >    Time to learn a little Haskell after all, I guess.
>> >    Still, a few questions first:
>> >    1. Would it make sense to cleanly separate code for bibtex vs. code
>> > for
>> >    biblatex? Handling of dates, and a number of other things, see below,
>> >    differs a lot.
>>
>> Since so much is the same, the thing to do is probably to add an
>> ifBibTeX conditional.
>>
>> >    2. (How) can the bibtex and biblatex "crossref" mechanism for
>> >    inheritance be implemented? (biblatex has introduced an alternative
>> >    cross-referencing mechanism using the xref field, but I'm not sure
>> > how
>> >    many people have actually started using this. I haven't seen any
>> >    advantage in using this instead of crossref yet.)
>>
>> I don't know.  Can you give an example of the use of crossref?
>>
>> >    3. Some fields will have to be concatenated,
>> >        - title + subtitle + titleaddon (same for booktitle, maintitle),
>> >    and journal/journaltitle + journalsubtitle. How will this be handled?
>> >    (titleaddon etc. are supposed to be printed "in a different font",
>> >    i.e., unitalicized or outside of speech marks, but this currently
>> > does
>> >    not seem to be possible in CSL.)
>> >        - note + addendum. biblatex has addendum at the end of an entry,
>> >    and note somewhere in front of that, but in CSL there's only one
>> >    suitable field: note. Differs from biblatex "annotation" (and its
>> > alias
>> >    "annote") which map to CSL "annote", and contain reader's notes.
>> >    4. Some fields have to be omitted if some certain other/newer
>> > field(s)
>> >    are present: In biblatex, "date", if present, is used instead of
>> >    "year", "month", and "day"; "location" instead of "address", etc. How
>> >    will this be handled?
>> >    5. Proper mapping of some fields or entry types is conditional upon
>> >    either the existence or the content of other fields. How can this be
>> >    made to work?
>> >        - for @book, @collection, @proceedings: title ==> title, unless
>> >    maintitle exists, in which case maintitle ==> title (and, once
>> >    consensus on new CSL variables is reached, title ==> volume-title; as
>> >    discussed earlier in this thread),
>> >        - for @inbook, @incollection, @proceedings: booktitle ==>
>> >    container-title, unless maintitle exists, in which case maintitle ==>
>> >    container-title (and booktitle ==> volume-title),
>> >        - mapping of editor, editora, editorb, editorc depends on the
>> >    content of editortype, editoratype, editorbtype, editorctype,
>> >        - An entrysubtype of "magazine" should change the CSL type from
>> >    "article-journal" to "article-magazine" (this is not from biblatex
>> >    itself but from biblatex-chicago, but I think it'd be very useful to
>> >    borrow certain bits from this package; an entrysubtype of "newspaper"
>> >    could of course map to CSL "article-newspaper").
>> >        - series maps to CSL collection-title, except for @article (and
>> >    @periodical?), in which case it should appear after the journal title
>> >    (the only solution in CSL is probably to concatenate the two fields).
>> >    6. "keys" are used liberally in biblatex fields, as shorthands, and
>> > for
>> >    localization (see biblatex manual 4.9). These should be translated to
>> >    natural language terms. Let's focus on English first - but what about
>> >    localization ?
>> >    7. In addition to "name lists" (authors etc.), biblatex has "literal
>> >    lists" with elements separated by "and", too: institution, location,
>> >    organization, origlocation, origpublisher, publisher, list[a-f],
>> >    address, school. Will routines used for names work unmodified?
>>
>> No, because this routine includes code for dividing a single name into
>> given, last, and prefix.  But it will be easy to write something that
>> will work for these.
>>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "pandoc-discuss" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/pandoc-discuss/-SajbqoPX8k/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pandoc-discuss/73c9fc49-d61e-457e-9304-cb42ca1ccac7%40googlegroups.com.
>
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAF-FPGPULneZLmjD9qC5W%2BnxiXzU4mGQCA6V%2BUfSkHTBSEPkmg%40mail.gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Nick,

This is going to be very helpful.

I've tinkered a bit more with bibtex2pandoc.hs.  My goal is to make
it possible for you to work on the source yourself, with minimal Haskell
knowledge.   I'm getting closer, but I still have a view things left.

In the end, this should be made part of pandoc-citeproc, and it should
replace bibutils for reading bibtex and biblatex files.

John

+++ Nick Bart [Sep 24 13 06:11 ]:
>    For what it's worth, this is my list of mappings from all biblatex
>    fields (as listed in the biblatex manual, 2.2.22.2.5) to matching CSL
>    variables, with a few comments thrown in. I hope it's useful both for
>    developing bibtex2pandoc and for testing existing conversion routines
>    such as biblatex -> MODS -> CSL.
>    Feedback on content and formatting is welcome, and maybe some improved
>    version could even be integrated into the general pandoc documentation.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Joseph Reagle]

On 09/24/2013 09:11 AM, Nick Bart wrote:
> For what it's worth, this is my list of mappings from all biblatex fields 
> (as listed in the biblatex manual, §§ 2.2.2–2.2.5) to matching CSL 
> variables, with a few comments thrown in. I hope it's useful both for 
> developing bibtex2pandoc and for testing existing conversion routines such 
> as biblatex -> MODS -> CSL.
> 
> Feedback on content and formatting is welcome, and maybe some improved 
> version could even be integrated into the general pandoc documentation.

Do you have a Web page or something for this?

I was working on something similar for my own purposes (as I likely
extend and migrate my data model towards CSL).

  http://reagle.org/joseph/2013/08/bib-mapping.html

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/52431C6C.1070003%40reagle.org.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Bruce D'Arcus]

Worth noting that the design of bibutils involves a common target
format, that might be analogous to something like pandocs native
format. It then has different readers and writers for the different
bib formats. This makes it easy to add new formats, and edit existing
ones (the C code has a bunch of simple mapping tables).

It seems a design like that would make sense here as well?

FWIW, I once tried to operationalize this idea in this effort to use
some simple json config files to set up the mapping:

https://github.com/bdarcus/bibmaps

But I never got very far, and am not sure how practical that is (it
might be; the tricky part would be how to configure parsing routines
for different name and date formats).

On Fri, Sep 13, 2013 at 10:26 AM, Nick Bart <nickbart1980-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> Fantastic. I'm amazed by your energy. And, yes, I'd love to contribute. Time
> to learn a little Haskell after all, I guess.
>
> Still, a few questions first:
>
> 1. Would it make sense to cleanly separate code for bibtex vs. code for
> biblatex? Handling of dates, and a number of other things, see below,
> differs a lot.
>
> 2. (How) can the bibtex and biblatex "crossref" mechanism for inheritance be
> implemented? (biblatex has introduced an alternative cross-referencing
> mechanism using the xref field, but I'm not sure how many people have
> actually started using this. I haven't seen any advantage in using this
> instead of crossref yet.)
>
> 3. Some fields will have to be concatenated,
>
>     - title + subtitle + titleaddon (same for booktitle, maintitle), and
> journal/journaltitle + journalsubtitle. How will this be handled?
> (titleaddon etc. are supposed to be printed "in a different font", i.e.,
> unitalicized or outside of speech marks, but this currently does not seem to
> be possible in CSL.)
>
>     - note + addendum. biblatex has addendum at the end of an entry, and
> note somewhere in front of that, but in CSL there's only one suitable field:
> note. – Differs from biblatex "annotation" (and its alias "annote") which
> map to CSL "annote", and contain reader's notes.
>
> 4. Some fields have to be omitted if some certain other/newer field(s) are
> present: In biblatex, "date", if present, is used instead of "year",
> "month", and "day"; "location" instead of "address", etc. How will this be
> handled?
>
> 5. Proper mapping of some fields or entry types is conditional upon either
> the existence or the content of other fields. How can this be made to work?
>
>     - for @book, @collection, @proceedings: title ==> title, unless
> maintitle exists, in which case maintitle ==> title (and, once consensus on
> new CSL variables is reached, title ==> volume-title; as discussed earlier
> in this thread),
>
>     - for @inbook, @incollection, @proceedings: booktitle ==>
> container-title, unless maintitle exists, in which case maintitle ==>
> container-title (and booktitle ==> volume-title),
>
>     - mapping of editor, editora, editorb, editorc depends on the content of
> editortype, editoratype, editorbtype, editorctype,
>
>     - An entrysubtype of "magazine" should change the CSL type from
> "article-journal" to "article-magazine" (this is not from biblatex itself
> but from biblatex-chicago, but I think it'd be very useful to borrow certain
> bits from this package; an entrysubtype of "newspaper" could of course map
> to CSL "article-newspaper").
>
>     - series maps to CSL collection-title, except for @article (and
> @periodical?), in which case it should appear after the journal title (the
> only solution in CSL is probably to concatenate the two fields).
>
> 6. "keys" are used liberally in biblatex fields, as shorthands, and for
> localization (see biblatex manual 4.9). These should be translated to
> natural language terms. Let's focus on English first - but what about
> localization …?
>
> 7. In addition to "name lists" (authors etc.), biblatex has "literal lists"
> with elements separated by "and", too: institution, location, organization,
> origlocation, origpublisher, publisher, list[a-f], address, school. Will
> routines used for names work unmodified?
>
>
>
> On Friday, 13 September 2013 08:24:45 UTC+2, fiddlosopher wrote:
>>
>> I've renamed to bibtex2pandoc.hs and made several improvements,
>> including parsing of non-dropping-particles.
>>
>> I've added all the biblatex entry types, and I've started adding support
>> for biblatex data fields.  (In the biblatex list of data fields,
>> 2.2.2 of http://www.math.upenn.edu/tex_docs/latex/biblatex/biblatex.pdf,
>> I've gotten as far as 'annotation'.)
>>
>> Nick, if you (or others) want to work on this, it should be pretty
>> self-explanatory how to add to the code, even if you don't know Haskell.
>> (Find the part of the code with all the arrows.  ==> means the field is
>> to be interpreted as latex; --> means it's a raw string; *=> is an
>> author list.)
>>
>> Recompile with 'ghc --make bibtex2pandoc.hs'
>>
>> John
>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "pandoc-discuss" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/pandoc-discuss/-SajbqoPX8k/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pandoc-discuss/b01350cd-72e7-49e5-aa50-6c25382376f6%40googlegroups.com.
>
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAF-FPGN4OCOGUY%2BzfcAAZmobp%3Da2YemLtkvppGPmWdW-KOiu1w%40mail.gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Nick,

I've made some more improvements to bibtex2pandoc.hs, including a kind
of monadic domain-specific language for doing the actual mapping.
This occurs in the function itemToMetaValue, which you can find in
the source.  I've tried to set this up so you can modify it without
knowing much Haskell.

(To compile, you'll need to make sure you've cabal installed pandoc
and pandoc-citeproc, and also the bibtex package and perhaps some
others.  Then:  ghc --make bibtex2pandoc)

Answers to questions below:

+++ Nick Bart [Sep 13 13 07:26 ]:
> Fantastic. I'm amazed by your energy. And, yes, I'd love to contribute. 
> Time to learn a little Haskell after all, I guess.
> 
> Still, a few questions first:
> 
> 1. Would it make sense to cleanly separate code for bibtex vs. code for 
> biblatex? Handling of dates, and a number of other things, see below, 
> differs a lot.

itemToMetaValue has a boolean parameter 'bibtex' which you can test
for in places where we need to distinguish.  Example:

  unless bibtex $ do
      getField "addendum" ==> appendField "note" (Space:)

> 2. (How) can the bibtex and biblatex "crossref" mechanism for inheritance 
> be implemented? (biblatex has introduced an alternative cross-referencing 
> mechanism using the xref field, but I'm not sure how many people have 
> actually started using this. I haven't seen any advantage in using this 
> instead of crossref yet.)

I have implemented this (including the biblatex mappings, if --bibtex
isn't used.)

> 3. Some fields will have to be concatenated,
> 
>     - title + subtitle + titleaddon (same for booktitle, maintitle), and 
> journal/journaltitle + journalsubtitle. How will this be handled? 
> (titleaddon etc. are supposed to be printed "in a different font", i.e., 
> unitalicized or outside of speech marks, but this currently does not seem 
> to be possible in CSL.)
> 
>     - note + addendum. biblatex has addendum at the end of an entry, and 
> note somewhere in front of that, but in CSL there's only one suitable 
> field: note. – Differs from biblatex "annotation" (and its alias "annote") 
> which map to CSL "annote", and contain reader's notes.

There's an 'appendField' function.  It takes as parameters the
target field name and a function of type [Inline] -> [Inline] that
says how to transform the thing you're appending (adding a colon,
a space, putting it in parentheses, etc.).  You can see how this is used
in the lines for "subtitle" and "titleaddon" and "addendum".

> 4. Some fields have to be omitted if some certain other/newer field(s) are 
> present: In biblatex, "date", if present, is used instead of "year", 
> "month", and "day"; "location" instead of "address", etc. How will this be 
> handled?

If you use setField, it will overwrite an existing value.  So,
for example, if you want "maintitle" to override "title" in setting
the "title" value:

  getField "title" ==> setField "title"
  getField "maintitle" ==> setField "title"

If "maintitle" is not found, the earlier value (from "title") will persist.
So we can handle these issues of priority by ordering.

> 5. Proper mapping of some fields or entry types is conditional upon either 
> the existence or the content of other fields. How can this be made to work?
> 
>     - for @book, @collection, @proceedings: title ==> title, unless 
> maintitle exists, in which case maintitle ==> title (and, once consensus on 
> new CSL variables is reached, title ==> volume-title; as discussed earlier 
> in this thread),

See above for this one.

>     - for @inbook, @incollection, @proceedings: booktitle ==> 
> container-title, unless maintitle exists, in which case maintitle ==> 
> container-title (and booktitle ==> volume-title),

  hasMaintitle <- isPresent "maintitle"
  getField "booktitle" ==> setField (if hasMaintitle &&
                                        et `elem` ["book","collection","proceedings"]
                                        then "volume-title"
                                        else "container-title")

>     - mapping of editor, editora, editorb, editorc depends on the content 
> of editortype, editoratype, editorbtype, editorctype,
> 
>     - An entrysubtype of "magazine" should change the CSL type from 
> "article-journal" to "article-magazine" (this is not from biblatex itself 
> but from biblatex-chicago, but I think it'd be very useful to borrow 
> certain bits from this package; an entrysubtype of "newspaper" could of 
> course map to CSL "article-newspaper").

  getRawField "magazine" ==> setRawField "type" "article-magazine"

>     - series maps to CSL collection-title, except for @article (and 
> @periodical?), in which case it should appear after the journal title (the 
> only solution in CSL is probably to concatenate the two fields).

  getField "series" ==> if et == "article" || et == "periodical"
                           then appendField "title" inParens
                           else setField "collection-title"

> 6. "keys" are used liberally in biblatex fields, as shorthands, and for 
> localization (see biblatex manual 4.9). These should be translated to 
> natural language terms. Let's focus on English first - but what about 
> localization …?

I'm ignoring this for now.  citeproc-hs has its own localization.

> 7. In addition to "name lists" (authors etc.), biblatex has "literal lists" 
> with elements separated by "and", too: institution, location, organization, 
> origlocation, origpublisher, publisher, list[a-f], address, school. Will 
> routines used for names work unmodified?

These are supported

  getLiteralList

  getAuthorList

Why don't you try adding to it and see what you can do?

John

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20130927182634.GA37542%40dhcp-128-32-252-11.lips.berkeley.edu.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 13459 bytes --]

Brilliant. But quite a challenge, too. Still, here's a diff with some of 
the more straightforward stuff hacked in. Other comments and questions 
below.

$ diff -u  ../pandoc-citeproc/bibtex2pandoc.hs bibtex2pandoc.hs
--- ../pandoc-citeproc/bibtex2pandoc.hs    2013-09-27 23:05:15.000000000 
+0200
+++ bibtex2pandoc.hs    2013-09-28 00:11:42.000000000 +0200
@@ -240,7 +240,7 @@
        "inproceedings"   -> setType "paper-conference"
        "manual"          -> setType "book"
        "mastersthesis"   -> setType "thesis" >>
-                             setRawField "genre" "Ph.D. thesis"
+                             setRawField "genre" "Master’s thesis"
        "misc"            -> setType "no-type"
        "mvbook"          -> setType "book"
        "mvcollection"    -> setType "book"
@@ -265,12 +265,24 @@
   getRawField "type" ==> setRawField "genre"
   getField "title" ==> setField "title"
   getField "subtitle" ==> appendField "title" addColon
-  getField "titleaddon" ==> appendField "title" inParens
+  getField "titleaddon" ==> appendField "title" addPeriod
+  getField "maintitle" ==> setField "container-title"
+  getField "mainsubtitle" ==> appendField "container-title" addColon
+  getField "maintitleaddon" ==> appendField "container-title" addPeriod
   hasMaintitle <- isPresent "maintitle"
   getField "booktitle" ==> setField (if hasMaintitle &&
-                                        et `elem` 
["book","collection","proceedings"]
+                                        et `elem` 
["inbook","incollection","inproceedings"]
                                         then "volume-title"
                                         else "container-title")
+  getField "booksubtitle" ==> appendField (if hasMaintitle &&
+                                        et `elem` 
["inbook","incollection","inproceedings"]
+                                        then "volume-title"
+                                        else "container-title") addColon
+  getField "booktitleaddon" ==> appendField (if hasMaintitle &&
+                                        et `elem` 
["inbook","incollection","inproceedings"]
+                                        then "volume-title"
+                                        else "container-title") addPeriod
+  getField "shorttitle" ==> setField "title-short"
   getField "series" ==> setField "collection-title"
   getField "pages" ==> setField "page"
   getField "volume" ==> setField "volume"
@@ -280,23 +292,53 @@
   getField "note" ==> setField "note"
   getRawField "url" ==> setRawField "url"
   getField "journal" ==> setField "container-title"
+  getField "journaltitle" ==> setField "container-title"
+  getField "journalsubtitle" ==> appendField "container-title" addColon
+  getField "shortjournal" ==> setField "container-title-short"
+  getField "howpublished" ==> setField "publisher"
   getField "school" ==> setField "publisher"
-  getLiteralList "institution" ==> setList "publisher"
+  unless bibtex $ do
+    getLiteralList "institution" ==> setList "publisher"
+  getField "publisher" ==> setField "publisher"
   getField "address" ==> setField "publisher-place"
+  unless bibtex $ do
+    getField "location" ==> setField "publisher-place"
   getAuthorList "author" ==> setList "author"
   getAuthorList "editor" ==> setList "editor"
-  getField "howpublished" ==> setField "note"
   getField "abstract" ==> setField "abstract"
   unless bibtex $ do
     getField "addendum" ==> appendField "note" (Space:)
   getField "annotation" ==> setField "annote"
+  getField "annote" ==> setField "annote"
   getField "year" ==> setSubField "issued" "year"
   getField "month" ==> setSubField "issued" "month"
-
+  getAuthorList "translator" ==> setList "translator"
+  getAuthorList "bookauthor" ==> setList "container-author"
+  getField "abstract" ==> setField "abstract"
+  getField "keywords" ==> setField "keyword"
+  getField "eventdate" ==> setField "event-date"
+  getField "eventtitle" ==> setField "event"
+  getLiteralList "venue" ==> setList "event-place"
+  getRawField "doi" ==> setRawField "DOI"
+  getRawField "isbn" ==> setRawField "ISBN"
+  getRawField "issn" ==> setRawField "ISSN"
+  getField "origdate" ==> setField "original-date"
+  getLiteralList "origlocation" ==> setList "original-publisher-place"
+  getLiteralList "origpublisher" ==> setList "original-publisher"
+  getField "origtitle" ==> setField "original-title"
+  getField "pagetotal" ==> setField "number-of-pages"
+  getField "pubstate" ==> setField "status"
+  getField "urldate" ==> setField "accessed"
+  getField "version" ==> setField "version"
+  getField "volumes" ==> setField "number-of-volumes"
+  getRawField "hyphenation" ==> setRawField "language"
 
 addColon :: [Inline] -> [Inline]
 addColon xs = [Str ":",Space] ++ xs
 
+addPeriod :: [Inline] -> [Inline]
+addPeriod xs = [Str ".",Space] ++ xs
+
 inParens :: [Inline] -> [Inline]
 inParens xs = [Space, Str "("] ++ xs ++ [Str ")"]

On Friday, 27 September 2013 18:26:34 UTC, fiddlosopher wrote:
>
> Nick, 
>
> I've made some more improvements to bibtex2pandoc.hs, including a kind 
> of monadic domain-specific language for doing the actual mapping. 
> This occurs in the function itemToMetaValue, which you can find in 
> the source.  I've tried to set this up so you can modify it without 
> knowing much Haskell. 
>
> (To compile, you'll need to make sure you've cabal installed pandoc 
> and pandoc-citeproc, and also the bibtex package and perhaps some 
> others.  Then:  ghc --make bibtex2pandoc) 
>
> Answers to questions below: 
>
> +++ Nick Bart [Sep 13 13 07:26 ]: 
> > Fantastic. I'm amazed by your energy. And, yes, I'd love to contribute. 
> > Time to learn a little Haskell after all, I guess. 
> > 
> > Still, a few questions first: 
> > 
> > 1. Would it make sense to cleanly separate code for bibtex vs. code for 
> > biblatex? Handling of dates, and a number of other things, see below, 
> > differs a lot. 
>
> itemToMetaValue has a boolean parameter 'bibtex' which you can test 
> for in places where we need to distinguish.  Example: 
>
>   unless bibtex $ do 
>       getField "addendum" ==> appendField "note" (Space:) 
>
> > 2. (How) can the bibtex and biblatex "crossref" mechanism for 
> inheritance 
> > be implemented? (biblatex has introduced an alternative 
> cross-referencing 
> > mechanism using the xref field, but I'm not sure how many people have 
> > actually started using this. I haven't seen any advantage in using this 
> > instead of crossref yet.) 
>
> I have implemented this (including the biblatex mappings, if --bibtex 
> isn't used.) 
>
> Hmm, my bib(la)tex crossref example from earlier in this thread gives:

---
references:
- editor:
  - family: Editor
    given:
    - Ed
  id: item2
  publisher: Publisher
  publisher-place: Location
  title: The Collection Title
  type: book
- author:
  - family: Doe
    given:
    - John
  editor:
  - family: Editor
    given:
    - Ed
  id: item1
  page: '12-34'
  publisher: Publisher
  publisher-place: Location
  title: The Incollection Title
  type: chapter
...

i.e., no container-title in item2. Can't quite see why not ...


 

> > 3. Some fields will have to be concatenated, 
> > 
> >     - title + subtitle + titleaddon (same for booktitle, maintitle), and 
> > journal/journaltitle + journalsubtitle. How will this be handled? 
> > (titleaddon etc. are supposed to be printed "in a different font", i.e., 
> > unitalicized or outside of speech marks, but this currently does not 
> seem 
> > to be possible in CSL.) 
> > 
> >     - note + addendum. biblatex has addendum at the end of an entry, and 
> > note somewhere in front of that, but in CSL there's only one suitable 
> > field: note. – Differs from biblatex "annotation" (and its alias 
> "annote") 
> > which map to CSL "annote", and contain reader's notes. 
>
> There's an 'appendField' function.  It takes as parameters the 
> target field name and a function of type [Inline] -> [Inline] that 
> says how to transform the thing you're appending (adding a colon, 
> a space, putting it in parentheses, etc.).  You can see how this is used 
> in the lines for "subtitle" and "titleaddon" and "addendum". 
>
> > 4. Some fields have to be omitted if some certain other/newer field(s) 
> are 
> > present: In biblatex, "date", if present, is used instead of "year", 
> > "month", and "day"; "location" instead of "address", etc. How will this 
> be 
> > handled? 
>
> If you use setField, it will overwrite an existing value.  So, 
> for example, if you want "maintitle" to override "title" in setting 
> the "title" value: 
>
>   getField "title" ==> setField "title" 
>   getField "maintitle" ==> setField "title" 
>
> If "maintitle" is not found, the earlier value (from "title") will 
> persist. 
> So we can handle these issues of priority by ordering. 
>

Fine, have to come to grips with this logic ... 

>
> > 5. Proper mapping of some fields or entry types is conditional upon 
> either 
> > the existence or the content of other fields. How can this be made to 
> work? 
> > 
> >     - for @book, @collection, @proceedings: title ==> title, unless 
> > maintitle exists, in which case maintitle ==> title (and, once consensus 
> on 
> > new CSL variables is reached, title ==> volume-title; as discussed 
> earlier 
> > in this thread), 
>
> See above for this one. 
>
> >     - for @inbook, @incollection, @proceedings: booktitle ==> 
> > container-title, unless maintitle exists, in which case maintitle ==> 
> > container-title (and booktitle ==> volume-title), 
>
>   hasMaintitle <- isPresent "maintitle" 
>   getField "booktitle" ==> setField (if hasMaintitle && 
>                                         et `elem` 
> ["book","collection","proceedings"] 
>                                         then "volume-title" 
>                                         else "container-title") 
>

Has to be "inbook" etc. here. books never have booktitles.
 

>
> >     - mapping of editor, editora, editorb, editorc depends on the 
> content 
> > of editortype, editoratype, editorbtype, editorctype, 
> > 
> >     - An entrysubtype of "magazine" should change the CSL type from 
> > "article-journal" to "article-magazine" (this is not from biblatex 
> itself 
> > but from biblatex-chicago, but I think it'd be very useful to borrow 
> > certain bits from this package; an entrysubtype of "newspaper" could of 
> > course map to CSL "article-newspaper"). 
>
>   getRawField "magazine" ==> setRawField "type" "article-magazine" 
>

Not sure I get this right: It would be
@article{key,
entrysubtype={magazine},
...
}
 

>
> >     - series maps to CSL collection-title, except for @article (and 
> > @periodical?), in which case it should appear after the journal title 
> (the 
> > only solution in CSL is probably to concatenate the two fields). 
>
>   getField "series" ==> if et == "article" || et == "periodical" 
>                            then appendField "title" inParens 
>                            else setField "collection-title" 
>
> > 6. "keys" are used liberally in biblatex fields, as shorthands, and for 
> > localization (see biblatex manual 4.9). These should be translated to 
> > natural language terms. Let's focus on English first - but what about 
> > localization …? 
>
> I'm ignoring this for now.  citeproc-hs has its own localization. 
>

Yes, but this is about, e.g., pubstate={inpress}, where the "inpress" 
should be transformed to 'in press'.
 

>
> > 7. In addition to "name lists" (authors etc.), biblatex has "literal 
> lists" 
> > with elements separated by "and", too: institution, location, 
> organization, 
> > origlocation, origpublisher, publisher, list[a-f], address, school. Will 
> > routines used for names work unmodified? 
>
> These are supported 
>
>   getLiteralList 
>
>   getAuthorList 
>
> Why don't you try adding to it and see what you can do? 
>
> John 
>
>
More questions:

- One of the most urgent issues is probably date handling. Currently only 
"year" works.

bibtex has, e.g.,
year={1990},
month=aug,

biblatex prefers the "date" field, and uses year and month only if no date 
is found.
Format is YYYY-MM-DD, YYYY-MM or YYYY.
Date ranges are specified by YYYY-MM-DD/YYYY-MM-DD.
(CSL specs mention date ranges, but I don't know about the JSON or YAML 
format.)
Partial dates are valid if date components are omitted at the end only.
Open ended date ranges have the range separator but no end date (e. g., 
YYYY/).

Again, thanks, and I'll continue to see what I can do.

Nick

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/cfd2f13e-48f4-4b0e-8415-a68597649e17%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 19374 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Excellent.  I've merged in your patch.

+++ Nick Bart [Sep 27 13 15:38 ]:
>    Brilliant. But quite a challenge, too. Still, here's a diff with some
>    of the more straightforward stuff hacked in. Other comments and
>    questions below.
 
>    Hmm, my bib(la)tex crossref example from earlier in this thread gives:
>    ---
>    references:
>    - editor:
>      - family: Editor
>        given:
>        - Ed
>      id: item2
>      publisher: Publisher
>      publisher-place: Location
>      title: The Collection Title
>      type: book
>    - author:
>      - family: Doe
>        given:
>        - John
>      editor:
>      - family: Editor
>        given:
>        - Ed
>      id: item1
>      page: '12-34'
>      publisher: Publisher
>      publisher-place: Location
>      title: The Incollection Title
>      type: chapter
>    ...
>    i.e., no container-title in item2. Can't quite see why not ...

I'll look into this.

>        getRawField "magazine" ==> setRawField "type" "article-magazine"
> 
>    Not sure I get this right: It would be
>    @article{key,
>    entrysubtype={magazine},
>    ...
>    }

Yes.  I was assumnig  that anything with entrysubtype = "magazine"
is a magazine article, hence did not check et (entryType).  But it
could be checked too in a conditional.


> 
>    Yes, but this is about, e.g., pubstate={inpress}, where the "inpress"
>    should be transformed to 'in press'.

So how does it work? Is it just a list of strings that are to be
expanded macro-like before further processing?

>    More questions:
>    - One of the most urgent issues is probably date handling. Currently
>    only "year" works.
>    bibtex has, e.g.,
>    year={1990},
>    month=aug,
>    biblatex prefers the "date" field, and uses year and month only if no
>    date is found.
>    Format is YYYY-MM-DD, YYYY-MM or YYYY.
>    Date ranges are specified by YYYY-MM-DD/YYYY-MM-DD.
>    (CSL specs mention date ranges, but I don't know about the JSON or YAML
>    format.)
>    Partial dates are valid if date components are omitted at the end only.
>    Open ended date ranges have the range separator but no end date (e. g.,
>    YYYY/).

I can add something for this, but I'd need to know more about how
citeproc dates ("issued") works.  Is there any way to indicate a range
there?

John

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ John MacFarlane [Sep 28 13 00:16 ]:
> Excellent.  I've merged in your patch.
> 
> +++ Nick Bart [Sep 27 13 15:38 ]:
> >    Brilliant. But quite a challenge, too. Still, here's a diff with some
> >    of the more straightforward stuff hacked in. Other comments and
> >    questions below.
>  
> >    Hmm, my bib(la)tex crossref example from earlier in this thread gives:
> >    ---
> >    references:
> >    - editor:
> >      - family: Editor
> >        given:
> >        - Ed
> >      id: item2
> >      publisher: Publisher
> >      publisher-place: Location
> >      title: The Collection Title
> >      type: book
> >    - author:
> >      - family: Doe
> >        given:
> >        - John
> >      editor:
> >      - family: Editor
> >        given:
> >        - Ed
> >      id: item1
> >      page: '12-34'
> >      publisher: Publisher
> >      publisher-place: Location
> >      title: The Incollection Title
> >      type: chapter
> >    ...
> >    i.e., no container-title in item2. Can't quite see why not ...
> 
> I'll look into this.

There was a typo in the code which I fixed.  But it didn't
affect this.  Looking at Appendix A of the biblatex manual,
I'm not seeing a rule for source = book, target = chapter.
(Rather, we have source = book, target = inbook, bookinbook,
or suppbook.) Am I misunderstanding something?

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2502 bytes --]

crossref
========

Do you mean "Appendix B: Default Inheritance Setup" (biblatex manual 
Version 2.7a)?

I think you're right, for source type "book" and target type = "inbook, 
bookinbook,
or suppbook"; source field "title" becomes target field "booktitle".

Now in my example, item1 should inherit "title" from item2 with the field 
designation changed to "booktitle".

"chapter" is not a biblatex field name, but just the CSL variable 
corresponding to inbook, bookinbook, suppbook, but also incollection and a 
few other in-s and supp-s.

Dates
=====

As to dates, including date ranges, and their format in what appears to be 
JSON, see the citeproc-js docs:

http://gsl-nagoya-u.net/http/pub/citeproc-doc.html#dates

magazine
========

If I insert

getRawField "magazine" ==> setRawField "type" "article-magazine"

I get the following error message:

$ ghc --make bibtex2pandoc
[1 of 1] Compiling Main             ( bibtex2pandoc.hs, bibtex2pandoc.o )

bibtex2pandoc.hs:267:30:
    Couldn't match type `RWST T () (M.Map String MetaValue) Maybe ()'
                  with `String -> BibM ()'
    Expected type: String -> BibM ()
      Actual type: BibM ()
    In the return type of a call of `setRawField'
    Probable cause: `setRawField' is applied to too many arguments
    In the second argument of `(==>)', namely
      `setRawField "type" "article-magazine"'
    In a stmt of a 'do' block:
      getRawField "magazine" ==> setRawField "type" "article-magazine"

Strictly speaking, it should be IF biblatex type is article (or periodical) 
AND entrysubtype is magazine THEN CSL type becomes article-magazine. One 
could probably get away with just checking the entrysubtype since it's not 
so likely entrysubtype=magazine will be used in other types, but stricter 
checking might be preferable. So far I haven't figured out how to test for 
multiple criteria using the bibtex2pandoc code though: is that possible?

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/35b0f493-82e9-486b-838d-09aa3b48a618%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3213 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Sep 28 13 01:43 ]:
>    crossref
>    ========
>    Do you mean "Appendix B: Default Inheritance Setup" (biblatex manual
>    Version 2.7a)?
>    I think you're right, for source type "book" and target type = "inbook,
>    bookinbook,
>    or suppbook"; source field "title" becomes target field "booktitle".
>    Now in my example, item1 should inherit "title" from item2 with the
>    field designation changed to "booktitle".
>    "chapter" is not a biblatex field name, but just the CSL variable
>    corresponding to inbook, bookinbook, suppbook, but also incollection
>    and a few other in-s and supp-s.

I was confused about a few things, but with the latest code,
crossrefs should be working (at least your example works).

>    Dates
>    =====
>    As to dates, including date ranges, and their format in what appears to
>    be JSON, see the citeproc-js docs:
>    http://gsl-nagoya-u.net/http/pub/citeproc-doc.html#dates

OK, I see.  That's a bit different than I thought.

>    magazine
>    ========
>    If I insert
>    getRawField "magazine" ==> setRawField "type" "article-magazine"
>    I get the following error message:
>    $ ghc --make bibtex2pandoc
>    [1 of 1] Compiling Main             ( bibtex2pandoc.hs, bibtex2pandoc.o
>    )
>    bibtex2pandoc.hs:267:30:
>        Couldn't match type `RWST T () (M.Map String MetaValue) Maybe ()'
>                      with `String -> BibM ()'
>        Expected type: String -> BibM ()
>          Actual type: BibM ()
>        In the return type of a call of `setRawField'
>        Probable cause: `setRawField' is applied to too many arguments
>        In the second argument of `(==>)', namely
>          `setRawField "type" "article-magazine"'
>        In a stmt of a 'do' block:
>          getRawField "magazine" ==> setRawField "type" "article-magazine"
>    Strictly speaking, it should be IF biblatex type is article (or
>    periodical) AND entrysubtype is magazine THEN CSL type becomes
>    article-magazine. One could probably get away with just checking the
>    entrysubtype since it's not so likely entrysubtype=magazine will be
>    used in other types, but stricter checking might be preferable. So far
>    I haven't figured out how to test for multiple criteria using the
>    bibtex2pandoc code though: is that possible?

I think I may have made things overly complicated by trying to make
them simpler with the artificial ==> operator.  Let me think about
how to simplify this so it's easier to do conditional logic.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ John MacFarlane [Sep 28 13 10:52 ]:
> I think I may have made things overly complicated by trying to make
> them simpler with the artificial ==> operator.  Let me think about
> how to simplify this so it's easier to do conditional logic.

OK, I've removed the ==> operator.  Now we just use the standard
monadic bind >>=, together with an 'opt' function.

Let me explain with a couple examples:

  getField "foo" >>= setField "bar"

will attempt to get the value of the "foo" field from the bibtex
entry.  If "foo" is not set, the action will fail, causing the
whole conversion to fail (so, this entry will not be translated).
If "foo" is set, the value is passed to the function

  setField "bar"

setField takes two arguments, a field name and a value, so

  setField "bar"

takes one argument, the value.

Normally we don't want the conversion to fail if the field is
not found (though there are exceptions, like the identifier).
So normally we'll put the whole combined action under the 'opt'
function (opt for optional).

  opt $ getField "foo" >>= setField "bar"

will attempt to get the "foo" field and set the "bar" field
of the new entry to its value, but if "foo" is not set, it will
just pass on silently rather than causing the whole conversion to
fail.

Note that $ is just a way of avoiding parentheses:  we could have
equivalently written

  opt (getField "foo" >>= setField "bar")

We could have also used "do" notation:

  opt $ do
    val <- getField "foo"
    setField "bar" val

And you may want to use 'do' notation for things that have complex logic.  So,

  opt $ do
    val <- getField "foo"
    if val == "zeb" || bibtex
    then setField "bar" val
    else do
      setField "baz" val
      getField "zim" >>= setField "zam"

Another operator that is useful to know about is ">>".  This is
sort of like ">>=", but it doesn't "feed" the output of the first
operand to the second one.  You might use this when you just want
to test for a value's presence.  So,

  opt $ getField "foo" >> getField "bar" >>= setField "baz"

would set the "baz" field in the pandoc entry to the value of the "bar"
field in the bibtex entry, IF the "foo" field is set; otherwise pass
on.

In 'do' notation this would be

  opt $ do
    getField "foo"
    val <- getField "bar"
    setField "baz" val

Hope this makes things easier.

John

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3954 bytes --]

Thanks a lot for the rundown. Seems I could try my hand at this now.

Still, when I insert 

  opt $ do
    val <- getField "entrysubtype"
    if val == "magazine"
    then setType "article-magazine"
    else setType "article-journal"

into bibtex2pandoc.hs,

ghc --make bibtex2pandoc

returns the error:

[1 of 1] Compiling Main             ( bibtex2pandoc.hs, bibtex2pandoc.o )

bibtex2pandoc.hs:277:15:
    Couldn't match expected type `MetaValue' with actual type `[Char]'
    In the second argument of `(==)', namely `"magazine"'
    In the expression: val == "magazine"
    In a stmt of a 'do' block:
      if val == "magazine" then
          setType "article-magazine"
      else
          setType "article-journal"

Any idea what would be wrong here?





On Saturday, September 28, 2013 6:35:16 PM UTC, fiddlosopher wrote:
>
> +++ John MacFarlane [Sep 28 13 10:52 ]: 
> > I think I may have made things overly complicated by trying to make 
> > them simpler with the artificial ==> operator.  Let me think about 
> > how to simplify this so it's easier to do conditional logic. 
>
> OK, I've removed the ==> operator.  Now we just use the standard 
> monadic bind >>=, together with an 'opt' function. 
>
> Let me explain with a couple examples: 
>
>   getField "foo" >>= setField "bar" 
>
> will attempt to get the value of the "foo" field from the bibtex 
> entry.  If "foo" is not set, the action will fail, causing the 
> whole conversion to fail (so, this entry will not be translated). 
> If "foo" is set, the value is passed to the function 
>
>   setField "bar" 
>
> setField takes two arguments, a field name and a value, so 
>
>   setField "bar" 
>
> takes one argument, the value. 
>
> Normally we don't want the conversion to fail if the field is 
> not found (though there are exceptions, like the identifier). 
> So normally we'll put the whole combined action under the 'opt' 
> function (opt for optional). 
>
>   opt $ getField "foo" >>= setField "bar" 
>
> will attempt to get the "foo" field and set the "bar" field 
> of the new entry to its value, but if "foo" is not set, it will 
> just pass on silently rather than causing the whole conversion to 
> fail. 
>
> Note that $ is just a way of avoiding parentheses:  we could have 
> equivalently written 
>
>   opt (getField "foo" >>= setField "bar") 
>
> We could have also used "do" notation: 
>
>   opt $ do 
>     val <- getField "foo" 
>     setField "bar" val 
>
> And you may want to use 'do' notation for things that have complex logic. 
>  So, 
>
>   opt $ do 
>     val <- getField "foo" 
>     if val == "zeb" || bibtex 
>     then setField "bar" val 
>     else do 
>       setField "baz" val 
>       getField "zim" >>= setField "zam" 
>
> Another operator that is useful to know about is ">>".  This is 
> sort of like ">>=", but it doesn't "feed" the output of the first 
> operand to the second one.  You might use this when you just want 
> to test for a value's presence.  So, 
>
>   opt $ getField "foo" >> getField "bar" >>= setField "baz" 
>
> would set the "baz" field in the pandoc entry to the value of the "bar" 
> field in the bibtex entry, IF the "foo" field is set; otherwise pass 
> on. 
>
> In 'do' notation this would be 
>
>   opt $ do 
>     getField "foo" 
>     val <- getField "bar" 
>     setField "baz" val 
>
> Hope this makes things easier. 
>
> John 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/93ff8b26-a132-4da8-a75a-09924f725d34%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5259 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

There are two problems with your code.
First, you need:

   setType "article-journal" val

not

  setType "article-journal"

Note:  the following are completely equivalent:

(1)  do val <- getField "foo"
        setField "bar" val

(2)  getField "foo" >>= setField "bar"

In the second case you don't need the second parameter on
setField, because it is "piped in" by the >>= operator.

Second problem (and this is what the message refers to).
getField returns a MetaValue, not a string.  IF you want
a string -- as you must if you want to compare it to
a string "magazine" -- then use getRawField instead.

+++ Nick Bart [Sep 28 13 13:09 ]:
>    Thanks a lot for the rundown. Seems I could try my hand at this now.
>    Still, when I insert
>      opt $ do
>        val <- getField "entrysubtype"
>        if val == "magazine"
>        then setType "article-magazine"
>        else setType "article-journal"
>    into bibtex2pandoc.hs,
>    ghc --make bibtex2pandoc
>    returns the error:
>    [1 of 1] Compiling Main             ( bibtex2pandoc.hs, bibtex2pandoc.o
>    )
>    bibtex2pandoc.hs:277:15:
>        Couldn't match expected type `MetaValue' with actual type `[Char]'
>        In the second argument of `(==)', namely `"magazine"'
>        In the expression: val == "magazine"
>        In a stmt of a 'do' block:
>          if val == "magazine" then
>              setType "article-magazine"
>          else
>              setType "article-journal"
>    Any idea what would be wrong here?
>    On Saturday, September 28, 2013 6:35:16 PM UTC, fiddlosopher wrote:
> 
>      +++ John MacFarlane [Sep 28 13 10:52 ]:
>      > I think I may have made things overly complicated by trying to
>      make
>      > them simpler with the artificial ==> operator.  Let me think about
>      > how to simplify this so it's easier to do conditional logic.
>      OK, I've removed the ==> operator.  Now we just use the standard
>      monadic bind >>=, together with an 'opt' function.
>      Let me explain with a couple examples:
>        getField "foo" >>= setField "bar"
>      will attempt to get the value of the "foo" field from the bibtex
>      entry.  If "foo" is not set, the action will fail, causing the
>      whole conversion to fail (so, this entry will not be translated).
>      If "foo" is set, the value is passed to the function
>        setField "bar"
>      setField takes two arguments, a field name and a value, so
>        setField "bar"
>      takes one argument, the value.
>      Normally we don't want the conversion to fail if the field is
>      not found (though there are exceptions, like the identifier).
>      So normally we'll put the whole combined action under the 'opt'
>      function (opt for optional).
>        opt $ getField "foo" >>= setField "bar"
>      will attempt to get the "foo" field and set the "bar" field
>      of the new entry to its value, but if "foo" is not set, it will
>      just pass on silently rather than causing the whole conversion to
>      fail.
>      Note that $ is just a way of avoiding parentheses:  we could have
>      equivalently written
>        opt (getField "foo" >>= setField "bar")
>      We could have also used "do" notation:
>        opt $ do
>          val <- getField "foo"
>          setField "bar" val
>      And you may want to use 'do' notation for things that have complex
>      logic.  So,
>        opt $ do
>          val <- getField "foo"
>          if val == "zeb" || bibtex
>          then setField "bar" val
>          else do
>            setField "baz" val
>            getField "zim" >>= setField "zam"
>      Another operator that is useful to know about is ">>".  This is
>      sort of like ">>=", but it doesn't "feed" the output of the first
>      operand to the second one.  You might use this when you just want
>      to test for a value's presence.  So,
>        opt $ getField "foo" >> getField "bar" >>= setField "baz"
>      would set the "baz" field in the pandoc entry to the value of the
>      "bar"
>      field in the bibtex entry, IF the "foo" field is set; otherwise pass
>      on.
>      In 'do' notation this would be
>        opt $ do
>          getField "foo"
>          val <- getField "bar"
>          setField "baz" val
>      Hope this makes things easier.
>      John
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/93ff8b26-a132-4da8-
>    a75a-09924f725d34%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/93ff8b26-a132-4da8-a75a-09924f725d34%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4825 bytes --]

A few additional ideas/issues:

Title vs. sentence case
=======================

bibtex and biblatex expect titles in title case in the database, converting 
these to sentence case if required by a particular style, except for 
strings protected by {}.
CSL does the opposite: It expects titles in sentence case, converting these 
to title case if required, except for certain stop words.

I think it make senses for bibtex2pandoc to convert bibtex=biblatex titles 
to sentence case for further consumption by CSL; i.e., to lower case, 
except strings protected by {}.
    - ex.:
    title = {{JFK}: The {CIA}, {Vietnam}, and the Plot to Assassinate {John 
F. Kennedy}}
    becomes
    title: 'JFK: The CIA, Vietnam, and the plot to assassinate John F. 
Kennedy'
    - to be discussed: should strings inside commands be converted, too?
     i.e., should
     title = {An Analysis of \textit{For Whom the Bell Tolls}}
     become
     title: 'An Analysis of <i>for whom the bell tolls</i>'
     and only extra protection
     title = {An Analysis of {\textit{For Whom the Bell Tolls}}}
     would yield
     title: 'An Analysis of <i>For Whom the Bell Tolls</i>'

Corporate authors
=================

Example: author = {{National Aeronautics and Space Administration}},
Current bibtex2pandoc output:
- author:
  - family: Aeronautics
    given:
    - National
  - family: Administration
    given:
    - Space
Expected:
- author:
  - literal: 'National Aeronautics and Space Administration'


Literal “and” in institution, organization, publisher, location, etc.
=====================================================================

If code includes 
 
  getLiteralList "publisher" ==> setList "publisher"

publisher = {Holt, Rinehart {and} Winston}

Current bibtex2pandoc output:

  publisher:
  - 'Holt, Rinehart'
  - Winston

Expected:

  publisher:
  - 'Holt, Rinehart and Winston'

Also noticed, however, that pandoc-citeproc does not seem to like it when 
actual multiple publishers occur:

  publisher:
  - 'Univ. of Toronto Press'
  - Routledge

throws an error:

pandoc-citeproc: Error parsing references: when expecting a String, 
encountered Array instead

This needs to be checked, but if CSL does not allow multiple publishers, 
the code above would need to be changed back to

  getField "publisher" ==> setField "publisher"

I see that multiple publishers (and locations) are being discussed at 
various Zotero/CSL/citeproc-js forums, but don't get a clear picture yet.

Names
=====

Can bibtex2pandoc be expected to distinguish all five name components 
(family, given, suffix, non-dropping-particle, dropping-particle)?

biblatex provides one useful additional bit of info for name parsing via 
the "useprefix" switch inside the "options" field:

From the biblatex manual (v 2.7a), "3.1.3 Entry Options"

> Whether the name prefix (von, van, of, da, de, della, etc.) is considered 
when printing the last name in citations. This also affects the sorting and 
formatting of the bibliography as well as the generation of certain types 
of labels. If this option is enabled, biblatex always precedes the last 
name with the prefix. For example, Ludwig van Beethoven would be cited as 
“Beethoven” and alphabetized as “Beethoven, Ludwig van” by default. If this 
option is enabled, he is cited as “van Beethoven” and alphabetized as “Van 
Beethoven, Ludwig” instead. With Biber, this option is also settable on a 
per-type basis.

Inline formatting
=================

It probably doesn't play a role for pandoc, but in the interest of 
portability, shouldn't inline formatting in a CSL-YAML file or metadata 
section rather use

    <i> and </i> for italics
    <b> and </b> for bold
    <sub> and </sub> for subscript
    <sup> and </sup> for superscript
    <span style="font-variant:small-caps;"> and </span> for smallcaps

instead of markdown formatting?

(see https://github.com/jgm/pandoc/issues/931, 
https://www.zotero.org/support/kb/rich_text_bibliography)


Small Caps
==========

Speaking of small caps: Any plans to add some markdownish formatting 
commands for small caps to pandoc?

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/09c521a8-1f44-474b-8829-60dd95cf0f94%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5781 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

This is tricky.  Right now I'm just using pandoc's latex reader
to read the contents of the bibtex/biblatex fields.

But this means that '{A}b' and 'Ab' get translated exactly the
same way, as Str "Ab", so I can't distinguish between them.

One possible solution would be to change the latex reader so
that it parses groups as Span elements -- that would not change
visible output in any format, but it would allow us to do the
"untitlecase" transformation.  A bit worried about the cart leading
the horse here, though.

John

+++ Nick Bart [Sep 28 13 03:52 ]:
>    A few additional ideas/issues:
>    Title vs. sentence case
>    =======================
>    bibtex and biblatex expect titles in title case in the database,
>    converting these to sentence case if required by a particular style,
>    except for strings protected by {}.
>    CSL does the opposite: It expects titles in sentence case, converting
>    these to title case if required, except for certain stop words.
>    I think it make senses for bibtex2pandoc to convert bibtex=biblatex
>    titles to sentence case for further consumption by CSL; i.e., to lower
>    case, except strings protected by {}.
>        - ex.:
>        title = {{JFK}: The {CIA}, {Vietnam}, and the Plot to Assassinate
>    {John F. Kennedy}}
>        becomes
>        title: 'JFK: The CIA, Vietnam, and the plot to assassinate John F.
>    Kennedy'
>        - to be discussed: should strings inside commands be converted,
>    too?
>         i.e., should
>         title = {An Analysis of \textit{For Whom the Bell Tolls}}
>         become
>         title: 'An Analysis of <i>for whom the bell tolls</i>'
>         and only extra protection
>         title = {An Analysis of {\textit{For Whom the Bell Tolls}}}
>         would yield
>         title: 'An Analysis of <i>For Whom the Bell Tolls</i>'
>    Corporate authors
>    =================
>    Example: author = {{National Aeronautics and Space Administration}},
>    Current bibtex2pandoc output:
>    - author:
>      - family: Aeronautics
>        given:
>        - National
>      - family: Administration
>        given:
>        - Space
>    Expected:
>    - author:
>      - literal: 'National Aeronautics and Space Administration'
>    Literal and in institution, organization, publisher, location, etc.
>    =====================================================================
>    If code includes
> 
>      getLiteralList "publisher" ==> setList "publisher"
>    publisher = {Holt, Rinehart {and} Winston}
>    Current bibtex2pandoc output:
>      publisher:
>      - 'Holt, Rinehart'
>      - Winston
>    Expected:
>      publisher:
>      - 'Holt, Rinehart and Winston'
>    Also noticed, however, that pandoc-citeproc does not seem to like it
>    when actual multiple publishers occur:
>      publisher:
>      - 'Univ. of Toronto Press'
>      - Routledge
>    throws an error:
>    pandoc-citeproc: Error parsing references: when expecting a String,
>    encountered Array instead
>    This needs to be checked, but if CSL does not allow multiple
>    publishers, the code above would need to be changed back to
>      getField "publisher" ==> setField "publisher"
>    I see that multiple publishers (and locations) are being discussed at
>    various Zotero/CSL/citeproc-js forums, but don't get a clear picture
>    yet.
>    Names
>    =====
>    Can bibtex2pandoc be expected to distinguish all five name components
>    (family, given, suffix, non-dropping-particle, dropping-particle)?
>    biblatex provides one useful additional bit of info for name parsing
>    via the "useprefix" switch inside the "options" field:
>    From the biblatex manual (v 2.7a), "3.1.3 Entry Options"
>    > Whether the name prefix (von, van, of, da, de, della, etc.) is
>    considered when printing the last name in citations. This also affects
>    the sorting and formatting of the bibliography as well as the
>    generation of certain types of labels. If this option is enabled,
>    biblatex always precedes the last name with the prefix. For example,
>    Ludwig van Beethoven would be cited as Beethoven and alphabetized as
>    Beethoven, Ludwig van by default. If this option is enabled, he is
>    cited as van Beethoven and alphabetized as Van Beethoven, Ludwig
>    instead. With Biber, this option is also settable on a per-type basis.
>    Inline formatting
>    =================
>    It probably doesn't play a role for pandoc, but in the interest of
>    portability, shouldn't inline formatting in a CSL-YAML file or metadata
>    section rather use
>        <i> and </i> for italics
>        <b> and </b> for bold
>        <sub> and </sub> for subscript
>        <sup> and </sup> for superscript
>        <span style="font-variant:small-caps;"> and </span> for smallcaps
>    instead of markdown formatting?
>    (see https://github.com/jgm/pandoc/issues/931,
>    https://www.zotero.org/support/kb/rich_text_bibliography)
>    Small Caps
>    ==========
>    Speaking of small caps: Any plans to add some markdownish formatting
>    commands for small caps to pandoc?
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/09c521a8-1f44-474b-
>    8829-60dd95cf0f94%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/09c521a8-1f44-474b-8829-60dd95cf0f94%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Sep 28 13 03:52 ]:
>    A few additional ideas/issues:
>    Title vs. sentence case
>    =======================
>    bibtex and biblatex expect titles in title case in the database,
>    converting these to sentence case if required by a particular style,
>    except for strings protected by {}.
>    CSL does the opposite: It expects titles in sentence case, converting
>    these to title case if required, except for certain stop words.
>    I think it make senses for bibtex2pandoc to convert bibtex=biblatex
>    titles to sentence case for further consumption by CSL; i.e., to lower
>    case, except strings protected by {}.
>        - ex.:
>        title = {{JFK}: The {CIA}, {Vietnam}, and the Plot to Assassinate
>    {John F. Kennedy}}
>        becomes
>        title: 'JFK: The CIA, Vietnam, and the plot to assassinate John F.
>    Kennedy'
>        - to be discussed: should strings inside commands be converted,
>    too?
>         i.e., should
>         title = {An Analysis of \textit{For Whom the Bell Tolls}}
>         become
>         title: 'An Analysis of <i>for whom the bell tolls</i>'
>         and only extra protection
>         title = {An Analysis of {\textit{For Whom the Bell Tolls}}}
>         would yield
>         title: 'An Analysis of <i>For Whom the Bell Tolls</i>'

I've implemented the unTitlecase transformation.  (This depends on a
recent patch to pandoc that causes the LaTeX reader to insert Span
when we have a bare group {Like This}.)

I don't know about the "to be discussed" -- currently strings inside
emphasis, etc., will retain their case.  I think that's probably right.

>    Corporate authors
>    =================
>    Example: author = {{National Aeronautics and Space Administration}},
>    Current bibtex2pandoc output:
>    - author:
>      - family: Aeronautics
>        given:
>        - National
>      - family: Administration
>        given:
>        - Space
>    Expected:
>    - author:
>      - literal: 'National Aeronautics and Space Administration'

I've added support for this.

>    Literal and in institution, organization, publisher, location, etc.
>    =====================================================================
>    If code includes
> 
>      getLiteralList "publisher" ==> setList "publisher"
>    publisher = {Holt, Rinehart {and} Winston}
>    Current bibtex2pandoc output:
>      publisher:
>      - 'Holt, Rinehart'
>      - Winston
>    Expected:
>      publisher:
>      - 'Holt, Rinehart and Winston'
>    Also noticed, however, that pandoc-citeproc does not seem to like it
>    when actual multiple publishers occur:
>      publisher:
>      - 'Univ. of Toronto Press'
>      - Routledge
>    throws an error:
>    pandoc-citeproc: Error parsing references: when expecting a String,
>    encountered Array instead
>    This needs to be checked, but if CSL does not allow multiple
>    publishers, the code above would need to be changed back to
>      getField "publisher" ==> setField "publisher"
>    I see that multiple publishers (and locations) are being discussed at
>    various Zotero/CSL/citeproc-js forums, but don't get a clear picture
>    yet.

The literal {and} will not cause breaking of author lists or literal
lists.

I'll leave the multiple publishers question open for now.  Either
bibtex2pandoc or pandoc-citeproc should be changed, not sure which.

>    Names
>    =====
>    Can bibtex2pandoc be expected to distinguish all five name components
>    (family, given, suffix, non-dropping-particle, dropping-particle)?
>    biblatex provides one useful additional bit of info for name parsing
>    via the "useprefix" switch inside the "options" field:
>    From the biblatex manual (v 2.7a), "3.1.3 Entry Options"
>    > Whether the name prefix (von, van, of, da, de, della, etc.) is
>    considered when printing the last name in citations. This also affects
>    the sorting and formatting of the bibliography as well as the
>    generation of certain types of labels. If this option is enabled,
>    biblatex always precedes the last name with the prefix. For example,
>    Ludwig van Beethoven would be cited as Beethoven and alphabetized as
>    Beethoven, Ludwig van by default. If this option is enabled, he is
>    cited as van Beethoven and alphabetized as Van Beethoven, Ludwig
>    instead. With Biber, this option is also settable on a per-type basis.

I used a rough-and-ready algorithm, probably not the same one biber
or biblatex uses.  Basically, the only time my code gives you a
non-dropping particle is when you have a name in format

   B A, D E

then B is the particle and A the given name.

I don't have any support now for dropping-particle or suffix.
This needs to be improved, but I'd need to know better what the
algorithm is.

>    Inline formatting
>    =================
>    It probably doesn't play a role for pandoc, but in the interest of
>    portability, shouldn't inline formatting in a CSL-YAML file or metadata
>    section rather use
>        <i> and </i> for italics
>        <b> and </b> for bold
>        <sub> and </sub> for subscript
>        <sup> and </sup> for superscript
>        <span style="font-variant:small-caps;"> and </span> for smallcaps
>    instead of markdown formatting?
>    (see https://github.com/jgm/pandoc/issues/931,
>    https://www.zotero.org/support/kb/rich_text_bibliography)

No, because the YAML file is standard pandoc metadata, formatted in
markdown.  When it is read by pandoc-citeproc, this will be converted
where possible to CSL metadata.

>    Small Caps
>    ==========
>    Speaking of small caps: Any plans to add some markdownish formatting
>    commands for small caps to pandoc?

No idea what a natural syntax would be.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 7373 bytes --]

unTitlecase: Brilliant. 

Forgot one important detail, though: unTitlecase should only be applied if 
the language is English, i.e., if an entry's biblatex hyphenation field is 
one of american, british, canadian, english, australian, newzealand, 
USenglish or UKenglish.  

From the biblatex manual (v 2.7a), "4.6.4 Miscellaneous Commands", item " 
\MakeSentenceCase":

> By default, converting to sentence case is enabled for the following 
language identifiers: american, british, canadian, english, australian, 
newzealand as well as the aliases USenglish and UKenglish.  

On Saturday, September 28, 2013 7:34:09 PM UTC, fiddlosopher wrote:
>
> +++ Nick Bart [Sep 28 13 03:52 ]: 
> >    A few additional ideas/issues: 
> >    Title vs. sentence case 
> >    ======================= 
> >    bibtex and biblatex expect titles in title case in the database, 
> >    converting these to sentence case if required by a particular style, 
> >    except for strings protected by {}. 
> >    CSL does the opposite: It expects titles in sentence case, converting 
> >    these to title case if required, except for certain stop words. 
> >    I think it make senses for bibtex2pandoc to convert bibtex=biblatex 
> >    titles to sentence case for further consumption by CSL; i.e., to 
> lower 
> >    case, except strings protected by {}. 
> >        - ex.: 
> >        title = {{JFK}: The {CIA}, {Vietnam}, and the Plot to Assassinate 
> >    {John F. Kennedy}} 
> >        becomes 
> >        title: 'JFK: The CIA, Vietnam, and the plot to assassinate John 
> F. 
> >    Kennedy' 
> >        - to be discussed: should strings inside commands be converted, 
> >    too? 
> >         i.e., should 
> >         title = {An Analysis of \textit{For Whom the Bell Tolls}} 
> >         become 
> >         title: 'An Analysis of <i>for whom the bell tolls</i>' 
> >         and only extra protection 
> >         title = {An Analysis of {\textit{For Whom the Bell Tolls}}} 
> >         would yield 
> >         title: 'An Analysis of <i>For Whom the Bell Tolls</i>' 
>
> I've implemented the unTitlecase transformation.  (This depends on a 
> recent patch to pandoc that causes the LaTeX reader to insert Span 
> when we have a bare group {Like This}.) 
>
> I don't know about the "to be discussed" -- currently strings inside 
> emphasis, etc., will retain their case.  I think that's probably right. 
>
> >    Corporate authors 
> >    ================= 
> >    Example: author = {{National Aeronautics and Space Administration}}, 
> >    Current bibtex2pandoc output: 
> >    - author: 
> >      - family: Aeronautics 
> >        given: 
> >        - National 
> >      - family: Administration 
> >        given: 
> >        - Space 
> >    Expected: 
> >    - author: 
> >      - literal: 'National Aeronautics and Space Administration' 
>
> I've added support for this. 
>
> >    Literal and in institution, organization, publisher, location, etc. 
> >    ===================================================================== 
> >    If code includes 
> > 
> >      getLiteralList "publisher" ==> setList "publisher" 
> >    publisher = {Holt, Rinehart {and} Winston} 
> >    Current bibtex2pandoc output: 
> >      publisher: 
> >      - 'Holt, Rinehart' 
> >      - Winston 
> >    Expected: 
> >      publisher: 
> >      - 'Holt, Rinehart and Winston' 
> >    Also noticed, however, that pandoc-citeproc does not seem to like it 
> >    when actual multiple publishers occur: 
> >      publisher: 
> >      - 'Univ. of Toronto Press' 
> >      - Routledge 
> >    throws an error: 
> >    pandoc-citeproc: Error parsing references: when expecting a String, 
> >    encountered Array instead 
> >    This needs to be checked, but if CSL does not allow multiple 
> >    publishers, the code above would need to be changed back to 
> >      getField "publisher" ==> setField "publisher" 
> >    I see that multiple publishers (and locations) are being discussed at 
> >    various Zotero/CSL/citeproc-js forums, but don't get a clear picture 
> >    yet. 
>
> The literal {and} will not cause breaking of author lists or literal 
> lists. 
>
> I'll leave the multiple publishers question open for now.  Either 
> bibtex2pandoc or pandoc-citeproc should be changed, not sure which. 
>
> >    Names 
> >    ===== 
> >    Can bibtex2pandoc be expected to distinguish all five name components 
> >    (family, given, suffix, non-dropping-particle, dropping-particle)? 
> >    biblatex provides one useful additional bit of info for name parsing 
> >    via the "useprefix" switch inside the "options" field: 
> >    From the biblatex manual (v 2.7a), "3.1.3 Entry Options" 
> >    > Whether the name prefix (von, van, of, da, de, della, etc.) is 
> >    considered when printing the last name in citations. This also 
> affects 
> >    the sorting and formatting of the bibliography as well as the 
> >    generation of certain types of labels. If this option is enabled, 
> >    biblatex always precedes the last name with the prefix. For example, 
> >    Ludwig van Beethoven would be cited as Beethoven and alphabetized as 
> >    Beethoven, Ludwig van by default. If this option is enabled, he is 
> >    cited as van Beethoven and alphabetized as Van Beethoven, Ludwig 
> >    instead. With Biber, this option is also settable on a per-type 
> basis. 
>
> I used a rough-and-ready algorithm, probably not the same one biber 
> or biblatex uses.  Basically, the only time my code gives you a 
> non-dropping particle is when you have a name in format 
>
>    B A, D E 
>
> then B is the particle and A the given name. 
>
> I don't have any support now for dropping-particle or suffix. 
> This needs to be improved, but I'd need to know better what the 
> algorithm is. 
>
> >    Inline formatting 
> >    ================= 
> >    It probably doesn't play a role for pandoc, but in the interest of 
> >    portability, shouldn't inline formatting in a CSL-YAML file or 
> metadata 
> >    section rather use 
> >        <i> and </i> for italics 
> >        <b> and </b> for bold 
> >        <sub> and </sub> for subscript 
> >        <sup> and </sup> for superscript 
> >        <span style="font-variant:small-caps;"> and </span> for smallcaps 
> >    instead of markdown formatting? 
> >    (see https://github.com/jgm/pandoc/issues/931, 
> >    https://www.zotero.org/support/kb/rich_text_bibliography) 
>
> No, because the YAML file is standard pandoc metadata, formatted in 
> markdown.  When it is read by pandoc-citeproc, this will be converted 
> where possible to CSL metadata. 
>
> >    Small Caps 
> >    ========== 
> >    Speaking of small caps: Any plans to add some markdownish formatting 
> >    commands for small caps to pandoc? 
>
> No idea what a natural syntax would be. 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 9995 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

How is the language represented in a biblatex file?  Or is it an
external parameter that comes from the locale or something else?

+++ Nick Bart [Sep 28 13 13:37 ]:
>    unTitlecase: Brilliant.
>    Forgot one important detail, though: unTitlecase should only be applied
>    if the language is English, i.e., if an entry's biblatex hyphenation
>    field is one of american, british, canadian, english, australian,
>    newzealand, USenglish or UKenglish.
>    From the biblatex manual (v 2.7a), "4.6.4 Miscellaneous Commands", item
>    " \MakeSentenceCase":
>    > By default, converting to sentence case is enabled for the following
>    language identifiers: american, british, canadian, english, australian,
>    newzealand as well as the aliases USenglish and UKenglish.
>    On Saturday, September 28, 2013 7:34:09 PM UTC, fiddlosopher wrote:
> 
>      +++ Nick Bart [Sep 28 13 03:52 ]:
>      >    A few additional ideas/issues:
>      >    Title vs. sentence case
>      >    =======================
>      >    bibtex and biblatex expect titles in title case in the
>      database,
>      >    converting these to sentence case if required by a particular
>      style,
>      >    except for strings protected by {}.
>      >    CSL does the opposite: It expects titles in sentence case,
>      converting
>      >    these to title case if required, except for certain stop words.
>      >    I think it make senses for bibtex2pandoc to convert
>      bibtex=biblatex
>      >    titles to sentence case for further consumption by CSL; i.e.,
>      to lower
>      >    case, except strings protected by {}.
>      >        - ex.:
>      >        title = {{JFK}: The {CIA}, {Vietnam}, and the Plot to
>      Assassinate
>      >    {John F. Kennedy}}
>      >        becomes
>      >        title: 'JFK: The CIA, Vietnam, and the plot to assassinate
>      John F.
>      >    Kennedy'
>      >        - to be discussed: should strings inside commands be
>      converted,
>      >    too?
>      >         i.e., should
>      >         title = {An Analysis of \textit{For Whom the Bell Tolls}}
>      >         become
>      >         title: 'An Analysis of <i>for whom the bell tolls</i>'
>      >         and only extra protection
>      >         title = {An Analysis of {\textit{For Whom the Bell
>      Tolls}}}
>      >         would yield
>      >         title: 'An Analysis of <i>For Whom the Bell Tolls</i>'
>      I've implemented the unTitlecase transformation.  (This depends on a
>      recent patch to pandoc that causes the LaTeX reader to insert Span
>      when we have a bare group {Like This}.)
>      I don't know about the "to be discussed" -- currently strings inside
>      emphasis, etc., will retain their case.  I think that's probably
>      right.
>      >    Corporate authors
>      >    =================
>      >    Example: author = {{National Aeronautics and Space
>      Administration}},
>      >    Current bibtex2pandoc output:
>      >    - author:
>      >      - family: Aeronautics
>      >        given:
>      >        - National
>      >      - family: Administration
>      >        given:
>      >        - Space
>      >    Expected:
>      >    - author:
>      >      - literal: 'National Aeronautics and Space Administration'
>      I've added support for this.
>      >    Literal and in institution, organization, publisher, location,
>      etc.
>      >    ===========================================================
>      ==========
>      >    If code includes
>      >
>      >      getLiteralList "publisher" ==> setList "publisher"
>      >    publisher = {Holt, Rinehart {and} Winston}
>      >    Current bibtex2pandoc output:
>      >      publisher:
>      >      - 'Holt, Rinehart'
>      >      - Winston
>      >    Expected:
>      >      publisher:
>      >      - 'Holt, Rinehart and Winston'
>      >    Also noticed, however, that pandoc-citeproc does not seem to
>      like it
>      >    when actual multiple publishers occur:
>      >      publisher:
>      >      - 'Univ. of Toronto Press'
>      >      - Routledge
>      >    throws an error:
>      >    pandoc-citeproc: Error parsing references: when expecting a
>      String,
>      >    encountered Array instead
>      >    This needs to be checked, but if CSL does not allow multiple
>      >    publishers, the code above would need to be changed back to
>      >      getField "publisher" ==> setField "publisher"
>      >    I see that multiple publishers (and locations) are being
>      discussed at
>      >    various Zotero/CSL/citeproc-js forums, but don't get a clear
>      picture
>      >    yet.
>      The literal {and} will not cause breaking of author lists or literal
>      lists.
>      I'll leave the multiple publishers question open for now.  Either
>      bibtex2pandoc or pandoc-citeproc should be changed, not sure which.
>      >    Names
>      >    =====
>      >    Can bibtex2pandoc be expected to distinguish all five name
>      components
>      >    (family, given, suffix, non-dropping-particle,
>      dropping-particle)?
>      >    biblatex provides one useful additional bit of info for name
>      parsing
>      >    via the "useprefix" switch inside the "options" field:
>      >    From the biblatex manual (v 2.7a), "3.1.3 Entry Options"
>      >    > Whether the name prefix (von, van, of, da, de, della, etc.)
>      is
>      >    considered when printing the last name in citations. This also
>      affects
>      >    the sorting and formatting of the bibliography as well as the
>      >    generation of certain types of labels. If this option is
>      enabled,
>      >    biblatex always precedes the last name with the prefix. For
>      example,
>      >    Ludwig van Beethoven would be cited as Beethoven and
>      alphabetized as
>      >    Beethoven, Ludwig van by default. If this option is enabled, he
>      is
>      >    cited as van Beethoven and alphabetized as Van Beethoven,
>      Ludwig
>      >    instead. With Biber, this option is also settable on a per-type
>      basis.
>      I used a rough-and-ready algorithm, probably not the same one biber
>      or biblatex uses.  Basically, the only time my code gives you a
>      non-dropping particle is when you have a name in format
>         B A, D E
>      then B is the particle and A the given name.
>      I don't have any support now for dropping-particle or suffix.
>      This needs to be improved, but I'd need to know better what the
>      algorithm is.
>      >    Inline formatting
>      >    =================
>      >    It probably doesn't play a role for pandoc, but in the interest
>      of
>      >    portability, shouldn't inline formatting in a CSL-YAML file or
>      metadata
>      >    section rather use
>      >        <i> and </i> for italics
>      >        <b> and </b> for bold
>      >        <sub> and </sub> for subscript
>      >        <sup> and </sup> for superscript
>      >        <span style="font-variant:small-caps;"> and </span> for
>      smallcaps
>      >    instead of markdown formatting?
>      >    (see [1]https://github.com/jgm/pandoc/issues/931,
>      >    [2]https://www.zotero.org/support/kb/rich_text_bibliography)
>      No, because the YAML file is standard pandoc metadata, formatted in
>      markdown.  When it is read by pandoc-citeproc, this will be
>      converted
>      where possible to CSL metadata.
>      >    Small Caps
>      >    ==========
>      >    Speaking of small caps: Any plans to add some markdownish
>      formatting
>      >    commands for small caps to pandoc?
>      No idea what a natural syntax would be.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [3]https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-
>    b2f0-3ee666209dae%40googlegroups.com.
>    For more options, visit [4]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://github.com/jgm/pandoc/issues/931
>    2. https://www.zotero.org/support/kb/rich_text_bibliography
>    3. https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%40googlegroups.com
>    4. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 10873 bytes --]

biblatex database entries may contain a "hyphenation" field, specifying the 
language of the bibliographic record, used for determining hyphenation 
patterns and whether title/sentence case conversion is to be applied. It 
must contain exactly one language term. The functional equivalent in CSL is 
the "language" variable, also used for determining whether title/sentence 
case conversion is to be applied.

biblatex database entries may also contain a "language" field, which 
describes the language or languages of the *content* of the work, e.g. 
language = {latin and english}. This has no counterpart in CSL.

unTitlecase should only be applied to titles in languages that do 
distinguish title from sentence case, and that's as far as I know, and in 
line with biblatex's default only English. In other words, unTitlecase 
should only be applied if an entry's hyphenation field contains one of the 
terms american, british, canadian, english, australian, newzealand, 
USenglish or UKenglish. 

On Saturday, September 28, 2013 9:34:33 PM UTC, fiddlosopher wrote:
>
> How is the language represented in a biblatex file?  Or is it an 
> external parameter that comes from the locale or something else? 
>
> +++ Nick Bart [Sep 28 13 13:37 ]: 
> >    unTitlecase: Brilliant. 
> >    Forgot one important detail, though: unTitlecase should only be 
> applied 
> >    if the language is English, i.e., if an entry's biblatex hyphenation 
> >    field is one of american, british, canadian, english, australian, 
> >    newzealand, USenglish or UKenglish. 
> >    From the biblatex manual (v 2.7a), "4.6.4 Miscellaneous Commands", 
> item 
> >    " \MakeSentenceCase": 
> >    > By default, converting to sentence case is enabled for the 
> following 
> >    language identifiers: american, british, canadian, english, 
> australian, 
> >    newzealand as well as the aliases USenglish and UKenglish. 
> >    On Saturday, September 28, 2013 7:34:09 PM UTC, fiddlosopher wrote: 
> > 
> >      +++ Nick Bart [Sep 28 13 03:52 ]: 
> >      >    A few additional ideas/issues: 
> >      >    Title vs. sentence case 
> >      >    ======================= 
> >      >    bibtex and biblatex expect titles in title case in the 
> >      database, 
> >      >    converting these to sentence case if required by a particular 
> >      style, 
> >      >    except for strings protected by {}. 
> >      >    CSL does the opposite: It expects titles in sentence case, 
> >      converting 
> >      >    these to title case if required, except for certain stop 
> words. 
> >      >    I think it make senses for bibtex2pandoc to convert 
> >      bibtex=biblatex 
> >      >    titles to sentence case for further consumption by CSL; i.e., 
> >      to lower 
> >      >    case, except strings protected by {}. 
> >      >        - ex.: 
> >      >        title = {{JFK}: The {CIA}, {Vietnam}, and the Plot to 
> >      Assassinate 
> >      >    {John F. Kennedy}} 
> >      >        becomes 
> >      >        title: 'JFK: The CIA, Vietnam, and the plot to assassinate 
> >      John F. 
> >      >    Kennedy' 
> >      >        - to be discussed: should strings inside commands be 
> >      converted, 
> >      >    too? 
> >      >         i.e., should 
> >      >         title = {An Analysis of \textit{For Whom the Bell Tolls}} 
> >      >         become 
> >      >         title: 'An Analysis of <i>for whom the bell tolls</i>' 
> >      >         and only extra protection 
> >      >         title = {An Analysis of {\textit{For Whom the Bell 
> >      Tolls}}} 
> >      >         would yield 
> >      >         title: 'An Analysis of <i>For Whom the Bell Tolls</i>' 
> >      I've implemented the unTitlecase transformation.  (This depends on 
> a 
> >      recent patch to pandoc that causes the LaTeX reader to insert Span 
> >      when we have a bare group {Like This}.) 
> >      I don't know about the "to be discussed" -- currently strings 
> inside 
> >      emphasis, etc., will retain their case.  I think that's probably 
> >      right. 
> >      >    Corporate authors 
> >      >    ================= 
> >      >    Example: author = {{National Aeronautics and Space 
> >      Administration}}, 
> >      >    Current bibtex2pandoc output: 
> >      >    - author: 
> >      >      - family: Aeronautics 
> >      >        given: 
> >      >        - National 
> >      >      - family: Administration 
> >      >        given: 
> >      >        - Space 
> >      >    Expected: 
> >      >    - author: 
> >      >      - literal: 'National Aeronautics and Space Administration' 
> >      I've added support for this. 
> >      >    Literal and in institution, organization, publisher, location, 
> >      etc. 
> >      >    =========================================================== 
> >      ========== 
> >      >    If code includes 
> >      > 
> >      >      getLiteralList "publisher" ==> setList "publisher" 
> >      >    publisher = {Holt, Rinehart {and} Winston} 
> >      >    Current bibtex2pandoc output: 
> >      >      publisher: 
> >      >      - 'Holt, Rinehart' 
> >      >      - Winston 
> >      >    Expected: 
> >      >      publisher: 
> >      >      - 'Holt, Rinehart and Winston' 
> >      >    Also noticed, however, that pandoc-citeproc does not seem to 
> >      like it 
> >      >    when actual multiple publishers occur: 
> >      >      publisher: 
> >      >      - 'Univ. of Toronto Press' 
> >      >      - Routledge 
> >      >    throws an error: 
> >      >    pandoc-citeproc: Error parsing references: when expecting a 
> >      String, 
> >      >    encountered Array instead 
> >      >    This needs to be checked, but if CSL does not allow multiple 
> >      >    publishers, the code above would need to be changed back to 
> >      >      getField "publisher" ==> setField "publisher" 
> >      >    I see that multiple publishers (and locations) are being 
> >      discussed at 
> >      >    various Zotero/CSL/citeproc-js forums, but don't get a clear 
> >      picture 
> >      >    yet. 
> >      The literal {and} will not cause breaking of author lists or 
> literal 
> >      lists. 
> >      I'll leave the multiple publishers question open for now.  Either 
> >      bibtex2pandoc or pandoc-citeproc should be changed, not sure which. 
> >      >    Names 
> >      >    ===== 
> >      >    Can bibtex2pandoc be expected to distinguish all five name 
> >      components 
> >      >    (family, given, suffix, non-dropping-particle, 
> >      dropping-particle)? 
> >      >    biblatex provides one useful additional bit of info for name 
> >      parsing 
> >      >    via the "useprefix" switch inside the "options" field: 
> >      >    From the biblatex manual (v 2.7a), "3.1.3 Entry Options" 
> >      >    > Whether the name prefix (von, van, of, da, de, della, etc.) 
> >      is 
> >      >    considered when printing the last name in citations. This also 
> >      affects 
> >      >    the sorting and formatting of the bibliography as well as the 
> >      >    generation of certain types of labels. If this option is 
> >      enabled, 
> >      >    biblatex always precedes the last name with the prefix. For 
> >      example, 
> >      >    Ludwig van Beethoven would be cited as Beethoven and 
> >      alphabetized as 
> >      >    Beethoven, Ludwig van by default. If this option is enabled, 
> he 
> >      is 
> >      >    cited as van Beethoven and alphabetized as Van Beethoven, 
> >      Ludwig 
> >      >    instead. With Biber, this option is also settable on a 
> per-type 
> >      basis. 
> >      I used a rough-and-ready algorithm, probably not the same one biber 
> >      or biblatex uses.  Basically, the only time my code gives you a 
> >      non-dropping particle is when you have a name in format 
> >         B A, D E 
> >      then B is the particle and A the given name. 
> >      I don't have any support now for dropping-particle or suffix. 
> >      This needs to be improved, but I'd need to know better what the 
> >      algorithm is. 
> >      >    Inline formatting 
> >      >    ================= 
> >      >    It probably doesn't play a role for pandoc, but in the 
> interest 
> >      of 
> >      >    portability, shouldn't inline formatting in a CSL-YAML file or 
> >      metadata 
> >      >    section rather use 
> >      >        <i> and </i> for italics 
> >      >        <b> and </b> for bold 
> >      >        <sub> and </sub> for subscript 
> >      >        <sup> and </sup> for superscript 
> >      >        <span style="font-variant:small-caps;"> and </span> for 
> >      smallcaps 
> >      >    instead of markdown formatting? 
> >      >    (see [1]https://github.com/jgm/pandoc/issues/931, 
> >      >    [2]https://www.zotero.org/support/kb/rich_text_bibliography) 
> >      No, because the YAML file is standard pandoc metadata, formatted in 
> >      markdown.  When it is read by pandoc-citeproc, this will be 
> >      converted 
> >      where possible to CSL metadata. 
> >      >    Small Caps 
> >      >    ========== 
> >      >    Speaking of small caps: Any plans to add some markdownish 
> >      formatting 
> >      >    commands for small caps to pandoc? 
> >      No idea what a natural syntax would be. 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [3]
> https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b- 
> >    b2f0-3ee666209dae%40googlegroups.com. 
> >    For more options, visit [4]https://groups.google.com/groups/opt_out. 
> > 
> > References 
> > 
> >    1. https://github.com/jgm/pandoc/issues/931 
> >    2. https://www.zotero.org/support/kb/rich_text_bibliography 
> >    3. 
> https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%40googlegroups.com 
> >    4. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9-8efe-0aa033c670da%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 17417 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

OK, I've added this feature.

In order to set the "language" variable properly, we need a mapping
from biblatex hyphenation field values to whatever fills the CSL
language slot.

Can you provide one?

+++ Nick Bart [Sep 28 13 15:19 ]:
>    biblatex database entries may contain a "hyphenation" field, specifying
>    the language of the bibliographic record, used for determining
>    hyphenation patterns and whether title/sentence case conversion is to
>    be applied. It must contain exactly one language term. The functional
>    equivalent in CSL is the "language" variable, also used for determining
>    whether title/sentence case conversion is to be applied.
>    biblatex database entries may also contain a "language" field, which
>    describes the language or languages of the *content* of the work, e.g.
>    language = {latin and english}. This has no counterpart in CSL.
>    unTitlecase should only be applied to titles in languages that do
>    distinguish title from sentence case, and that's as far as I know, and
>    in line with biblatex's default only English. In other words,
>    unTitlecase should only be applied if an entry's hyphenation field
>    contains one of the terms american, british, canadian, english,
>    australian, newzealand, USenglish or UKenglish.
>    On Saturday, September 28, 2013 9:34:33 PM UTC, fiddlosopher wrote:
> 
>      How is the language represented in a biblatex file?  Or is it an
>      external parameter that comes from the locale or something else?
>      +++ Nick Bart [Sep 28 13 13:37 ]:
>      >    unTitlecase: Brilliant.
>      >    Forgot one important detail, though: unTitlecase should only be
>      applied
>      >    if the language is English, i.e., if an entry's biblatex
>      hyphenation
>      >    field is one of american, british, canadian, english,
>      australian,
>      >    newzealand, USenglish or UKenglish.
>      >    From the biblatex manual (v 2.7a), "4.6.4 Miscellaneous
>      Commands", item
>      >    " \MakeSentenceCase":
>      >    > By default, converting to sentence case is enabled for the
>      following
>      >    language identifiers: american, british, canadian, english,
>      australian,
>      >    newzealand as well as the aliases USenglish and UKenglish.
>      >    On Saturday, September 28, 2013 7:34:09 PM UTC, fiddlosopher
>      wrote:
>      >
>      >      +++ Nick Bart [Sep 28 13 03:52 ]:
>      >      >    A few additional ideas/issues:
>      >      >    Title vs. sentence case
>      >      >    =======================
>      >      >    bibtex and biblatex expect titles in title case in the
>      >      database,
>      >      >    converting these to sentence case if required by a
>      particular
>      >      style,
>      >      >    except for strings protected by {}.
>      >      >    CSL does the opposite: It expects titles in sentence
>      case,
>      >      converting
>      >      >    these to title case if required, except for certain stop
>      words.
>      >      >    I think it make senses for bibtex2pandoc to convert
>      >      bibtex=biblatex
>      >      >    titles to sentence case for further consumption by CSL;
>      i.e.,
>      >      to lower
>      >      >    case, except strings protected by {}.
>      >      >        - ex.:
>      >      >        title = {{JFK}: The {CIA}, {Vietnam}, and the Plot
>      to
>      >      Assassinate
>      >      >    {John F. Kennedy}}
>      >      >        becomes
>      >      >        title: 'JFK: The CIA, Vietnam, and the plot to
>      assassinate
>      >      John F.
>      >      >    Kennedy'
>      >      >        - to be discussed: should strings inside commands be
>      >      converted,
>      >      >    too?
>      >      >         i.e., should
>      >      >         title = {An Analysis of \textit{For Whom the Bell
>      Tolls}}
>      >      >         become
>      >      >         title: 'An Analysis of <i>for whom the bell
>      tolls</i>'
>      >      >         and only extra protection
>      >      >         title = {An Analysis of {\textit{For Whom the Bell
>      >      Tolls}}}
>      >      >         would yield
>      >      >         title: 'An Analysis of <i>For Whom the Bell
>      Tolls</i>'
>      >      I've implemented the unTitlecase transformation.  (This
>      depends on a
>      >      recent patch to pandoc that causes the LaTeX reader to insert
>      Span
>      >      when we have a bare group {Like This}.)
>      >      I don't know about the "to be discussed" -- currently strings
>      inside
>      >      emphasis, etc., will retain their case.  I think that's
>      probably
>      >      right.
>      >      >    Corporate authors
>      >      >    =================
>      >      >    Example: author = {{National Aeronautics and Space
>      >      Administration}},
>      >      >    Current bibtex2pandoc output:
>      >      >    - author:
>      >      >      - family: Aeronautics
>      >      >        given:
>      >      >        - National
>      >      >      - family: Administration
>      >      >        given:
>      >      >        - Space
>      >      >    Expected:
>      >      >    - author:
>      >      >      - literal: 'National Aeronautics and Space
>      Administration'
>      >      I've added support for this.
>      >      >    Literal and in institution, organization, publisher,
>      location,
>      >      etc.
>      >      >    =============================
>      ==============================
>      >      ==========
>      >      >    If code includes
>      >      >
>      >      >      getLiteralList "publisher" ==> setList "publisher"
>      >      >    publisher = {Holt, Rinehart {and} Winston}
>      >      >    Current bibtex2pandoc output:
>      >      >      publisher:
>      >      >      - 'Holt, Rinehart'
>      >      >      - Winston
>      >      >    Expected:
>      >      >      publisher:
>      >      >      - 'Holt, Rinehart and Winston'
>      >      >    Also noticed, however, that pandoc-citeproc does not
>      seem to
>      >      like it
>      >      >    when actual multiple publishers occur:
>      >      >      publisher:
>      >      >      - 'Univ. of Toronto Press'
>      >      >      - Routledge
>      >      >    throws an error:
>      >      >    pandoc-citeproc: Error parsing references: when
>      expecting a
>      >      String,
>      >      >    encountered Array instead
>      >      >    This needs to be checked, but if CSL does not allow
>      multiple
>      >      >    publishers, the code above would need to be changed back
>      to
>      >      >      getField "publisher" ==> setField "publisher"
>      >      >    I see that multiple publishers (and locations) are being
>      >      discussed at
>      >      >    various Zotero/CSL/citeproc-js forums, but don't get a
>      clear
>      >      picture
>      >      >    yet.
>      >      The literal {and} will not cause breaking of author lists or
>      literal
>      >      lists.
>      >      I'll leave the multiple publishers question open for now.
>      Either
>      >      bibtex2pandoc or pandoc-citeproc should be changed, not sure
>      which.
>      >      >    Names
>      >      >    =====
>      >      >    Can bibtex2pandoc be expected to distinguish all five
>      name
>      >      components
>      >      >    (family, given, suffix, non-dropping-particle,
>      >      dropping-particle)?
>      >      >    biblatex provides one useful additional bit of info for
>      name
>      >      parsing
>      >      >    via the "useprefix" switch inside the "options" field:
>      >      >    From the biblatex manual (v 2.7a), "3.1.3 Entry Options"
>      >      >    > Whether the name prefix (von, van, of, da, de, della,
>      etc.)
>      >      is
>      >      >    considered when printing the last name in citations.
>      This also
>      >      affects
>      >      >    the sorting and formatting of the bibliography as well
>      as the
>      >      >    generation of certain types of labels. If this option is
>      >      enabled,
>      >      >    biblatex always precedes the last name with the prefix.
>      For
>      >      example,
>      >      >    Ludwig van Beethoven would be cited as Beethoven and
>      >      alphabetized as
>      >      >    Beethoven, Ludwig van by default. If this option is
>      enabled, he
>      >      is
>      >      >    cited as van Beethoven and alphabetized as Van
>      Beethoven,
>      >      Ludwig
>      >      >    instead. With Biber, this option is also settable on a
>      per-type
>      >      basis.
>      >      I used a rough-and-ready algorithm, probably not the same one
>      biber
>      >      or biblatex uses.  Basically, the only time my code gives you
>      a
>      >      non-dropping particle is when you have a name in format
>      >         B A, D E
>      >      then B is the particle and A the given name.
>      >      I don't have any support now for dropping-particle or suffix.
>      >      This needs to be improved, but I'd need to know better what
>      the
>      >      algorithm is.
>      >      >    Inline formatting
>      >      >    =================
>      >      >    It probably doesn't play a role for pandoc, but in the
>      interest
>      >      of
>      >      >    portability, shouldn't inline formatting in a CSL-YAML
>      file or
>      >      metadata
>      >      >    section rather use
>      >      >        <i> and </i> for italics
>      >      >        <b> and </b> for bold
>      >      >        <sub> and </sub> for subscript
>      >      >        <sup> and </sup> for superscript
>      >      >        <span style="font-variant:small-caps;"> and </span>
>      for
>      >      smallcaps
>      >      >    instead of markdown formatting?
>      >      >    (see [1][1]https://github.com/jgm/pandoc/issues/931,
>      >      >    [2][2]https://www.zotero.org/support/kb/rich_text_
>      bibliography)
>      >      No, because the YAML file is standard pandoc metadata,
>      formatted in
>      >      markdown.  When it is read by pandoc-citeproc, this will be
>      >      converted
>      >      where possible to CSL metadata.
>      >      >    Small Caps
>      >      >    ==========
>      >      >    Speaking of small caps: Any plans to add some
>      markdownish
>      >      formatting
>      >      >    commands for small caps to pandoc?
>      >      No idea what a natural syntax would be.
>      >
>      >    --
>      >    You received this message because you are subscribed to the
>      Google
>      >    Groups "pandoc-discuss" group.
>      >    To unsubscribe from this group and stop receiving emails from
>      it, send
>      >    an email to [3]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To post to this group, send email to
>      [4]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To view this discussion on the web visit
>      >    [3][5]https://groups.google.com/d/msgid/pandoc-discuss/
>      a86067ec-6b4a-437b-
>      >    b2f0-3ee666209dae%[6]40googlegroups.com.
>      >    For more options, visit [4][7]https://groups.google.com/
>      groups/opt_out.
>      >
>      > References
>      >
>      >    1. [8]https://github.com/jgm/pandoc/issues/931
>      >    2. [9]https://www.zotero.org/support/kb/rich_text_bibliography
>      >    3. [10]https://groups.google.com/d/
>      msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%
>      40googlegroups.com
>      >    4. [11]https://groups.google.com/groups/opt_out
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [12]https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9
>    -8efe-0aa033c670da%40googlegroups.com.
>    For more options, visit [13]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://github.com/jgm/pandoc/issues/931
>    2. https://www.zotero.org/support/kb/rich_text_bibliography
>    3. javascript:/
>    4. javascript:/
>    5. https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-
>    6. http://40googlegroups.com/
>    7. https://groups.google.com/groups/opt_out
>    8. https://github.com/jgm/pandoc/issues/931
>    9. https://www.zotero.org/support/kb/rich_text_bibliography
>   10. https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%40googlegroups.com
>   11. https://groups.google.com/groups/opt_out
>   12. https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9-8efe-0aa033c670da%40googlegroups.com
>   13. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 14861 bytes --]

I already added this earlier:

  opt $ getRawField "hyphenation" >>= setRawField "language"

Seems to work nicely ...

On Saturday, September 28, 2013 10:59:47 PM UTC, fiddlosopher wrote:
>
> OK, I've added this feature. 
>
> In order to set the "language" variable properly, we need a mapping 
> from biblatex hyphenation field values to whatever fills the CSL 
> language slot. 
>
> Can you provide one? 
>
> +++ Nick Bart [Sep 28 13 15:19 ]: 
> >    biblatex database entries may contain a "hyphenation" field, 
> specifying 
> >    the language of the bibliographic record, used for determining 
> >    hyphenation patterns and whether title/sentence case conversion is to 
> >    be applied. It must contain exactly one language term. The functional 
> >    equivalent in CSL is the "language" variable, also used for 
> determining 
> >    whether title/sentence case conversion is to be applied. 
> >    biblatex database entries may also contain a "language" field, which 
> >    describes the language or languages of the *content* of the work, 
> e.g. 
> >    language = {latin and english}. This has no counterpart in CSL. 
> >    unTitlecase should only be applied to titles in languages that do 
> >    distinguish title from sentence case, and that's as far as I know, 
> and 
> >    in line with biblatex's default only English. In other words, 
> >    unTitlecase should only be applied if an entry's hyphenation field 
> >    contains one of the terms american, british, canadian, english, 
> >    australian, newzealand, USenglish or UKenglish. 
> >    On Saturday, September 28, 2013 9:34:33 PM UTC, fiddlosopher wrote: 
> > 
> >      How is the language represented in a biblatex file?  Or is it an 
> >      external parameter that comes from the locale or something else? 
> >      +++ Nick Bart [Sep 28 13 13:37 ]: 
> >      >    unTitlecase: Brilliant. 
> >      >    Forgot one important detail, though: unTitlecase should only 
> be 
> >      applied 
> >      >    if the language is English, i.e., if an entry's biblatex 
> >      hyphenation 
> >      >    field is one of american, british, canadian, english, 
> >      australian, 
> >      >    newzealand, USenglish or UKenglish. 
> >      >    From the biblatex manual (v 2.7a), "4.6.4 Miscellaneous 
> >      Commands", item 
> >      >    " \MakeSentenceCase": 
> >      >    > By default, converting to sentence case is enabled for the 
> >      following 
> >      >    language identifiers: american, british, canadian, english, 
> >      australian, 
> >      >    newzealand as well as the aliases USenglish and UKenglish. 
> >      >    On Saturday, September 28, 2013 7:34:09 PM UTC, fiddlosopher 
> >      wrote: 
> >      > 
> >      >      +++ Nick Bart [Sep 28 13 03:52 ]: 
> >      >      >    A few additional ideas/issues: 
> >      >      >    Title vs. sentence case 
> >      >      >    ======================= 
> >      >      >    bibtex and biblatex expect titles in title case in the 
> >      >      database, 
> >      >      >    converting these to sentence case if required by a 
> >      particular 
> >      >      style, 
> >      >      >    except for strings protected by {}. 
> >      >      >    CSL does the opposite: It expects titles in sentence 
> >      case, 
> >      >      converting 
> >      >      >    these to title case if required, except for certain 
> stop 
> >      words. 
> >      >      >    I think it make senses for bibtex2pandoc to convert 
> >      >      bibtex=biblatex 
> >      >      >    titles to sentence case for further consumption by CSL; 
> >      i.e., 
> >      >      to lower 
> >      >      >    case, except strings protected by {}. 
> >      >      >        - ex.: 
> >      >      >        title = {{JFK}: The {CIA}, {Vietnam}, and the Plot 
> >      to 
> >      >      Assassinate 
> >      >      >    {John F. Kennedy}} 
> >      >      >        becomes 
> >      >      >        title: 'JFK: The CIA, Vietnam, and the plot to 
> >      assassinate 
> >      >      John F. 
> >      >      >    Kennedy' 
> >      >      >        - to be discussed: should strings inside commands 
> be 
> >      >      converted, 
> >      >      >    too? 
> >      >      >         i.e., should 
> >      >      >         title = {An Analysis of \textit{For Whom the Bell 
> >      Tolls}} 
> >      >      >         become 
> >      >      >         title: 'An Analysis of <i>for whom the bell 
> >      tolls</i>' 
> >      >      >         and only extra protection 
> >      >      >         title = {An Analysis of {\textit{For Whom the Bell 
> >      >      Tolls}}} 
> >      >      >         would yield 
> >      >      >         title: 'An Analysis of <i>For Whom the Bell 
> >      Tolls</i>' 
> >      >      I've implemented the unTitlecase transformation.  (This 
> >      depends on a 
> >      >      recent patch to pandoc that causes the LaTeX reader to 
> insert 
> >      Span 
> >      >      when we have a bare group {Like This}.) 
> >      >      I don't know about the "to be discussed" -- currently 
> strings 
> >      inside 
> >      >      emphasis, etc., will retain their case.  I think that's 
> >      probably 
> >      >      right. 
> >      >      >    Corporate authors 
> >      >      >    ================= 
> >      >      >    Example: author = {{National Aeronautics and Space 
> >      >      Administration}}, 
> >      >      >    Current bibtex2pandoc output: 
> >      >      >    - author: 
> >      >      >      - family: Aeronautics 
> >      >      >        given: 
> >      >      >        - National 
> >      >      >      - family: Administration 
> >      >      >        given: 
> >      >      >        - Space 
> >      >      >    Expected: 
> >      >      >    - author: 
> >      >      >      - literal: 'National Aeronautics and Space 
> >      Administration' 
> >      >      I've added support for this. 
> >      >      >    Literal and in institution, organization, publisher, 
> >      location, 
> >      >      etc. 
> >      >      >    ============================= 
> >      ============================== 
> >      >      ========== 
> >      >      >    If code includes 
> >      >      > 
> >      >      >      getLiteralList "publisher" ==> setList "publisher" 
> >      >      >    publisher = {Holt, Rinehart {and} Winston} 
> >      >      >    Current bibtex2pandoc output: 
> >      >      >      publisher: 
> >      >      >      - 'Holt, Rinehart' 
> >      >      >      - Winston 
> >      >      >    Expected: 
> >      >      >      publisher: 
> >      >      >      - 'Holt, Rinehart and Winston' 
> >      >      >    Also noticed, however, that pandoc-citeproc does not 
> >      seem to 
> >      >      like it 
> >      >      >    when actual multiple publishers occur: 
> >      >      >      publisher: 
> >      >      >      - 'Univ. of Toronto Press' 
> >      >      >      - Routledge 
> >      >      >    throws an error: 
> >      >      >    pandoc-citeproc: Error parsing references: when 
> >      expecting a 
> >      >      String, 
> >      >      >    encountered Array instead 
> >      >      >    This needs to be checked, but if CSL does not allow 
> >      multiple 
> >      >      >    publishers, the code above would need to be changed 
> back 
> >      to 
> >      >      >      getField "publisher" ==> setField "publisher" 
> >      >      >    I see that multiple publishers (and locations) are 
> being 
> >      >      discussed at 
> >      >      >    various Zotero/CSL/citeproc-js forums, but don't get a 
> >      clear 
> >      >      picture 
> >      >      >    yet. 
> >      >      The literal {and} will not cause breaking of author lists or 
> >      literal 
> >      >      lists. 
> >      >      I'll leave the multiple publishers question open for now. 
> >      Either 
> >      >      bibtex2pandoc or pandoc-citeproc should be changed, not sure 
> >      which. 
> >      >      >    Names 
> >      >      >    ===== 
> >      >      >    Can bibtex2pandoc be expected to distinguish all five 
> >      name 
> >      >      components 
> >      >      >    (family, given, suffix, non-dropping-particle, 
> >      >      dropping-particle)? 
> >      >      >    biblatex provides one useful additional bit of info for 
> >      name 
> >      >      parsing 
> >      >      >    via the "useprefix" switch inside the "options" field: 
> >      >      >    From the biblatex manual (v 2.7a), "3.1.3 Entry 
> Options" 
> >      >      >    > Whether the name prefix (von, van, of, da, de, della, 
> >      etc.) 
> >      >      is 
> >      >      >    considered when printing the last name in citations. 
> >      This also 
> >      >      affects 
> >      >      >    the sorting and formatting of the bibliography as well 
> >      as the 
> >      >      >    generation of certain types of labels. If this option 
> is 
> >      >      enabled, 
> >      >      >    biblatex always precedes the last name with the prefix. 
> >      For 
> >      >      example, 
> >      >      >    Ludwig van Beethoven would be cited as Beethoven and 
> >      >      alphabetized as 
> >      >      >    Beethoven, Ludwig van by default. If this option is 
> >      enabled, he 
> >      >      is 
> >      >      >    cited as van Beethoven and alphabetized as Van 
> >      Beethoven, 
> >      >      Ludwig 
> >      >      >    instead. With Biber, this option is also settable on a 
> >      per-type 
> >      >      basis. 
> >      >      I used a rough-and-ready algorithm, probably not the same 
> one 
> >      biber 
> >      >      or biblatex uses.  Basically, the only time my code gives 
> you 
> >      a 
> >      >      non-dropping particle is when you have a name in format 
> >      >         B A, D E 
> >      >      then B is the particle and A the given name. 
> >      >      I don't have any support now for dropping-particle or 
> suffix. 
> >      >      This needs to be improved, but I'd need to know better what 
> >      the 
> >      >      algorithm is. 
> >      >      >    Inline formatting 
> >      >      >    ================= 
> >      >      >    It probably doesn't play a role for pandoc, but in the 
> >      interest 
> >      >      of 
> >      >      >    portability, shouldn't inline formatting in a CSL-YAML 
> >      file or 
> >      >      metadata 
> >      >      >    section rather use 
> >      >      >        <i> and </i> for italics 
> >      >      >        <b> and </b> for bold 
> >      >      >        <sub> and </sub> for subscript 
> >      >      >        <sup> and </sup> for superscript 
> >      >      >        <span style="font-variant:small-caps;"> and </span> 
> >      for 
> >      >      smallcaps 
> >      >      >    instead of markdown formatting? 
> >      >      >    (see [1][1]https://github.com/jgm/pandoc/issues/931, 
> >      >      >    [2][2]https://www.zotero.org/support/kb/rich_text_ 
> >      bibliography) 
> >      >      No, because the YAML file is standard pandoc metadata, 
> >      formatted in 
> >      >      markdown.  When it is read by pandoc-citeproc, this will be 
> >      >      converted 
> >      >      where possible to CSL metadata. 
> >      >      >    Small Caps 
> >      >      >    ========== 
> >      >      >    Speaking of small caps: Any plans to add some 
> >      markdownish 
> >      >      formatting 
> >      >      >    commands for small caps to pandoc? 
> >      >      No idea what a natural syntax would be. 
> >      > 
> >      >    -- 
> >      >    You received this message because you are subscribed to the 
> >      Google 
> >      >    Groups "pandoc-discuss" group. 
> >      >    To unsubscribe from this group and stop receiving emails from 
> >      it, send 
> >      >    an email to [3]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >    To post to this group, send email to 
> >      [4]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >    To view this discussion on the web visit 
> >      >    [3][5]https://groups.google.com/d/msgid/pandoc-discuss/ 
> >      a86067ec-6b4a-437b- 
> >      >    b2f0-3ee666209dae%[6]40googlegroups.com. 
> >      >    For more options, visit [4][7]https://groups.google.com/ 
> >      groups/opt_out. 
> >      > 
> >      > References 
> >      > 
> >      >    1. [8]https://github.com/jgm/pandoc/issues/931 
> >      >    2. [9]https://www.zotero.org/support/kb/rich_text_bibliography 
> >      >    3. [10]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae% 
> >      40googlegroups.com 
> >      >    4. [11]https://groups.google.com/groups/opt_out 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [12]
> https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9 
> >    -8efe-0aa033c670da%40googlegroups.com. 
> >    For more options, visit [13]https://groups.google.com/groups/opt_out. 
>
> > 
> > References 
> > 
> >    1. https://github.com/jgm/pandoc/issues/931 
> >    2. https://www.zotero.org/support/kb/rich_text_bibliography 
> >    3. javascript:/ 
> >    4. javascript:/ 
> >    5. 
> https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b- 
> >    6. http://40googlegroups.com/ 
> >    7. https://groups.google.com/groups/opt_out 
> >    8. https://github.com/jgm/pandoc/issues/931 
> >    9. https://www.zotero.org/support/kb/rich_text_bibliography 
> >   10. 
> https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%40googlegroups.com 
> >   11. https://groups.google.com/groups/opt_out 
> >   12. 
> https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9-8efe-0aa033c670da%40googlegroups.com 
> >   13. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/b268960d-ddc1-4346-8535-f90124e1e890%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 27719 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 15718 bytes --]

On second thoughts, however: Works for me, since I have only "english" in 
my hyphenation fields.

However, CSL's title cases routine is simply triggered if the CSL language 
field starts with "en", but not by, e.g., "american"; so we'll probably 
have to map hyphenation = american, british, canadian, australian, 
newzealand, USenglish or UKenglish either all to "english", or individually 
to "en-UK", "en-US", etc.

On Saturday, September 28, 2013 11:16:42 PM UTC, Nick Bart wrote:
>
> I already added this earlier:
>
>   opt $ getRawField "hyphenation" >>= setRawField "language"
>
> Seems to work nicely ...
>
> On Saturday, September 28, 2013 10:59:47 PM UTC, fiddlosopher wrote:
>>
>> OK, I've added this feature. 
>>
>> In order to set the "language" variable properly, we need a mapping 
>> from biblatex hyphenation field values to whatever fills the CSL 
>> language slot. 
>>
>> Can you provide one? 
>>
>> +++ Nick Bart [Sep 28 13 15:19 ]: 
>> >    biblatex database entries may contain a "hyphenation" field, 
>> specifying 
>> >    the language of the bibliographic record, used for determining 
>> >    hyphenation patterns and whether title/sentence case conversion is 
>> to 
>> >    be applied. It must contain exactly one language term. The 
>> functional 
>> >    equivalent in CSL is the "language" variable, also used for 
>> determining 
>> >    whether title/sentence case conversion is to be applied. 
>> >    biblatex database entries may also contain a "language" field, which 
>> >    describes the language or languages of the *content* of the work, 
>> e.g. 
>> >    language = {latin and english}. This has no counterpart in CSL. 
>> >    unTitlecase should only be applied to titles in languages that do 
>> >    distinguish title from sentence case, and that's as far as I know, 
>> and 
>> >    in line with biblatex's default only English. In other words, 
>> >    unTitlecase should only be applied if an entry's hyphenation field 
>> >    contains one of the terms american, british, canadian, english, 
>> >    australian, newzealand, USenglish or UKenglish. 
>> >    On Saturday, September 28, 2013 9:34:33 PM UTC, fiddlosopher wrote: 
>> > 
>> >      How is the language represented in a biblatex file?  Or is it an 
>> >      external parameter that comes from the locale or something else? 
>> >      +++ Nick Bart [Sep 28 13 13:37 ]: 
>> >      >    unTitlecase: Brilliant. 
>> >      >    Forgot one important detail, though: unTitlecase should only 
>> be 
>> >      applied 
>> >      >    if the language is English, i.e., if an entry's biblatex 
>> >      hyphenation 
>> >      >    field is one of american, british, canadian, english, 
>> >      australian, 
>> >      >    newzealand, USenglish or UKenglish. 
>> >      >    From the biblatex manual (v 2.7a), "4.6.4 Miscellaneous 
>> >      Commands", item 
>> >      >    " \MakeSentenceCase": 
>> >      >    > By default, converting to sentence case is enabled for the 
>> >      following 
>> >      >    language identifiers: american, british, canadian, english, 
>> >      australian, 
>> >      >    newzealand as well as the aliases USenglish and UKenglish. 
>> >      >    On Saturday, September 28, 2013 7:34:09 PM UTC, fiddlosopher 
>> >      wrote: 
>> >      > 
>> >      >      +++ Nick Bart [Sep 28 13 03:52 ]: 
>> >      >      >    A few additional ideas/issues: 
>> >      >      >    Title vs. sentence case 
>> >      >      >    ======================= 
>> >      >      >    bibtex and biblatex expect titles in title case in the 
>> >      >      database, 
>> >      >      >    converting these to sentence case if required by a 
>> >      particular 
>> >      >      style, 
>> >      >      >    except for strings protected by {}. 
>> >      >      >    CSL does the opposite: It expects titles in sentence 
>> >      case, 
>> >      >      converting 
>> >      >      >    these to title case if required, except for certain 
>> stop 
>> >      words. 
>> >      >      >    I think it make senses for bibtex2pandoc to convert 
>> >      >      bibtex=biblatex 
>> >      >      >    titles to sentence case for further consumption by 
>> CSL; 
>> >      i.e., 
>> >      >      to lower 
>> >      >      >    case, except strings protected by {}. 
>> >      >      >        - ex.: 
>> >      >      >        title = {{JFK}: The {CIA}, {Vietnam}, and the Plot 
>> >      to 
>> >      >      Assassinate 
>> >      >      >    {John F. Kennedy}} 
>> >      >      >        becomes 
>> >      >      >        title: 'JFK: The CIA, Vietnam, and the plot to 
>> >      assassinate 
>> >      >      John F. 
>> >      >      >    Kennedy' 
>> >      >      >        - to be discussed: should strings inside commands 
>> be 
>> >      >      converted, 
>> >      >      >    too? 
>> >      >      >         i.e., should 
>> >      >      >         title = {An Analysis of \textit{For Whom the Bell 
>> >      Tolls}} 
>> >      >      >         become 
>> >      >      >         title: 'An Analysis of <i>for whom the bell 
>> >      tolls</i>' 
>> >      >      >         and only extra protection 
>> >      >      >         title = {An Analysis of {\textit{For Whom the 
>> Bell 
>> >      >      Tolls}}} 
>> >      >      >         would yield 
>> >      >      >         title: 'An Analysis of <i>For Whom the Bell 
>> >      Tolls</i>' 
>> >      >      I've implemented the unTitlecase transformation.  (This 
>> >      depends on a 
>> >      >      recent patch to pandoc that causes the LaTeX reader to 
>> insert 
>> >      Span 
>> >      >      when we have a bare group {Like This}.) 
>> >      >      I don't know about the "to be discussed" -- currently 
>> strings 
>> >      inside 
>> >      >      emphasis, etc., will retain their case.  I think that's 
>> >      probably 
>> >      >      right. 
>> >      >      >    Corporate authors 
>> >      >      >    ================= 
>> >      >      >    Example: author = {{National Aeronautics and Space 
>> >      >      Administration}}, 
>> >      >      >    Current bibtex2pandoc output: 
>> >      >      >    - author: 
>> >      >      >      - family: Aeronautics 
>> >      >      >        given: 
>> >      >      >        - National 
>> >      >      >      - family: Administration 
>> >      >      >        given: 
>> >      >      >        - Space 
>> >      >      >    Expected: 
>> >      >      >    - author: 
>> >      >      >      - literal: 'National Aeronautics and Space 
>> >      Administration' 
>> >      >      I've added support for this. 
>> >      >      >    Literal and in institution, organization, publisher, 
>> >      location, 
>> >      >      etc. 
>> >      >      >    ============================= 
>> >      ============================== 
>> >      >      ========== 
>> >      >      >    If code includes 
>> >      >      > 
>> >      >      >      getLiteralList "publisher" ==> setList "publisher" 
>> >      >      >    publisher = {Holt, Rinehart {and} Winston} 
>> >      >      >    Current bibtex2pandoc output: 
>> >      >      >      publisher: 
>> >      >      >      - 'Holt, Rinehart' 
>> >      >      >      - Winston 
>> >      >      >    Expected: 
>> >      >      >      publisher: 
>> >      >      >      - 'Holt, Rinehart and Winston' 
>> >      >      >    Also noticed, however, that pandoc-citeproc does not 
>> >      seem to 
>> >      >      like it 
>> >      >      >    when actual multiple publishers occur: 
>> >      >      >      publisher: 
>> >      >      >      - 'Univ. of Toronto Press' 
>> >      >      >      - Routledge 
>> >      >      >    throws an error: 
>> >      >      >    pandoc-citeproc: Error parsing references: when 
>> >      expecting a 
>> >      >      String, 
>> >      >      >    encountered Array instead 
>> >      >      >    This needs to be checked, but if CSL does not allow 
>> >      multiple 
>> >      >      >    publishers, the code above would need to be changed 
>> back 
>> >      to 
>> >      >      >      getField "publisher" ==> setField "publisher" 
>> >      >      >    I see that multiple publishers (and locations) are 
>> being 
>> >      >      discussed at 
>> >      >      >    various Zotero/CSL/citeproc-js forums, but don't get a 
>> >      clear 
>> >      >      picture 
>> >      >      >    yet. 
>> >      >      The literal {and} will not cause breaking of author lists 
>> or 
>> >      literal 
>> >      >      lists. 
>> >      >      I'll leave the multiple publishers question open for now. 
>> >      Either 
>> >      >      bibtex2pandoc or pandoc-citeproc should be changed, not 
>> sure 
>> >      which. 
>> >      >      >    Names 
>> >      >      >    ===== 
>> >      >      >    Can bibtex2pandoc be expected to distinguish all five 
>> >      name 
>> >      >      components 
>> >      >      >    (family, given, suffix, non-dropping-particle, 
>> >      >      dropping-particle)? 
>> >      >      >    biblatex provides one useful additional bit of info 
>> for 
>> >      name 
>> >      >      parsing 
>> >      >      >    via the "useprefix" switch inside the "options" field: 
>> >      >      >    From the biblatex manual (v 2.7a), "3.1.3 Entry 
>> Options" 
>> >      >      >    > Whether the name prefix (von, van, of, da, de, 
>> della, 
>> >      etc.) 
>> >      >      is 
>> >      >      >    considered when printing the last name in citations. 
>> >      This also 
>> >      >      affects 
>> >      >      >    the sorting and formatting of the bibliography as well 
>> >      as the 
>> >      >      >    generation of certain types of labels. If this option 
>> is 
>> >      >      enabled, 
>> >      >      >    biblatex always precedes the last name with the 
>> prefix. 
>> >      For 
>> >      >      example, 
>> >      >      >    Ludwig van Beethoven would be cited as Beethoven and 
>> >      >      alphabetized as 
>> >      >      >    Beethoven, Ludwig van by default. If this option is 
>> >      enabled, he 
>> >      >      is 
>> >      >      >    cited as van Beethoven and alphabetized as Van 
>> >      Beethoven, 
>> >      >      Ludwig 
>> >      >      >    instead. With Biber, this option is also settable on a 
>> >      per-type 
>> >      >      basis. 
>> >      >      I used a rough-and-ready algorithm, probably not the same 
>> one 
>> >      biber 
>> >      >      or biblatex uses.  Basically, the only time my code gives 
>> you 
>> >      a 
>> >      >      non-dropping particle is when you have a name in format 
>> >      >         B A, D E 
>> >      >      then B is the particle and A the given name. 
>> >      >      I don't have any support now for dropping-particle or 
>> suffix. 
>> >      >      This needs to be improved, but I'd need to know better what 
>> >      the 
>> >      >      algorithm is. 
>> >      >      >    Inline formatting 
>> >      >      >    ================= 
>> >      >      >    It probably doesn't play a role for pandoc, but in the 
>> >      interest 
>> >      >      of 
>> >      >      >    portability, shouldn't inline formatting in a CSL-YAML 
>> >      file or 
>> >      >      metadata 
>> >      >      >    section rather use 
>> >      >      >        <i> and </i> for italics 
>> >      >      >        <b> and </b> for bold 
>> >      >      >        <sub> and </sub> for subscript 
>> >      >      >        <sup> and </sup> for superscript 
>> >      >      >        <span style="font-variant:small-caps;"> and 
>> </span> 
>> >      for 
>> >      >      smallcaps 
>> >      >      >    instead of markdown formatting? 
>> >      >      >    (see [1][1]https://github.com/jgm/pandoc/issues/931, 
>> >      >      >    [2][2]https://www.zotero.org/support/kb/rich_text_ 
>> >      bibliography) 
>> >      >      No, because the YAML file is standard pandoc metadata, 
>> >      formatted in 
>> >      >      markdown.  When it is read by pandoc-citeproc, this will be 
>> >      >      converted 
>> >      >      where possible to CSL metadata. 
>> >      >      >    Small Caps 
>> >      >      >    ========== 
>> >      >      >    Speaking of small caps: Any plans to add some 
>> >      markdownish 
>> >      >      formatting 
>> >      >      >    commands for small caps to pandoc? 
>> >      >      No idea what a natural syntax would be. 
>> >      > 
>> >      >    -- 
>> >      >    You received this message because you are subscribed to the 
>> >      Google 
>> >      >    Groups "pandoc-discuss" group. 
>> >      >    To unsubscribe from this group and stop receiving emails from 
>> >      it, send 
>> >      >    an email to [3]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>> >      >    To post to this group, send email to 
>> >      [4]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>> >      >    To view this discussion on the web visit 
>> >      >    [3][5]https://groups.google.com/d/msgid/pandoc-discuss/ 
>> >      a86067ec-6b4a-437b- 
>> >      >    b2f0-3ee666209dae%[6]40googlegroups.com. 
>> >      >    For more options, visit [4][7]https://groups.google.com/ 
>> >      groups/opt_out. 
>> >      > 
>> >      > References 
>> >      > 
>> >      >    1. [8]https://github.com/jgm/pandoc/issues/931 
>> >      >    2. [9]
>> https://www.zotero.org/support/kb/rich_text_bibliography 
>> >      >    3. [10]https://groups.google.com/d/ 
>> >      msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae% 
>> >      40googlegroups.com 
>> >      >    4. [11]https://groups.google.com/groups/opt_out 
>> > 
>> >    -- 
>> >    You received this message because you are subscribed to the Google 
>> >    Groups "pandoc-discuss" group. 
>> >    To unsubscribe from this group and stop receiving emails from it, 
>> send 
>> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>> >    To view this discussion on the web visit 
>> >    [12]
>> https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9 
>> >    -8efe-0aa033c670da%40googlegroups.com. 
>> >    For more options, visit [13]https://groups.google.com/groups/opt_out. 
>>
>> > 
>> > References 
>> > 
>> >    1. https://github.com/jgm/pandoc/issues/931 
>> >    2. https://www.zotero.org/support/kb/rich_text_bibliography 
>> >    3. javascript:/ 
>> >    4. javascript:/ 
>> >    5. 
>> https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b- 
>> >    6. http://40googlegroups.com/ 
>> >    7. https://groups.google.com/groups/opt_out 
>> >    8. https://github.com/jgm/pandoc/issues/931 
>> >    9. https://www.zotero.org/support/kb/rich_text_bibliography 
>> >   10. 
>> https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%40googlegroups.com 
>> >   11. https://groups.google.com/groups/opt_out 
>> >   12. 
>> https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9-8efe-0aa033c670da%40googlegroups.com 
>> >   13. https://groups.google.com/groups/opt_out 
>>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/2e3d65a5-01f8-4be5-94b5-2346160886ac%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 28209 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 16856 bytes --]

OK, this should do the trick:

  opt $ do
    val <- getRawField "hyphenation"
    case val of 
      "english"    -> setRawField "language" "en"
      "american"   -> setRawField "language" "en-US"
      "british"    -> setRawField "language" "en-UK"
      "canadian"   -> setRawField "language" "en-CA"
      "australian" -> setRawField "language" "en-AU"
      "newzealand" -> setRawField "language" "en-NZ"
      "USenglish"  -> setRawField "language" "en-US"
      "UKenglish"  -> setRawField "language" "en-UK"
      _            -> setRawField "language" val


And remove 

opt $ getRawField "hyphenation" >>= setRawField "language"



On Saturday, September 28, 2013 11:30:24 PM UTC, Nick Bart wrote:
>
> On second thoughts, however: Works for me, since I have only "english" in 
> my hyphenation fields.
>
> However, CSL's title cases routine is simply triggered if the CSL language 
> field starts with "en", but not by, e.g., "american"; so we'll probably 
> have to map hyphenation = american, british, canadian, australian, 
> newzealand, USenglish or UKenglish either all to "english", or individually 
> to "en-UK", "en-US", etc.
>
> On Saturday, September 28, 2013 11:16:42 PM UTC, Nick Bart wrote:
>>
>> I already added this earlier:
>>
>>   opt $ getRawField "hyphenation" >>= setRawField "language"
>>
>> Seems to work nicely ...
>>
>> On Saturday, September 28, 2013 10:59:47 PM UTC, fiddlosopher wrote:
>>>
>>> OK, I've added this feature. 
>>>
>>> In order to set the "language" variable properly, we need a mapping 
>>> from biblatex hyphenation field values to whatever fills the CSL 
>>> language slot. 
>>>
>>> Can you provide one? 
>>>
>>> +++ Nick Bart [Sep 28 13 15:19 ]: 
>>> >    biblatex database entries may contain a "hyphenation" field, 
>>> specifying 
>>> >    the language of the bibliographic record, used for determining 
>>> >    hyphenation patterns and whether title/sentence case conversion is 
>>> to 
>>> >    be applied. It must contain exactly one language term. The 
>>> functional 
>>> >    equivalent in CSL is the "language" variable, also used for 
>>> determining 
>>> >    whether title/sentence case conversion is to be applied. 
>>> >    biblatex database entries may also contain a "language" field, 
>>> which 
>>> >    describes the language or languages of the *content* of the work, 
>>> e.g. 
>>> >    language = {latin and english}. This has no counterpart in CSL. 
>>> >    unTitlecase should only be applied to titles in languages that do 
>>> >    distinguish title from sentence case, and that's as far as I know, 
>>> and 
>>> >    in line with biblatex's default only English. In other words, 
>>> >    unTitlecase should only be applied if an entry's hyphenation field 
>>> >    contains one of the terms american, british, canadian, english, 
>>> >    australian, newzealand, USenglish or UKenglish. 
>>> >    On Saturday, September 28, 2013 9:34:33 PM UTC, fiddlosopher wrote: 
>>> > 
>>> >      How is the language represented in a biblatex file?  Or is it an 
>>> >      external parameter that comes from the locale or something else? 
>>> >      +++ Nick Bart [Sep 28 13 13:37 ]: 
>>> >      >    unTitlecase: Brilliant. 
>>> >      >    Forgot one important detail, though: unTitlecase should only 
>>> be 
>>> >      applied 
>>> >      >    if the language is English, i.e., if an entry's biblatex 
>>> >      hyphenation 
>>> >      >    field is one of american, british, canadian, english, 
>>> >      australian, 
>>> >      >    newzealand, USenglish or UKenglish. 
>>> >      >    From the biblatex manual (v 2.7a), "4.6.4 Miscellaneous 
>>> >      Commands", item 
>>> >      >    " \MakeSentenceCase": 
>>> >      >    > By default, converting to sentence case is enabled for the 
>>> >      following 
>>> >      >    language identifiers: american, british, canadian, english, 
>>> >      australian, 
>>> >      >    newzealand as well as the aliases USenglish and UKenglish. 
>>> >      >    On Saturday, September 28, 2013 7:34:09 PM UTC, fiddlosopher 
>>> >      wrote: 
>>> >      > 
>>> >      >      +++ Nick Bart [Sep 28 13 03:52 ]: 
>>> >      >      >    A few additional ideas/issues: 
>>> >      >      >    Title vs. sentence case 
>>> >      >      >    ======================= 
>>> >      >      >    bibtex and biblatex expect titles in title case in 
>>> the 
>>> >      >      database, 
>>> >      >      >    converting these to sentence case if required by a 
>>> >      particular 
>>> >      >      style, 
>>> >      >      >    except for strings protected by {}. 
>>> >      >      >    CSL does the opposite: It expects titles in sentence 
>>> >      case, 
>>> >      >      converting 
>>> >      >      >    these to title case if required, except for certain 
>>> stop 
>>> >      words. 
>>> >      >      >    I think it make senses for bibtex2pandoc to convert 
>>> >      >      bibtex=biblatex 
>>> >      >      >    titles to sentence case for further consumption by 
>>> CSL; 
>>> >      i.e., 
>>> >      >      to lower 
>>> >      >      >    case, except strings protected by {}. 
>>> >      >      >        - ex.: 
>>> >      >      >        title = {{JFK}: The {CIA}, {Vietnam}, and the 
>>> Plot 
>>> >      to 
>>> >      >      Assassinate 
>>> >      >      >    {John F. Kennedy}} 
>>> >      >      >        becomes 
>>> >      >      >        title: 'JFK: The CIA, Vietnam, and the plot to 
>>> >      assassinate 
>>> >      >      John F. 
>>> >      >      >    Kennedy' 
>>> >      >      >        - to be discussed: should strings inside commands 
>>> be 
>>> >      >      converted, 
>>> >      >      >    too? 
>>> >      >      >         i.e., should 
>>> >      >      >         title = {An Analysis of \textit{For Whom the 
>>> Bell 
>>> >      Tolls}} 
>>> >      >      >         become 
>>> >      >      >         title: 'An Analysis of <i>for whom the bell 
>>> >      tolls</i>' 
>>> >      >      >         and only extra protection 
>>> >      >      >         title = {An Analysis of {\textit{For Whom the 
>>> Bell 
>>> >      >      Tolls}}} 
>>> >      >      >         would yield 
>>> >      >      >         title: 'An Analysis of <i>For Whom the Bell 
>>> >      Tolls</i>' 
>>> >      >      I've implemented the unTitlecase transformation.  (This 
>>> >      depends on a 
>>> >      >      recent patch to pandoc that causes the LaTeX reader to 
>>> insert 
>>> >      Span 
>>> >      >      when we have a bare group {Like This}.) 
>>> >      >      I don't know about the "to be discussed" -- currently 
>>> strings 
>>> >      inside 
>>> >      >      emphasis, etc., will retain their case.  I think that's 
>>> >      probably 
>>> >      >      right. 
>>> >      >      >    Corporate authors 
>>> >      >      >    ================= 
>>> >      >      >    Example: author = {{National Aeronautics and Space 
>>> >      >      Administration}}, 
>>> >      >      >    Current bibtex2pandoc output: 
>>> >      >      >    - author: 
>>> >      >      >      - family: Aeronautics 
>>> >      >      >        given: 
>>> >      >      >        - National 
>>> >      >      >      - family: Administration 
>>> >      >      >        given: 
>>> >      >      >        - Space 
>>> >      >      >    Expected: 
>>> >      >      >    - author: 
>>> >      >      >      - literal: 'National Aeronautics and Space 
>>> >      Administration' 
>>> >      >      I've added support for this. 
>>> >      >      >    Literal and in institution, organization, publisher, 
>>> >      location, 
>>> >      >      etc. 
>>> >      >      >    ============================= 
>>> >      ============================== 
>>> >      >      ========== 
>>> >      >      >    If code includes 
>>> >      >      > 
>>> >      >      >      getLiteralList "publisher" ==> setList "publisher" 
>>> >      >      >    publisher = {Holt, Rinehart {and} Winston} 
>>> >      >      >    Current bibtex2pandoc output: 
>>> >      >      >      publisher: 
>>> >      >      >      - 'Holt, Rinehart' 
>>> >      >      >      - Winston 
>>> >      >      >    Expected: 
>>> >      >      >      publisher: 
>>> >      >      >      - 'Holt, Rinehart and Winston' 
>>> >      >      >    Also noticed, however, that pandoc-citeproc does not 
>>> >      seem to 
>>> >      >      like it 
>>> >      >      >    when actual multiple publishers occur: 
>>> >      >      >      publisher: 
>>> >      >      >      - 'Univ. of Toronto Press' 
>>> >      >      >      - Routledge 
>>> >      >      >    throws an error: 
>>> >      >      >    pandoc-citeproc: Error parsing references: when 
>>> >      expecting a 
>>> >      >      String, 
>>> >      >      >    encountered Array instead 
>>> >      >      >    This needs to be checked, but if CSL does not allow 
>>> >      multiple 
>>> >      >      >    publishers, the code above would need to be changed 
>>> back 
>>> >      to 
>>> >      >      >      getField "publisher" ==> setField "publisher" 
>>> >      >      >    I see that multiple publishers (and locations) are 
>>> being 
>>> >      >      discussed at 
>>> >      >      >    various Zotero/CSL/citeproc-js forums, but don't get 
>>> a 
>>> >      clear 
>>> >      >      picture 
>>> >      >      >    yet. 
>>> >      >      The literal {and} will not cause breaking of author lists 
>>> or 
>>> >      literal 
>>> >      >      lists. 
>>> >      >      I'll leave the multiple publishers question open for now. 
>>> >      Either 
>>> >      >      bibtex2pandoc or pandoc-citeproc should be changed, not 
>>> sure 
>>> >      which. 
>>> >      >      >    Names 
>>> >      >      >    ===== 
>>> >      >      >    Can bibtex2pandoc be expected to distinguish all five 
>>> >      name 
>>> >      >      components 
>>> >      >      >    (family, given, suffix, non-dropping-particle, 
>>> >      >      dropping-particle)? 
>>> >      >      >    biblatex provides one useful additional bit of info 
>>> for 
>>> >      name 
>>> >      >      parsing 
>>> >      >      >    via the "useprefix" switch inside the "options" 
>>> field: 
>>> >      >      >    From the biblatex manual (v 2.7a), "3.1.3 Entry 
>>> Options" 
>>> >      >      >    > Whether the name prefix (von, van, of, da, de, 
>>> della, 
>>> >      etc.) 
>>> >      >      is 
>>> >      >      >    considered when printing the last name in citations. 
>>> >      This also 
>>> >      >      affects 
>>> >      >      >    the sorting and formatting of the bibliography as 
>>> well 
>>> >      as the 
>>> >      >      >    generation of certain types of labels. If this option 
>>> is 
>>> >      >      enabled, 
>>> >      >      >    biblatex always precedes the last name with the 
>>> prefix. 
>>> >      For 
>>> >      >      example, 
>>> >      >      >    Ludwig van Beethoven would be cited as Beethoven and 
>>> >      >      alphabetized as 
>>> >      >      >    Beethoven, Ludwig van by default. If this option is 
>>> >      enabled, he 
>>> >      >      is 
>>> >      >      >    cited as van Beethoven and alphabetized as Van 
>>> >      Beethoven, 
>>> >      >      Ludwig 
>>> >      >      >    instead. With Biber, this option is also settable on 
>>> a 
>>> >      per-type 
>>> >      >      basis. 
>>> >      >      I used a rough-and-ready algorithm, probably not the same 
>>> one 
>>> >      biber 
>>> >      >      or biblatex uses.  Basically, the only time my code gives 
>>> you 
>>> >      a 
>>> >      >      non-dropping particle is when you have a name in format 
>>> >      >         B A, D E 
>>> >      >      then B is the particle and A the given name. 
>>> >      >      I don't have any support now for dropping-particle or 
>>> suffix. 
>>> >      >      This needs to be improved, but I'd need to know better 
>>> what 
>>> >      the 
>>> >      >      algorithm is. 
>>> >      >      >    Inline formatting 
>>> >      >      >    ================= 
>>> >      >      >    It probably doesn't play a role for pandoc, but in 
>>> the 
>>> >      interest 
>>> >      >      of 
>>> >      >      >    portability, shouldn't inline formatting in a 
>>> CSL-YAML 
>>> >      file or 
>>> >      >      metadata 
>>> >      >      >    section rather use 
>>> >      >      >        <i> and </i> for italics 
>>> >      >      >        <b> and </b> for bold 
>>> >      >      >        <sub> and </sub> for subscript 
>>> >      >      >        <sup> and </sup> for superscript 
>>> >      >      >        <span style="font-variant:small-caps;"> and 
>>> </span> 
>>> >      for 
>>> >      >      smallcaps 
>>> >      >      >    instead of markdown formatting? 
>>> >      >      >    (see [1][1]https://github.com/jgm/pandoc/issues/931, 
>>> >      >      >    [2][2]https://www.zotero.org/support/kb/rich_text_ 
>>> >      bibliography) 
>>> >      >      No, because the YAML file is standard pandoc metadata, 
>>> >      formatted in 
>>> >      >      markdown.  When it is read by pandoc-citeproc, this will 
>>> be 
>>> >      >      converted 
>>> >      >      where possible to CSL metadata. 
>>> >      >      >    Small Caps 
>>> >      >      >    ========== 
>>> >      >      >    Speaking of small caps: Any plans to add some 
>>> >      markdownish 
>>> >      >      formatting 
>>> >      >      >    commands for small caps to pandoc? 
>>> >      >      No idea what a natural syntax would be. 
>>> >      > 
>>> >      >    -- 
>>> >      >    You received this message because you are subscribed to the 
>>> >      Google 
>>> >      >    Groups "pandoc-discuss" group. 
>>> >      >    To unsubscribe from this group and stop receiving emails 
>>> from 
>>> >      it, send 
>>> >      >    an email to [3]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>>> >      >    To post to this group, send email to 
>>> >      [4]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>>> >      >    To view this discussion on the web visit 
>>> >      >    [3][5]https://groups.google.com/d/msgid/pandoc-discuss/ 
>>> >      a86067ec-6b4a-437b- 
>>> >      >    b2f0-3ee666209dae%[6]40googlegroups.com. 
>>> >      >    For more options, visit [4][7]https://groups.google.com/ 
>>> >      groups/opt_out. 
>>> >      > 
>>> >      > References 
>>> >      > 
>>> >      >    1. [8]https://github.com/jgm/pandoc/issues/931 
>>> >      >    2. [9]
>>> https://www.zotero.org/support/kb/rich_text_bibliography 
>>> >      >    3. [10]https://groups.google.com/d/ 
>>> >      msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae% 
>>> >      40googlegroups.com 
>>> >      >    4. [11]https://groups.google.com/groups/opt_out 
>>> > 
>>> >    -- 
>>> >    You received this message because you are subscribed to the Google 
>>> >    Groups "pandoc-discuss" group. 
>>> >    To unsubscribe from this group and stop receiving emails from it, 
>>> send 
>>> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>>> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>>> >    To view this discussion on the web visit 
>>> >    [12]
>>> https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9 
>>> >    -8efe-0aa033c670da%40googlegroups.com. 
>>> >    For more options, visit [13]
>>> https://groups.google.com/groups/opt_out. 
>>> > 
>>> > References 
>>> > 
>>> >    1. https://github.com/jgm/pandoc/issues/931 
>>> >    2. https://www.zotero.org/support/kb/rich_text_bibliography 
>>> >    3. javascript:/ 
>>> >    4. javascript:/ 
>>> >    5. 
>>> https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b- 
>>> >    6. http://40googlegroups.com/ 
>>> >    7. https://groups.google.com/groups/opt_out 
>>> >    8. https://github.com/jgm/pandoc/issues/931 
>>> >    9. https://www.zotero.org/support/kb/rich_text_bibliography 
>>> >   10. 
>>> https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%40googlegroups.com 
>>> >   11. https://groups.google.com/groups/opt_out 
>>> >   12. 
>>> https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9-8efe-0aa033c670da%40googlegroups.com 
>>> >   13. https://groups.google.com/groups/opt_out 
>>>
>>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/ebae1a26-c6cb-44c4-add2-5dc57f620a4c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 29539 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

But this only covers English locales.  We also want to map
Spanish to "es-SP" or whatever it is, and so on....

Is there a list of the valid values for the biblatex
"hyphenation" field?

I also think it makes more sense to define a function

toLocale :: String -> String
toLocale "english" = "en"
toLocale "australian" = "en-AU"
...etc...

and then

    opt $ getRawField "hyphenation" >>= setRawField "language" . toLocale

The . is function composition, so this says:  apply to toLocale
function to the value before passing it to setRawField "language".

+++ Nick Bart [Sep 28 13 16:48 ]:
>    OK, this should do the trick:
>      opt $ do
>        val <- getRawField "hyphenation"
>        case val of
>          "english"    -> setRawField "language" "en"
>          "american"   -> setRawField "language" "en-US"
>          "british"    -> setRawField "language" "en-UK"
>          "canadian"   -> setRawField "language" "en-CA"
>          "australian" -> setRawField "language" "en-AU"
>          "newzealand" -> setRawField "language" "en-NZ"
>          "USenglish"  -> setRawField "language" "en-US"
>          "UKenglish"  -> setRawField "language" "en-UK"
>          _            -> setRawField "language" val
>    And remove
>    opt $ getRawField "hyphenation" >>= setRawField "language"
>    On Saturday, September 28, 2013 11:30:24 PM UTC, Nick Bart wrote:
> 
>    On second thoughts, however: Works for me, since I have only "english"
>    in my hyphenation fields.
>    However, CSL's title cases routine is simply triggered if the CSL
>    language field starts with "en", but not by, e.g., "american"; so we'll
>    probably have to map hyphenation = american, british, canadian,
>    australian, newzealand, USenglish or UKenglish either all to "english",
>    or individually to "en-UK", "en-US", etc.
>    On Saturday, September 28, 2013 11:16:42 PM UTC, Nick Bart wrote:
> 
>    I already added this earlier:
>      opt $ getRawField "hyphenation" >>= setRawField "language"
>    Seems to work nicely ...
>    On Saturday, September 28, 2013 10:59:47 PM UTC, fiddlosopher wrote:
> 
>      OK, I've added this feature.
>      In order to set the "language" variable properly, we need a mapping
>      from biblatex hyphenation field values to whatever fills the CSL
>      language slot.
>      Can you provide one?
>      +++ Nick Bart [Sep 28 13 15:19 ]:
>      >    biblatex database entries may contain a "hyphenation" field,
>      specifying
>      >    the language of the bibliographic record, used for determining
>      >    hyphenation patterns and whether title/sentence case conversion
>      is to
>      >    be applied. It must contain exactly one language term. The
>      functional
>      >    equivalent in CSL is the "language" variable, also used for
>      determining
>      >    whether title/sentence case conversion is to be applied.
>      >    biblatex database entries may also contain a "language" field,
>      which
>      >    describes the language or languages of the *content* of the
>      work, e.g.
>      >    language = {latin and english}. This has no counterpart in CSL.
>      >    unTitlecase should only be applied to titles in languages that
>      do
>      >    distinguish title from sentence case, and that's as far as I
>      know, and
>      >    in line with biblatex's default only English. In other words,
>      >    unTitlecase should only be applied if an entry's hyphenation
>      field
>      >    contains one of the terms american, british, canadian, english,
>      >    australian, newzealand, USenglish or UKenglish.
>      >    On Saturday, September 28, 2013 9:34:33 PM UTC, fiddlosopher
>      wrote:
>      >
>      >      How is the language represented in a biblatex file?  Or is it
>      an
>      >      external parameter that comes from the locale or something
>      else?
>      >      +++ Nick Bart [Sep 28 13 13:37 ]:
>      >      >    unTitlecase: Brilliant.
>      >      >    Forgot one important detail, though: unTitlecase should
>      only be
>      >      applied
>      >      >    if the language is English, i.e., if an entry's biblatex
>      >      hyphenation
>      >      >    field is one of american, british, canadian, english,
>      >      australian,
>      >      >    newzealand, USenglish or UKenglish.
>      >      >    From the biblatex manual (v 2.7a), "4.6.4 Miscellaneous
>      >      Commands", item
>      >      >    " \MakeSentenceCase":
>      >      >    > By default, converting to sentence case is enabled for
>      the
>      >      following
>      >      >    language identifiers: american, british, canadian,
>      english,
>      >      australian,
>      >      >    newzealand as well as the aliases USenglish and
>      UKenglish.
>      >      >    On Saturday, September 28, 2013 7:34:09 PM UTC,
>      fiddlosopher
>      >      wrote:
>      >      >
>      >      >      +++ Nick Bart [Sep 28 13 03:52 ]:
>      >      >      >    A few additional ideas/issues:
>      >      >      >    Title vs. sentence case
>      >      >      >    =======================
>      >      >      >    bibtex and biblatex expect titles in title case
>      in the
>      >      >      database,
>      >      >      >    converting these to sentence case if required by
>      a
>      >      particular
>      >      >      style,
>      >      >      >    except for strings protected by {}.
>      >      >      >    CSL does the opposite: It expects titles in
>      sentence
>      >      case,
>      >      >      converting
>      >      >      >    these to title case if required, except for
>      certain stop
>      >      words.
>      >      >      >    I think it make senses for bibtex2pandoc to
>      convert
>      >      >      bibtex=biblatex
>      >      >      >    titles to sentence case for further consumption
>      by CSL;
>      >      i.e.,
>      >      >      to lower
>      >      >      >    case, except strings protected by {}.
>      >      >      >        - ex.:
>      >      >      >        title = {{JFK}: The {CIA}, {Vietnam}, and the
>      Plot
>      >      to
>      >      >      Assassinate
>      >      >      >    {John F. Kennedy}}
>      >      >      >        becomes
>      >      >      >        title: 'JFK: The CIA, Vietnam, and the plot
>      to
>      >      assassinate
>      >      >      John F.
>      >      >      >    Kennedy'
>      >      >      >        - to be discussed: should strings inside
>      commands be
>      >      >      converted,
>      >      >      >    too?
>      >      >      >         i.e., should
>      >      >      >         title = {An Analysis of \textit{For Whom the
>      Bell
>      >      Tolls}}
>      >      >      >         become
>      >      >      >         title: 'An Analysis of <i>for whom the bell
>      >      tolls</i>'
>      >      >      >         and only extra protection
>      >      >      >         title = {An Analysis of {\textit{For Whom
>      the Bell
>      >      >      Tolls}}}
>      >      >      >         would yield
>      >      >      >         title: 'An Analysis of <i>For Whom the Bell
>      >      Tolls</i>'
>      >      >      I've implemented the unTitlecase transformation.
>      (This
>      >      depends on a
>      >      >      recent patch to pandoc that causes the LaTeX reader to
>      insert
>      >      Span
>      >      >      when we have a bare group {Like This}.)
>      >      >      I don't know about the "to be discussed" -- currently
>      strings
>      >      inside
>      >      >      emphasis, etc., will retain their case.  I think
>      that's
>      >      probably
>      >      >      right.
>      >      >      >    Corporate authors
>      >      >      >    =================
>      >      >      >    Example: author = {{National Aeronautics and
>      Space
>      >      >      Administration}},
>      >      >      >    Current bibtex2pandoc output:
>      >      >      >    - author:
>      >      >      >      - family: Aeronautics
>      >      >      >        given:
>      >      >      >        - National
>      >      >      >      - family: Administration
>      >      >      >        given:
>      >      >      >        - Space
>      >      >      >    Expected:
>      >      >      >    - author:
>      >      >      >      - literal: 'National Aeronautics and Space
>      >      Administration'
>      >      >      I've added support for this.
>      >      >      >    Literal and in institution, organization,
>      publisher,
>      >      location,
>      >      >      etc.
>      >      >      >    =============================
>      >      ==============================
>      >      >      ==========
>      >      >      >    If code includes
>      >      >      >
>      >      >      >      getLiteralList "publisher" ==> setList
>      "publisher"
>      >      >      >    publisher = {Holt, Rinehart {and} Winston}
>      >      >      >    Current bibtex2pandoc output:
>      >      >      >      publisher:
>      >      >      >      - 'Holt, Rinehart'
>      >      >      >      - Winston
>      >      >      >    Expected:
>      >      >      >      publisher:
>      >      >      >      - 'Holt, Rinehart and Winston'
>      >      >      >    Also noticed, however, that pandoc-citeproc does
>      not
>      >      seem to
>      >      >      like it
>      >      >      >    when actual multiple publishers occur:
>      >      >      >      publisher:
>      >      >      >      - 'Univ. of Toronto Press'
>      >      >      >      - Routledge
>      >      >      >    throws an error:
>      >      >      >    pandoc-citeproc: Error parsing references: when
>      >      expecting a
>      >      >      String,
>      >      >      >    encountered Array instead
>      >      >      >    This needs to be checked, but if CSL does not
>      allow
>      >      multiple
>      >      >      >    publishers, the code above would need to be
>      changed back
>      >      to
>      >      >      >      getField "publisher" ==> setField "publisher"
>      >      >      >    I see that multiple publishers (and locations)
>      are being
>      >      >      discussed at
>      >      >      >    various Zotero/CSL/citeproc-js forums, but don't
>      get a
>      >      clear
>      >      >      picture
>      >      >      >    yet.
>      >      >      The literal {and} will not cause breaking of author
>      lists or
>      >      literal
>      >      >      lists.
>      >      >      I'll leave the multiple publishers question open for
>      now.
>      >      Either
>      >      >      bibtex2pandoc or pandoc-citeproc should be changed,
>      not sure
>      >      which.
>      >      >      >    Names
>      >      >      >    =====
>      >      >      >    Can bibtex2pandoc be expected to distinguish all
>      five
>      >      name
>      >      >      components
>      >      >      >    (family, given, suffix, non-dropping-particle,
>      >      >      dropping-particle)?
>      >      >      >    biblatex provides one useful additional bit of
>      info for
>      >      name
>      >      >      parsing
>      >      >      >    via the "useprefix" switch inside the "options"
>      field:
>      >      >      >    From the biblatex manual (v 2.7a), "3.1.3 Entry
>      Options"
>      >      >      >    > Whether the name prefix (von, van, of, da, de,
>      della,
>      >      etc.)
>      >      >      is
>      >      >      >    considered when printing the last name in
>      citations.
>      >      This also
>      >      >      affects
>      >      >      >    the sorting and formatting of the bibliography as
>      well
>      >      as the
>      >      >      >    generation of certain types of labels. If this
>      option is
>      >      >      enabled,
>      >      >      >    biblatex always precedes the last name with the
>      prefix.
>      >      For
>      >      >      example,
>      >      >      >    Ludwig van Beethoven would be cited as Beethoven
>      and
>      >      >      alphabetized as
>      >      >      >    Beethoven, Ludwig van by default. If this option
>      is
>      >      enabled, he
>      >      >      is
>      >      >      >    cited as van Beethoven and alphabetized as Van
>      >      Beethoven,
>      >      >      Ludwig
>      >      >      >    instead. With Biber, this option is also settable
>      on a
>      >      per-type
>      >      >      basis.
>      >      >      I used a rough-and-ready algorithm, probably not the
>      same one
>      >      biber
>      >      >      or biblatex uses.  Basically, the only time my code
>      gives you
>      >      a
>      >      >      non-dropping particle is when you have a name in
>      format
>      >      >         B A, D E
>      >      >      then B is the particle and A the given name.
>      >      >      I don't have any support now for dropping-particle or
>      suffix.
>      >      >      This needs to be improved, but I'd need to know better
>      what
>      >      the
>      >      >      algorithm is.
>      >      >      >    Inline formatting
>      >      >      >    =================
>      >      >      >    It probably doesn't play a role for pandoc, but
>      in the
>      >      interest
>      >      >      of
>      >      >      >    portability, shouldn't inline formatting in a
>      CSL-YAML
>      >      file or
>      >      >      metadata
>      >      >      >    section rather use
>      >      >      >        <i> and </i> for italics
>      >      >      >        <b> and </b> for bold
>      >      >      >        <sub> and </sub> for subscript
>      >      >      >        <sup> and </sup> for superscript
>      >      >      >        <span style="font-variant:small-caps;"> and
>      </span>
>      >      for
>      >      >      smallcaps
>      >      >      >    instead of markdown formatting?
>      >      >      >    (see [1][1][1]https://github.com/jgm/
>      pandoc/issues/931,
>      >      >      >    [2][2][2]https://www.zotero.org/
>      support/kb/rich_text_
>      >      bibliography)
>      >      >      No, because the YAML file is standard pandoc metadata,
>      >      formatted in
>      >      >      markdown.  When it is read by pandoc-citeproc, this
>      will be
>      >      >      converted
>      >      >      where possible to CSL metadata.
>      >      >      >    Small Caps
>      >      >      >    ==========
>      >      >      >    Speaking of small caps: Any plans to add some
>      >      markdownish
>      >      >      formatting
>      >      >      >    commands for small caps to pandoc?
>      >      >      No idea what a natural syntax would be.
>      >      >
>      >      >    --
>      >      >    You received this message because you are subscribed to
>      the
>      >      Google
>      >      >    Groups "pandoc-discuss" group.
>      >      >    To unsubscribe from this group and stop receiving emails
>      from
>      >      it, send
>      >      >    an email to [3]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >    To post to this group, send email to
>      >      [4]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >    To view this discussion on the web visit
>      >      >    [3][5][3]https://groups.google.
>      com/d/msgid/pandoc-discuss/
>      >      a86067ec-6b4a-437b-
>      >      >    b2f0-3ee666209dae%[6][4]40googlegroups.com.
>      >      >    For more options, visit [4][7][5]https://groups.google.
>      com/
>      >      groups/opt_out.
>      >      >
>      >      > References
>      >      >
>      >      >    1. [8][6]https://github.com/jgm/pandoc/issues/931
>      >      >    2. [9][7]https://www.zotero.org/support/kb/rich_text_
>      bibliography
>      >      >    3. [10][8]https://groups.google.com/d/
>      >      msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%
>      >      [9]40googlegroups.com
>      >      >    4. [11][10]https://groups.google.com/groups/opt_out
>      >
>      >    --
>      >    You received this message because you are subscribed to the
>      Google
>      >    Groups "pandoc-discuss" group.
>      >    To unsubscribe from this group and stop receiving emails from
>      it, send
>      >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To post to this group, send email to
>      pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To view this discussion on the web visit
>      >    [12][11]https://groups.google.com/d/msgid/pandoc-discuss/
>      605aad90-4967-42b9
>      >    -8efe-0aa033c670da%[12]40googlegroups.com.
>      >    For more options, visit [13][13]https://groups.google.com/
>      groups/opt_out.
>      >
>      > References
>      >
>      >    1. [14]https://github.com/jgm/pandoc/issues/931
>      >    2. [15]https://www.zotero.org/support/kb/rich_text_bibliography
>      >    3. javascript:/
>      >    4. javascript:/
>      >    5. [16]https://groups.google.com/d/
>      msgid/pandoc-discuss/a86067ec-6b4a-437b-
>      >    6. [17]http://40googlegroups.com/
>      >    7. [18]https://groups.google.com/groups/opt_out
>      >    8. [19]https://github.com/jgm/pandoc/issues/931
>      >    9. [20]https://www.zotero.org/support/kb/rich_text_bibliography
>      >   10. [21]https://groups.google.com/d/
>      msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%
>      40googlegroups.com
>      >   11. [22]https://groups.google.com/groups/opt_out
>      >   12. [23]https://groups.google.com/d/
>      msgid/pandoc-discuss/605aad90-4967-42b9-8efe-0aa033c670da%
>      40googlegroups.com
>      >   13. [24]https://groups.google.com/groups/opt_out
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [25]https://groups.google.com/d/msgid/pandoc-discuss/ebae1a26-c6cb-44c4
>    -add2-5dc57f620a4c%40googlegroups.com.
>    For more options, visit [26]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://github.com/jgm/pandoc/issues/931
>    2. https://www.zotero.org/support/kb/rich_text_
>    3. https://groups.google.com/d/msgid/pandoc-discuss/
>    4. http://40googlegroups.com/
>    5. https://groups.google.com/
>    6. https://github.com/jgm/pandoc/issues/931
>    7. https://www.zotero.org/support/kb/rich_text_bibliography
>    8. https://groups.google.com/d/
>    9. http://40googlegroups.com/
>   10. https://groups.google.com/groups/opt_out
>   11. https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9
>   12. http://40googlegroups.com/
>   13. https://groups.google.com/groups/opt_out
>   14. https://github.com/jgm/pandoc/issues/931
>   15. https://www.zotero.org/support/kb/rich_text_bibliography
>   16. https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-
>   17. http://40googlegroups.com/
>   18. https://groups.google.com/groups/opt_out
>   19. https://github.com/jgm/pandoc/issues/931
>   20. https://www.zotero.org/support/kb/rich_text_bibliography
>   21. https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%40googlegroups.com
>   22. https://groups.google.com/groups/opt_out
>   23. https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9-8efe-0aa033c670da%40googlegroups.com
>   24. https://groups.google.com/groups/opt_out
>   25. https://groups.google.com/d/msgid/pandoc-discuss/ebae1a26-c6cb-44c4-add2-5dc57f620a4c%40googlegroups.com
>   26. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I guess they are Babel language names.

So here's the list from the Babel manual.  We just need to
correlate these with things like en-US:

Afrikaans afrikaans
Bahasa bahasa
Basque basque
Breton breton
Bulgarian bulgarian
Catalan catalan
Croatian croatian
Czech czech
Danish danish
Dutch dutch
English english, USenglish, american, UKenglish, british,
canadian, australian, newzealand
Esperanto esperanto
Estonian estonian
Finnish finnish
French french, francais, canadien, acadian
Galician galician
German austrian, german, germanb, ngerman, naustrian
Greek greek, polutonikogreek
Hebrew hebrew
Hungarian magyar, hungarian
Icelandic icelandic
Interlingua interlingua
Irish Gaelic irish
Italian italian
Latin latin
Lower Sorbian lowersorbian
North Sami samin
Norwegian norsk, nynorsk
Polish polish
Portuguese portuges, portuguese, brazilian, brazil
Romanian romanian
Russian russian
Scottish Gaelic scottish
Spanish spanish
Slovakian slovak
Slovenian slovene
Swedish swedish
Serbian serbian
Turkish turkish
Ukrainian ukrainian
Upper Sorbian uppersorbian
Welsh welsh

This page (http://demo.icu-project.org/icu-bin/locexp) has
teh information.


+++ John MacFarlane [Sep 28 13 18:25 ]:
> But this only covers English locales.  We also want to map
> Spanish to "es-SP" or whatever it is, and so on....
> 
> Is there a list of the valid values for the biblatex
> "hyphenation" field?
> 
> I also think it makes more sense to define a function
> 
> toLocale :: String -> String
> toLocale "english" = "en"
> toLocale "australian" = "en-AU"
> ...etc...
> 
> and then
> 
>     opt $ getRawField "hyphenation" >>= setRawField "language" . toLocale
> 
> The . is function composition, so this says:  apply to toLocale
> function to the value before passing it to setRawField "language".
> 
> +++ Nick Bart [Sep 28 13 16:48 ]:
> >    OK, this should do the trick:
> >      opt $ do
> >        val <- getRawField "hyphenation"
> >        case val of
> >          "english"    -> setRawField "language" "en"
> >          "american"   -> setRawField "language" "en-US"
> >          "british"    -> setRawField "language" "en-UK"
> >          "canadian"   -> setRawField "language" "en-CA"
> >          "australian" -> setRawField "language" "en-AU"
> >          "newzealand" -> setRawField "language" "en-NZ"
> >          "USenglish"  -> setRawField "language" "en-US"
> >          "UKenglish"  -> setRawField "language" "en-UK"
> >          _            -> setRawField "language" val
> >    And remove
> >    opt $ getRawField "hyphenation" >>= setRawField "language"
> >    On Saturday, September 28, 2013 11:30:24 PM UTC, Nick Bart wrote:
> > 
> >    On second thoughts, however: Works for me, since I have only "english"
> >    in my hyphenation fields.
> >    However, CSL's title cases routine is simply triggered if the CSL
> >    language field starts with "en", but not by, e.g., "american"; so we'll
> >    probably have to map hyphenation = american, british, canadian,
> >    australian, newzealand, USenglish or UKenglish either all to "english",
> >    or individually to "en-UK", "en-US", etc.
> >    On Saturday, September 28, 2013 11:16:42 PM UTC, Nick Bart wrote:
> > 
> >    I already added this earlier:
> >      opt $ getRawField "hyphenation" >>= setRawField "language"
> >    Seems to work nicely ...
> >    On Saturday, September 28, 2013 10:59:47 PM UTC, fiddlosopher wrote:
> > 
> >      OK, I've added this feature.
> >      In order to set the "language" variable properly, we need a mapping
> >      from biblatex hyphenation field values to whatever fills the CSL
> >      language slot.
> >      Can you provide one?
> >      +++ Nick Bart [Sep 28 13 15:19 ]:
> >      >    biblatex database entries may contain a "hyphenation" field,
> >      specifying
> >      >    the language of the bibliographic record, used for determining
> >      >    hyphenation patterns and whether title/sentence case conversion
> >      is to
> >      >    be applied. It must contain exactly one language term. The
> >      functional
> >      >    equivalent in CSL is the "language" variable, also used for
> >      determining
> >      >    whether title/sentence case conversion is to be applied.
> >      >    biblatex database entries may also contain a "language" field,
> >      which
> >      >    describes the language or languages of the *content* of the
> >      work, e.g.
> >      >    language = {latin and english}. This has no counterpart in CSL.
> >      >    unTitlecase should only be applied to titles in languages that
> >      do
> >      >    distinguish title from sentence case, and that's as far as I
> >      know, and
> >      >    in line with biblatex's default only English. In other words,
> >      >    unTitlecase should only be applied if an entry's hyphenation
> >      field
> >      >    contains one of the terms american, british, canadian, english,
> >      >    australian, newzealand, USenglish or UKenglish.
> >      >    On Saturday, September 28, 2013 9:34:33 PM UTC, fiddlosopher
> >      wrote:
> >      >
> >      >      How is the language represented in a biblatex file?  Or is it
> >      an
> >      >      external parameter that comes from the locale or something
> >      else?
> >      >      +++ Nick Bart [Sep 28 13 13:37 ]:
> >      >      >    unTitlecase: Brilliant.
> >      >      >    Forgot one important detail, though: unTitlecase should
> >      only be
> >      >      applied
> >      >      >    if the language is English, i.e., if an entry's biblatex
> >      >      hyphenation
> >      >      >    field is one of american, british, canadian, english,
> >      >      australian,
> >      >      >    newzealand, USenglish or UKenglish.
> >      >      >    From the biblatex manual (v 2.7a), "4.6.4 Miscellaneous
> >      >      Commands", item
> >      >      >    " \MakeSentenceCase":
> >      >      >    > By default, converting to sentence case is enabled for
> >      the
> >      >      following
> >      >      >    language identifiers: american, british, canadian,
> >      english,
> >      >      australian,
> >      >      >    newzealand as well as the aliases USenglish and
> >      UKenglish.
> >      >      >    On Saturday, September 28, 2013 7:34:09 PM UTC,
> >      fiddlosopher
> >      >      wrote:
> >      >      >
> >      >      >      +++ Nick Bart [Sep 28 13 03:52 ]:
> >      >      >      >    A few additional ideas/issues:
> >      >      >      >    Title vs. sentence case
> >      >      >      >    =======================
> >      >      >      >    bibtex and biblatex expect titles in title case
> >      in the
> >      >      >      database,
> >      >      >      >    converting these to sentence case if required by
> >      a
> >      >      particular
> >      >      >      style,
> >      >      >      >    except for strings protected by {}.
> >      >      >      >    CSL does the opposite: It expects titles in
> >      sentence
> >      >      case,
> >      >      >      converting
> >      >      >      >    these to title case if required, except for
> >      certain stop
> >      >      words.
> >      >      >      >    I think it make senses for bibtex2pandoc to
> >      convert
> >      >      >      bibtex=biblatex
> >      >      >      >    titles to sentence case for further consumption
> >      by CSL;
> >      >      i.e.,
> >      >      >      to lower
> >      >      >      >    case, except strings protected by {}.
> >      >      >      >        - ex.:
> >      >      >      >        title = {{JFK}: The {CIA}, {Vietnam}, and the
> >      Plot
> >      >      to
> >      >      >      Assassinate
> >      >      >      >    {John F. Kennedy}}
> >      >      >      >        becomes
> >      >      >      >        title: 'JFK: The CIA, Vietnam, and the plot
> >      to
> >      >      assassinate
> >      >      >      John F.
> >      >      >      >    Kennedy'
> >      >      >      >        - to be discussed: should strings inside
> >      commands be
> >      >      >      converted,
> >      >      >      >    too?
> >      >      >      >         i.e., should
> >      >      >      >         title = {An Analysis of \textit{For Whom the
> >      Bell
> >      >      Tolls}}
> >      >      >      >         become
> >      >      >      >         title: 'An Analysis of <i>for whom the bell
> >      >      tolls</i>'
> >      >      >      >         and only extra protection
> >      >      >      >         title = {An Analysis of {\textit{For Whom
> >      the Bell
> >      >      >      Tolls}}}
> >      >      >      >         would yield
> >      >      >      >         title: 'An Analysis of <i>For Whom the Bell
> >      >      Tolls</i>'
> >      >      >      I've implemented the unTitlecase transformation.
> >      (This
> >      >      depends on a
> >      >      >      recent patch to pandoc that causes the LaTeX reader to
> >      insert
> >      >      Span
> >      >      >      when we have a bare group {Like This}.)
> >      >      >      I don't know about the "to be discussed" -- currently
> >      strings
> >      >      inside
> >      >      >      emphasis, etc., will retain their case.  I think
> >      that's
> >      >      probably
> >      >      >      right.
> >      >      >      >    Corporate authors
> >      >      >      >    =================
> >      >      >      >    Example: author = {{National Aeronautics and
> >      Space
> >      >      >      Administration}},
> >      >      >      >    Current bibtex2pandoc output:
> >      >      >      >    - author:
> >      >      >      >      - family: Aeronautics
> >      >      >      >        given:
> >      >      >      >        - National
> >      >      >      >      - family: Administration
> >      >      >      >        given:
> >      >      >      >        - Space
> >      >      >      >    Expected:
> >      >      >      >    - author:
> >      >      >      >      - literal: 'National Aeronautics and Space
> >      >      Administration'
> >      >      >      I've added support for this.
> >      >      >      >    Literal and in institution, organization,
> >      publisher,
> >      >      location,
> >      >      >      etc.
> >      >      >      >    =============================
> >      >      ==============================
> >      >      >      ==========
> >      >      >      >    If code includes
> >      >      >      >
> >      >      >      >      getLiteralList "publisher" ==> setList
> >      "publisher"
> >      >      >      >    publisher = {Holt, Rinehart {and} Winston}
> >      >      >      >    Current bibtex2pandoc output:
> >      >      >      >      publisher:
> >      >      >      >      - 'Holt, Rinehart'
> >      >      >      >      - Winston
> >      >      >      >    Expected:
> >      >      >      >      publisher:
> >      >      >      >      - 'Holt, Rinehart and Winston'
> >      >      >      >    Also noticed, however, that pandoc-citeproc does
> >      not
> >      >      seem to
> >      >      >      like it
> >      >      >      >    when actual multiple publishers occur:
> >      >      >      >      publisher:
> >      >      >      >      - 'Univ. of Toronto Press'
> >      >      >      >      - Routledge
> >      >      >      >    throws an error:
> >      >      >      >    pandoc-citeproc: Error parsing references: when
> >      >      expecting a
> >      >      >      String,
> >      >      >      >    encountered Array instead
> >      >      >      >    This needs to be checked, but if CSL does not
> >      allow
> >      >      multiple
> >      >      >      >    publishers, the code above would need to be
> >      changed back
> >      >      to
> >      >      >      >      getField "publisher" ==> setField "publisher"
> >      >      >      >    I see that multiple publishers (and locations)
> >      are being
> >      >      >      discussed at
> >      >      >      >    various Zotero/CSL/citeproc-js forums, but don't
> >      get a
> >      >      clear
> >      >      >      picture
> >      >      >      >    yet.
> >      >      >      The literal {and} will not cause breaking of author
> >      lists or
> >      >      literal
> >      >      >      lists.
> >      >      >      I'll leave the multiple publishers question open for
> >      now.
> >      >      Either
> >      >      >      bibtex2pandoc or pandoc-citeproc should be changed,
> >      not sure
> >      >      which.
> >      >      >      >    Names
> >      >      >      >    =====
> >      >      >      >    Can bibtex2pandoc be expected to distinguish all
> >      five
> >      >      name
> >      >      >      components
> >      >      >      >    (family, given, suffix, non-dropping-particle,
> >      >      >      dropping-particle)?
> >      >      >      >    biblatex provides one useful additional bit of
> >      info for
> >      >      name
> >      >      >      parsing
> >      >      >      >    via the "useprefix" switch inside the "options"
> >      field:
> >      >      >      >    From the biblatex manual (v 2.7a), "3.1.3 Entry
> >      Options"
> >      >      >      >    > Whether the name prefix (von, van, of, da, de,
> >      della,
> >      >      etc.)
> >      >      >      is
> >      >      >      >    considered when printing the last name in
> >      citations.
> >      >      This also
> >      >      >      affects
> >      >      >      >    the sorting and formatting of the bibliography as
> >      well
> >      >      as the
> >      >      >      >    generation of certain types of labels. If this
> >      option is
> >      >      >      enabled,
> >      >      >      >    biblatex always precedes the last name with the
> >      prefix.
> >      >      For
> >      >      >      example,
> >      >      >      >    Ludwig van Beethoven would be cited as Beethoven
> >      and
> >      >      >      alphabetized as
> >      >      >      >    Beethoven, Ludwig van by default. If this option
> >      is
> >      >      enabled, he
> >      >      >      is
> >      >      >      >    cited as van Beethoven and alphabetized as Van
> >      >      Beethoven,
> >      >      >      Ludwig
> >      >      >      >    instead. With Biber, this option is also settable
> >      on a
> >      >      per-type
> >      >      >      basis.
> >      >      >      I used a rough-and-ready algorithm, probably not the
> >      same one
> >      >      biber
> >      >      >      or biblatex uses.  Basically, the only time my code
> >      gives you
> >      >      a
> >      >      >      non-dropping particle is when you have a name in
> >      format
> >      >      >         B A, D E
> >      >      >      then B is the particle and A the given name.
> >      >      >      I don't have any support now for dropping-particle or
> >      suffix.
> >      >      >      This needs to be improved, but I'd need to know better
> >      what
> >      >      the
> >      >      >      algorithm is.
> >      >      >      >    Inline formatting
> >      >      >      >    =================
> >      >      >      >    It probably doesn't play a role for pandoc, but
> >      in the
> >      >      interest
> >      >      >      of
> >      >      >      >    portability, shouldn't inline formatting in a
> >      CSL-YAML
> >      >      file or
> >      >      >      metadata
> >      >      >      >    section rather use
> >      >      >      >        <i> and </i> for italics
> >      >      >      >        <b> and </b> for bold
> >      >      >      >        <sub> and </sub> for subscript
> >      >      >      >        <sup> and </sup> for superscript
> >      >      >      >        <span style="font-variant:small-caps;"> and
> >      </span>
> >      >      for
> >      >      >      smallcaps
> >      >      >      >    instead of markdown formatting?
> >      >      >      >    (see [1][1][1]https://github.com/jgm/
> >      pandoc/issues/931,
> >      >      >      >    [2][2][2]https://www.zotero.org/
> >      support/kb/rich_text_
> >      >      bibliography)
> >      >      >      No, because the YAML file is standard pandoc metadata,
> >      >      formatted in
> >      >      >      markdown.  When it is read by pandoc-citeproc, this
> >      will be
> >      >      >      converted
> >      >      >      where possible to CSL metadata.
> >      >      >      >    Small Caps
> >      >      >      >    ==========
> >      >      >      >    Speaking of small caps: Any plans to add some
> >      >      markdownish
> >      >      >      formatting
> >      >      >      >    commands for small caps to pandoc?
> >      >      >      No idea what a natural syntax would be.
> >      >      >
> >      >      >    --
> >      >      >    You received this message because you are subscribed to
> >      the
> >      >      Google
> >      >      >    Groups "pandoc-discuss" group.
> >      >      >    To unsubscribe from this group and stop receiving emails
> >      from
> >      >      it, send
> >      >      >    an email to [3]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >      >    To post to this group, send email to
> >      >      [4]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >      >    To view this discussion on the web visit
> >      >      >    [3][5][3]https://groups.google.
> >      com/d/msgid/pandoc-discuss/
> >      >      a86067ec-6b4a-437b-
> >      >      >    b2f0-3ee666209dae%[6][4]40googlegroups.com.
> >      >      >    For more options, visit [4][7][5]https://groups.google.
> >      com/
> >      >      groups/opt_out.
> >      >      >
> >      >      > References
> >      >      >
> >      >      >    1. [8][6]https://github.com/jgm/pandoc/issues/931
> >      >      >    2. [9][7]https://www.zotero.org/support/kb/rich_text_
> >      bibliography
> >      >      >    3. [10][8]https://groups.google.com/d/
> >      >      msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%
> >      >      [9]40googlegroups.com
> >      >      >    4. [11][10]https://groups.google.com/groups/opt_out
> >      >
> >      >    --
> >      >    You received this message because you are subscribed to the
> >      Google
> >      >    Groups "pandoc-discuss" group.
> >      >    To unsubscribe from this group and stop receiving emails from
> >      it, send
> >      >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >    To post to this group, send email to
> >      pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >    To view this discussion on the web visit
> >      >    [12][11]https://groups.google.com/d/msgid/pandoc-discuss/
> >      605aad90-4967-42b9
> >      >    -8efe-0aa033c670da%[12]40googlegroups.com.
> >      >    For more options, visit [13][13]https://groups.google.com/
> >      groups/opt_out.
> >      >
> >      > References
> >      >
> >      >    1. [14]https://github.com/jgm/pandoc/issues/931
> >      >    2. [15]https://www.zotero.org/support/kb/rich_text_bibliography
> >      >    3. javascript:/
> >      >    4. javascript:/
> >      >    5. [16]https://groups.google.com/d/
> >      msgid/pandoc-discuss/a86067ec-6b4a-437b-
> >      >    6. [17]http://40googlegroups.com/
> >      >    7. [18]https://groups.google.com/groups/opt_out
> >      >    8. [19]https://github.com/jgm/pandoc/issues/931
> >      >    9. [20]https://www.zotero.org/support/kb/rich_text_bibliography
> >      >   10. [21]https://groups.google.com/d/
> >      msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%
> >      40googlegroups.com
> >      >   11. [22]https://groups.google.com/groups/opt_out
> >      >   12. [23]https://groups.google.com/d/
> >      msgid/pandoc-discuss/605aad90-4967-42b9-8efe-0aa033c670da%
> >      40googlegroups.com
> >      >   13. [24]https://groups.google.com/groups/opt_out
> > 
> >    --
> >    You received this message because you are subscribed to the Google
> >    Groups "pandoc-discuss" group.
> >    To unsubscribe from this group and stop receiving emails from it, send
> >    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >    To view this discussion on the web visit
> >    [25]https://groups.google.com/d/msgid/pandoc-discuss/ebae1a26-c6cb-44c4
> >    -add2-5dc57f620a4c%40googlegroups.com.
> >    For more options, visit [26]https://groups.google.com/groups/opt_out.
> > 
> > References
> > 
> >    1. https://github.com/jgm/pandoc/issues/931
> >    2. https://www.zotero.org/support/kb/rich_text_
> >    3. https://groups.google.com/d/msgid/pandoc-discuss/
> >    4. http://40googlegroups.com/
> >    5. https://groups.google.com/
> >    6. https://github.com/jgm/pandoc/issues/931
> >    7. https://www.zotero.org/support/kb/rich_text_bibliography
> >    8. https://groups.google.com/d/
> >    9. http://40googlegroups.com/
> >   10. https://groups.google.com/groups/opt_out
> >   11. https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9
> >   12. http://40googlegroups.com/
> >   13. https://groups.google.com/groups/opt_out
> >   14. https://github.com/jgm/pandoc/issues/931
> >   15. https://www.zotero.org/support/kb/rich_text_bibliography
> >   16. https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-
> >   17. http://40googlegroups.com/
> >   18. https://groups.google.com/groups/opt_out
> >   19. https://github.com/jgm/pandoc/issues/931
> >   20. https://www.zotero.org/support/kb/rich_text_bibliography
> >   21. https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%40googlegroups.com
> >   22. https://groups.google.com/groups/opt_out
> >   23. https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9-8efe-0aa033c670da%40googlegroups.com
> >   24. https://groups.google.com/groups/opt_out
> >   25. https://groups.google.com/d/msgid/pandoc-discuss/ebae1a26-c6cb-44c4-add2-5dc57f620a4c%40googlegroups.com
> >   26. https://groups.google.com/groups/opt_out
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20130929012541.GC45785%40Johns-MacBook-Pro.local.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 25814 bytes --]

John, you are prefectly right, biblatex uses Babel language terms. In a 
comprehensive mapping from biblatex to CSL they should surely be 
implemented, and I can look into this. Still, to get title/sentence case 
right, it just a question of English vs. other languages, and this seems to 
work well with the small set of languages included so far.

Are you aware of anything CSL/citeproc would actually be doing with 
entry-specific language information (except for dealing with title/sentence 
case)?

On a related note, how is per-document localization done in 
pandoc(-citeproc)? 

On Sunday, 29 September 2013 01:41:30 UTC, fiddlosopher wrote:
>
> I guess they are Babel language names. 
>
> So here's the list from the Babel manual.  We just need to 
> correlate these with things like en-US: 
>
> Afrikaans afrikaans 
> Bahasa bahasa 
> Basque basque 
> Breton breton 
> Bulgarian bulgarian 
> Catalan catalan 
> Croatian croatian 
> Czech czech 
> Danish danish 
> Dutch dutch 
> English english, USenglish, american, UKenglish, british, 
> canadian, australian, newzealand 
> Esperanto esperanto 
> Estonian estonian 
> Finnish finnish 
> French french, francais, canadien, acadian 
> Galician galician 
> German austrian, german, germanb, ngerman, naustrian 
> Greek greek, polutonikogreek 
> Hebrew hebrew 
> Hungarian magyar, hungarian 
> Icelandic icelandic 
> Interlingua interlingua 
> Irish Gaelic irish 
> Italian italian 
> Latin latin 
> Lower Sorbian lowersorbian 
> North Sami samin 
> Norwegian norsk, nynorsk 
> Polish polish 
> Portuguese portuges, portuguese, brazilian, brazil 
> Romanian romanian 
> Russian russian 
> Scottish Gaelic scottish 
> Spanish spanish 
> Slovakian slovak 
> Slovenian slovene 
> Swedish swedish 
> Serbian serbian 
> Turkish turkish 
> Ukrainian ukrainian 
> Upper Sorbian uppersorbian 
> Welsh welsh 
>
> This page (http://demo.icu-project.org/icu-bin/locexp) has 
> teh information. 
>
>
> +++ John MacFarlane [Sep 28 13 18:25 ]: 
> > But this only covers English locales.  We also want to map 
> > Spanish to "es-SP" or whatever it is, and so on.... 
> > 
> > Is there a list of the valid values for the biblatex 
> > "hyphenation" field? 
> > 
> > I also think it makes more sense to define a function 
> > 
> > toLocale :: String -> String 
> > toLocale "english" = "en" 
> > toLocale "australian" = "en-AU" 
> > ...etc... 
> > 
> > and then 
> > 
> >     opt $ getRawField "hyphenation" >>= setRawField "language" . 
> toLocale 
> > 
> > The . is function composition, so this says:  apply to toLocale 
> > function to the value before passing it to setRawField "language". 
> > 
> > +++ Nick Bart [Sep 28 13 16:48 ]: 
> > >    OK, this should do the trick: 
> > >      opt $ do 
> > >        val <- getRawField "hyphenation" 
> > >        case val of 
> > >          "english"    -> setRawField "language" "en" 
> > >          "american"   -> setRawField "language" "en-US" 
> > >          "british"    -> setRawField "language" "en-UK" 
> > >          "canadian"   -> setRawField "language" "en-CA" 
> > >          "australian" -> setRawField "language" "en-AU" 
> > >          "newzealand" -> setRawField "language" "en-NZ" 
> > >          "USenglish"  -> setRawField "language" "en-US" 
> > >          "UKenglish"  -> setRawField "language" "en-UK" 
> > >          _            -> setRawField "language" val 
> > >    And remove 
> > >    opt $ getRawField "hyphenation" >>= setRawField "language" 
> > >    On Saturday, September 28, 2013 11:30:24 PM UTC, Nick Bart wrote: 
> > > 
> > >    On second thoughts, however: Works for me, since I have only 
> "english" 
> > >    in my hyphenation fields. 
> > >    However, CSL's title cases routine is simply triggered if the CSL 
> > >    language field starts with "en", but not by, e.g., "american"; so 
> we'll 
> > >    probably have to map hyphenation = american, british, canadian, 
> > >    australian, newzealand, USenglish or UKenglish either all to 
> "english", 
> > >    or individually to "en-UK", "en-US", etc. 
> > >    On Saturday, September 28, 2013 11:16:42 PM UTC, Nick Bart wrote: 
> > > 
> > >    I already added this earlier: 
> > >      opt $ getRawField "hyphenation" >>= setRawField "language" 
> > >    Seems to work nicely ... 
> > >    On Saturday, September 28, 2013 10:59:47 PM UTC, fiddlosopher 
> wrote: 
> > > 
> > >      OK, I've added this feature. 
> > >      In order to set the "language" variable properly, we need a 
> mapping 
> > >      from biblatex hyphenation field values to whatever fills the CSL 
> > >      language slot. 
> > >      Can you provide one? 
> > >      +++ Nick Bart [Sep 28 13 15:19 ]: 
> > >      >    biblatex database entries may contain a "hyphenation" field, 
> > >      specifying 
> > >      >    the language of the bibliographic record, used for 
> determining 
> > >      >    hyphenation patterns and whether title/sentence case 
> conversion 
> > >      is to 
> > >      >    be applied. It must contain exactly one language term. The 
> > >      functional 
> > >      >    equivalent in CSL is the "language" variable, also used for 
> > >      determining 
> > >      >    whether title/sentence case conversion is to be applied. 
> > >      >    biblatex database entries may also contain a "language" 
> field, 
> > >      which 
> > >      >    describes the language or languages of the *content* of the 
> > >      work, e.g. 
> > >      >    language = {latin and english}. This has no counterpart in 
> CSL. 
> > >      >    unTitlecase should only be applied to titles in languages 
> that 
> > >      do 
> > >      >    distinguish title from sentence case, and that's as far as I 
> > >      know, and 
> > >      >    in line with biblatex's default only English. In other 
> words, 
> > >      >    unTitlecase should only be applied if an entry's hyphenation 
> > >      field 
> > >      >    contains one of the terms american, british, canadian, 
> english, 
> > >      >    australian, newzealand, USenglish or UKenglish. 
> > >      >    On Saturday, September 28, 2013 9:34:33 PM UTC, fiddlosopher 
> > >      wrote: 
> > >      > 
> > >      >      How is the language represented in a biblatex file?  Or is 
> it 
> > >      an 
> > >      >      external parameter that comes from the locale or something 
> > >      else? 
> > >      >      +++ Nick Bart [Sep 28 13 13:37 ]: 
> > >      >      >    unTitlecase: Brilliant. 
> > >      >      >    Forgot one important detail, though: unTitlecase 
> should 
> > >      only be 
> > >      >      applied 
> > >      >      >    if the language is English, i.e., if an entry's 
> biblatex 
> > >      >      hyphenation 
> > >      >      >    field is one of american, british, canadian, english, 
> > >      >      australian, 
> > >      >      >    newzealand, USenglish or UKenglish. 
> > >      >      >    From the biblatex manual (v 2.7a), "4.6.4 
> Miscellaneous 
> > >      >      Commands", item 
> > >      >      >    " \MakeSentenceCase": 
> > >      >      >    > By default, converting to sentence case is enabled 
> for 
> > >      the 
> > >      >      following 
> > >      >      >    language identifiers: american, british, canadian, 
> > >      english, 
> > >      >      australian, 
> > >      >      >    newzealand as well as the aliases USenglish and 
> > >      UKenglish. 
> > >      >      >    On Saturday, September 28, 2013 7:34:09 PM UTC, 
> > >      fiddlosopher 
> > >      >      wrote: 
> > >      >      > 
> > >      >      >      +++ Nick Bart [Sep 28 13 03:52 ]: 
> > >      >      >      >    A few additional ideas/issues: 
> > >      >      >      >    Title vs. sentence case 
> > >      >      >      >    ======================= 
> > >      >      >      >    bibtex and biblatex expect titles in title 
> case 
> > >      in the 
> > >      >      >      database, 
> > >      >      >      >    converting these to sentence case if required 
> by 
> > >      a 
> > >      >      particular 
> > >      >      >      style, 
> > >      >      >      >    except for strings protected by {}. 
> > >      >      >      >    CSL does the opposite: It expects titles in 
> > >      sentence 
> > >      >      case, 
> > >      >      >      converting 
> > >      >      >      >    these to title case if required, except for 
> > >      certain stop 
> > >      >      words. 
> > >      >      >      >    I think it make senses for bibtex2pandoc to 
> > >      convert 
> > >      >      >      bibtex=biblatex 
> > >      >      >      >    titles to sentence case for further 
> consumption 
> > >      by CSL; 
> > >      >      i.e., 
> > >      >      >      to lower 
> > >      >      >      >    case, except strings protected by {}. 
> > >      >      >      >        - ex.: 
> > >      >      >      >        title = {{JFK}: The {CIA}, {Vietnam}, and 
> the 
> > >      Plot 
> > >      >      to 
> > >      >      >      Assassinate 
> > >      >      >      >    {John F. Kennedy}} 
> > >      >      >      >        becomes 
> > >      >      >      >        title: 'JFK: The CIA, Vietnam, and the 
> plot 
> > >      to 
> > >      >      assassinate 
> > >      >      >      John F. 
> > >      >      >      >    Kennedy' 
> > >      >      >      >        - to be discussed: should strings inside 
> > >      commands be 
> > >      >      >      converted, 
> > >      >      >      >    too? 
> > >      >      >      >         i.e., should 
> > >      >      >      >         title = {An Analysis of \textit{For Whom 
> the 
> > >      Bell 
> > >      >      Tolls}} 
> > >      >      >      >         become 
> > >      >      >      >         title: 'An Analysis of <i>for whom the 
> bell 
> > >      >      tolls</i>' 
> > >      >      >      >         and only extra protection 
> > >      >      >      >         title = {An Analysis of {\textit{For Whom 
> > >      the Bell 
> > >      >      >      Tolls}}} 
> > >      >      >      >         would yield 
> > >      >      >      >         title: 'An Analysis of <i>For Whom the 
> Bell 
> > >      >      Tolls</i>' 
> > >      >      >      I've implemented the unTitlecase transformation. 
> > >      (This 
> > >      >      depends on a 
> > >      >      >      recent patch to pandoc that causes the LaTeX reader 
> to 
> > >      insert 
> > >      >      Span 
> > >      >      >      when we have a bare group {Like This}.) 
> > >      >      >      I don't know about the "to be discussed" -- 
> currently 
> > >      strings 
> > >      >      inside 
> > >      >      >      emphasis, etc., will retain their case.  I think 
> > >      that's 
> > >      >      probably 
> > >      >      >      right. 
> > >      >      >      >    Corporate authors 
> > >      >      >      >    ================= 
> > >      >      >      >    Example: author = {{National Aeronautics and 
> > >      Space 
> > >      >      >      Administration}}, 
> > >      >      >      >    Current bibtex2pandoc output: 
> > >      >      >      >    - author: 
> > >      >      >      >      - family: Aeronautics 
> > >      >      >      >        given: 
> > >      >      >      >        - National 
> > >      >      >      >      - family: Administration 
> > >      >      >      >        given: 
> > >      >      >      >        - Space 
> > >      >      >      >    Expected: 
> > >      >      >      >    - author: 
> > >      >      >      >      - literal: 'National Aeronautics and Space 
> > >      >      Administration' 
> > >      >      >      I've added support for this. 
> > >      >      >      >    Literal and in institution, organization, 
> > >      publisher, 
> > >      >      location, 
> > >      >      >      etc. 
> > >      >      >      >    ============================= 
> > >      >      ============================== 
> > >      >      >      ========== 
> > >      >      >      >    If code includes 
> > >      >      >      > 
> > >      >      >      >      getLiteralList "publisher" ==> setList 
> > >      "publisher" 
> > >      >      >      >    publisher = {Holt, Rinehart {and} Winston} 
> > >      >      >      >    Current bibtex2pandoc output: 
> > >      >      >      >      publisher: 
> > >      >      >      >      - 'Holt, Rinehart' 
> > >      >      >      >      - Winston 
> > >      >      >      >    Expected: 
> > >      >      >      >      publisher: 
> > >      >      >      >      - 'Holt, Rinehart and Winston' 
> > >      >      >      >    Also noticed, however, that pandoc-citeproc 
> does 
> > >      not 
> > >      >      seem to 
> > >      >      >      like it 
> > >      >      >      >    when actual multiple publishers occur: 
> > >      >      >      >      publisher: 
> > >      >      >      >      - 'Univ. of Toronto Press' 
> > >      >      >      >      - Routledge 
> > >      >      >      >    throws an error: 
> > >      >      >      >    pandoc-citeproc: Error parsing references: 
> when 
> > >      >      expecting a 
> > >      >      >      String, 
> > >      >      >      >    encountered Array instead 
> > >      >      >      >    This needs to be checked, but if CSL does not 
> > >      allow 
> > >      >      multiple 
> > >      >      >      >    publishers, the code above would need to be 
> > >      changed back 
> > >      >      to 
> > >      >      >      >      getField "publisher" ==> setField 
> "publisher" 
> > >      >      >      >    I see that multiple publishers (and locations) 
> > >      are being 
> > >      >      >      discussed at 
> > >      >      >      >    various Zotero/CSL/citeproc-js forums, but 
> don't 
> > >      get a 
> > >      >      clear 
> > >      >      >      picture 
> > >      >      >      >    yet. 
> > >      >      >      The literal {and} will not cause breaking of author 
> > >      lists or 
> > >      >      literal 
> > >      >      >      lists. 
> > >      >      >      I'll leave the multiple publishers question open 
> for 
> > >      now. 
> > >      >      Either 
> > >      >      >      bibtex2pandoc or pandoc-citeproc should be changed, 
> > >      not sure 
> > >      >      which. 
> > >      >      >      >    Names 
> > >      >      >      >    ===== 
> > >      >      >      >    Can bibtex2pandoc be expected to distinguish 
> all 
> > >      five 
> > >      >      name 
> > >      >      >      components 
> > >      >      >      >    (family, given, suffix, non-dropping-particle, 
> > >      >      >      dropping-particle)? 
> > >      >      >      >    biblatex provides one useful additional bit of 
> > >      info for 
> > >      >      name 
> > >      >      >      parsing 
> > >      >      >      >    via the "useprefix" switch inside the 
> "options" 
> > >      field: 
> > >      >      >      >    From the biblatex manual (v 2.7a), "3.1.3 
> Entry 
> > >      Options" 
> > >      >      >      >    > Whether the name prefix (von, van, of, da, 
> de, 
> > >      della, 
> > >      >      etc.) 
> > >      >      >      is 
> > >      >      >      >    considered when printing the last name in 
> > >      citations. 
> > >      >      This also 
> > >      >      >      affects 
> > >      >      >      >    the sorting and formatting of the bibliography 
> as 
> > >      well 
> > >      >      as the 
> > >      >      >      >    generation of certain types of labels. If this 
> > >      option is 
> > >      >      >      enabled, 
> > >      >      >      >    biblatex always precedes the last name with 
> the 
> > >      prefix. 
> > >      >      For 
> > >      >      >      example, 
> > >      >      >      >    Ludwig van Beethoven would be cited as 
> Beethoven 
> > >      and 
> > >      >      >      alphabetized as 
> > >      >      >      >    Beethoven, Ludwig van by default. If this 
> option 
> > >      is 
> > >      >      enabled, he 
> > >      >      >      is 
> > >      >      >      >    cited as van Beethoven and alphabetized as Van 
> > >      >      Beethoven, 
> > >      >      >      Ludwig 
> > >      >      >      >    instead. With Biber, this option is also 
> settable 
> > >      on a 
> > >      >      per-type 
> > >      >      >      basis. 
> > >      >      >      I used a rough-and-ready algorithm, probably not 
> the 
> > >      same one 
> > >      >      biber 
> > >      >      >      or biblatex uses.  Basically, the only time my code 
> > >      gives you 
> > >      >      a 
> > >      >      >      non-dropping particle is when you have a name in 
> > >      format 
> > >      >      >         B A, D E 
> > >      >      >      then B is the particle and A the given name. 
> > >      >      >      I don't have any support now for dropping-particle 
> or 
> > >      suffix. 
> > >      >      >      This needs to be improved, but I'd need to know 
> better 
> > >      what 
> > >      >      the 
> > >      >      >      algorithm is. 
> > >      >      >      >    Inline formatting 
> > >      >      >      >    ================= 
> > >      >      >      >    It probably doesn't play a role for pandoc, 
> but 
> > >      in the 
> > >      >      interest 
> > >      >      >      of 
> > >      >      >      >    portability, shouldn't inline formatting in a 
> > >      CSL-YAML 
> > >      >      file or 
> > >      >      >      metadata 
> > >      >      >      >    section rather use 
> > >      >      >      >        <i> and </i> for italics 
> > >      >      >      >        <b> and </b> for bold 
> > >      >      >      >        <sub> and </sub> for subscript 
> > >      >      >      >        <sup> and </sup> for superscript 
> > >      >      >      >        <span style="font-variant:small-caps;"> 
> and 
> > >      </span> 
> > >      >      for 
> > >      >      >      smallcaps 
> > >      >      >      >    instead of markdown formatting? 
> > >      >      >      >    (see [1][1][1]https://github.com/jgm/ 
> > >      pandoc/issues/931, 
> > >      >      >      >    [2][2][2]https://www.zotero.org/ 
> > >      support/kb/rich_text_ 
> > >      >      bibliography) 
> > >      >      >      No, because the YAML file is standard pandoc 
> metadata, 
> > >      >      formatted in 
> > >      >      >      markdown.  When it is read by pandoc-citeproc, this 
> > >      will be 
> > >      >      >      converted 
> > >      >      >      where possible to CSL metadata. 
> > >      >      >      >    Small Caps 
> > >      >      >      >    ========== 
> > >      >      >      >    Speaking of small caps: Any plans to add some 
> > >      >      markdownish 
> > >      >      >      formatting 
> > >      >      >      >    commands for small caps to pandoc? 
> > >      >      >      No idea what a natural syntax would be. 
> > >      >      > 
> > >      >      >    -- 
> > >      >      >    You received this message because you are subscribed 
> to 
> > >      the 
> > >      >      Google 
> > >      >      >    Groups "pandoc-discuss" group. 
> > >      >      >    To unsubscribe from this group and stop receiving 
> emails 
> > >      from 
> > >      >      it, send 
> > >      >      >    an email to [3]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >      >    To post to this group, send email to 
> > >      >      [4]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >      >    To view this discussion on the web visit 
> > >      >      >    [3][5][3]https://groups.google. 
> > >      com/d/msgid/pandoc-discuss/ 
> > >      >      a86067ec-6b4a-437b- 
> > >      >      >    b2f0-3ee666209dae%[6][4]40googlegroups.com. 
> > >      >      >    For more options, visit [4][7][5]
> https://groups.google. 
> > >      com/ 
> > >      >      groups/opt_out. 
> > >      >      > 
> > >      >      > References 
> > >      >      > 
> > >      >      >    1. [8][6]https://github.com/jgm/pandoc/issues/931 
> > >      >      >    2. [9][7]https://www.zotero.org/support/kb/rich_text_ 
> > >      bibliography 
> > >      >      >    3. [10][8]https://groups.google.com/d/ 
> > >      >      msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae% 
> > >      >      [9]40googlegroups.com 
> > >      >      >    4. [11][10]https://groups.google.com/groups/opt_out 
> > >      > 
> > >      >    -- 
> > >      >    You received this message because you are subscribed to the 
> > >      Google 
> > >      >    Groups "pandoc-discuss" group. 
> > >      >    To unsubscribe from this group and stop receiving emails 
> from 
> > >      it, send 
> > >      >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >    To post to this group, send email to 
> > >      pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >    To view this discussion on the web visit 
> > >      >    [12][11]https://groups.google.com/d/msgid/pandoc-discuss/ 
> > >      605aad90-4967-42b9 
> > >      >    -8efe-0aa033c670da%[12]40googlegroups.com. 
> > >      >    For more options, visit [13][13]https://groups.google.com/ 
> > >      groups/opt_out. 
> > >      > 
> > >      > References 
> > >      > 
> > >      >    1. [14]https://github.com/jgm/pandoc/issues/931 
> > >      >    2. [15]
> https://www.zotero.org/support/kb/rich_text_bibliography 
> > >      >    3. javascript:/ 
> > >      >    4. javascript:/ 
> > >      >    5. [16]https://groups.google.com/d/ 
> > >      msgid/pandoc-discuss/a86067ec-6b4a-437b- 
> > >      >    6. [17]http://40googlegroups.com/ 
> > >      >    7. [18]https://groups.google.com/groups/opt_out 
> > >      >    8. [19]https://github.com/jgm/pandoc/issues/931 
> > >      >    9. [20]
> https://www.zotero.org/support/kb/rich_text_bibliography 
> > >      >   10. [21]https://groups.google.com/d/ 
> > >      msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae% 
> > >      40googlegroups.com 
> > >      >   11. [22]https://groups.google.com/groups/opt_out 
> > >      >   12. [23]https://groups.google.com/d/ 
> > >      msgid/pandoc-discuss/605aad90-4967-42b9-8efe-0aa033c670da% 
> > >      40googlegroups.com 
> > >      >   13. [24]https://groups.google.com/groups/opt_out 
> > > 
> > >    -- 
> > >    You received this message because you are subscribed to the Google 
> > >    Groups "pandoc-discuss" group. 
> > >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> > >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > >    To view this discussion on the web visit 
> > >    [25]
> https://groups.google.com/d/msgid/pandoc-discuss/ebae1a26-c6cb-44c4 
> > >    -add2-5dc57f620a4c%40googlegroups.com. 
> > >    For more options, visit [26]
> https://groups.google.com/groups/opt_out. 
> > > 
> > > References 
> > > 
> > >    1. https://github.com/jgm/pandoc/issues/931 
> > >    2. https://www.zotero.org/support/kb/rich_text_ 
> > >    3. https://groups.google.com/d/msgid/pandoc-discuss/ 
> > >    4. http://40googlegroups.com/ 
> > >    5. https://groups.google.com/ 
> > >    6. https://github.com/jgm/pandoc/issues/931 
> > >    7. https://www.zotero.org/support/kb/rich_text_bibliography 
> > >    8. https://groups.google.com/d/ 
> > >    9. http://40googlegroups.com/ 
> > >   10. https://groups.google.com/groups/opt_out 
> > >   11. 
> https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9 
> > >   12. http://40googlegroups.com/ 
> > >   13. https://groups.google.com/groups/opt_out 
> > >   14. https://github.com/jgm/pandoc/issues/931 
> > >   15. https://www.zotero.org/support/kb/rich_text_bibliography 
> > >   16. 
> https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b- 
> > >   17. http://40googlegroups.com/ 
> > >   18. https://groups.google.com/groups/opt_out 
> > >   19. https://github.com/jgm/pandoc/issues/931 
> > >   20. https://www.zotero.org/support/kb/rich_text_bibliography 
> > >   21. 
> https://groups.google.com/d/msgid/pandoc-discuss/a86067ec-6b4a-437b-b2f0-3ee666209dae%40googlegroups.com 
> > >   22. https://groups.google.com/groups/opt_out 
> > >   23. 
> https://groups.google.com/d/msgid/pandoc-discuss/605aad90-4967-42b9-8efe-0aa033c670da%40googlegroups.com 
> > >   24. https://groups.google.com/groups/opt_out 
> > >   25. 
> https://groups.google.com/d/msgid/pandoc-discuss/ebae1a26-c6cb-44c4-add2-5dc57f620a4c%40googlegroups.com 
> > >   26. https://groups.google.com/groups/opt_out 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "pandoc-discuss" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/20130929012541.GC45785%40Johns-MacBook-Pro.local. 
>
> > For more options, visit https://groups.google.com/groups/opt_out. 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/fd944048-488f-4d68-a27e-5db0f5d41723%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 50928 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Sep 29 13 04:39 ]:
>    John, you are prefectly right, biblatex uses Babel language terms. In a
>    comprehensive mapping from biblatex to CSL they should surely be
>    implemented, and I can look into this. Still, to get title/sentence
>    case right, it just a question of English vs. other languages, and this
>    seems to work well with the small set of languages included so far.
>    Are you aware of anything CSL/citeproc would actually be doing with
>    entry-specific language information (except for dealing with
>    title/sentence case)?

That may be true, but it seems worth doing it right; otherwise
we end up with 'language' fields that are invalid.

>    On a related note, how is per-document localization done in
>    pandoc(-citeproc)?

I'm going to look into this more, but I think it just takes the
locale from the environment, and uses it to find appropriate
terms for things like "editor".  (These can be overridden, I
believe, in the CSL files.)  In any case, the per-entry 'language'
field wouldn't be used for this; rather, we'd want the language
for the document containing the bibliography.

John

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

On localization keys in biblatex:  I looked at the manual.
Most of the keys look like things that are handled automatically
by citeproc-hs.  For example:  editor.  So we really need to focus on
keys that might be used as shorthand in fields.

Perhaps things like 'inpreparation' qualify.

Nick, could you come up with a list of the keys that are likely
to be used in this way, which we must support?

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3590 bytes --]

These are the biblatex fields specified as "field(key)", each with possible 
values

- authortype
- bookpagination
    - page
    - column
    - line
    - verse
    - section
    - paragraph
- editortype
    - editor
    - compiler
    - founder
    - continuator
    - redactor
    - reviser
    - collaborator
- editoratype - see editortype
- editorbtype - see editortype
- editorctype - see editortype
- language
    - langamerican langbrazilian langcatalan langcroatian langczech 
langdanish langdutch langenglish langfinnish langfrench langgerman 
langgreek langitalian langlatin langnorwegian langpolish langportuguese 
langrussian langspanish langswedish
- origlanguage
    - see language
- pagination
    - page
    - column
    - line
    - verse
    - section
    - paragraph
- pubstate
    - inpreparation -> 'in preparation'
    - submitted -> 'submitted'
    - forthcoming -> 'forthcoming'
    - inpress -> 'in press'
    - prepublished -> 'pre-published'
- type
    - mathesis -> ‘Master’s thesis’
    - phdthesis -> ‘PhD thesis’
    - candthesis -> ‘Candidate thesis’
    - techreport -> ‘technical report’
    - resreport -> ‘research report’
    - software -> ‘computer software’
    - datacd -> ‘data CD’
    - audiocd -> audio CD
- name[a–c]type

What does this mean?
    - authortype, name[a–c]type are not used in standard styles.
    - editortype is to be used for determining the CSL item type, though 
straightforward mapping is mostly difficult; CSL has only "editor", 
"editorial-director", "director", "composer", "illustrator", "interviewer", 
or "series editor" (some are rather types of author, though)
    - pagination, bookpagination: CSL has similar "locator" terms, but I 
have neither seen them used nor been able to figure out how they work yet.
    - language, origlanguage describe language(s) of content; have no 
counterpart in CSL
        - they are independent of the biblatex field hyphenation which, 
confusingly, needs to be mapped to the CSL "language" variable and 
specifies the language (exactly one) of the bibliographic record.
        - language can describe multilingual works, e.g., language = 
{langlatin and langgerman},
        - In a pinch, bibtex2pandoc could append language and origlanguage 
info to the note field, but I'm not sure it's worth it.
    - pubstate and type should be mapped as indicated -- however, unless it 
is possible to map to a CSL "term" which in turn can be localized 
(currently CSL offers only "forthcoming", and "in press"), this will be a 
solution that is limited to English only. Still, short of providing 
bibtex2pandoc with a language switch and having it produce localized output 
I have no better idea. Anyway, English only it still better than to have a 
key such as "mathesis" appear as is in your printed output.
    

On Sunday, 29 September 2013 17:02:10 UTC, fiddlosopher wrote:
>
> On localization keys in biblatex:  
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/2e05de82-6208-49bf-be01-1f5f710c89de%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4782 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've added some preliminary support for keys, doing the two translations
you indicate below.  The code has a hook for locale (which is read from
the LANG environment variable), but currently only values for en-US
are given (and those are used for unknown locales).

+++ Nick Bart [Sep 29 13 13:33 ]:
>    These are the biblatex fields specified as "field(key)", each with
>    possible values
>    - authortype
>    - bookpagination
>        - page
>        - column
>        - line
>        - verse
>        - section
>        - paragraph
>    - editortype
>        - editor
>        - compiler
>        - founder
>        - continuator
>        - redactor
>        - reviser
>        - collaborator
>    - editoratype - see editortype
>    - editorbtype - see editortype
>    - editorctype - see editortype
>    - language
>        - langamerican langbrazilian langcatalan langcroatian langczech
>    langdanish langdutch langenglish langfinnish langfrench langgerman
>    langgreek langitalian langlatin langnorwegian langpolish langportuguese
>    langrussian langspanish langswedish
>    - origlanguage
>        - see language
>    - pagination
>        - page
>        - column
>        - line
>        - verse
>        - section
>        - paragraph
>    - pubstate
>        - inpreparation -> 'in preparation'
>        - submitted -> 'submitted'
>        - forthcoming -> 'forthcoming'
>        - inpress -> 'in press'
>        - prepublished -> 'pre-published'
>    - type
>        - mathesis -> Masters thesis
>        - phdthesis -> PhD thesis
>        - candthesis -> Candidate thesis
>        - techreport -> technical report
>        - resreport -> research report
>        - software -> computer software
>        - datacd -> data CD
>        - audiocd -> audio CD
>    - name[ac]type
>    What does this mean?
>        - authortype, name[ac]type are not used in standard styles.
>        - editortype is to be used for determining the CSL item type,
>    though straightforward mapping is mostly difficult; CSL has only
>    "editor", "editorial-director", "director", "composer", "illustrator",
>    "interviewer", or "series editor" (some are rather types of author,
>    though)
>        - pagination, bookpagination: CSL has similar "locator" terms, but
>    I have neither seen them used nor been able to figure out how they work
>    yet.
>        - language, origlanguage describe language(s) of content; have no
>    counterpart in CSL
>            - they are independent of the biblatex field hyphenation which,
>    confusingly, needs to be mapped to the CSL "language" variable and
>    specifies the language (exactly one) of the bibliographic record.
>            - language can describe multilingual works, e.g., language =
>    {langlatin and langgerman},
>            - In a pinch, bibtex2pandoc could append language and
>    origlanguage info to the note field, but I'm not sure it's worth it.
>        - pubstate and type should be mapped as indicated -- however,
>    unless it is possible to map to a CSL "term" which in turn can be
>    localized (currently CSL offers only "forthcoming", and "in press"),
>    this will be a solution that is limited to English only. Still, short
>    of providing bibtex2pandoc with a language switch and having it produce
>    localized output I have no better idea. Anyway, English only it still
>    better than to have a key such as "mathesis" appear as is in your
>    printed output.
> 
>    On Sunday, 29 September 2013 17:02:10 UTC, fiddlosopher wrote:
> 
>      On localization keys in biblatex:
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/2e05de82-6208-49bf-
>    be01-1f5f710c89de%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/2e05de82-6208-49bf-be01-1f5f710c89de%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 12375 bytes --]

Here's another patch. Thank you, Rintze, for the link to the CSL language 
codes.
I've added or fixed code for 
- series, number, and issue fields,
- article with entrysubtype= magazine
- editortype (unfortunately, the overlap between biblatex and CSL is 
minimal here; editoratype, editorbtype, editorctype would look exactly the 
same and can be added easily)
- language codes


--- ../pandoc-citeproc/bibtex2pandoc.hs    2013-09-30 07:08:21.000000000 
+0100
+++ bibtex2pandoc.hs    2013-09-30 16:42:20.000000000 +0100
@@ -241,6 +241,69 @@
   where go (Str ys)     = Str $ map toLower ys
         go z            = z
 
+toLocale :: String -> String
+toLocale "english"    = "en-US" -- "en-EN" unavailable in CSL
+toLocale "USenglish"  = "en-US"
+toLocale "american"   = "en-US"
+toLocale "british"    = "en-GB"
+toLocale "UKenglish"  = "en-GB"
+toLocale "canadian"   = "en-US" -- "en-CA" unavailable in CSL
+toLocale "australian" = "en-GB" -- "en-AU" unavailable in CSL
+toLocale "newzealand" = "en-GB" -- "en-NZ" unavailable in CSL
+toLocale "afrikaans"  = "af-ZA"
+toLocale "arabic"     = "ar-AR"
+toLocale "basque"     = "eu"
+toLocale "bulgarian"  = "bg-BG"
+toLocale "catalan"    = "ca-AD"
+toLocale "croatian"   = "hr-HR"
+toLocale "czech"      = "cs-CZ"
+toLocale "danish"     = "da-DK"
+toLocale "dutch"      = "nl-NL"
+toLocale "estonian"   = "et-EE"
+toLocale "finnish"    = "fi-FI"
+toLocale "canadien"   = "fr-CA"
+toLocale "acadian"    = "fr-CA"
+toLocale "french"     = "fr-FR"
+toLocale "francais"   = "fr-FR"
+toLocale "austrian"   = "de-AT"
+toLocale "naustrian"  = "de-AT"
+toLocale "german"     = "de-DE"
+toLocale "germanb"    = "de-DE"
+toLocale "ngerman"    = "de-DE"
+toLocale "greek"      = "el-GR"
+toLocale "polutonikogreek" = "el-GR"
+toLocale "hebrew"     = "he-IL"
+toLocale "hungarian"  = "hu-HU"
+toLocale "icelandic"  = "is-IS"
+toLocale "italian"    = "it-IT"
+toLocale "japanese"   = "ja-JP"
+toLocale "Khmer"      = "km-KH"
+toLocale "latvian"    = "lv-LV"
+toLocale "lithuanian" = "lt-LT"
+toLocale "magyar"     = "hu-HU"
+toLocale "mongolian"  = "mn-MN"
+toLocale "norsk"      = "nb-NO"
+toLocale "nynorsk"    = "nn-NO"
+toLocale "Persian"    = "fa-IR"
+toLocale "polish"     = "pl-PL"
+toLocale "brazil"     = "pt-BR"
+toLocale "brazilian"  = "pt-BR"
+toLocale "portugues"  = "pt-PT"
+toLocale "portuguese" = "pt-PT"
+toLocale "romanian"   = "ro-RO"
+toLocale "russian"    = "ru-RU"
+toLocale "serbian"    = "sr-RS"
+toLocale "serbianc"   = "sr-RS"
+toLocale "slovak"     = "sk-SK"
+toLocale "slovene"    = "sl-SL"
+toLocale "spanish"    = "es-ES"
+toLocale "swedish"    = "sv-SE"
+toLocale "thai"       = "th-TH"
+toLocale "turkish"    = "tr-TR"
+toLocale "ukrainian"  = "uk-UA"
+toLocale "vietnamese" = "vi-VN"
+toLocale _            = ""  
+
 itemToMetaValue :: Lang -> Bool -> T -> MetaValue
 itemToMetaValue lang bibtex = bibItem $ do
   getId >>= setRawField "id"
@@ -282,6 +345,14 @@
        "unpublished"     -> setType "manuscript"
        "www"             -> setType "webpage"
        _                 -> setType "no-type"
+       
+  opt $ do
+    val <- getRawField "entrysubtype"
+    if  et == "article" && val == "magazine" then
+          setType "article-magazine"
+    else
+          return ()
+       
   opt $ getRawField "type" >>= setRawField "genre" . resolveKey lang
   hyphenation <- getRawField "hyphenation" <|> return "english"
   let processTitle = if (map toLower hyphenation) `elem`
@@ -317,7 +388,13 @@
   opt $ getField "series" >>= setField "collection-title" . processTitle
   opt $ getField "pages" >>= setField "page"
   opt $ getField "volume" >>= setField "volume"
-  opt $ getField "number" >>= setField "number"
+  opt $ getField "number" >>=
+             setField (if et `elem` ["report"]
+                       then "number"
+                       else if et `elem` ["article", "periodical"]
+                       then "issue"
+                       else "collection-number")
+  opt $ getField "issue" >>= appendField "issue" addComma
   opt $ getField "chapter" >>= setField "chapter-number"
   opt $ getField "edition" >>= setField "edition"
   opt $ getField "note" >>= setField "note"
@@ -334,8 +411,36 @@
   opt $ getField "address" >>= setField "publisher-place"
   unless bibtex $ do
     opt $ getField "location" >>= setField "publisher-place"
+
   opt $ getAuthorList "author" >>= setList "author"
-  opt $ getAuthorList "editor" >>= setList "editor"
+  opt $ do
+    val <- getRawField "editortype"
+    getAuthorList "editor" >>=  setList (
+      case val of 
+      ""             -> "editor"             -- from here on biblatex & CSL
+      "editor"       -> "editor"
+      "compiler"     -> "editor"             -- from here on biblatex only
+      "founder"      -> "editor"
+      "continuator"  -> "editor"
+      "redactor"     -> "editor"
+      "reviser"      -> "editor"
+      "collaborator" -> "editor"
+      "director"     -> "director"           -- from here on 
biblatex-chicago & CSL
+--    "conductor"    -> ""                   -- from here on 
biblatex-chicago only
+--    "producer"     -> ""
+--    "none"         -> ""                   -- = performer
+--    ""             -> "editorial-director" -- from here on CSL only
+--    ""             -> "composer"
+--    ""             -> "illustrator"
+--    ""             -> "interviewer"
+--    ""             -> "collection-editor"
+      _              -> "editor")
+
+-- FIXME: add same for editora, editorb, editorc
+
+  opt $ getAuthorList "director" >>= setList "director"
+  -- director from biblatex-apa, which has also producer, writer, 
execproducer (FIXME?)
+
   opt $ getField "abstract" >>= setField "abstract"
   unless bibtex $ do
     opt $ getField "addendum" >>= appendField "note" (Space:)
@@ -363,11 +468,14 @@
   opt $ getField "urldate" >>= setField "accessed"
   opt $ getField "version" >>= setField "version"
   opt $ getField "volumes" >>= setField "number-of-volumes"
-  opt $ getRawField "hyphenation" >>= setRawField "language"
+  opt $ getRawField "hyphenation" >>= setRawField "language" . toLocale
 
 addColon :: [Inline] -> [Inline]
 addColon xs = [Str ":",Space] ++ xs
 
+addComma :: [Inline] -> [Inline]
+addComma xs = [Str ",",Space] ++ xs
+
 addPeriod :: [Inline] -> [Inline]
 addPeriod xs = [Str ".",Space] ++ xs
 
@@ -440,7 +548,7 @@
        "forthcoming"   -> "forthcoming"
        "inpress"       -> "in press"
        "prepublished"  -> "pre-published"
-       "mathesis"      -> "Masters thesis"
+       "mathesis"      -> "Master’s thesis"
        "phdthesis"     -> "PhD thesis"
        "candthesis"    -> "Candidate thesis"
        "techreport"    -> "technical report"


On Monday, September 30, 2013 2:51:39 AM UTC, fiddlosopher wrote:
>
> I've added some preliminary support for keys, doing the two translations 
> you indicate below.  The code has a hook for locale (which is read from 
> the LANG environment variable), but currently only values for en-US 
> are given (and those are used for unknown locales). 
>
> +++ Nick Bart [Sep 29 13 13:33 ]: 
> >    These are the biblatex fields specified as "field(key)", each with 
> >    possible values 
> >    - authortype 
> >    - bookpagination 
> >        - page 
> >        - column 
> >        - line 
> >        - verse 
> >        - section 
> >        - paragraph 
> >    - editortype 
> >        - editor 
> >        - compiler 
> >        - founder 
> >        - continuator 
> >        - redactor 
> >        - reviser 
> >        - collaborator 
> >    - editoratype - see editortype 
> >    - editorbtype - see editortype 
> >    - editorctype - see editortype 
> >    - language 
> >        - langamerican langbrazilian langcatalan langcroatian langczech 
> >    langdanish langdutch langenglish langfinnish langfrench langgerman 
> >    langgreek langitalian langlatin langnorwegian langpolish 
> langportuguese 
> >    langrussian langspanish langswedish 
> >    - origlanguage 
> >        - see language 
> >    - pagination 
> >        - page 
> >        - column 
> >        - line 
> >        - verse 
> >        - section 
> >        - paragraph 
> >    - pubstate 
> >        - inpreparation -> 'in preparation' 
> >        - submitted -> 'submitted' 
> >        - forthcoming -> 'forthcoming' 
> >        - inpress -> 'in press' 
> >        - prepublished -> 'pre-published' 
> >    - type 
> >        - mathesis -> Masters thesis 
> >        - phdthesis -> PhD thesis 
> >        - candthesis -> Candidate thesis 
> >        - techreport -> technical report 
> >        - resreport -> research report 
> >        - software -> computer software 
> >        - datacd -> data CD 
> >        - audiocd -> audio CD 
> >    - name[ac]type 
> >    What does this mean? 
> >        - authortype, name[ac]type are not used in standard styles. 
> >        - editortype is to be used for determining the CSL item type, 
> >    though straightforward mapping is mostly difficult; CSL has only 
> >    "editor", "editorial-director", "director", "composer", 
> "illustrator", 
> >    "interviewer", or "series editor" (some are rather types of author, 
> >    though) 
> >        - pagination, bookpagination: CSL has similar "locator" terms, 
> but 
> >    I have neither seen them used nor been able to figure out how they 
> work 
> >    yet. 
> >        - language, origlanguage describe language(s) of content; have no 
> >    counterpart in CSL 
> >            - they are independent of the biblatex field hyphenation 
> which, 
> >    confusingly, needs to be mapped to the CSL "language" variable and 
> >    specifies the language (exactly one) of the bibliographic record. 
> >            - language can describe multilingual works, e.g., language = 
> >    {langlatin and langgerman}, 
> >            - In a pinch, bibtex2pandoc could append language and 
> >    origlanguage info to the note field, but I'm not sure it's worth it. 
> >        - pubstate and type should be mapped as indicated -- however, 
> >    unless it is possible to map to a CSL "term" which in turn can be 
> >    localized (currently CSL offers only "forthcoming", and "in press"), 
> >    this will be a solution that is limited to English only. Still, short 
> >    of providing bibtex2pandoc with a language switch and having it 
> produce 
> >    localized output I have no better idea. Anyway, English only it still 
> >    better than to have a key such as "mathesis" appear as is in your 
> >    printed output. 
> > 
> >    On Sunday, 29 September 2013 17:02:10 UTC, fiddlosopher wrote: 
> > 
> >      On localization keys in biblatex: 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [1]
> https://groups.google.com/d/msgid/pandoc-discuss/2e05de82-6208-49bf- 
> >    be01-1f5f710c89de%40googlegroups.com. 
> >    For more options, visit [2]https://groups.google.com/groups/opt_out. 
> > 
> > References 
> > 
> >    1. 
> https://groups.google.com/d/msgid/pandoc-discuss/2e05de82-6208-49bf-be01-1f5f710c89de%40googlegroups.com 
> >    2. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/e39a1944-d7c8-46d1-80ce-1ee66b8c12bd%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 19826 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Applied, thanks.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Benct Philip jonsson]

2013-09-30 17:48, Nick Bart skrev:
> +toLocale "Persian"    = "fa-IR"

Is uppercase 'P' a typo here?
All the codes except
'USenglish', 'UKenglish', 'Khmer' and 'Persian'
are all-lowercase.

/bpj

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 6420 bytes --]

Another patch. Thank you, bpj, for catching the typo: should be "farsi". 
And Khmer is known to CSL, but not to babel.
Other fixes:
- fixed editortype routine which did not reliably set CSL editor variable 
if editortype was nonexistent. (If there's a simpler way, please fix.)
- doi, isbn, issn are shown uppercase in 
http://citationstyles.org/downloads/specification.html#appendix-iv-variables, 
but didn't work unless lowercase here.
- decided to try concatenating school, institution, organization, 
howpublished, publisher (which all go to CSL "publisher"). Most don't 
co-occur, but a few do, and if so, no info will be lost. Comments on 
whether this works well in the wild are welcome.


--- ../pandoc-citeproc/bibtex2pandoc.hs    2013-09-30 18:04:12.000000000 
+0100
+++ bibtex2pandoc.hs    2013-09-30 20:51:48.000000000 +0100
@@ -277,14 +277,13 @@
 toLocale "icelandic"  = "is-IS"
 toLocale "italian"    = "it-IT"
 toLocale "japanese"   = "ja-JP"
-toLocale "Khmer"      = "km-KH"
 toLocale "latvian"    = "lv-LV"
 toLocale "lithuanian" = "lt-LT"
 toLocale "magyar"     = "hu-HU"
 toLocale "mongolian"  = "mn-MN"
 toLocale "norsk"      = "nb-NO"
 toLocale "nynorsk"    = "nn-NO"
-toLocale "Persian"    = "fa-IR"
+toLocale "farsi"      = "fa-IR"
 toLocale "polish"     = "pl-PL"
 toLocale "brazil"     = "pt-BR"
 toLocale "brazilian"  = "pt-BR"
@@ -403,45 +402,56 @@
   opt $ getField "journaltitle" >>= setField "container-title"
   opt $ getField "journalsubtitle" >>= appendField "container-title" 
addColon
   opt $ getField "shortjournal" >>= setField "container-title-short"
-  opt $ getField "howpublished" >>= setField "publisher"
-  opt $ getField "school" >>= setField "publisher"
-  unless bibtex $ do
-    opt $ getLiteralList "institution" >>= setList "publisher"
-  opt $ getField "publisher" >>= setField "publisher"
+
+--   opt $ getField "school" >>= setField "publisher"
+--   opt $ getField "institution" >>= setField "publisher"
+--   opt $ getField "organization" >>= setField "publisher"
+--   opt $ getField "howpublished" >>= setField "publisher"
+--   opt $ getField "publisher" >>= setField "publisher"
+
+  opt $ getField "school" >>= appendField "publisher" addComma
+  opt $ getField "institution" >>= appendField "publisher" addComma
+  opt $ getField "organization" >>= appendField "publisher" addComma
+  opt $ getField "howpublished" >>= appendField "publisher" addComma
+  opt $ getField "publisher" >>= appendField "publisher" addComma
+
   opt $ getField "address" >>= setField "publisher-place"
   unless bibtex $ do
     opt $ getField "location" >>= setField "publisher-place"
 
   opt $ getAuthorList "author" >>= setList "author"
-  opt $ do
-    val <- getRawField "editortype"
-    getAuthorList "editor" >>=  setList (
-      case val of 
-      ""             -> "editor"             -- from here on biblatex & CSL
-      "editor"       -> "editor"
-      "compiler"     -> "editor"             -- from here on biblatex only
-      "founder"      -> "editor"
-      "continuator"  -> "editor"
-      "redactor"     -> "editor"
-      "reviser"      -> "editor"
-      "collaborator" -> "editor"
-      "director"     -> "director"           -- from here on 
biblatex-chicago & CSL
---    "conductor"    -> ""                   -- from here on 
biblatex-chicago only
---    "producer"     -> ""
---    "none"         -> ""                   -- = performer
---    ""             -> "editorial-director" -- from here on CSL only
---    ""             -> "composer"
---    ""             -> "illustrator"
---    ""             -> "interviewer"
---    ""             -> "collection-editor"
-      _              -> "editor")
 
+  hasEditortype <- isPresent "editortype"
+  if hasEditortype then
+    opt $ do
+      val <- getRawField "editortype"
+      getAuthorList "editor" >>=  setList (
+        case val of 
+        "editor"       -> "editor"             -- from here on biblatex & 
CSL
+        "compiler"     -> "editor"             -- from here on biblatex 
only
+        "founder"      -> "editor"
+        "continuator"  -> "editor"
+        "redactor"     -> "editor"
+        "reviser"      -> "editor"
+        "collaborator" -> "editor"
+        "director"     -> "director"           -- from here on 
biblatex-chicago & CSL
+  --    "conductor"    -> ""                   -- from here on 
biblatex-chicago only
+  --    "producer"     -> ""
+  --    "none"         -> ""                   -- = performer
+  --    ""             -> "editorial-director" -- from here on CSL only
+  --    ""             -> "composer"
+  --    ""             -> "illustrator"
+  --    ""             -> "interviewer"
+  --    ""             -> "collection-editor"
+        _              -> "editor")
+    else
+      getAuthorList "editor" >>= setList "editor"
+ 
 -- FIXME: add same for editora, editorb, editorc
 
   opt $ getAuthorList "director" >>= setList "director"
   -- director from biblatex-apa, which has also producer, writer, 
execproducer (FIXME?)
 
-  opt $ getField "abstract" >>= setField "abstract"
   unless bibtex $ do
     opt $ getField "addendum" >>= appendField "note" (Space:)
   opt $ getField "annotation" >>= setField "annote"
@@ -455,9 +465,9 @@
   opt $ getField "eventdate" >>= setField "event-date"
   opt $ getField "eventtitle" >>= setField "event"
   opt $ getLiteralList "venue" >>= setList "event-place"
-  opt $ getRawField "doi" >>= setRawField "DOI"
-  opt $ getRawField "isbn" >>= setRawField "ISBN"
-  opt $ getRawField "issn" >>= setRawField "ISSN"
+  opt $ getRawField "doi" >>= setRawField "doi"
+  opt $ getRawField "isbn" >>= setRawField "isbn"
+  opt $ getRawField "issn" >>= setRawField "issn"
   opt $ getField "origdate" >>= setField "original-date"
   opt $ getLiteralList "origlocation" >>=
              setList "original-publisher-place"

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/f61e532e-8d2b-4264-83ce-41e15ca2adea%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 11122 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 790 bytes --]

Note that in the patch above,

  hasEditortype <- isPresent "editortype"
  if hasEditortype then
  opt $ do

should read:

  hasEditortype <- isPresent "editortype"
  opt $ if hasEditortype
  then do

Apologies.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/66a61c32-a3b4-45c2-b548-442592eab56d%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1126 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Thanks, patch applied.

I'm beginning to rethink the general approach a bit.

Currently bibtex2pandoc reads the bibtex file, converts it
to a pandoc 'MetaValue', and uses the markdown writer to write
it out as YAML metadata.

This is fairly nice if you want to convert a bibtex file to
a format in which it can be used inline in pandoc, but it makes
less sense if you're using an external bibtex file as your
bibliography.

For that, it would make more sense to convert directly to a
list of citeproc Reference objects.  We've defined ToJSON
and FromJSON instances for these, so it's easy to convert to
JSON or YAML (that's what biblio2yaml does).  However, the
YAML produced will be slightly different:  it will have
<i>this</i> kind of emphasis instead of *standard markdown*
emphasis, for example.

It seems that ultimately we're going to want to integrate the
code we're writing into pandoc-citeproc itself, so that we can
avoid the bibutils conversion step when using bibtex/biblatex
bibliographies.  That suggests targeting [Reference] rather
than MetaValue.  (So, we'd fold the code into a new module
in pandoc-citeproc library, and biblio2yaml could be used for
the conversion.)

Thoughts?


+++ Nick Bart [Sep 30 13 14:31 ]:
>    Note that in the patch above,
>      hasEditortype <- isPresent "editortype"
>      if hasEditortype then
>      opt $ do
>    should read:
>      hasEditortype <- isPresent "editortype"
>      opt $ if hasEditortype
>      then do
>    Apologies.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/66a61c32-a3b4-45c2-
>    b548-442592eab56d%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/66a61c32-a3b4-45c2-b548-442592eab56d%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4062 bytes --]

Sounds very good.

Ideally, I feel, bibtex2pandoc would do three things:

(1) convert biblatex/bibtex to YAML metadata ready for insertion into a 
pandoc document (as it currently does),
(2) convert biblatex/bibtex to YAML written to a file that can be read in 
again by, e.g., `pandoc --bibliography bibfile.yaml`, and
(3) convert biblatex/bibtex to "a list of citeproc Reference objects" fit 
for immediate consumption by pandoc-citeproc.

Except for file headers etc., I imagine all of these could use essentially 
the very same format.

As to inline formatting, I would not mind having <i>emph</i> instead of 
*emph* in these data at all.

One suggestion, though: Readability would improve if (1) empty lines could 
be added between entries and (2) id, title, author, editor, and date could 
be moved to the top of each entry.

And a command line option could be added for suppressing fields such as 
abstract, keywords, annote/annotation which are usually not needed for 
formatting bibliographies (except annotated ones, that is).


Apart from that, at this stage a few issues remain that I cannot fix alone:

- date handling (date, urldate, etc. are essential fields in biblatex; many 
biblatex users do not even use "year" fields any longer).

- handling of name prefixes and suffixes (I'll see what I can contribute)

- handling of multiple publishers, locations, and other "literal list" 
field components: Having "getLiteralList", as intended, split such fields 
into components would be the cleanest solution, but pandoc-citeproc 
currently does not accept arrays (thus I've mostly replaced 
"getLiteralList" by "getField" for now), so it would have to be modified.


In addition, I came across two issues that are most likely pandoc-citeproc 
issues:

- Extra strings such as "Edited & translated by." in pandoc output. This 
has been reported here in detail: 
https://code.google.com/p/citeproc-hs/issues/detail?id=92

- Very long titles in non-English languages:

Spanish (French, German, too): Linebreak after title, before '.”'

$ cat vlt-es.md
Foo [@item1, p. 12].

# References

---
references:
- author:
- family: Author
given:
- Al
container-title: 'Journaltitle'
id: item1
issue: 3
issued:
year: 2012
language: 'es-ES'
page: 'S12-S34'
title: |
A very very very very very very very very very very very very very very 
very long title
type: 'article-journal'
volume: 5
...

$ pandoc --filter pandoc-citeproc vlt-es.md -t markdown-citations
Foo (Author 2012, 12).

<div class="references">

References
==========

Author, Al. 2012. “A very very very very very very very very very very
very very very very very long title
.” *Journaltitle* 5 (3): S12–S34.

</div>

English: no such linebreak

$ cat vlt-en.md
Foo [@item1, p. 12].

# References

---
references:
- author:
- family: Author
given:
- Al
container-title: 'Journaltitle'
id: item1
issue: 3
issued:
year: 2012
language: 'en-EN'
page: 'S12-S34'
title: |
A very very very very very very very very very very very very very very 
very long title
type: 'article-journal'
volume: 5
...

$ pandoc --filter pandoc-citeproc vlt-en.md -t markdown-citations
Foo (Author 2012, 12).

<div class="references">

References
==========

Author, Al. 2012. “A Very Very Very Very Very Very Very Very Very Very
Very Very Very Very Very Long Title.” *Journaltitle* 5 (3): S12–S34.

</div>


-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/15bfac5e-d7df-476c-8a42-193962e6bca3%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4544 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1756 bytes --]

Issue: Term “vols.” after number-of-volumes missing
============================================

This might be a pandoc-citeproc issue:

$ cat test.md
Foo [@item1, p. 12].

# References

---
references:
- author:
  - family: Author
    given:
    - Al
  id: item1
  issued:
    year: 2013
  language: 'en-US'
  number-of-volumes: 5
  publisher: Publisher
  publisher-place: Location
  title: Title of the Book
  type: book
...
$ pandoc -S --filter pandoc-citeproc -t markdown-citations test.md
Foo (Author 2013, 12).

<div class="references">

References
==========

Author, Al. 2013. *Title of the Book*. 5. Location: Publisher.

</div>

Expected output is:

Author, Al. 2013. *Title of the Book*. 5 vols. Location: Publisher.

Same effect with other CSL style files. Does work as expected in Zotero.

chicago-author-date.csl contains the following, which looks ok:
  <number variable="number-of-volumes" form="numeric"/>
  <text term="volume" form="short" prefix=" " plural="true"/>


By the way, I've been wondering, is it intentional that <div 
class="references"> ... </div> appears in the markdown output format?

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/f93e4825-005b-42e0-85f1-e7964da2887d%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2248 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3967 bytes --]

On name parsing (2013-10-01 15:33)
===============================

First, the target format. CSL gives the following specs, see 
<http://citationstyles.org/downloads/specification.html>:

>Personal names require a "family" name-part, and may also contain "given", 
"suffix", "non-dropping-particle" and "dropping-particle" name-parts. These 
name-parts are defined as:

>- "family" - surname minus any particles and suffixes
>- "given" - given names, either full ("John Edward") or initialized ("J. 
E.")
>- "suffix" - name suffix, e.g. "Jr." in "John Smith Jr." and "III" in 
"Bill Gates III"
>- "non-dropping-particle" - name particles that are not dropped when only 
the surname is shown ("de" in the Dutch surname "de Koning") but which may 
be treated separately from the family name, e.g. for sorting
>- "dropping-particle" - name particles that are dropped when only the 
surname is shown ("van" in "Ludwig van Beethoven", which becomes 
"Beethoven")

Second, bibtex. Its name parsing algorithms are described rather fully in 
“Tame the BeaST”, 
<http://mirrors.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf>.

In a nutshell, valid formats are:
- First von Last
- von Last, First
- von Last, Jr, First
... which can be identified by counting the number of commas.

It seems that an uppercase "von" part, or prefix is recognized as a prefix 
if it is enclosed in curly braces, or at least its first letter(s) are -- 
Example: `author = "{D}e {L}a Cruz, Maria"` where `{D}e {L}a` is the prefix 
(“Tame the BeaST”, § 13.2)

It appears - though I must admit I haven't really checked - that biblatex 
uses the same algorithms.

The only extension for biblatex appears to be that it has one additional 
way of specifying the character of the prefix in the "options" field. From 
the biblatex manual (v 2.7a):

> useprefix=true, false default: false

> Whether the name prefix (von, van, of, da, de, della, etc.) is considered 
when printing the last name in citations. [...] If this option is enabled, 
biblatex always precedes the last name with the prefix. For example, Ludwig 
van Beethoven would be cited as “Beethoven” and alphabetized as “Beethoven, 
Ludwig van” by default. If this option is enabled, he is cited as “van 
Beethoven” and alphabetized as “Van Beethoven, Ludwig” instead.

I take this to mean that if a biblatex entry contains 
`options={useprefix=false}`, the "von part" should be considered to be a 
"dropping-particle"; if it contains `options={useprefix=true}`, the "von 
part" should be considered to be a "non-dropping-particle".

In addition, biblatex’s `sortname` field might contain some useful 
information as well:

For example, `author={Jean de La Fontaine}` would be parsed into
- First = "Jean"
- von = "de"
- Last = "La Fontaine"
However, assuming this particular entry also contained 
`sortname={Fontaine}`, the conclusion could be drawn that the "La" from "La 
Fontaine" is to be understood as a "non-dropping-particle".

The useprefix option is of course set per entry, so it cannot be expected 
to handle - hopefully rare - cases with two or more authors with 
conflicting prefix requirements. Neither would the sortname heuristic work 
for any name except for the first one, but then again, the others would not 
be used for sorting anyway.

Comments are welcome!

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/6e1172e2-04e5-43e2-b77a-baca29fc5d88%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4281 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3923 bytes --]

On name parsing
===============

First, the target format. CSL gives the following specs, see 
<http://citationstyles.org/downloads/specification.html>:

>Personal names require a "family" name-part, and may also contain "given", 
"suffix", "non-dropping-particle" and "dropping-particle" name-parts. These 
name-parts are defined as:

>- "family" - surname minus any particles and suffixes
>- "given" - given names, either full ("John Edward") or initialized ("J. 
E.")
>- "suffix" - name suffix, e.g. "Jr." in "John Smith Jr." and "III" in 
"Bill Gates III"
>- "non-dropping-particle" - name particles that are not dropped when only 
the surname is shown ("de" in the Dutch surname "de Koning") but which may 
be treated separately from the family name, e.g. for sorting
>- "dropping-particle" - name particles that are dropped when only the 
surname is shown ("van" in "Ludwig van Beethoven", which becomes 
"Beethoven")

Second, bibtex. Its name parsing algorithms are described rather fully in 
“Tame the BeaST”, 
<http://mirrors.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf>.

In a nutshell, valid formats are:
- First von Last
- von Last, First
- von Last, Jr, First
... which can be identified by counting the number of commas.

It seems that an uppercase "von" part, or prefix is recognized as a prefix 
if it is enclosed in curly braces, or at least its first letter(s) are -- 
Example: `author = "{D}e {L}a Cruz, Maria"` where `{D}e {L}a` is the prefix 
(“Tame the BeaST”, § 13.2)

It appears - though I must admit I haven't really checked - that biblatex 
uses the same algorithms.

The only extension for biblatex appears to be that it has one additional 
way of specifying the character of the prefix in the "options" field. From 
the biblatex manual (v 2.7a):

> useprefix=true, false default: false

> Whether the name prefix (von, van, of, da, de, della, etc.) is considered 
when printing the last name in citations. [...] If this option is enabled, 
biblatex always precedes the last name with the prefix. For example, Ludwig 
van Beethoven would be cited as “Beethoven” and alphabetized as “Beethoven, 
Ludwig van” by default. If this option is enabled, he is cited as “van 
Beethoven” and alphabetized as “Van Beethoven, Ludwig” instead.

I take this to mean that if a biblatex entry contains 
`options={useprefix=false}`, the "von part" should be considered to be a 
"dropping-particle"; if it contains `options={useprefix=true}`, the "von 
part" should be considered to be a "non-dropping-particle".

In addition, biblatex’s `sortname` field might contain some useful 
information as well:

For example, `author={Jean de La Fontaine}` would be parsed into
- First = "Jean"
- von = "de"
- Last = "La Fontaine"
However, assuming this particular entry also contained 
`sortname={Fontaine}`, the conclusion could be drawn that the "La" from "La 
Fontaine" is to be understood as a "non-dropping-particle".

The useprefix option is of course set per entry, so it cannot be expected 
to handle - hopefully rare - cases with two or more authors with 
conflicting prefix requirements. Neither would the sortname heuristic work 
for any name except for the first - but then gain, these would not be used 
for sorting anyway.

Comments are welcome!

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/fa1df90f-6ca5-43f2-a0d8-ec413aab1bb8%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4237 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[fiddlosopher]

[-- Attachment #1: Type: text/plain, Size: 1859 bytes --]



On Tuesday, October 1, 2013 4:06:40 AM UTC-7, Nick Bart wrote:
>
> Issue: Term “vols.” after number-of-volumes missing
> ============================================
>
> This might be a pandoc-citeproc issue:
>
> $ cat test.md
> Foo [@item1, p. 12].
>
> # References
>
> ---
> references:
> - author:
>   - family: Author
>     given:
>     - Al
>   id: item1
>   issued:
>     year: 2013
>   language: 'en-US'
>   number-of-volumes: 5
>   publisher: Publisher
>   publisher-place: Location
>   title: Title of the Book
>   type: book
> ...
>

Yes, this is another one it would be good to have Andrea look into.

By the way, I've been wondering, is it intentional that <div 
> class="references"> ... </div> appears in the markdown output format?
>

It's intentional to put the bibliography in a pandoc Div element.  That 
allows people to style it differently if they like, plus scripts can see 
it, slide shows know what is a bibliography, etc.  Normally Div elements 
are rendered using
HTML div tags in markdown (though you can turn this off by disabling the 
'markdown_in_html_blocks' extension).  I'm somewhat tempted to special-case 
this one and have the markdown writer ignore this particular Div, but I'm 
not sure.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/d52d9518-9832-45c3-96a0-c3d5b9de3090%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2640 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1982 bytes --]

On the redone bibtex2pandoc: Nice. Glad to see that dates are implemented 
now.

I noticed a few regressions, however:

-    , collectionTitle = volumeTitle' ++ volumeSubtitle' ++ 
volumeTitleAddon'

collection-title (I'll be using the CSL variable names here) must receive 
the content of biblatex "series" field (unless the biblatex entry type is 
"article", "periodical" or "suppperiodical", in which case series is 
appended to CSL container-title)

"volumetitle'" etc. goes to the (not yet official) CSL variable 
"volume-title"

-     --, collectionNumber = undefined --:: String --Int
  and
      -- , issue               = undefined -- :: String

The biblatex "number" field is mapped to CSL "number",
- unless the entry type is "book", "collection", "proceedings", 
"reference", "mvbook", "mvcollection", "mvproceedings", "mvreference", 
"bookinbook", "inbook", "incollection", "inproceedings", "inreference", 
"suppbook", "suppcollection", in which case it is mapped to CSL 
"collection-number",
- or the entry type is "article", "periodical" or "suppperiodical", in 
which case the best solution, used in the previous version of 
bibtex2pandoc, is to map the biblatex fields number *and* issue (in that 
sequence, separated by a comma) together to CSL "issue".

- When Journal and Journalsubtitle are combined, the ": " between them is 
missing now.

  container-title: JournalJournalsubtitle

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/e590b84d-7a2f-40b7-9b4d-26d1644b8a53%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2436 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2551 bytes --]

As to the issues with the biblatex "series" field, though I cannot claim to 
fully understand the new code yet, what I see is that while *sometimes* the 
output is as expected (test-book-tms.bib below), in other cases it clearly 
isn't (test-inbook-tbms.bib):

$ cat test-book-tms.bib 
@book{item1,
    Author = {Author, Al},
    Date = {2013},
    Hyphenation = {french},
    Location = {Location},
    Mainsubtitle = {Mainsubtitle},
    Maintitle = {Maintitle},
    Maintitleaddon = {Maintitleaddon},
    Number = {3},
    Publisher = {Publisher},
    Series = {Series},
    Subtitle = {Subtitle},
    Title = {Title of the Book},
    Titleaddon = {Titleaddon}}
$ bibtex2pandoc test-book-tms.bib 
---
references:
- title: ! 'Title of the book: Subtitle. Titleaddon'
  id: item1
  issued:
    year: 2013
  author:
    given:
    - Al
    family: Author
  container-title: ! 'Maintitle: Mainsubtitle. Maintitleaddon'
  publisher: Publisher
  collection-title: Series
  type: book
  publisher-place: Location
  language: fr-FR
...

$ cat test-inbook-tbms.bib 
@inbook{item1,
    Author = {Author, Al},
    Booksubtitle = {Booksubtitle},
    Booktitle = {Booktitle},
    Booktitleaddon = {Booktitleaddon},
    Date = {2011},
    Hyphenation = {french},
    Location = {Location},
    Mainsubtitle = {Mainsubtitle},
    Maintitle = {Maintitle},
    Maintitleaddon = {Maintitleaddon},
    Number = {3},
    Publisher = {Publisher},
    Series = {Series},
    Subtitle = {Subtitle},
    Title = {Title of the "@inbook" Entry},
    Titleaddon = {Titleaddon}}
$ bibtex2pandoc test-inbook-tbms.bib 
---
references:
- title: ! 'Title of the "@inbook" entry: Subtitle. Titleaddon'
  id: item1
  issued:
    year: 2011
  author:
    given:
    - Al
    family: Author
  container-title: ! 'Maintitle: Mainsubtitle. Maintitleaddon'
  publisher: Publisher
  collection-title: ! 'Booktitle: Booksubtitle. Booktitleaddon'
  type: chapter
  publisher-place: Location
  language: fr-FR
...

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/967cb63f-a612-4ded-bc2b-6273c0239f68%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3674 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

The difference is due to the line

   (getTitle lang "maintitle" >> guard hasVolumes >> getTitle lang "booktitle")

which matches in test-inbook-tbms.bib. Because we have "maintitle" here,
collectionTitle is populated by "booktitle" in this case.

I was trying to follow the logic of the old version here -- may have
misunderstood something.

+++ Nick Bart [Oct 05 13 02:32 ]:
>    As to the issues with the biblatex "series" field, though I cannot
>    claim to fully understand the new code yet, what I see is that while
>    *sometimes* the output is as expected (test-book-tms.bib below), in
>    other cases it clearly isn't (test-inbook-tbms.bib):
>    $ cat test-book-tms.bib
>    @book{item1,
>        Author = {Author, Al},
>        Date = {2013},
>        Hyphenation = {french},
>        Location = {Location},
>        Mainsubtitle = {Mainsubtitle},
>        Maintitle = {Maintitle},
>        Maintitleaddon = {Maintitleaddon},
>        Number = {3},
>        Publisher = {Publisher},
>        Series = {Series},
>        Subtitle = {Subtitle},
>        Title = {Title of the Book},
>        Titleaddon = {Titleaddon}}
>    $ bibtex2pandoc test-book-tms.bib
>    ---
>    references:
>    - title: ! 'Title of the book: Subtitle. Titleaddon'
>      id: item1
>      issued:
>        year: 2013
>      author:
>        given:
>        - Al
>        family: Author
>      container-title: ! 'Maintitle: Mainsubtitle. Maintitleaddon'
>      publisher: Publisher
>      collection-title: Series
>      type: book
>      publisher-place: Location
>      language: fr-FR
>    ...
>    $ cat test-inbook-tbms.bib
>    @inbook{item1,
>        Author = {Author, Al},
>        Booksubtitle = {Booksubtitle},
>        Booktitle = {Booktitle},
>        Booktitleaddon = {Booktitleaddon},
>        Date = {2011},
>        Hyphenation = {french},
>        Location = {Location},
>        Mainsubtitle = {Mainsubtitle},
>        Maintitle = {Maintitle},
>        Maintitleaddon = {Maintitleaddon},
>        Number = {3},
>        Publisher = {Publisher},
>        Series = {Series},
>        Subtitle = {Subtitle},
>        Title = {Title of the "@inbook" Entry},
>        Titleaddon = {Titleaddon}}
>    $ bibtex2pandoc test-inbook-tbms.bib
>    ---
>    references:
>    - title: ! 'Title of the "@inbook" entry: Subtitle. Titleaddon'
>      id: item1
>      issued:
>        year: 2011
>      author:
>        given:
>        - Al
>        family: Author
>      container-title: ! 'Maintitle: Mainsubtitle. Maintitleaddon'
>      publisher: Publisher
>      collection-title: ! 'Booktitle: Booksubtitle. Booktitleaddon'
>      type: chapter
>      publisher-place: Location
>      language: fr-FR
>    ...
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/967cb63f-a612-4ded-
>    bc2b-6273c0239f68%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/967cb63f-a612-4ded-bc2b-6273c0239f68%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4839 bytes --]

Well, "collectionTitle" should never be populated by anything but the 
biblatex "series" field.

biblatex "title", "booktitle", and "maintitle" can only ever map to CSL 
"title", "container-title", and the new "volume-title", and if you do not 
want to use the latter yet (I would!), "booktitle" must be discarded until 
"volume-title" has become part of the CSL specs. 

On Sunday, October 6, 2013 12:58:56 AM UTC, fiddlosopher wrote:
>
> The difference is due to the line 
>
>    (getTitle lang "maintitle" >> guard hasVolumes >> getTitle lang 
> "booktitle") 
>
> which matches in test-inbook-tbms.bib. Because we have "maintitle" here, 
> collectionTitle is populated by "booktitle" in this case. 
>
> I was trying to follow the logic of the old version here -- may have 
> misunderstood something. 
>
> +++ Nick Bart [Oct 05 13 02:32 ]: 
> >    As to the issues with the biblatex "series" field, though I cannot 
> >    claim to fully understand the new code yet, what I see is that while 
> >    *sometimes* the output is as expected (test-book-tms.bib below), in 
> >    other cases it clearly isn't (test-inbook-tbms.bib): 
> >    $ cat test-book-tms.bib 
> >    @book{item1, 
> >        Author = {Author, Al}, 
> >        Date = {2013}, 
> >        Hyphenation = {french}, 
> >        Location = {Location}, 
> >        Mainsubtitle = {Mainsubtitle}, 
> >        Maintitle = {Maintitle}, 
> >        Maintitleaddon = {Maintitleaddon}, 
> >        Number = {3}, 
> >        Publisher = {Publisher}, 
> >        Series = {Series}, 
> >        Subtitle = {Subtitle}, 
> >        Title = {Title of the Book}, 
> >        Titleaddon = {Titleaddon}} 
> >    $ bibtex2pandoc test-book-tms.bib 
> >    --- 
> >    references: 
> >    - title: ! 'Title of the book: Subtitle. Titleaddon' 
> >      id: item1 
> >      issued: 
> >        year: 2013 
> >      author: 
> >        given: 
> >        - Al 
> >        family: Author 
> >      container-title: ! 'Maintitle: Mainsubtitle. Maintitleaddon' 
> >      publisher: Publisher 
> >      collection-title: Series 
> >      type: book 
> >      publisher-place: Location 
> >      language: fr-FR 
> >    ... 
> >    $ cat test-inbook-tbms.bib 
> >    @inbook{item1, 
> >        Author = {Author, Al}, 
> >        Booksubtitle = {Booksubtitle}, 
> >        Booktitle = {Booktitle}, 
> >        Booktitleaddon = {Booktitleaddon}, 
> >        Date = {2011}, 
> >        Hyphenation = {french}, 
> >        Location = {Location}, 
> >        Mainsubtitle = {Mainsubtitle}, 
> >        Maintitle = {Maintitle}, 
> >        Maintitleaddon = {Maintitleaddon}, 
> >        Number = {3}, 
> >        Publisher = {Publisher}, 
> >        Series = {Series}, 
> >        Subtitle = {Subtitle}, 
> >        Title = {Title of the "@inbook" Entry}, 
> >        Titleaddon = {Titleaddon}} 
> >    $ bibtex2pandoc test-inbook-tbms.bib 
> >    --- 
> >    references: 
> >    - title: ! 'Title of the "@inbook" entry: Subtitle. Titleaddon' 
> >      id: item1 
> >      issued: 
> >        year: 2011 
> >      author: 
> >        given: 
> >        - Al 
> >        family: Author 
> >      container-title: ! 'Maintitle: Mainsubtitle. Maintitleaddon' 
> >      publisher: Publisher 
> >      collection-title: ! 'Booktitle: Booksubtitle. Booktitleaddon' 
> >      type: chapter 
> >      publisher-place: Location 
> >      language: fr-FR 
> >    ... 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [1]
> https://groups.google.com/d/msgid/pandoc-discuss/967cb63f-a612-4ded- 
> >    bc2b-6273c0239f68%40googlegroups.com. 
> >    For more options, visit [2]https://groups.google.com/groups/opt_out. 
> > 
> > References 
> > 
> >    1. 
> https://groups.google.com/d/msgid/pandoc-discuss/967cb63f-a612-4ded-bc2b-6273c0239f68%40googlegroups.com 
> >    2. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/733d7104-44b1-440a-a386-ec356d2197bd%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 7595 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

OK, I've merged in a number of changes Andrea made to citeproc-hs
after I split off pandoc-citeproc.  These fix some bugs, including
#90, #93, #95, #86.  volumeTitle has been added to Reference, so
I think I'm now doing the right thing with volumeTitle and
containerTitle, but as always your testing will be invaluable!

+++ Nick Bart [Oct 05 13 23:29 ]:
>    Well, "collectionTitle" should never be populated by anything but the
>    biblatex "series" field.
>    biblatex "title", "booktitle", and "maintitle" can only ever map to CSL
>    "title", "container-title", and the new "volume-title", and if you do
>    not want to use the latter yet (I would!), "booktitle" must be
>    discarded until "volume-title" has become part of the CSL specs.
>    On Sunday, October 6, 2013 12:58:56 AM UTC, fiddlosopher wrote:
> 
>      The difference is due to the line
>         (getTitle lang "maintitle" >> guard hasVolumes >> getTitle lang
>      "booktitle")
>      which matches in test-inbook-tbms.bib. Because we have "maintitle"
>      here,
>      collectionTitle is populated by "booktitle" in this case.
>      I was trying to follow the logic of the old version here -- may have
>      misunderstood something.
>      +++ Nick Bart [Oct 05 13 02:32 ]:
>      >    As to the issues with the biblatex "series" field, though I
>      cannot
>      >    claim to fully understand the new code yet, what I see is that
>      while
>      >    *sometimes* the output is as expected (test-book-tms.bib
>      below), in
>      >    other cases it clearly isn't (test-inbook-tbms.bib):
>      >    $ cat test-book-tms.bib
>      >    @book{item1,
>      >        Author = {Author, Al},
>      >        Date = {2013},
>      >        Hyphenation = {french},
>      >        Location = {Location},
>      >        Mainsubtitle = {Mainsubtitle},
>      >        Maintitle = {Maintitle},
>      >        Maintitleaddon = {Maintitleaddon},
>      >        Number = {3},
>      >        Publisher = {Publisher},
>      >        Series = {Series},
>      >        Subtitle = {Subtitle},
>      >        Title = {Title of the Book},
>      >        Titleaddon = {Titleaddon}}
>      >    $ bibtex2pandoc test-book-tms.bib
>      >    ---
>      >    references:
>      >    - title: ! 'Title of the book: Subtitle. Titleaddon'
>      >      id: item1
>      >      issued:
>      >        year: 2013
>      >      author:
>      >        given:
>      >        - Al
>      >        family: Author
>      >      container-title: ! 'Maintitle: Mainsubtitle. Maintitleaddon'
>      >      publisher: Publisher
>      >      collection-title: Series
>      >      type: book
>      >      publisher-place: Location
>      >      language: fr-FR
>      >    ...
>      >    $ cat test-inbook-tbms.bib
>      >    @inbook{item1,
>      >        Author = {Author, Al},
>      >        Booksubtitle = {Booksubtitle},
>      >        Booktitle = {Booktitle},
>      >        Booktitleaddon = {Booktitleaddon},
>      >        Date = {2011},
>      >        Hyphenation = {french},
>      >        Location = {Location},
>      >        Mainsubtitle = {Mainsubtitle},
>      >        Maintitle = {Maintitle},
>      >        Maintitleaddon = {Maintitleaddon},
>      >        Number = {3},
>      >        Publisher = {Publisher},
>      >        Series = {Series},
>      >        Subtitle = {Subtitle},
>      >        Title = {Title of the "@inbook" Entry},
>      >        Titleaddon = {Titleaddon}}
>      >    $ bibtex2pandoc test-inbook-tbms.bib
>      >    ---
>      >    references:
>      >    - title: ! 'Title of the "@inbook" entry: Subtitle. Titleaddon'
>      >      id: item1
>      >      issued:
>      >        year: 2011
>      >      author:
>      >        given:
>      >        - Al
>      >        family: Author
>      >      container-title: ! 'Maintitle: Mainsubtitle. Maintitleaddon'
>      >      publisher: Publisher
>      >      collection-title: ! 'Booktitle: Booksubtitle. Booktitleaddon'
>      >      type: chapter
>      >      publisher-place: Location
>      >      language: fr-FR
>      >    ...
>      >
>      >    --
>      >    You received this message because you are subscribed to the
>      Google
>      >    Groups "pandoc-discuss" group.
>      >    To unsubscribe from this group and stop receiving emails from
>      it, send
>      >    an email to [1]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To post to this group, send email to
>      [2]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To view this discussion on the web visit
>      >    [1][3]https://groups.google.com/d/msgid/pandoc-discuss/
>      967cb63f-a612-4ded-
>      >    bc2b-6273c0239f68%[4]40googlegroups.com.
>      >    For more options, visit [2][5]https://groups.google.com/
>      groups/opt_out.
>      >
>      > References
>      >
>      >    1. [6]https://groups.google.com/d/
>      msgid/pandoc-discuss/967cb63f-a612-4ded-bc2b-6273c0239f68%
>      40googlegroups.com
>      >    2. [7]https://groups.google.com/groups/opt_out
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [8]https://groups.google.com/d/msgid/pandoc-discuss/733d7104-44b1-440a-
>    a386-ec356d2197bd%40googlegroups.com.
>    For more options, visit [9]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. javascript:/
>    2. javascript:/
>    3. https://groups.google.com/d/msgid/pandoc-discuss/967cb63f-a612-4ded-
>    4. http://40googlegroups.com/
>    5. https://groups.google.com/groups/opt_out
>    6. https://groups.google.com/d/msgid/pandoc-discuss/967cb63f-a612-4ded-bc2b-6273c0239f68%40googlegroups.com
>    7. https://groups.google.com/groups/opt_out
>    8. https://groups.google.com/d/msgid/pandoc-discuss/733d7104-44b1-440a-a386-ec356d2197bd%40googlegroups.com
>    9. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've removed bibtex2pandoc and merged the code into a new library
module, Text.CSL.Input.Bibtex.

This is now used by pandoc-citeproc instead of bibutils, when
you specify a .bib or .bibtex bibliography.

You can use biblio2yaml to do what bibtex2pandoc used to do (convert
bibtex or biblatex to yaml).

John

+++ John MacFarlane [Oct 06 13 21:43 ]:
> OK, I've merged in a number of changes Andrea made to citeproc-hs
> after I split off pandoc-citeproc.  These fix some bugs, including
> #90, #93, #95, #86.  volumeTitle has been added to Reference, so
> I think I'm now doing the right thing with volumeTitle and
> containerTitle, but as always your testing will be invaluable!
> 
> +++ Nick Bart [Oct 05 13 23:29 ]:
> >    Well, "collectionTitle" should never be populated by anything but the
> >    biblatex "series" field.
> >    biblatex "title", "booktitle", and "maintitle" can only ever map to CSL
> >    "title", "container-title", and the new "volume-title", and if you do
> >    not want to use the latter yet (I would!), "booktitle" must be
> >    discarded until "volume-title" has become part of the CSL specs.
> >    On Sunday, October 6, 2013 12:58:56 AM UTC, fiddlosopher wrote:
> > 
> >      The difference is due to the line
> >         (getTitle lang "maintitle" >> guard hasVolumes >> getTitle lang
> >      "booktitle")
> >      which matches in test-inbook-tbms.bib. Because we have "maintitle"
> >      here,
> >      collectionTitle is populated by "booktitle" in this case.
> >      I was trying to follow the logic of the old version here -- may have
> >      misunderstood something.
> >      +++ Nick Bart [Oct 05 13 02:32 ]:
> >      >    As to the issues with the biblatex "series" field, though I
> >      cannot
> >      >    claim to fully understand the new code yet, what I see is that
> >      while
> >      >    *sometimes* the output is as expected (test-book-tms.bib
> >      below), in
> >      >    other cases it clearly isn't (test-inbook-tbms.bib):
> >      >    $ cat test-book-tms.bib
> >      >    @book{item1,
> >      >        Author = {Author, Al},
> >      >        Date = {2013},
> >      >        Hyphenation = {french},
> >      >        Location = {Location},
> >      >        Mainsubtitle = {Mainsubtitle},
> >      >        Maintitle = {Maintitle},
> >      >        Maintitleaddon = {Maintitleaddon},
> >      >        Number = {3},
> >      >        Publisher = {Publisher},
> >      >        Series = {Series},
> >      >        Subtitle = {Subtitle},
> >      >        Title = {Title of the Book},
> >      >        Titleaddon = {Titleaddon}}
> >      >    $ bibtex2pandoc test-book-tms.bib
> >      >    ---
> >      >    references:
> >      >    - title: ! 'Title of the book: Subtitle. Titleaddon'
> >      >      id: item1
> >      >      issued:
> >      >        year: 2013
> >      >      author:
> >      >        given:
> >      >        - Al
> >      >        family: Author
> >      >      container-title: ! 'Maintitle: Mainsubtitle. Maintitleaddon'
> >      >      publisher: Publisher
> >      >      collection-title: Series
> >      >      type: book
> >      >      publisher-place: Location
> >      >      language: fr-FR
> >      >    ...
> >      >    $ cat test-inbook-tbms.bib
> >      >    @inbook{item1,
> >      >        Author = {Author, Al},
> >      >        Booksubtitle = {Booksubtitle},
> >      >        Booktitle = {Booktitle},
> >      >        Booktitleaddon = {Booktitleaddon},
> >      >        Date = {2011},
> >      >        Hyphenation = {french},
> >      >        Location = {Location},
> >      >        Mainsubtitle = {Mainsubtitle},
> >      >        Maintitle = {Maintitle},
> >      >        Maintitleaddon = {Maintitleaddon},
> >      >        Number = {3},
> >      >        Publisher = {Publisher},
> >      >        Series = {Series},
> >      >        Subtitle = {Subtitle},
> >      >        Title = {Title of the "@inbook" Entry},
> >      >        Titleaddon = {Titleaddon}}
> >      >    $ bibtex2pandoc test-inbook-tbms.bib
> >      >    ---
> >      >    references:
> >      >    - title: ! 'Title of the "@inbook" entry: Subtitle. Titleaddon'
> >      >      id: item1
> >      >      issued:
> >      >        year: 2011
> >      >      author:
> >      >        given:
> >      >        - Al
> >      >        family: Author
> >      >      container-title: ! 'Maintitle: Mainsubtitle. Maintitleaddon'
> >      >      publisher: Publisher
> >      >      collection-title: ! 'Booktitle: Booksubtitle. Booktitleaddon'
> >      >      type: chapter
> >      >      publisher-place: Location
> >      >      language: fr-FR
> >      >    ...
> >      >
> >      >    --
> >      >    You received this message because you are subscribed to the
> >      Google
> >      >    Groups "pandoc-discuss" group.
> >      >    To unsubscribe from this group and stop receiving emails from
> >      it, send
> >      >    an email to [1]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >    To post to this group, send email to
> >      [2]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >    To view this discussion on the web visit
> >      >    [1][3]https://groups.google.com/d/msgid/pandoc-discuss/
> >      967cb63f-a612-4ded-
> >      >    bc2b-6273c0239f68%[4]40googlegroups.com.
> >      >    For more options, visit [2][5]https://groups.google.com/
> >      groups/opt_out.
> >      >
> >      > References
> >      >
> >      >    1. [6]https://groups.google.com/d/
> >      msgid/pandoc-discuss/967cb63f-a612-4ded-bc2b-6273c0239f68%
> >      40googlegroups.com
> >      >    2. [7]https://groups.google.com/groups/opt_out
> > 
> >    --
> >    You received this message because you are subscribed to the Google
> >    Groups "pandoc-discuss" group.
> >    To unsubscribe from this group and stop receiving emails from it, send
> >    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >    To view this discussion on the web visit
> >    [8]https://groups.google.com/d/msgid/pandoc-discuss/733d7104-44b1-440a-
> >    a386-ec356d2197bd%40googlegroups.com.
> >    For more options, visit [9]https://groups.google.com/groups/opt_out.
> > 
> > References
> > 
> >    1. javascript:/
> >    2. javascript:/
> >    3. https://groups.google.com/d/msgid/pandoc-discuss/967cb63f-a612-4ded-
> >    4. http://40googlegroups.com/
> >    5. https://groups.google.com/groups/opt_out
> >    6. https://groups.google.com/d/msgid/pandoc-discuss/967cb63f-a612-4ded-bc2b-6273c0239f68%40googlegroups.com
> >    7. https://groups.google.com/groups/opt_out
> >    8. https://groups.google.com/d/msgid/pandoc-discuss/733d7104-44b1-440a-a386-ec356d2197bd%40googlegroups.com
> >    9. https://groups.google.com/groups/opt_out
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131007044345.GA41791%40Johns-MacBook-Pro.local.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 704 bytes --]

Another regression: Titles of *non*-English entries are not preserved as 
is, but converted to lower case in the new version.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/cc118a6c-260a-4ed1-9eb9-4f5dab8bb879%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 967 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1405 bytes --]

Something curious: The family name "True" is causing problems:

$ cat test.md
Foo [@item1, p. 12].

# References

---
references:
- title: Title
  id: item1
  issued:
    year: 2012
  author:
    given:
    - Michael
    family: True
  type: article-journal
...

$ pandoc --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks test.md
pandoc-citeproc: Error parsing references: when expecting a String, 
encountered Boolean instead
pandoc: Error running filter pandoc-citeproc


In addition, are there other strings values besides "True" and "False" that 
would be interpreted as Boolean here? I get the same error message when 
using a larger bibfile.yaml containing 6000+ entries, successfully 
converted from a biblatex file, even with the "Michael True" entry, and a 
title beginning with "False" removed.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/928cefbc-dcf7-411c-8dee-e555e7cfaf6f%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1818 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

You can force interpretation as a YAML string:

   family: 'True'

+++ Nick Bart [Oct 06 13 05:26 ]:
>    Something curious: The family name "True" is causing problems:
>    $ cat test.md
>    Foo [@item1, p. 12].
>    # References
>    ---
>    references:
>    - title: Title
>      id: item1
>      issued:
>        year: 2012
>      author:
>        given:
>        - Michael
>        family: True
>      type: article-journal
>    ...
>    $ pandoc --filter pandoc-citeproc -t
>    markdown-citations-markdown_in_html_blocks test.md
>    pandoc-citeproc: Error parsing references: when expecting a String,
>    encountered Boolean instead
>    pandoc: Error running filter pandoc-citeproc
>    In addition, are there other strings values besides "True" and "False"
>    that would be interpreted as Boolean here? I get the same error message
>    when using a larger bibfile.yaml containing 6000+ entries, successfully
>    converted from a biblatex file, even with the "Michael True" entry, and
>    a title beginning with "False" removed.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/928cefbc-dcf7-411c-
>    8dee-e555e7cfaf6f%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/928cefbc-dcf7-411c-8dee-e555e7cfaf6f%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3322 bytes --]

Thank you – but isn't there a possibility to fix pandoc2bibtex (e.g., to 
enclose *all* names in single quotes) or, probably better, pandoc-citeproc 
(to prevent it from interpreting any fields as Boolean that are quite 
clearly always to be understood as non-Boolean; which would essentially 
mean all fields: I don't even think there *are* any Boolean fields 
currently in use)?

Also, are there any *other* strings except True/False that could be 
(mis-)understood as Boolean, and that could be causing trouble with my 
large database? Or any other ideas? Capacity issues, perhaps?

On Sunday, October 6, 2013 2:35:49 PM UTC, fiddlosopher wrote:
>
> You can force interpretation as a YAML string: 
>
>    family: 'True' 
>
> +++ Nick Bart [Oct 06 13 05:26 ]: 
> >    Something curious: The family name "True" is causing problems: 
> >    $ cat test.md 
> >    Foo [@item1, p. 12]. 
> >    # References 
> >    --- 
> >    references: 
> >    - title: Title 
> >      id: item1 
> >      issued: 
> >        year: 2012 
> >      author: 
> >        given: 
> >        - Michael 
> >        family: True 
> >      type: article-journal 
> >    ... 
> >    $ pandoc --filter pandoc-citeproc -t 
> >    markdown-citations-markdown_in_html_blocks test.md 
> >    pandoc-citeproc: Error parsing references: when expecting a String, 
> >    encountered Boolean instead 
> >    pandoc: Error running filter pandoc-citeproc 
> >    In addition, are there other strings values besides "True" and 
> "False" 
> >    that would be interpreted as Boolean here? I get the same error 
> message 
> >    when using a larger bibfile.yaml containing 6000+ entries, 
> successfully 
> >    converted from a biblatex file, even with the "Michael True" entry, 
> and 
> >    a title beginning with "False" removed. 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [1]
> https://groups.google.com/d/msgid/pandoc-discuss/928cefbc-dcf7-411c- 
> >    8dee-e555e7cfaf6f%40googlegroups.com. 
> >    For more options, visit [2]https://groups.google.com/groups/opt_out. 
> > 
> > References 
> > 
> >    1. 
> https://groups.google.com/d/msgid/pandoc-discuss/928cefbc-dcf7-411c-8dee-e555e7cfaf6f%40googlegroups.com 
> >    2. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/bfbc1a45-3031-4f2e-bf7f-0c14537011bf%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4952 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 06 13 08:11 ]:
>    Thank you but isn't there a possibility to fix pandoc2bibtex (e.g., to
>    enclose *all* names in single quotes) or, probably better,
>    pandoc-citeproc (to prevent it from interpreting any fields as Boolean
>    that are quite clearly always to be understood as non-Boolean; which
>    would essentially mean all fields: I don't even think there *are* any
>    Boolean fields currently in use)?
>    Also, are there any *other* strings except True/False that could be
>    (mis-)understood as Boolean, and that could be causing trouble with my
>    large database?

y|Y|yes|Yes|YES|n|N|no|No|NO
|true|True|TRUE|false|False|FALSE
|on|On|ON|off|Off|OFF

will all be interpreted as Boolean. Since this is standard YAML, there's
no way to prevent this (nor would it be desirable to add ad hoc exceptions.)

pandoc2bibtex ought to put the strings that can be interpreted as Booleans
in quotes.  But this work is done by the yaml library, so maybe I need to
file a bug report there.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Issue report here:
https://github.com/snoyberg/yaml/issues/31

+++ John MacFarlane [Oct 06 13 08:40 ]:
> +++ Nick Bart [Oct 06 13 08:11 ]:
> >    Thank you but isn't there a possibility to fix pandoc2bibtex (e.g., to
> >    enclose *all* names in single quotes) or, probably better,
> >    pandoc-citeproc (to prevent it from interpreting any fields as Boolean
> >    that are quite clearly always to be understood as non-Boolean; which
> >    would essentially mean all fields: I don't even think there *are* any
> >    Boolean fields currently in use)?
> >    Also, are there any *other* strings except True/False that could be
> >    (mis-)understood as Boolean, and that could be causing trouble with my
> >    large database?
> 
> y|Y|yes|Yes|YES|n|N|no|No|NO
> |true|True|TRUE|false|False|FALSE
> |on|On|ON|off|Off|OFF
> 
> will all be interpreted as Boolean. Since this is standard YAML, there's
> no way to prevent this (nor would it be desirable to add ad hoc exceptions.)
> 
> pandoc2bibtex ought to put the strings that can be interpreted as Booleans
> in quotes.  But this work is done by the yaml library, so maybe I need to
> file a bug report there.
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131006154016.GE89729%40Johns-MacBook-Pro.local.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2938 bytes --]

Thank you. The list of strings matching the pattern for Booleans was 
helpful: After enclosing all titles starting with "true", "false", and "on" 
in single quotes, and finally hunting down one

  - given:
    - Stephen
    - N
    family: Miller

and adding a period after the "N", pandoc-citeproc no longer throws the 
error described above.

But of course I'm looking forward to a more permanent fix on the yaml side.

On Sunday, October 6, 2013 3:47:47 PM UTC, fiddlosopher wrote:
>
> Issue report here: 
> https://github.com/snoyberg/yaml/issues/31 
>
> +++ John MacFarlane [Oct 06 13 08:40 ]: 
> > +++ Nick Bart [Oct 06 13 08:11 ]: 
> > >    Thank you but isn't there a possibility to fix pandoc2bibtex (e.g., 
> to 
> > >    enclose *all* names in single quotes) or, probably better, 
> > >    pandoc-citeproc (to prevent it from interpreting any fields as 
> Boolean 
> > >    that are quite clearly always to be understood as non-Boolean; 
> which 
> > >    would essentially mean all fields: I don't even think there *are* 
> any 
> > >    Boolean fields currently in use)? 
> > >    Also, are there any *other* strings except True/False that could be 
> > >    (mis-)understood as Boolean, and that could be causing trouble with 
> my 
> > >    large database? 
> > 
> > y|Y|yes|Yes|YES|n|N|no|No|NO 
> > |true|True|TRUE|false|False|FALSE 
> > |on|On|ON|off|Off|OFF 
> > 
> > will all be interpreted as Boolean. Since this is standard YAML, there's 
> > no way to prevent this (nor would it be desirable to add ad hoc 
> exceptions.) 
> > 
> > pandoc2bibtex ought to put the strings that can be interpreted as 
> Booleans 
> > in quotes.  But this work is done by the yaml library, so maybe I need 
> to 
> > file a bug report there. 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "pandoc-discuss" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/20131006154016.GE89729%40Johns-MacBook-Pro.local. 
>
> > For more options, visit https://groups.google.com/groups/opt_out. 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/7b1f2b56-a682-4026-930a-8d8956b7bb14%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4040 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

This is now fixed.

+++ Nick Bart [Oct 06 13 04:29 ]:
>    Another regression: Titles of *non*-English entries are not preserved
>    as is, but converted to lower case in the new version.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/cc118a6c-260a-4ed1-
>    9eb9-4f5dab8bb879%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/cc118a6c-260a-4ed1-9eb9-4f5dab8bb879%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 05 13 00:41 ]:
>    On the redone bibtex2pandoc: Nice. Glad to see that dates are
>    implemented now.
>    I noticed a few regressions, however:
>    -    , collectionTitle = volumeTitle' ++ volumeSubtitle' ++
>    volumeTitleAddon'
>    collection-title (I'll be using the CSL variable names here) must
>    receive the content of biblatex "series" field (unless the biblatex
>    entry type is "article", "periodical" or "suppperiodical", in which
>    case series is appended to CSL container-title)

It seems to me that the code should work this way already:

  volumeTitle' <- (getTitle lang "maintitle" >> guard hasVolumes
                    >> getTitle lang "booktitle")
                  <|> (guard (not isArticle) >> getTitle lang "series")
                  <|> return ""

Not that isArticle abbreviates "is article, periodical, or suppperiodical".

So the second clause should populate volumeTitle' with "series",
unless (a) the type is article, periodical, or suppperiodical,
or (b) "maintitle" is set, the type is inbook, incollection,
inproceedings, or bookinbook, and "booktitle" is set, in which
case this value is taken.

>    "volumetitle'" etc. goes to the (not yet official) CSL variable
>    "volume-title"

All we have in Reference is collectionTitle, so we have to make
do with that until Reference is expanded.

>    -     --, collectionNumber = undefined --:: String --Int
>      and
>          -- , issue               = undefined -- :: String
>    The biblatex "number" field is mapped to CSL "number",
>    - unless the entry type is "book", "collection", "proceedings",
>    "reference", "mvbook", "mvcollection", "mvproceedings", "mvreference",
>    "bookinbook", "inbook", "incollection", "inproceedings", "inreference",
>    "suppbook", "suppcollection", in which case it is mapped to CSL
>    "collection-number",

Should now be fixed.

>    - or the entry type is "article", "periodical" or "suppperiodical", in
>    which case the best solution, used in the previous version of
>    bibtex2pandoc, is to map the biblatex fields number *and* issue (in
>    that sequence, separated by a comma) together to CSL "issue".

Should now be fixed.

>    - When Journal and Journalsubtitle are combined, the ": " between them
>    is missing now.
>      container-title: JournalJournalsubtitle

Should now be fixed.

Re: Decoupling citeproc and highlighting-kate from pandoc

[fiddlosopher]

[-- Attachment #1: Type: text/plain, Size: 2635 bytes --]

I have redone bibtex2pandoc.hs so that it creates a list of Reference 
objects.  This will be easy to integrate into the pandoc-citeproc library, 
allowing use of bibtex databases without creating a YAML intermediary.

One suggestion, though: Readability would improve if (1) empty lines could 
> be added between entries and (2) id, title, author, editor, and date could 
> be moved to the top of each entry.
>

I think this could be done later with some postprocessing of the YAML. The 
yaml library doesn't have much configurability. 
 

> Apart from that, at this stage a few issues remain that I cannot fix alone:
>
> - date handling (date, urldate, etc. are essential fields in biblatex; 
> many biblatex users do not even use "year" fields any longer).
>

This has been implemented, hopefully correctly, in bibtex2pandoc.hs. 
 

> - handling of name prefixes and suffixes (I'll see what I can contribute)
>

Not currently correct, and a bit complex.  I'll see what I can do when I 
can find time.
 

> - handling of multiple publishers, locations, and other "literal list" 
> field components: Having "getLiteralList", as intended, split such fields 
> into components would be the cleanest solution, but pandoc-citeproc 
> currently does not accept arrays (thus I've mostly replaced 
> "getLiteralList" by "getField" for now), so it would have to be modified.
>

I'm not sure what the ramifications would be.  (Does CSL allow these things 
to be multiple?)

In addition, I came across two issues that are most likely pandoc-citeproc 
> issues:
>
> - Extra strings such as "Edited & translated by." in pandoc output. This 
> has been reported here in detail: 
> https://code.google.com/p/citeproc-hs/issues/detail?id=92
>

No idea yet, haven't really delved into it.  Maybe now that Andrea Rossato 
is back, he can see what is going wrong here.

- Very long titles in non-English languages:
>
> Spanish (French, German, too): Linebreak after title, before '.”
>

I was able to fix this.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/30423993-afdf-4a47-bb2b-8ae27c6130b6%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4072 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Bruce D'Arcus]

[-- Attachment #1: Type: text/plain, Size: 1431 bytes --]



On Saturday, October 5, 2013 1:19:55 AM UTC-4, fiddlosopher wrote:

...
 

> - handling of multiple publishers, locations, and other "literal list" 
>> field components: Having "getLiteralList", as intended, split such fields 
>> into components would be the cleanest solution, but pandoc-citeproc 
>> currently does not accept arrays (thus I've mostly replaced 
>> "getLiteralList" by "getField" for now), so it would have to be modified.
>>
>
> I'm not sure what the ramifications would be.  (Does CSL allow these 
> things to be multiple?)
>

The csl-data json schema is here (though with my previous caveat that it's 
not set in stone, and subject to change):

https://github.com/citation-style-language/schema/blob/master/csl-data.json

The answer: no.

There's a cost to complexity, and so where things can be simple strings, 
they are.

Bruce 

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/615c9846-d27b-4fa0-a549-40748f524a35%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2246 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2373 bytes --]

Thank you, Bruce, for clarifying.

Though I hope the CSL schema can be extended at some point - I'd like to 
remind of the fact that style guides differ in their requirements here, 
CMoS and APA demanding to print the first publisher and/or place only, ISO 
690 and MLA wanting all of them - the sensible thing for bibtex2pandoc, for 
now, would be to map all "and"s in bibtex/biblatex "literal list" fields, 
i.e., address, location, origlocation, school, institution, organization, 
howpublished, publisher, origpublisher - leaving proteced "{and}"s alone, 
of course - to something that is closer to the average formatting 
conventions, neutral with regard to language, and does not conflict with 
commas in place names, either:

Hence I would recommend - for now - mapping any "and" in the fields listed 
above to a semicolon.


On Saturday, October 5, 2013 4:52:48 PM UTC, Bruce D'Arcus wrote:
>
>
>
> On Saturday, October 5, 2013 1:19:55 AM UTC-4, fiddlosopher wrote:
>
> ...
>  
>
>> - handling of multiple publishers, locations, and other "literal list" 
>>> field components: Having "getLiteralList", as intended, split such fields 
>>> into components would be the cleanest solution, but pandoc-citeproc 
>>> currently does not accept arrays (thus I've mostly replaced 
>>> "getLiteralList" by "getField" for now), so it would have to be modified.
>>>
>>
>> I'm not sure what the ramifications would be.  (Does CSL allow these 
>> things to be multiple?)
>>
>
> The csl-data json schema is here (though with my previous caveat that it's 
> not set in stone, and subject to change):
>
> https://github.com/citation-style-language/schema/blob/master/csl-data.json
>
> The answer: no.
>
> There's a cost to complexity, and so where things can be simple strings, 
> they are.
>
> Bruce 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/b717ff1f-ab65-4f1f-85b9-05b7b5bcb9b6%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3326 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 06 13 01:35 ]:
>    Thank you, Bruce, for clarifying.
>    Though I hope the CSL schema can be extended at some point - I'd like
>    to remind of the fact that style guides differ in their requirements
>    here, CMoS and APA demanding to print the first publisher and/or place
>    only, ISO 690 and MLA wanting all of them - the sensible thing for
>    bibtex2pandoc, for now, would be to map all "and"s in bibtex/biblatex
>    "literal list" fields, i.e., address, location, origlocation, school,
>    institution, organization, howpublished, publisher, origpublisher -
>    leaving proteced "{and}"s alone, of course - to something that is
>    closer to the average formatting conventions, neutral with regard to
>    language, and does not conflict with commas in place names, either:
>    Hence I would recommend - for now - mapping any "and" in the fields
>    listed above to a semicolon.

Not so sure.  Do you really want

   publisher = {Harcourt, Brace, and Howe}

to be mapped to

   publisher: "Harcourt, Brace; Howe"?

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3060 bytes --]

Well, yes, if it's biblatex. It's a literal list field, and according to 
the biblatex specs, any unprotected "and" *separates* two list elements, 
it's not part of any of them (see biblatex manual, v2.7a, § 2.3.4, "Literal 
Lists")

If it's actually one publisher with an "and" in their name, the "and" needs 
to be protected by either wrapping it in braces:

   publisher = {Harcourt, Brace {and} Howe}

or by wrapping the entire name in braces:

   publisher = {{Harcourt, Brace and Howe}}

Both forms should of course be mapped to

   publisher: "Harcourt, Brace and Howe"

If it's indeed two publishers, we'd have:

   publisher = {Paternoster Press and Eerdmans}
   publisher = {Univ. of Toronto Press and Routledge {and} Kegan Paul}
   publisher = {Univ. of Toronto Press and {Routledge and Kegan Paul}}

which should come out as

   publisher: Paternoster Press; Eerdmans
   publisher: Univ. of Toronto Press; Routledge and Kegan Paul

Same for institution, organization, location are literal lists in terms of 
§ 2.2. 􏰁is also applies to origlocation, origpublisher and to the field 
aliases address and school.

   location = {Exeter and Grand Rapids}

is mapped to

   publisher-place = {Exeter; Grand Rapids}


On Monday, October 7, 2013 4:45:16 AM UTC, fiddlosopher wrote:
>
> +++ Nick Bart [Oct 06 13 01:35 ]: 
> >    Thank you, Bruce, for clarifying. 
> >    Though I hope the CSL schema can be extended at some point - I'd like 
> >    to remind of the fact that style guides differ in their requirements 
> >    here, CMoS and APA demanding to print the first publisher and/or 
> place 
> >    only, ISO 690 and MLA wanting all of them - the sensible thing for 
> >    bibtex2pandoc, for now, would be to map all "and"s in bibtex/biblatex 
> >    "literal list" fields, i.e., address, location, origlocation, school, 
> >    institution, organization, howpublished, publisher, origpublisher - 
> >    leaving proteced "{and}"s alone, of course - to something that is 
> >    closer to the average formatting conventions, neutral with regard to 
> >    language, and does not conflict with commas in place names, either: 
> >    Hence I would recommend - for now - mapping any "and" in the fields 
> >    listed above to a semicolon. 
>
> Not so sure.  Do you really want 
>
>    publisher = {Harcourt, Brace, and Howe} 
>
> to be mapped to 
>
>    publisher: "Harcourt, Brace; Howe"? 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a72be8d7-27fe-4560-ad38-739e4f1287d7%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3811 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

OK, done.

+++ Nick Bart [Oct 06 13 23:15 ]:
>    Well, yes, if it's biblatex. It's a literal list field, and according
>    to the biblatex specs, any unprotected "and" *separates* two list
>    elements, it's not part of any of them (see biblatex manual, v2.7a, §
>    2.3.4, "Literal Lists")
>    If it's actually one publisher with an "and" in their name, the "and"
>    needs to be protected by either wrapping it in braces:
>       publisher = {Harcourt, Brace {and} Howe}
>    or by wrapping the entire name in braces:
>       publisher = {{Harcourt, Brace and Howe}}
>    Both forms should of course be mapped to
>       publisher: "Harcourt, Brace and Howe"
>    If it's indeed two publishers, we'd have:
>       publisher = {Paternoster Press and Eerdmans}
>       publisher = {Univ. of Toronto Press and Routledge {and} Kegan Paul}
>       publisher = {Univ. of Toronto Press and {Routledge and Kegan Paul}}
>    which should come out as
>       publisher: Paternoster Press; Eerdmans
>       publisher: Univ. of Toronto Press; Routledge and Kegan Paul
>    Same for institution, organization, location are literal lists in terms
>    of § 2.2. 􏰁is also applies to origlocation, origpublisher and to the
>    field aliases address and school.
>       location = {Exeter and Grand Rapids}
>    is mapped to
>       publisher-place = {Exeter; Grand Rapids}
>    On Monday, October 7, 2013 4:45:16 AM UTC, fiddlosopher wrote:
> 
>      +++ Nick Bart [Oct 06 13 01:35 ]:
>      >    Thank you, Bruce, for clarifying.
>      >    Though I hope the CSL schema can be extended at some point -
>      I'd like
>      >    to remind of the fact that style guides differ in their
>      requirements
>      >    here, CMoS and APA demanding to print the first publisher
>      and/or place
>      >    only, ISO 690 and MLA wanting all of them - the sensible thing
>      for
>      >    bibtex2pandoc, for now, would be to map all "and"s in
>      bibtex/biblatex
>      >    "literal list" fields, i.e., address, location, origlocation,
>      school,
>      >    institution, organization, howpublished, publisher,
>      origpublisher -
>      >    leaving proteced "{and}"s alone, of course - to something that
>      is
>      >    closer to the average formatting conventions, neutral with
>      regard to
>      >    language, and does not conflict with commas in place names,
>      either:
>      >    Hence I would recommend - for now - mapping any "and" in the
>      fields
>      >    listed above to a semicolon.
>      Not so sure.  Do you really want
>         publisher = {Harcourt, Brace, and Howe}
>      to be mapped to
>         publisher: "Harcourt, Brace; Howe"?
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/a72be8d7-27fe-4560-
>    ad38-739e4f1287d7%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/a72be8d7-27fe-4560-ad38-739e4f1287d7%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131008031419.GA49273%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4489 bytes --]

Great!

On Tuesday, October 8, 2013 3:14:19 AM UTC, fiddlosopher wrote:
>
> OK, done. 
>
> +++ Nick Bart [Oct 06 13 23:15 ]: 
> >    Well, yes, if it's biblatex. It's a literal list field, and according 
> >    to the biblatex specs, any unprotected "and" *separates* two list 
> >    elements, it's not part of any of them (see biblatex manual, v2.7a, § 
> >    2.3.4, "Literal Lists") 
> >    If it's actually one publisher with an "and" in their name, the "and" 
> >    needs to be protected by either wrapping it in braces: 
> >       publisher = {Harcourt, Brace {and} Howe} 
> >    or by wrapping the entire name in braces: 
> >       publisher = {{Harcourt, Brace and Howe}} 
> >    Both forms should of course be mapped to 
> >       publisher: "Harcourt, Brace and Howe" 
> >    If it's indeed two publishers, we'd have: 
> >       publisher = {Paternoster Press and Eerdmans} 
> >       publisher = {Univ. of Toronto Press and Routledge {and} Kegan 
> Paul} 
> >       publisher = {Univ. of Toronto Press and {Routledge and Kegan 
> Paul}} 
> >    which should come out as 
> >       publisher: Paternoster Press; Eerdmans 
> >       publisher: Univ. of Toronto Press; Routledge and Kegan Paul 
> >    Same for institution, organization, location are literal lists in 
> terms 
> >    of § 2.2. 􏰁is also applies to origlocation, origpublisher and to the 
> >    field aliases address and school. 
> >       location = {Exeter and Grand Rapids} 
> >    is mapped to 
> >       publisher-place = {Exeter; Grand Rapids} 
> >    On Monday, October 7, 2013 4:45:16 AM UTC, fiddlosopher wrote: 
> > 
> >      +++ Nick Bart [Oct 06 13 01:35 ]: 
> >      >    Thank you, Bruce, for clarifying. 
> >      >    Though I hope the CSL schema can be extended at some point - 
> >      I'd like 
> >      >    to remind of the fact that style guides differ in their 
> >      requirements 
> >      >    here, CMoS and APA demanding to print the first publisher 
> >      and/or place 
> >      >    only, ISO 690 and MLA wanting all of them - the sensible thing 
> >      for 
> >      >    bibtex2pandoc, for now, would be to map all "and"s in 
> >      bibtex/biblatex 
> >      >    "literal list" fields, i.e., address, location, origlocation, 
> >      school, 
> >      >    institution, organization, howpublished, publisher, 
> >      origpublisher - 
> >      >    leaving proteced "{and}"s alone, of course - to something that 
> >      is 
> >      >    closer to the average formatting conventions, neutral with 
> >      regard to 
> >      >    language, and does not conflict with commas in place names, 
> >      either: 
> >      >    Hence I would recommend - for now - mapping any "and" in the 
> >      fields 
> >      >    listed above to a semicolon. 
> >      Not so sure.  Do you really want 
> >         publisher = {Harcourt, Brace, and Howe} 
> >      to be mapped to 
> >         publisher: "Harcourt, Brace; Howe"? 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [1]
> https://groups.google.com/d/msgid/pandoc-discuss/a72be8d7-27fe-4560- 
> >    ad38-739e4f1287d7%40googlegroups.com. 
> >    For more options, visit [2]https://groups.google.com/groups/opt_out. 
> > 
> > References 
> > 
> >    1. 
> https://groups.google.com/d/msgid/pandoc-discuss/a72be8d7-27fe-4560-ad38-739e4f1287d7%40googlegroups.com 
> >    2. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/310ae93b-889b-44af-9da7-97e071c23936%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 6740 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1213 bytes --]

Now, since yesterday pandoc doesn't build any longer:

pandoc $ cabal install --enable-tests --force-reinstalls
Resolving dependencies...
Warning: The following packages are likely to be broken by the reinstalls:
pandoc-citeproc-0.1.1.2
Continuing even though the plan contains dangerous reinstalls.
Configuring pandoc-1.12.0.2...
Building pandoc-1.12.0.2...
setup: internal error: componentsInBuildOrder: no such component: CExeName 
"make-pandoc-man-pages"
Failed to install pandoc-1.12.0.2
cabal: Error: some packages failed to install:
pandoc-1.12.0.2 failed during the building phase. The exception was:
ExitFailure 1

Any ideas?

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1521 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1560 bytes --]

And another two issues, with one example to demonstrate both:
- Corporate names are, again, broken
- Title: Strings wrapped in braces are not protected from being converted 
to lowercase

$ cat test-corporate-author.bib 

@report{item1,
    Author = {{National Aeronautics and Space Administration} and Author, 
Alfred and {Foo and Bar}},
    Date = {2013},
    Hyphenation = {USenglish},
    Location = {Place1 and Place2},
    Title = {The Title, Including a {Proper Name} and an {ACRNM}}}
$ ../pandoc-citeproc/dist/build/biblio2yaml/biblio2yaml 
test-corporate-author.bib 
---
references:
- title: The title, including a proper name and an acrnm
  id: item1
  issued:
    year: 2013
  author:
  - given:
    - National
    family: Aeronautics
  - given:
    - Space
    family: Administration
  - given:
    - Alfred
    family: Author
  - given:
    - ''
    family: Foo
  - given:
    - ''
    family: Bar
  type: report
  publisher-place: Place1; Place2
  language: en-US
...

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/3e0e0fc9-ea6c-454e-a12a-5562d48c94cb%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2234 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

[-- Attachment #1: Type: text/plain, Size: 2437 bytes --]

You need to install the latest pandoc from the github repo for this to work.  Then reinstall pandoc-citeproc


> On Oct 7, 2013, at 11:39 PM, Nick Bart <nickbart1980-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> 
> And another two issues, with one example to demonstrate both:
> - Corporate names are, again, broken
> - Title: Strings wrapped in braces are not protected from being converted to lowercase
> 
> $ cat test-corporate-author.bib 
> 
> @report{item1,
>     Author = {{National Aeronautics and Space Administration} and Author, Alfred and {Foo and Bar}},
>     Date = {2013},
>     Hyphenation = {USenglish},
>     Location = {Place1 and Place2},
>     Title = {The Title, Including a {Proper Name} and an {ACRNM}}}
> $ ../pandoc-citeproc/dist/build/biblio2yaml/biblio2yaml test-corporate-author.bib 
> ---
> references:
> - title: The title, including a proper name and an acrnm
>   id: item1
>   issued:
>     year: 2013
>   author:
>   - given:
>     - National
>     family: Aeronautics
>   - given:
>     - Space
>     family: Administration
>   - given:
>     - Alfred
>     family: Author
>   - given:
>     - ''
>     family: Foo
>   - given:
>     - ''
>     family: Bar
>   type: report
>   publisher-place: Place1; Place2
>   language: en-US
> ...
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/3e0e0fc9-ea6c-454e-a12a-5562d48c94cb%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/432132EE-5C00-4E9A-AB31-63A3E95ED949%40gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3639 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

[-- Attachment #1: Type: text/plain, Size: 1977 bytes --]

Do cabal clean first, then it should build


> On Oct 7, 2013, at 11:26 PM, Nick Bart <nickbart1980-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> 
> Now, since yesterday pandoc doesn't build any longer:
> 
> pandoc $ cabal install --enable-tests --force-reinstalls
> Resolving dependencies...
> Warning: The following packages are likely to be broken by the reinstalls:
> pandoc-citeproc-0.1.1.2
> Continuing even though the plan contains dangerous reinstalls.
> Configuring pandoc-1.12.0.2...
> Building pandoc-1.12.0.2...
> setup: internal error: componentsInBuildOrder: no such component: CExeName "make-pandoc-man-pages"
> Failed to install pandoc-1.12.0.2
> cabal: Error: some packages failed to install:
> pandoc-1.12.0.2 failed during the building phase. The exception was:
> ExitFailure 1
> 
> Any ideas?
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/9910B96C-195A-4D16-B944-1A1314A94F7F%40gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2855 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2111 bytes --]

Hmm, no. Any other ideas?

On Tuesday, October 8, 2013 1:39:05 PM UTC, fiddlosopher wrote:
>
> Do cabal clean first, then it should build
>
>
> On Oct 7, 2013, at 11:26 PM, Nick Bart <nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org <javascript:>> 
> wrote:
>
> Now, since yesterday pandoc doesn't build any longer:
>
> pandoc $ cabal install --enable-tests --force-reinstalls
> Resolving dependencies...
> Warning: The following packages are likely to be broken by the reinstalls:
> pandoc-citeproc-0.1.1.2
> Continuing even though the plan contains dangerous reinstalls.
> Configuring pandoc-1.12.0.2...
> Building pandoc-1.12.0.2...
> setup: internal error: componentsInBuildOrder: no such component: CExeName 
> "make-pandoc-man-pages"
> Failed to install pandoc-1.12.0.2
> cabal: Error: some packages failed to install:
> pandoc-1.12.0.2 failed during the building phase. The exception was:
> ExitFailure 1
>
> Any ideas?
>
> -- 
> You received this message because you are subscribed to the Google Groups 
> "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>.
> To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>
> .
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com
> .
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3041 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Try removing the dist subdirectory entirely, if 'cabal clean'
doesn't do that.

+++ Nick Bart [Oct 08 13 10:45 ]:
>    Hmm, no. Any other ideas?
>    On Tuesday, October 8, 2013 1:39:05 PM UTC, fiddlosopher wrote:
> 
>    Do cabal clean first, then it should build
>    On Oct 7, 2013, at 11:26 PM, Nick Bart <[1]nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> 
>    Now, since yesterday pandoc doesn't build any longer:
>    pandoc $ cabal install --enable-tests --force-reinstalls
>    Resolving dependencies...
>    Warning: The following packages are likely to be broken by the
>    reinstalls:
>    pandoc-citeproc-0.1.1.2
>    Continuing even though the plan contains dangerous reinstalls.
>    Configuring pandoc-1.12.0.2...
>    Building pandoc-1.12.0.2...
>    setup: internal error: componentsInBuildOrder: no such component:
>    CExeName "make-pandoc-man-pages"
>    Failed to install pandoc-1.12.0.2
>    cabal: Error: some packages failed to install:
>    pandoc-1.12.0.2 failed during the building phase. The exception was:
>    ExitFailure 1
>    Any ideas?
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [4]https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-
>    99a6-448c-8967-6d37bdba58ad%40googlegroups.com.
>    For more options, visit [5]https://groups.google.com/groups/opt_out.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [6]https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-
>    a60b-6bba4777cbf4%40googlegroups.com.
>    For more options, visit [7]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. javascript:/
>    2. javascript:/
>    3. javascript:/
>    4. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com
>    5. https://groups.google.com/groups/opt_out
>    6. https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%40googlegroups.com
>    7. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3870 bytes --]

Sorry but no, not even removing the complete pandoc, pandoc-citeproc, and 
pandoc-types directories and a fresh

git clone git://github.com/jgm/pandoc-types
git clone git://github.com/jgm/pandoc-citeproc
git clone git://github.com/jgm/pandoc

did the trick. Anything else I could try?


On Tuesday, October 8, 2013 6:10:41 PM UTC, fiddlosopher wrote:
>
> Try removing the dist subdirectory entirely, if 'cabal clean' 
> doesn't do that. 
>
> +++ Nick Bart [Oct 08 13 10:45 ]: 
> >    Hmm, no. Any other ideas? 
> >    On Tuesday, October 8, 2013 1:39:05 PM UTC, fiddlosopher wrote: 
> > 
> >    Do cabal clean first, then it should build 
> >    On Oct 7, 2013, at 11:26 PM, Nick Bart <[1]nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 
> wrote: 
> > 
> >    Now, since yesterday pandoc doesn't build any longer: 
> >    pandoc $ cabal install --enable-tests --force-reinstalls 
> >    Resolving dependencies... 
> >    Warning: The following packages are likely to be broken by the 
> >    reinstalls: 
> >    pandoc-citeproc-0.1.1.2 
> >    Continuing even though the plan contains dangerous reinstalls. 
> >    Configuring pandoc-1.12.0.2... 
> >    Building pandoc-1.12.0.2... 
> >    setup: internal error: componentsInBuildOrder: no such component: 
> >    CExeName "make-pandoc-man-pages" 
> >    Failed to install pandoc-1.12.0.2 
> >    cabal: Error: some packages failed to install: 
> >    pandoc-1.12.0.2 failed during the building phase. The exception was: 
> >    ExitFailure 1 
> >    Any ideas? 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >    To post to this group, send email to [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >    To view this discussion on the web visit 
> >    [4]https://groups.google.com/d/msgid/pandoc-discuss/28779dd0- 
> >    99a6-448c-8967-6d37bdba58ad%40googlegroups.com. 
> >    For more options, visit [5]https://groups.google.com/groups/opt_out. 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [6]
> https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a- 
> >    a60b-6bba4777cbf4%40googlegroups.com. 
> >    For more options, visit [7]https://groups.google.com/groups/opt_out. 
> > 
> > References 
> > 
> >    1. javascript:/ 
> >    2. javascript:/ 
> >    3. javascript:/ 
> >    4. 
> https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com 
> >    5. https://groups.google.com/groups/opt_out 
> >    6. 
> https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%40googlegroups.com 
> >    7. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 6019 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Very strange.  I just installed with no problems.  What OS
are you using?  What does 'cabal --version' say?

+++ Nick Bart [Oct 08 13 11:33 ]:
>    Sorry but no, not even removing the complete pandoc, pandoc-citeproc,
>    and pandoc-types directories and a fresh
>    git clone git://github.com/jgm/pandoc-types
>    git clone git://github.com/jgm/pandoc-citeproc
>    git clone git://github.com/jgm/pandoc
>    did the trick. Anything else I could try?
>    On Tuesday, October 8, 2013 6:10:41 PM UTC, fiddlosopher wrote:
> 
>      Try removing the dist subdirectory entirely, if 'cabal clean'
>      doesn't do that.
>      +++ Nick Bart [Oct 08 13 10:45 ]:
>      >    Hmm, no. Any other ideas?
>      >    On Tuesday, October 8, 2013 1:39:05 PM UTC, fiddlosopher wrote:
>      >
>      >    Do cabal clean first, then it should build
>      >    On Oct 7, 2013, at 11:26 PM, Nick Bart <[1]nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
>      wrote:
>      >
>      >    Now, since yesterday pandoc doesn't build any longer:
>      >    pandoc $ cabal install --enable-tests --force-reinstalls
>      >    Resolving dependencies...
>      >    Warning: The following packages are likely to be broken by the
>      >    reinstalls:
>      >    pandoc-citeproc-0.1.1.2
>      >    Continuing even though the plan contains dangerous reinstalls.
>      >    Configuring pandoc-1.12.0.2...
>      >    Building pandoc-1.12.0.2...
>      >    setup: internal error: componentsInBuildOrder: no such
>      component:
>      >    CExeName "make-pandoc-man-pages"
>      >    Failed to install pandoc-1.12.0.2
>      >    cabal: Error: some packages failed to install:
>      >    pandoc-1.12.0.2 failed during the building phase. The exception
>      was:
>      >    ExitFailure 1
>      >    Any ideas?
>      >
>      >    --
>      >    You received this message because you are subscribed to the
>      Google
>      >    Groups "pandoc-discuss" group.
>      >    To unsubscribe from this group and stop receiving emails from
>      it, send
>      >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To post to this group, send email to
>      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To view this discussion on the web visit
>      >    [4][1]https://groups.google.com/d/msgid/pandoc-discuss/
>      28779dd0-
>      >    99a6-448c-8967-6d37bdba58ad%[2]40googlegroups.com.
>      >    For more options, visit [5][3]https://groups.google.com/
>      groups/opt_out.
>      >
>      >    --
>      >    You received this message because you are subscribed to the
>      Google
>      >    Groups "pandoc-discuss" group.
>      >    To unsubscribe from this group and stop receiving emails from
>      it, send
>      >    an email to [4]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To post to this group, send email to
>      [5]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To view this discussion on the web visit
>      >    [6][6]https://groups.google.com/d/msgid/pandoc-discuss/
>      63edb53b-b03c-468a-
>      >    a60b-6bba4777cbf4%[7]40googlegroups.com.
>      >    For more options, visit [7][8]https://groups.google.com/
>      groups/opt_out.
>      >
>      > References
>      >
>      >    1. javascript:/
>      >    2. javascript:/
>      >    3. javascript:/
>      >    4. [9]https://groups.google.com/d/
>      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%
>      40googlegroups.com
>      >    5. [10]https://groups.google.com/groups/opt_out
>      >    6. [11]https://groups.google.com/d/
>      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%
>      40googlegroups.com
>      >    7. [12]https://groups.google.com/groups/opt_out
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [13]https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d
>    -8269-80981156bddc%40googlegroups.com.
>    For more options, visit [14]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-
>    2. http://40googlegroups.com/
>    3. https://groups.google.com/groups/opt_out
>    4. javascript:/
>    5. javascript:/
>    6. https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-
>    7. http://40googlegroups.com/
>    8. https://groups.google.com/groups/opt_out
>    9. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com
>   10. https://groups.google.com/groups/opt_out
>   11. https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%40googlegroups.com
>   12. https://groups.google.com/groups/opt_out
>   13. https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%40googlegroups.com
>   14. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 6441 bytes --]

Mac OS X 10.7.5

$ cabal --version
cabal-install version 1.18.0.2
using version 1.18.1 of the Cabal library 

On Tuesday, October 8, 2013 6:56:02 PM UTC, fiddlosopher wrote:
>
> Very strange.  I just installed with no problems.  What OS 
> are you using?  What does 'cabal --version' say? 
>
> +++ Nick Bart [Oct 08 13 11:33 ]: 
> >    Sorry but no, not even removing the complete pandoc, pandoc-citeproc, 
> >    and pandoc-types directories and a fresh 
> >    git clone git://github.com/jgm/pandoc-types 
> >    git clone git://github.com/jgm/pandoc-citeproc 
> >    git clone git://github.com/jgm/pandoc 
> >    did the trick. Anything else I could try? 
> >    On Tuesday, October 8, 2013 6:10:41 PM UTC, fiddlosopher wrote: 
> > 
> >      Try removing the dist subdirectory entirely, if 'cabal clean' 
> >      doesn't do that. 
> >      +++ Nick Bart [Oct 08 13 10:45 ]: 
> >      >    Hmm, no. Any other ideas? 
> >      >    On Tuesday, October 8, 2013 1:39:05 PM UTC, fiddlosopher 
> wrote: 
> >      > 
> >      >    Do cabal clean first, then it should build 
> >      >    On Oct 7, 2013, at 11:26 PM, Nick Bart <[1]nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 
>
> >      wrote: 
> >      > 
> >      >    Now, since yesterday pandoc doesn't build any longer: 
> >      >    pandoc $ cabal install --enable-tests --force-reinstalls 
> >      >    Resolving dependencies... 
> >      >    Warning: The following packages are likely to be broken by the 
> >      >    reinstalls: 
> >      >    pandoc-citeproc-0.1.1.2 
> >      >    Continuing even though the plan contains dangerous reinstalls. 
> >      >    Configuring pandoc-1.12.0.2... 
> >      >    Building pandoc-1.12.0.2... 
> >      >    setup: internal error: componentsInBuildOrder: no such 
> >      component: 
> >      >    CExeName "make-pandoc-man-pages" 
> >      >    Failed to install pandoc-1.12.0.2 
> >      >    cabal: Error: some packages failed to install: 
> >      >    pandoc-1.12.0.2 failed during the building phase. The 
> exception 
> >      was: 
> >      >    ExitFailure 1 
> >      >    Any ideas? 
> >      > 
> >      >    -- 
> >      >    You received this message because you are subscribed to the 
> >      Google 
> >      >    Groups "pandoc-discuss" group. 
> >      >    To unsubscribe from this group and stop receiving emails from 
> >      it, send 
> >      >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >    To post to this group, send email to 
> >      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >    To view this discussion on the web visit 
> >      >    [4][1]https://groups.google.com/d/msgid/pandoc-discuss/ 
> >      28779dd0- 
> >      >    99a6-448c-8967-6d37bdba58ad%[2]40googlegroups.com. 
> >      >    For more options, visit [5][3]https://groups.google.com/ 
> >      groups/opt_out. 
> >      > 
> >      >    -- 
> >      >    You received this message because you are subscribed to the 
> >      Google 
> >      >    Groups "pandoc-discuss" group. 
> >      >    To unsubscribe from this group and stop receiving emails from 
> >      it, send 
> >      >    an email to [4]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >    To post to this group, send email to 
> >      [5]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >    To view this discussion on the web visit 
> >      >    [6][6]https://groups.google.com/d/msgid/pandoc-discuss/ 
> >      63edb53b-b03c-468a- 
> >      >    a60b-6bba4777cbf4%[7]40googlegroups.com. 
> >      >    For more options, visit [7][8]https://groups.google.com/ 
> >      groups/opt_out. 
> >      > 
> >      > References 
> >      > 
> >      >    1. javascript:/ 
> >      >    2. javascript:/ 
> >      >    3. javascript:/ 
> >      >    4. [9]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad% 
> >      40googlegroups.com 
> >      >    5. [10]https://groups.google.com/groups/opt_out 
> >      >    6. [11]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4% 
> >      40googlegroups.com 
> >      >    7. [12]https://groups.google.com/groups/opt_out 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [13]
> https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d 
> >    -8269-80981156bddc%40googlegroups.com. 
> >    For more options, visit [14]https://groups.google.com/groups/opt_out. 
>
> > 
> > References 
> > 
> >    1. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0- 
> >    2. http://40googlegroups.com/ 
> >    3. https://groups.google.com/groups/opt_out 
> >    4. javascript:/ 
> >    5. javascript:/ 
> >    6. 
> https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a- 
> >    7. http://40googlegroups.com/ 
> >    8. https://groups.google.com/groups/opt_out 
> >    9. 
> https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com 
> >   10. https://groups.google.com/groups/opt_out 
> >   11. 
> https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%40googlegroups.com 
> >   12. https://groups.google.com/groups/opt_out 
> >   13. 
> https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%40googlegroups.com 
> >   14. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/8470226a-c385-42bd-a6ba-71037729c707%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 11870 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Even stranger.  I use roughly the same setup at home.

Why don't you send the output of

cabal clean && cabal install -v2


+++ Nick Bart [Oct 08 13 12:06 ]:
>    Mac OS X 10.7.5
>    $ cabal --version
>    cabal-install version 1.18.0.2
>    using version 1.18.1 of the Cabal library
>    On Tuesday, October 8, 2013 6:56:02 PM UTC, fiddlosopher wrote:
> 
>      Very strange.  I just installed with no problems.  What OS
>      are you using?  What does 'cabal --version' say?
>      +++ Nick Bart [Oct 08 13 11:33 ]:
>      >    Sorry but no, not even removing the complete pandoc,
>      pandoc-citeproc,
>      >    and pandoc-types directories and a fresh
>      >    git clone git://[1]github.com/jgm/pandoc-types
>      >    git clone git://[2]github.com/jgm/pandoc-citeproc
>      >    git clone git://[3]github.com/jgm/pandoc
>      >    did the trick. Anything else I could try?
>      >    On Tuesday, October 8, 2013 6:10:41 PM UTC, fiddlosopher wrote:
>      >
>      >      Try removing the dist subdirectory entirely, if 'cabal clean'
>      >      doesn't do that.
>      >      +++ Nick Bart [Oct 08 13 10:45 ]:
>      >      >    Hmm, no. Any other ideas?
>      >      >    On Tuesday, October 8, 2013 1:39:05 PM UTC, fiddlosopher
>      wrote:
>      >      >
>      >      >    Do cabal clean first, then it should build
>      >      >    On Oct 7, 2013, at 11:26 PM, Nick Bart
>      <[1]nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
>      >      wrote:
>      >      >
>      >      >    Now, since yesterday pandoc doesn't build any longer:
>      >      >    pandoc $ cabal install --enable-tests --force-reinstalls
>      >      >    Resolving dependencies...
>      >      >    Warning: The following packages are likely to be broken
>      by the
>      >      >    reinstalls:
>      >      >    pandoc-citeproc-0.1.1.2
>      >      >    Continuing even though the plan contains dangerous
>      reinstalls.
>      >      >    Configuring pandoc-1.12.0.2...
>      >      >    Building pandoc-1.12.0.2...
>      >      >    setup: internal error: componentsInBuildOrder: no such
>      >      component:
>      >      >    CExeName "make-pandoc-man-pages"
>      >      >    Failed to install pandoc-1.12.0.2
>      >      >    cabal: Error: some packages failed to install:
>      >      >    pandoc-1.12.0.2 failed during the building phase. The
>      exception
>      >      was:
>      >      >    ExitFailure 1
>      >      >    Any ideas?
>      >      >
>      >      >    --
>      >      >    You received this message because you are subscribed to
>      the
>      >      Google
>      >      >    Groups "pandoc-discuss" group.
>      >      >    To unsubscribe from this group and stop receiving emails
>      from
>      >      it, send
>      >      >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >    To post to this group, send email to
>      >      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >    To view this discussion on the web visit
>      >      >    [4][1][4]https://groups.google.
>      com/d/msgid/pandoc-discuss/
>      >      28779dd0-
>      >      >    99a6-448c-8967-6d37bdba58ad%[2][5]40googlegroups.com.
>      >      >    For more options, visit [5][3][6]https://groups.google.
>      com/
>      >      groups/opt_out.
>      >      >
>      >      >    --
>      >      >    You received this message because you are subscribed to
>      the
>      >      Google
>      >      >    Groups "pandoc-discuss" group.
>      >      >    To unsubscribe from this group and stop receiving emails
>      from
>      >      it, send
>      >      >    an email to [4]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >    To post to this group, send email to
>      >      [5]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >    To view this discussion on the web visit
>      >      >    [6][6][7]https://groups.google.
>      com/d/msgid/pandoc-discuss/
>      >      63edb53b-b03c-468a-
>      >      >    a60b-6bba4777cbf4%[7][8]40googlegroups.com.
>      >      >    For more options, visit [7][8][9]https://groups.google.
>      com/
>      >      groups/opt_out.
>      >      >
>      >      > References
>      >      >
>      >      >    1. javascript:/
>      >      >    2. javascript:/
>      >      >    3. javascript:/
>      >      >    4. [9][10]https://groups.google.com/d/
>      >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%
>      >      [11]40googlegroups.com
>      >      >    5. [10][12]https://groups.google.com/groups/opt_out
>      >      >    6. [11][13]https://groups.google.com/d/
>      >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%
>      >      [14]40googlegroups.com
>      >      >    7. [12][15]https://groups.google.com/groups/opt_out
>      >
>      >    --
>      >    You received this message because you are subscribed to the
>      Google
>      >    Groups "pandoc-discuss" group.
>      >    To unsubscribe from this group and stop receiving emails from
>      it, send
>      >    an email to [16]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To post to this group, send email to
>      [17]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To view this discussion on the web visit
>      >    [13][18]https://groups.google.com/d/msgid/pandoc-discuss/
>      233c21c4-cff5-452d
>      >    -8269-80981156bddc%[19]40googlegroups.com.
>      >    For more options, visit [14][20]https://groups.google.com/
>      groups/opt_out.
>      >
>      > References
>      >
>      >    1. [21]https://groups.google.com/d/
>      msgid/pandoc-discuss/28779dd0-
>      >    2. [22]http://40googlegroups.com/
>      >    3. [23]https://groups.google.com/groups/opt_out
>      >    4. javascript:/
>      >    5. javascript:/
>      >    6. [24]https://groups.google.com/d/
>      msgid/pandoc-discuss/63edb53b-b03c-468a-
>      >    7. [25]http://40googlegroups.com/
>      >    8. [26]https://groups.google.com/groups/opt_out
>      >    9. [27]https://groups.google.com/d/
>      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%
>      40googlegroups.com
>      >   10. [28]https://groups.google.com/groups/opt_out
>      >   11. [29]https://groups.google.com/d/
>      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%
>      40googlegroups.com
>      >   12. [30]https://groups.google.com/groups/opt_out
>      >   13. [31]https://groups.google.com/d/
>      msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%
>      40googlegroups.com
>      >   14. [32]https://groups.google.com/groups/opt_out
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [33]https://groups.google.com/d/msgid/pandoc-discuss/8470226a-c385-42bd
>    -a6ba-71037729c707%40googlegroups.com.
>    For more options, visit [34]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. http://github.com/jgm/pandoc-types
>    2. http://github.com/jgm/pandoc-citeproc
>    3. http://github.com/jgm/pandoc
>    4. https://groups.google.com/d/msgid/pandoc-discuss/
>    5. http://40googlegroups.com/
>    6. https://groups.google.com/
>    7. https://groups.google.com/d/msgid/pandoc-discuss/
>    8. http://40googlegroups.com/
>    9. https://groups.google.com/
>   10. https://groups.google.com/d/
>   11. http://40googlegroups.com/
>   12. https://groups.google.com/groups/opt_out
>   13. https://groups.google.com/d/
>   14. http://40googlegroups.com/
>   15. https://groups.google.com/groups/opt_out
>   16. javascript:/
>   17. javascript:/
>   18. https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d
>   19. http://40googlegroups.com/
>   20. https://groups.google.com/groups/opt_out
>   21. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-
>   22. http://40googlegroups.com/
>   23. https://groups.google.com/groups/opt_out
>   24. https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-
>   25. http://40googlegroups.com/
>   26. https://groups.google.com/groups/opt_out
>   27. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com
>   28. https://groups.google.com/groups/opt_out
>   29. https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%40googlegroups.com
>   30. https://groups.google.com/groups/opt_out
>   31. https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%40googlegroups.com
>   32. https://groups.google.com/groups/opt_out
>   33. https://groups.google.com/d/msgid/pandoc-discuss/8470226a-c385-42bd-a6ba-71037729c707%40googlegroups.com
>   34. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 10916 bytes --]

Alright: Output of cabal clean && cabal install -v2 and cabal clean && 
cabal install -v2 --force-reinstalls is at 
https://gist.github.com/anonymous/6892387.

A few more details: Haskell installed via homebrew:
brew install haskell-platform
brew install ghc
brew install icu4c

Also tried, as recommended by Joseph,
rm -rf ~/.pandoc ~/.cabal ~/.ghc ~/src/{pandoc,pandoc-types,pandoc-citeproc}
before reinstalling.

On Tuesday, 8 October 2013 20:49:14 UTC, fiddlosopher wrote:
>
> Even stranger.  I use roughly the same setup at home. 
>
> Why don't you send the output of 
>
> cabal clean && cabal install -v2 
>
>
> +++ Nick Bart [Oct 08 13 12:06 ]: 
> >    Mac OS X 10.7.5 
> >    $ cabal --version 
> >    cabal-install version 1.18.0.2 
> >    using version 1.18.1 of the Cabal library 
> >    On Tuesday, October 8, 2013 6:56:02 PM UTC, fiddlosopher wrote: 
> > 
> >      Very strange.  I just installed with no problems.  What OS 
> >      are you using?  What does 'cabal --version' say? 
> >      +++ Nick Bart [Oct 08 13 11:33 ]: 
> >      >    Sorry but no, not even removing the complete pandoc, 
> >      pandoc-citeproc, 
> >      >    and pandoc-types directories and a fresh 
> >      >    git clone git://[1]github.com/jgm/pandoc-types 
> >      >    git clone git://[2]github.com/jgm/pandoc-citeproc 
> >      >    git clone git://[3]github.com/jgm/pandoc 
> >      >    did the trick. Anything else I could try? 
> >      >    On Tuesday, October 8, 2013 6:10:41 PM UTC, fiddlosopher 
> wrote: 
> >      > 
> >      >      Try removing the dist subdirectory entirely, if 'cabal 
> clean' 
> >      >      doesn't do that. 
> >      >      +++ Nick Bart [Oct 08 13 10:45 ]: 
> >      >      >    Hmm, no. Any other ideas? 
> >      >      >    On Tuesday, October 8, 2013 1:39:05 PM UTC, 
> fiddlosopher 
> >      wrote: 
> >      >      > 
> >      >      >    Do cabal clean first, then it should build 
> >      >      >    On Oct 7, 2013, at 11:26 PM, Nick Bart 
> >      <[1]nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 
> >      >      wrote: 
> >      >      > 
> >      >      >    Now, since yesterday pandoc doesn't build any longer: 
> >      >      >    pandoc $ cabal install --enable-tests 
> --force-reinstalls 
> >      >      >    Resolving dependencies... 
> >      >      >    Warning: The following packages are likely to be broken 
> >      by the 
> >      >      >    reinstalls: 
> >      >      >    pandoc-citeproc-0.1.1.2 
> >      >      >    Continuing even though the plan contains dangerous 
> >      reinstalls. 
> >      >      >    Configuring pandoc-1.12.0.2... 
> >      >      >    Building pandoc-1.12.0.2... 
> >      >      >    setup: internal error: componentsInBuildOrder: no such 
> >      >      component: 
> >      >      >    CExeName "make-pandoc-man-pages" 
> >      >      >    Failed to install pandoc-1.12.0.2 
> >      >      >    cabal: Error: some packages failed to install: 
> >      >      >    pandoc-1.12.0.2 failed during the building phase. The 
> >      exception 
> >      >      was: 
> >      >      >    ExitFailure 1 
> >      >      >    Any ideas? 
> >      >      > 
> >      >      >    -- 
> >      >      >    You received this message because you are subscribed to 
> >      the 
> >      >      Google 
> >      >      >    Groups "pandoc-discuss" group. 
> >      >      >    To unsubscribe from this group and stop receiving 
> emails 
> >      from 
> >      >      it, send 
> >      >      >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >      >    To post to this group, send email to 
> >      >      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >      >    To view this discussion on the web visit 
> >      >      >    [4][1][4]https://groups.google. 
> >      com/d/msgid/pandoc-discuss/ 
> >      >      28779dd0- 
> >      >      >    99a6-448c-8967-6d37bdba58ad%[2][5]40googlegroups.com. 
> >      >      >    For more options, visit [5][3][6]https://groups.google. 
>
> >      com/ 
> >      >      groups/opt_out. 
> >      >      > 
> >      >      >    -- 
> >      >      >    You received this message because you are subscribed to 
> >      the 
> >      >      Google 
> >      >      >    Groups "pandoc-discuss" group. 
> >      >      >    To unsubscribe from this group and stop receiving 
> emails 
> >      from 
> >      >      it, send 
> >      >      >    an email to [4]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >      >    To post to this group, send email to 
> >      >      [5]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >      >    To view this discussion on the web visit 
> >      >      >    [6][6][7]https://groups.google. 
> >      com/d/msgid/pandoc-discuss/ 
> >      >      63edb53b-b03c-468a- 
> >      >      >    a60b-6bba4777cbf4%[7][8]40googlegroups.com. 
> >      >      >    For more options, visit [7][8][9]https://groups.google. 
>
> >      com/ 
> >      >      groups/opt_out. 
> >      >      > 
> >      >      > References 
> >      >      > 
> >      >      >    1. javascript:/ 
> >      >      >    2. javascript:/ 
> >      >      >    3. javascript:/ 
> >      >      >    4. [9][10]https://groups.google.com/d/ 
> >      >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad% 
> >      >      [11]40googlegroups.com 
> >      >      >    5. [10][12]https://groups.google.com/groups/opt_out 
> >      >      >    6. [11][13]https://groups.google.com/d/ 
> >      >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4% 
> >      >      [14]40googlegroups.com 
> >      >      >    7. [12][15]https://groups.google.com/groups/opt_out 
> >      > 
> >      >    -- 
> >      >    You received this message because you are subscribed to the 
> >      Google 
> >      >    Groups "pandoc-discuss" group. 
> >      >    To unsubscribe from this group and stop receiving emails from 
> >      it, send 
> >      >    an email to [16]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >    To post to this group, send email to 
> >      [17]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >    To view this discussion on the web visit 
> >      >    [13][18]https://groups.google.com/d/msgid/pandoc-discuss/ 
> >      233c21c4-cff5-452d 
> >      >    -8269-80981156bddc%[19]40googlegroups.com. 
> >      >    For more options, visit [14][20]https://groups.google.com/ 
> >      groups/opt_out. 
> >      > 
> >      > References 
> >      > 
> >      >    1. [21]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/28779dd0- 
> >      >    2. [22]http://40googlegroups.com/ 
> >      >    3. [23]https://groups.google.com/groups/opt_out 
> >      >    4. javascript:/ 
> >      >    5. javascript:/ 
> >      >    6. [24]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/63edb53b-b03c-468a- 
> >      >    7. [25]http://40googlegroups.com/ 
> >      >    8. [26]https://groups.google.com/groups/opt_out 
> >      >    9. [27]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad% 
> >      40googlegroups.com 
> >      >   10. [28]https://groups.google.com/groups/opt_out 
> >      >   11. [29]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4% 
> >      40googlegroups.com 
> >      >   12. [30]https://groups.google.com/groups/opt_out 
> >      >   13. [31]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc% 
> >      40googlegroups.com 
> >      >   14. [32]https://groups.google.com/groups/opt_out 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [33]
> https://groups.google.com/d/msgid/pandoc-discuss/8470226a-c385-42bd 
> >    -a6ba-71037729c707%40googlegroups.com. 
> >    For more options, visit [34]https://groups.google.com/groups/opt_out. 
>
> > 
> > References 
> > 
> >    1. http://github.com/jgm/pandoc-types 
> >    2. http://github.com/jgm/pandoc-citeproc 
> >    3. http://github.com/jgm/pandoc 
> >    4. https://groups.google.com/d/msgid/pandoc-discuss/ 
> >    5. http://40googlegroups.com/ 
> >    6. https://groups.google.com/ 
> >    7. https://groups.google.com/d/msgid/pandoc-discuss/ 
> >    8. http://40googlegroups.com/ 
> >    9. https://groups.google.com/ 
> >   10. https://groups.google.com/d/ 
> >   11. http://40googlegroups.com/ 
> >   12. https://groups.google.com/groups/opt_out 
> >   13. https://groups.google.com/d/ 
> >   14. http://40googlegroups.com/ 
> >   15. https://groups.google.com/groups/opt_out 
> >   16. javascript:/ 
> >   17. javascript:/ 
> >   18. 
> https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d 
> >   19. http://40googlegroups.com/ 
> >   20. https://groups.google.com/groups/opt_out 
> >   21. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0- 
> >   22. http://40googlegroups.com/ 
> >   23. https://groups.google.com/groups/opt_out 
> >   24. 
> https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a- 
> >   25. http://40googlegroups.com/ 
> >   26. https://groups.google.com/groups/opt_out 
> >   27. 
> https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com 
> >   28. https://groups.google.com/groups/opt_out 
> >   29. 
> https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%40googlegroups.com 
> >   30. https://groups.google.com/groups/opt_out 
> >   31. 
> https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%40googlegroups.com 
> >   32. https://groups.google.com/groups/opt_out 
> >   33. 
> https://groups.google.com/d/msgid/pandoc-discuss/8470226a-c385-42bd-a6ba-71037729c707%40googlegroups.com 
> >   34. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/86ec8122-f24b-4251-a8f3-89953c0cf7a5%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 22177 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Sure enough -- I had cabal-install 1.16 with Cabal 1.16.  When
I upgraded to 1.18 I got the same error you did.

+++ Nick Bart [Oct 08 13 14:58 ]:
>    Alright: Output of cabal clean && cabal install -v2 and cabal clean &&
>    cabal install -v2 --force-reinstalls is at
>    https://gist.github.com/anonymous/6892387.
>    A few more details: Haskell installed via homebrew:
>    brew install haskell-platform
>    brew install ghc
>    brew install icu4c
>    Also tried, as recommended by Joseph,
>    rm -rf ~/.pandoc ~/.cabal ~/.ghc
>    ~/src/{pandoc,pandoc-types,pandoc-citeproc}
>    before reinstalling.
>    On Tuesday, 8 October 2013 20:49:14 UTC, fiddlosopher wrote:
> 
>      Even stranger.  I use roughly the same setup at home.
>      Why don't you send the output of
>      cabal clean && cabal install -v2
>      +++ Nick Bart [Oct 08 13 12:06 ]:
>      >    Mac OS X 10.7.5
>      >    $ cabal --version
>      >    cabal-install version 1.18.0.2
>      >    using version 1.18.1 of the Cabal library
>      >    On Tuesday, October 8, 2013 6:56:02 PM UTC, fiddlosopher wrote:
>      >
>      >      Very strange.  I just installed with no problems.  What OS
>      >      are you using?  What does 'cabal --version' say?
>      >      +++ Nick Bart [Oct 08 13 11:33 ]:
>      >      >    Sorry but no, not even removing the complete pandoc,
>      >      pandoc-citeproc,
>      >      >    and pandoc-types directories and a fresh
>      >      >    git clone git://[1][1]github.com/jgm/pandoc-types
>      >      >    git clone git://[2][2]github.com/jgm/pandoc-citeproc
>      >      >    git clone git://[3][3]github.com/jgm/pandoc
>      >      >    did the trick. Anything else I could try?
>      >      >    On Tuesday, October 8, 2013 6:10:41 PM UTC, fiddlosopher
>      wrote:
>      >      >
>      >      >      Try removing the dist subdirectory entirely, if 'cabal
>      clean'
>      >      >      doesn't do that.
>      >      >      +++ Nick Bart [Oct 08 13 10:45 ]:
>      >      >      >    Hmm, no. Any other ideas?
>      >      >      >    On Tuesday, October 8, 2013 1:39:05 PM UTC,
>      fiddlosopher
>      >      wrote:
>      >      >      >
>      >      >      >    Do cabal clean first, then it should build
>      >      >      >    On Oct 7, 2013, at 11:26 PM, Nick Bart
>      >      <[1]nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
>      >      >      wrote:
>      >      >      >
>      >      >      >    Now, since yesterday pandoc doesn't build any
>      longer:
>      >      >      >    pandoc $ cabal install --enable-tests
>      --force-reinstalls
>      >      >      >    Resolving dependencies...
>      >      >      >    Warning: The following packages are likely to be
>      broken
>      >      by the
>      >      >      >    reinstalls:
>      >      >      >    pandoc-citeproc-0.1.1.2
>      >      >      >    Continuing even though the plan contains
>      dangerous
>      >      reinstalls.
>      >      >      >    Configuring pandoc-1.12.0.2...
>      >      >      >    Building pandoc-1.12.0.2...
>      >      >      >    setup: internal error: componentsInBuildOrder: no
>      such
>      >      >      component:
>      >      >      >    CExeName "make-pandoc-man-pages"
>      >      >      >    Failed to install pandoc-1.12.0.2
>      >      >      >    cabal: Error: some packages failed to install:
>      >      >      >    pandoc-1.12.0.2 failed during the building phase.
>      The
>      >      exception
>      >      >      was:
>      >      >      >    ExitFailure 1
>      >      >      >    Any ideas?
>      >      >      >
>      >      >      >    --
>      >      >      >    You received this message because you are
>      subscribed to
>      >      the
>      >      >      Google
>      >      >      >    Groups "pandoc-discuss" group.
>      >      >      >    To unsubscribe from this group and stop receiving
>      emails
>      >      from
>      >      >      it, send
>      >      >      >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >      >    To post to this group, send email to
>      >      >      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >      >    To view this discussion on the web visit
>      >      >      >    [4][1][4][4]https://groups.google.
>      >      com/d/msgid/pandoc-discuss/
>      >      >      28779dd0-
>      >      >      >    99a6-448c-8967-6d37bdba58ad%[
>      2][5][5]40googlegroups.com.
>      >      >      >    For more options, visit
>      [5][3][6][6]https://groups.google.
>      >      com/
>      >      >      groups/opt_out.
>      >      >      >
>      >      >      >    --
>      >      >      >    You received this message because you are
>      subscribed to
>      >      the
>      >      >      Google
>      >      >      >    Groups "pandoc-discuss" group.
>      >      >      >    To unsubscribe from this group and stop receiving
>      emails
>      >      from
>      >      >      it, send
>      >      >      >    an email to [4]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >      >    To post to this group, send email to
>      >      >      [5]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >      >    To view this discussion on the web visit
>      >      >      >    [6][6][7][7]https://groups.google.
>      >      com/d/msgid/pandoc-discuss/
>      >      >      63edb53b-b03c-468a-
>      >      >      >    a60b-6bba4777cbf4%[7][8][8]40googlegroups.com.
>      >      >      >    For more options, visit
>      [7][8][9][9]https://groups.google.
>      >      com/
>      >      >      groups/opt_out.
>      >      >      >
>      >      >      > References
>      >      >      >
>      >      >      >    1. javascript:/
>      >      >      >    2. javascript:/
>      >      >      >    3. javascript:/
>      >      >      >    4. [9][10][10]https://groups.google.com/d/
>      >      >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-
>      6d37bdba58ad%
>      >      >      [11][11]40googlegroups.com
>      >      >      >    5. [10][12][12]https://groups.google.
>      com/groups/opt_out
>      >      >      >    6. [11][13][13]https://groups.google.com/d/
>      >      >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-
>      6bba4777cbf4%
>      >      >      [14][14]40googlegroups.com
>      >      >      >    7. [12][15][15]https://groups.google.
>      com/groups/opt_out
>      >      >
>      >      >    --
>      >      >    You received this message because you are subscribed to
>      the
>      >      Google
>      >      >    Groups "pandoc-discuss" group.
>      >      >    To unsubscribe from this group and stop receiving emails
>      from
>      >      it, send
>      >      >    an email to [16]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >    To post to this group, send email to
>      >      [17]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >    To view this discussion on the web visit
>      >      >    [13][18][16]https://groups.google.com/d/msgid/pandoc-
>      discuss/
>      >      233c21c4-cff5-452d
>      >      >    -8269-80981156bddc%[19][17]40googlegroups.com.
>      >      >    For more options, visit
>      [14][20][18]https://groups.google.com/
>      >      groups/opt_out.
>      >      >
>      >      > References
>      >      >
>      >      >    1. [21][19]https://groups.google.com/d/
>      >      msgid/pandoc-discuss/28779dd0-
>      >      >    2. [22][20]http://40googlegroups.com/
>      >      >    3. [23][21]https://groups.google.com/groups/opt_out
>      >      >    4. javascript:/
>      >      >    5. javascript:/
>      >      >    6. [24][22]https://groups.google.com/d/
>      >      msgid/pandoc-discuss/63edb53b-b03c-468a-
>      >      >    7. [25][23]http://40googlegroups.com/
>      >      >    8. [26][24]https://groups.google.com/groups/opt_out
>      >      >    9. [27][25]https://groups.google.com/d/
>      >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%
>      >      [26]40googlegroups.com
>      >      >   10. [28][27]https://groups.google.com/groups/opt_out
>      >      >   11. [29][28]https://groups.google.com/d/
>      >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%
>      >      [29]40googlegroups.com
>      >      >   12. [30][30]https://groups.google.com/groups/opt_out
>      >      >   13. [31][31]https://groups.google.com/d/
>      >      msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%
>      >      [32]40googlegroups.com
>      >      >   14. [32][33]https://groups.google.com/groups/opt_out
>      >
>      >    --
>      >    You received this message because you are subscribed to the
>      Google
>      >    Groups "pandoc-discuss" group.
>      >    To unsubscribe from this group and stop receiving emails from
>      it, send
>      >    an email to [34]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To post to this group, send email to
>      [35]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To view this discussion on the web visit
>      >    [33][36]https://groups.google.com/d/msgid/pandoc-discuss/
>      8470226a-c385-42bd
>      >    -a6ba-71037729c707%[37]40googlegroups.com.
>      >    For more options, visit [34][38]https://groups.google.com/
>      groups/opt_out.
>      >
>      > References
>      >
>      >    1. [39]http://github.com/jgm/pandoc-types
>      >    2. [40]http://github.com/jgm/pandoc-citeproc
>      >    3. [41]http://github.com/jgm/pandoc
>      >    4. [42]https://groups.google.com/d/msgid/pandoc-discuss/
>      >    5. [43]http://40googlegroups.com/
>      >    6. [44]https://groups.google.com/
>      >    7. [45]https://groups.google.com/d/msgid/pandoc-discuss/
>      >    8. [46]http://40googlegroups.com/
>      >    9. [47]https://groups.google.com/
>      >   10. [48]https://groups.google.com/d/
>      >   11. [49]http://40googlegroups.com/
>      >   12. [50]https://groups.google.com/groups/opt_out
>      >   13. [51]https://groups.google.com/d/
>      >   14. [52]http://40googlegroups.com/
>      >   15. [53]https://groups.google.com/groups/opt_out
>      >   16. javascript:/
>      >   17. javascript:/
>      >   18. [54]https://groups.google.com/d/
>      msgid/pandoc-discuss/233c21c4-cff5-452d
>      >   19. [55]http://40googlegroups.com/
>      >   20. [56]https://groups.google.com/groups/opt_out
>      >   21. [57]https://groups.google.com/d/
>      msgid/pandoc-discuss/28779dd0-
>      >   22. [58]http://40googlegroups.com/
>      >   23. [59]https://groups.google.com/groups/opt_out
>      >   24. [60]https://groups.google.com/d/
>      msgid/pandoc-discuss/63edb53b-b03c-468a-
>      >   25. [61]http://40googlegroups.com/
>      >   26. [62]https://groups.google.com/groups/opt_out
>      >   27. [63]https://groups.google.com/d/
>      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%
>      40googlegroups.com
>      >   28. [64]https://groups.google.com/groups/opt_out
>      >   29. [65]https://groups.google.com/d/
>      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%
>      40googlegroups.com
>      >   30. [66]https://groups.google.com/groups/opt_out
>      >   31. [67]https://groups.google.com/d/
>      msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%
>      40googlegroups.com
>      >   32. [68]https://groups.google.com/groups/opt_out
>      >   33. [69]https://groups.google.com/d/
>      msgid/pandoc-discuss/8470226a-c385-42bd-a6ba-71037729c707%
>      40googlegroups.com
>      >   34. [70]https://groups.google.com/groups/opt_out
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [71]https://groups.google.com/d/msgid/pandoc-discuss/86ec8122-f24b-4251
>    -a8f3-89953c0cf7a5%40googlegroups.com.
>    For more options, visit [72]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. http://github.com/jgm/pandoc-types
>    2. http://github.com/jgm/pandoc-citeproc
>    3. http://github.com/jgm/pandoc
>    4. https://groups.google/
>    5. http://40googlegroups.com/
>    6. https://groups.google/
>    7. https://groups.google/
>    8. http://40googlegroups.com/
>    9. https://groups.google/
>   10. https://groups.google.com/d/
>   11. http://40googlegroups.com/
>   12. https://groups.google.com/groups/opt_out
>   13. https://groups.google.com/d/
>   14. http://40googlegroups.com/
>   15. https://groups.google.com/groups/opt_out
>   16. https://groups.google.com/d/msgid/pandoc-discuss/
>   17. http://40googlegroups.com/
>   18. https://groups.google.com/
>   19. https://groups.google.com/d/
>   20. http://40googlegroups.com/
>   21. https://groups.google.com/groups/opt_out
>   22. https://groups.google.com/d/
>   23. http://40googlegroups.com/
>   24. https://groups.google.com/groups/opt_out
>   25. https://groups.google.com/d/
>   26. http://40googlegroups.com/
>   27. https://groups.google.com/groups/opt_out
>   28. https://groups.google.com/d/
>   29. http://40googlegroups.com/
>   30. https://groups.google.com/groups/opt_out
>   31. https://groups.google.com/d/
>   32. http://40googlegroups.com/
>   33. https://groups.google.com/groups/opt_out
>   34. javascript:/
>   35. javascript:/
>   36. https://groups.google.com/d/msgid/pandoc-discuss/8470226a-c385-42bd
>   37. http://40googlegroups.com/
>   38. https://groups.google.com/groups/opt_out
>   39. http://github.com/jgm/pandoc-types
>   40. http://github.com/jgm/pandoc-citeproc
>   41. http://github.com/jgm/pandoc
>   42. https://groups.google.com/d/msgid/pandoc-discuss/
>   43. http://40googlegroups.com/
>   44. https://groups.google.com/
>   45. https://groups.google.com/d/msgid/pandoc-discuss/
>   46. http://40googlegroups.com/
>   47. https://groups.google.com/
>   48. https://groups.google.com/d/
>   49. http://40googlegroups.com/
>   50. https://groups.google.com/groups/opt_out
>   51. https://groups.google.com/d/
>   52. http://40googlegroups.com/
>   53. https://groups.google.com/groups/opt_out
>   54. https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d
>   55. http://40googlegroups.com/
>   56. https://groups.google.com/groups/opt_out
>   57. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-
>   58. http://40googlegroups.com/
>   59. https://groups.google.com/groups/opt_out
>   60. https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-
>   61. http://40googlegroups.com/
>   62. https://groups.google.com/groups/opt_out
>   63. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com
>   64. https://groups.google.com/groups/opt_out
>   65. https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%40googlegroups.com
>   66. https://groups.google.com/groups/opt_out
>   67. https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%40googlegroups.com
>   68. https://groups.google.com/groups/opt_out
>   69. https://groups.google.com/d/msgid/pandoc-discuss/8470226a-c385-42bd-a6ba-71037729c707%40googlegroups.com
>   70. https://groups.google.com/groups/opt_out
>   71. https://groups.google.com/d/msgid/pandoc-discuss/86ec8122-f24b-4251-a8f3-89953c0cf7a5%40googlegroups.com
>   72. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Try the latest.

+++ John MacFarlane [Oct 08 13 20:12 ]:
> Sure enough -- I had cabal-install 1.16 with Cabal 1.16.  When
> I upgraded to 1.18 I got the same error you did.
> 
> +++ Nick Bart [Oct 08 13 14:58 ]:
> >    Alright: Output of cabal clean && cabal install -v2 and cabal clean &&
> >    cabal install -v2 --force-reinstalls is at
> >    https://gist.github.com/anonymous/6892387.
> >    A few more details: Haskell installed via homebrew:
> >    brew install haskell-platform
> >    brew install ghc
> >    brew install icu4c
> >    Also tried, as recommended by Joseph,
> >    rm -rf ~/.pandoc ~/.cabal ~/.ghc
> >    ~/src/{pandoc,pandoc-types,pandoc-citeproc}
> >    before reinstalling.
> >    On Tuesday, 8 October 2013 20:49:14 UTC, fiddlosopher wrote:
> > 
> >      Even stranger.  I use roughly the same setup at home.
> >      Why don't you send the output of
> >      cabal clean && cabal install -v2
> >      +++ Nick Bart [Oct 08 13 12:06 ]:
> >      >    Mac OS X 10.7.5
> >      >    $ cabal --version
> >      >    cabal-install version 1.18.0.2
> >      >    using version 1.18.1 of the Cabal library
> >      >    On Tuesday, October 8, 2013 6:56:02 PM UTC, fiddlosopher wrote:
> >      >
> >      >      Very strange.  I just installed with no problems.  What OS
> >      >      are you using?  What does 'cabal --version' say?
> >      >      +++ Nick Bart [Oct 08 13 11:33 ]:
> >      >      >    Sorry but no, not even removing the complete pandoc,
> >      >      pandoc-citeproc,
> >      >      >    and pandoc-types directories and a fresh
> >      >      >    git clone git://[1][1]github.com/jgm/pandoc-types
> >      >      >    git clone git://[2][2]github.com/jgm/pandoc-citeproc
> >      >      >    git clone git://[3][3]github.com/jgm/pandoc
> >      >      >    did the trick. Anything else I could try?
> >      >      >    On Tuesday, October 8, 2013 6:10:41 PM UTC, fiddlosopher
> >      wrote:
> >      >      >
> >      >      >      Try removing the dist subdirectory entirely, if 'cabal
> >      clean'
> >      >      >      doesn't do that.
> >      >      >      +++ Nick Bart [Oct 08 13 10:45 ]:
> >      >      >      >    Hmm, no. Any other ideas?
> >      >      >      >    On Tuesday, October 8, 2013 1:39:05 PM UTC,
> >      fiddlosopher
> >      >      wrote:
> >      >      >      >
> >      >      >      >    Do cabal clean first, then it should build
> >      >      >      >    On Oct 7, 2013, at 11:26 PM, Nick Bart
> >      >      <[1]nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
> >      >      >      wrote:
> >      >      >      >
> >      >      >      >    Now, since yesterday pandoc doesn't build any
> >      longer:
> >      >      >      >    pandoc $ cabal install --enable-tests
> >      --force-reinstalls
> >      >      >      >    Resolving dependencies...
> >      >      >      >    Warning: The following packages are likely to be
> >      broken
> >      >      by the
> >      >      >      >    reinstalls:
> >      >      >      >    pandoc-citeproc-0.1.1.2
> >      >      >      >    Continuing even though the plan contains
> >      dangerous
> >      >      reinstalls.
> >      >      >      >    Configuring pandoc-1.12.0.2...
> >      >      >      >    Building pandoc-1.12.0.2...
> >      >      >      >    setup: internal error: componentsInBuildOrder: no
> >      such
> >      >      >      component:
> >      >      >      >    CExeName "make-pandoc-man-pages"
> >      >      >      >    Failed to install pandoc-1.12.0.2
> >      >      >      >    cabal: Error: some packages failed to install:
> >      >      >      >    pandoc-1.12.0.2 failed during the building phase.
> >      The
> >      >      exception
> >      >      >      was:
> >      >      >      >    ExitFailure 1
> >      >      >      >    Any ideas?
> >      >      >      >
> >      >      >      >    --
> >      >      >      >    You received this message because you are
> >      subscribed to
> >      >      the
> >      >      >      Google
> >      >      >      >    Groups "pandoc-discuss" group.
> >      >      >      >    To unsubscribe from this group and stop receiving
> >      emails
> >      >      from
> >      >      >      it, send
> >      >      >      >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >      >      >    To post to this group, send email to
> >      >      >      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >      >      >    To view this discussion on the web visit
> >      >      >      >    [4][1][4][4]https://groups.google.
> >      >      com/d/msgid/pandoc-discuss/
> >      >      >      28779dd0-
> >      >      >      >    99a6-448c-8967-6d37bdba58ad%[
> >      2][5][5]40googlegroups.com.
> >      >      >      >    For more options, visit
> >      [5][3][6][6]https://groups.google.
> >      >      com/
> >      >      >      groups/opt_out.
> >      >      >      >
> >      >      >      >    --
> >      >      >      >    You received this message because you are
> >      subscribed to
> >      >      the
> >      >      >      Google
> >      >      >      >    Groups "pandoc-discuss" group.
> >      >      >      >    To unsubscribe from this group and stop receiving
> >      emails
> >      >      from
> >      >      >      it, send
> >      >      >      >    an email to [4]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >      >      >    To post to this group, send email to
> >      >      >      [5]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >      >      >    To view this discussion on the web visit
> >      >      >      >    [6][6][7][7]https://groups.google.
> >      >      com/d/msgid/pandoc-discuss/
> >      >      >      63edb53b-b03c-468a-
> >      >      >      >    a60b-6bba4777cbf4%[7][8][8]40googlegroups.com.
> >      >      >      >    For more options, visit
> >      [7][8][9][9]https://groups.google.
> >      >      com/
> >      >      >      groups/opt_out.
> >      >      >      >
> >      >      >      > References
> >      >      >      >
> >      >      >      >    1. javascript:/
> >      >      >      >    2. javascript:/
> >      >      >      >    3. javascript:/
> >      >      >      >    4. [9][10][10]https://groups.google.com/d/
> >      >      >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-
> >      6d37bdba58ad%
> >      >      >      [11][11]40googlegroups.com
> >      >      >      >    5. [10][12][12]https://groups.google.
> >      com/groups/opt_out
> >      >      >      >    6. [11][13][13]https://groups.google.com/d/
> >      >      >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-
> >      6bba4777cbf4%
> >      >      >      [14][14]40googlegroups.com
> >      >      >      >    7. [12][15][15]https://groups.google.
> >      com/groups/opt_out
> >      >      >
> >      >      >    --
> >      >      >    You received this message because you are subscribed to
> >      the
> >      >      Google
> >      >      >    Groups "pandoc-discuss" group.
> >      >      >    To unsubscribe from this group and stop receiving emails
> >      from
> >      >      it, send
> >      >      >    an email to [16]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >      >    To post to this group, send email to
> >      >      [17]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >      >    To view this discussion on the web visit
> >      >      >    [13][18][16]https://groups.google.com/d/msgid/pandoc-
> >      discuss/
> >      >      233c21c4-cff5-452d
> >      >      >    -8269-80981156bddc%[19][17]40googlegroups.com.
> >      >      >    For more options, visit
> >      [14][20][18]https://groups.google.com/
> >      >      groups/opt_out.
> >      >      >
> >      >      > References
> >      >      >
> >      >      >    1. [21][19]https://groups.google.com/d/
> >      >      msgid/pandoc-discuss/28779dd0-
> >      >      >    2. [22][20]http://40googlegroups.com/
> >      >      >    3. [23][21]https://groups.google.com/groups/opt_out
> >      >      >    4. javascript:/
> >      >      >    5. javascript:/
> >      >      >    6. [24][22]https://groups.google.com/d/
> >      >      msgid/pandoc-discuss/63edb53b-b03c-468a-
> >      >      >    7. [25][23]http://40googlegroups.com/
> >      >      >    8. [26][24]https://groups.google.com/groups/opt_out
> >      >      >    9. [27][25]https://groups.google.com/d/
> >      >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%
> >      >      [26]40googlegroups.com
> >      >      >   10. [28][27]https://groups.google.com/groups/opt_out
> >      >      >   11. [29][28]https://groups.google.com/d/
> >      >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%
> >      >      [29]40googlegroups.com
> >      >      >   12. [30][30]https://groups.google.com/groups/opt_out
> >      >      >   13. [31][31]https://groups.google.com/d/
> >      >      msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%
> >      >      [32]40googlegroups.com
> >      >      >   14. [32][33]https://groups.google.com/groups/opt_out
> >      >
> >      >    --
> >      >    You received this message because you are subscribed to the
> >      Google
> >      >    Groups "pandoc-discuss" group.
> >      >    To unsubscribe from this group and stop receiving emails from
> >      it, send
> >      >    an email to [34]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >    To post to this group, send email to
> >      [35]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >    To view this discussion on the web visit
> >      >    [33][36]https://groups.google.com/d/msgid/pandoc-discuss/
> >      8470226a-c385-42bd
> >      >    -a6ba-71037729c707%[37]40googlegroups.com.
> >      >    For more options, visit [34][38]https://groups.google.com/
> >      groups/opt_out.
> >      >
> >      > References
> >      >
> >      >    1. [39]http://github.com/jgm/pandoc-types
> >      >    2. [40]http://github.com/jgm/pandoc-citeproc
> >      >    3. [41]http://github.com/jgm/pandoc
> >      >    4. [42]https://groups.google.com/d/msgid/pandoc-discuss/
> >      >    5. [43]http://40googlegroups.com/
> >      >    6. [44]https://groups.google.com/
> >      >    7. [45]https://groups.google.com/d/msgid/pandoc-discuss/
> >      >    8. [46]http://40googlegroups.com/
> >      >    9. [47]https://groups.google.com/
> >      >   10. [48]https://groups.google.com/d/
> >      >   11. [49]http://40googlegroups.com/
> >      >   12. [50]https://groups.google.com/groups/opt_out
> >      >   13. [51]https://groups.google.com/d/
> >      >   14. [52]http://40googlegroups.com/
> >      >   15. [53]https://groups.google.com/groups/opt_out
> >      >   16. javascript:/
> >      >   17. javascript:/
> >      >   18. [54]https://groups.google.com/d/
> >      msgid/pandoc-discuss/233c21c4-cff5-452d
> >      >   19. [55]http://40googlegroups.com/
> >      >   20. [56]https://groups.google.com/groups/opt_out
> >      >   21. [57]https://groups.google.com/d/
> >      msgid/pandoc-discuss/28779dd0-
> >      >   22. [58]http://40googlegroups.com/
> >      >   23. [59]https://groups.google.com/groups/opt_out
> >      >   24. [60]https://groups.google.com/d/
> >      msgid/pandoc-discuss/63edb53b-b03c-468a-
> >      >   25. [61]http://40googlegroups.com/
> >      >   26. [62]https://groups.google.com/groups/opt_out
> >      >   27. [63]https://groups.google.com/d/
> >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%
> >      40googlegroups.com
> >      >   28. [64]https://groups.google.com/groups/opt_out
> >      >   29. [65]https://groups.google.com/d/
> >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%
> >      40googlegroups.com
> >      >   30. [66]https://groups.google.com/groups/opt_out
> >      >   31. [67]https://groups.google.com/d/
> >      msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%
> >      40googlegroups.com
> >      >   32. [68]https://groups.google.com/groups/opt_out
> >      >   33. [69]https://groups.google.com/d/
> >      msgid/pandoc-discuss/8470226a-c385-42bd-a6ba-71037729c707%
> >      40googlegroups.com
> >      >   34. [70]https://groups.google.com/groups/opt_out
> > 
> >    --
> >    You received this message because you are subscribed to the Google
> >    Groups "pandoc-discuss" group.
> >    To unsubscribe from this group and stop receiving emails from it, send
> >    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >    To view this discussion on the web visit
> >    [71]https://groups.google.com/d/msgid/pandoc-discuss/86ec8122-f24b-4251
> >    -a8f3-89953c0cf7a5%40googlegroups.com.
> >    For more options, visit [72]https://groups.google.com/groups/opt_out.
> > 
> > References
> > 
> >    1. http://github.com/jgm/pandoc-types
> >    2. http://github.com/jgm/pandoc-citeproc
> >    3. http://github.com/jgm/pandoc
> >    4. https://groups.google/
> >    5. http://40googlegroups.com/
> >    6. https://groups.google/
> >    7. https://groups.google/
> >    8. http://40googlegroups.com/
> >    9. https://groups.google/
> >   10. https://groups.google.com/d/
> >   11. http://40googlegroups.com/
> >   12. https://groups.google.com/groups/opt_out
> >   13. https://groups.google.com/d/
> >   14. http://40googlegroups.com/
> >   15. https://groups.google.com/groups/opt_out
> >   16. https://groups.google.com/d/msgid/pandoc-discuss/
> >   17. http://40googlegroups.com/
> >   18. https://groups.google.com/
> >   19. https://groups.google.com/d/
> >   20. http://40googlegroups.com/
> >   21. https://groups.google.com/groups/opt_out
> >   22. https://groups.google.com/d/
> >   23. http://40googlegroups.com/
> >   24. https://groups.google.com/groups/opt_out
> >   25. https://groups.google.com/d/
> >   26. http://40googlegroups.com/
> >   27. https://groups.google.com/groups/opt_out
> >   28. https://groups.google.com/d/
> >   29. http://40googlegroups.com/
> >   30. https://groups.google.com/groups/opt_out
> >   31. https://groups.google.com/d/
> >   32. http://40googlegroups.com/
> >   33. https://groups.google.com/groups/opt_out
> >   34. javascript:/
> >   35. javascript:/
> >   36. https://groups.google.com/d/msgid/pandoc-discuss/8470226a-c385-42bd
> >   37. http://40googlegroups.com/
> >   38. https://groups.google.com/groups/opt_out
> >   39. http://github.com/jgm/pandoc-types
> >   40. http://github.com/jgm/pandoc-citeproc
> >   41. http://github.com/jgm/pandoc
> >   42. https://groups.google.com/d/msgid/pandoc-discuss/
> >   43. http://40googlegroups.com/
> >   44. https://groups.google.com/
> >   45. https://groups.google.com/d/msgid/pandoc-discuss/
> >   46. http://40googlegroups.com/
> >   47. https://groups.google.com/
> >   48. https://groups.google.com/d/
> >   49. http://40googlegroups.com/
> >   50. https://groups.google.com/groups/opt_out
> >   51. https://groups.google.com/d/
> >   52. http://40googlegroups.com/
> >   53. https://groups.google.com/groups/opt_out
> >   54. https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d
> >   55. http://40googlegroups.com/
> >   56. https://groups.google.com/groups/opt_out
> >   57. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-
> >   58. http://40googlegroups.com/
> >   59. https://groups.google.com/groups/opt_out
> >   60. https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-
> >   61. http://40googlegroups.com/
> >   62. https://groups.google.com/groups/opt_out
> >   63. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com
> >   64. https://groups.google.com/groups/opt_out
> >   65. https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%40googlegroups.com
> >   66. https://groups.google.com/groups/opt_out
> >   67. https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%40googlegroups.com
> >   68. https://groups.google.com/groups/opt_out
> >   69. https://groups.google.com/d/msgid/pandoc-discuss/8470226a-c385-42bd-a6ba-71037729c707%40googlegroups.com
> >   70. https://groups.google.com/groups/opt_out
> >   71. https://groups.google.com/d/msgid/pandoc-discuss/86ec8122-f24b-4251-a8f3-89953c0cf7a5%40googlegroups.com
> >   72. https://groups.google.com/groups/opt_out
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131009031248.GA51945%40Johns-MacBook-Pro.local.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 19028 bytes --]

Now that sure fixed it. Thank you!

On Wednesday, October 9, 2013 4:04:47 AM UTC, fiddlosopher wrote:
>
> Try the latest. 
>
> +++ John MacFarlane [Oct 08 13 20:12 ]: 
> > Sure enough -- I had cabal-install 1.16 with Cabal 1.16.  When 
> > I upgraded to 1.18 I got the same error you did. 
> > 
> > +++ Nick Bart [Oct 08 13 14:58 ]: 
> > >    Alright: Output of cabal clean && cabal install -v2 and cabal clean 
> && 
> > >    cabal install -v2 --force-reinstalls is at 
> > >    https://gist.github.com/anonymous/6892387. 
> > >    A few more details: Haskell installed via homebrew: 
> > >    brew install haskell-platform 
> > >    brew install ghc 
> > >    brew install icu4c 
> > >    Also tried, as recommended by Joseph, 
> > >    rm -rf ~/.pandoc ~/.cabal ~/.ghc 
> > >    ~/src/{pandoc,pandoc-types,pandoc-citeproc} 
> > >    before reinstalling. 
> > >    On Tuesday, 8 October 2013 20:49:14 UTC, fiddlosopher wrote: 
> > > 
> > >      Even stranger.  I use roughly the same setup at home. 
> > >      Why don't you send the output of 
> > >      cabal clean && cabal install -v2 
> > >      +++ Nick Bart [Oct 08 13 12:06 ]: 
> > >      >    Mac OS X 10.7.5 
> > >      >    $ cabal --version 
> > >      >    cabal-install version 1.18.0.2 
> > >      >    using version 1.18.1 of the Cabal library 
> > >      >    On Tuesday, October 8, 2013 6:56:02 PM UTC, fiddlosopher 
> wrote: 
> > >      > 
> > >      >      Very strange.  I just installed with no problems.  What OS 
> > >      >      are you using?  What does 'cabal --version' say? 
> > >      >      +++ Nick Bart [Oct 08 13 11:33 ]: 
> > >      >      >    Sorry but no, not even removing the complete pandoc, 
> > >      >      pandoc-citeproc, 
> > >      >      >    and pandoc-types directories and a fresh 
> > >      >      >    git clone git://[1][1]github.com/jgm/pandoc-types 
> > >      >      >    git clone git://[2][2]github.com/jgm/pandoc-citeproc 
> > >      >      >    git clone git://[3][3]github.com/jgm/pandoc 
> > >      >      >    did the trick. Anything else I could try? 
> > >      >      >    On Tuesday, October 8, 2013 6:10:41 PM UTC, 
> fiddlosopher 
> > >      wrote: 
> > >      >      > 
> > >      >      >      Try removing the dist subdirectory entirely, if 
> 'cabal 
> > >      clean' 
> > >      >      >      doesn't do that. 
> > >      >      >      +++ Nick Bart [Oct 08 13 10:45 ]: 
> > >      >      >      >    Hmm, no. Any other ideas? 
> > >      >      >      >    On Tuesday, October 8, 2013 1:39:05 PM UTC, 
> > >      fiddlosopher 
> > >      >      wrote: 
> > >      >      >      > 
> > >      >      >      >    Do cabal clean first, then it should build 
> > >      >      >      >    On Oct 7, 2013, at 11:26 PM, Nick Bart 
> > >      >      <[1]nickba...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 
> > >      >      >      wrote: 
> > >      >      >      > 
> > >      >      >      >    Now, since yesterday pandoc doesn't build any 
> > >      longer: 
> > >      >      >      >    pandoc $ cabal install --enable-tests 
> > >      --force-reinstalls 
> > >      >      >      >    Resolving dependencies... 
> > >      >      >      >    Warning: The following packages are likely to 
> be 
> > >      broken 
> > >      >      by the 
> > >      >      >      >    reinstalls: 
> > >      >      >      >    pandoc-citeproc-0.1.1.2 
> > >      >      >      >    Continuing even though the plan contains 
> > >      dangerous 
> > >      >      reinstalls. 
> > >      >      >      >    Configuring pandoc-1.12.0.2... 
> > >      >      >      >    Building pandoc-1.12.0.2... 
> > >      >      >      >    setup: internal error: componentsInBuildOrder: 
> no 
> > >      such 
> > >      >      >      component: 
> > >      >      >      >    CExeName "make-pandoc-man-pages" 
> > >      >      >      >    Failed to install pandoc-1.12.0.2 
> > >      >      >      >    cabal: Error: some packages failed to install: 
> > >      >      >      >    pandoc-1.12.0.2 failed during the building 
> phase. 
> > >      The 
> > >      >      exception 
> > >      >      >      was: 
> > >      >      >      >    ExitFailure 1 
> > >      >      >      >    Any ideas? 
> > >      >      >      > 
> > >      >      >      >    -- 
> > >      >      >      >    You received this message because you are 
> > >      subscribed to 
> > >      >      the 
> > >      >      >      Google 
> > >      >      >      >    Groups "pandoc-discuss" group. 
> > >      >      >      >    To unsubscribe from this group and stop 
> receiving 
> > >      emails 
> > >      >      from 
> > >      >      >      it, send 
> > >      >      >      >    an email to [2]
> pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >      >      >    To post to this group, send email to 
> > >      >      >      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >      >      >    To view this discussion on the web visit 
> > >      >      >      >    [4][1][4][4]https://groups.google. 
> > >      >      com/d/msgid/pandoc-discuss/ 
> > >      >      >      28779dd0- 
> > >      >      >      >    99a6-448c-8967-6d37bdba58ad%[ 
> > >      2][5][5]40googlegroups.com. 
> > >      >      >      >    For more options, visit 
> > >      [5][3][6][6]https://groups.google. 
> > >      >      com/ 
> > >      >      >      groups/opt_out. 
> > >      >      >      > 
> > >      >      >      >    -- 
> > >      >      >      >    You received this message because you are 
> > >      subscribed to 
> > >      >      the 
> > >      >      >      Google 
> > >      >      >      >    Groups "pandoc-discuss" group. 
> > >      >      >      >    To unsubscribe from this group and stop 
> receiving 
> > >      emails 
> > >      >      from 
> > >      >      >      it, send 
> > >      >      >      >    an email to [4]
> pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >      >      >    To post to this group, send email to 
> > >      >      >      [5]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >      >      >    To view this discussion on the web visit 
> > >      >      >      >    [6][6][7][7]https://groups.google. 
> > >      >      com/d/msgid/pandoc-discuss/ 
> > >      >      >      63edb53b-b03c-468a- 
> > >      >      >      >    a60b-6bba4777cbf4%[7][8][8]40googlegroups.com. 
>
> > >      >      >      >    For more options, visit 
> > >      [7][8][9][9]https://groups.google. 
> > >      >      com/ 
> > >      >      >      groups/opt_out. 
> > >      >      >      > 
> > >      >      >      > References 
> > >      >      >      > 
> > >      >      >      >    1. javascript:/ 
> > >      >      >      >    2. javascript:/ 
> > >      >      >      >    3. javascript:/ 
> > >      >      >      >    4. [9][10][10]https://groups.google.com/d/ 
> > >      >      >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967- 
> > >      6d37bdba58ad% 
> > >      >      >      [11][11]40googlegroups.com 
> > >      >      >      >    5. [10][12][12]https://groups.google. 
> > >      com/groups/opt_out 
> > >      >      >      >    6. [11][13][13]https://groups.google.com/d/ 
> > >      >      >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b- 
> > >      6bba4777cbf4% 
> > >      >      >      [14][14]40googlegroups.com 
> > >      >      >      >    7. [12][15][15]https://groups.google. 
> > >      com/groups/opt_out 
> > >      >      > 
> > >      >      >    -- 
> > >      >      >    You received this message because you are subscribed 
> to 
> > >      the 
> > >      >      Google 
> > >      >      >    Groups "pandoc-discuss" group. 
> > >      >      >    To unsubscribe from this group and stop receiving 
> emails 
> > >      from 
> > >      >      it, send 
> > >      >      >    an email to [16]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >      >    To post to this group, send email to 
> > >      >      [17]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >      >    To view this discussion on the web visit 
> > >      >      >    [13][18][16]https://groups.google.com/d/msgid/pandoc- 
> > >      discuss/ 
> > >      >      233c21c4-cff5-452d 
> > >      >      >    -8269-80981156bddc%[19][17]40googlegroups.com. 
> > >      >      >    For more options, visit 
> > >      [14][20][18]https://groups.google.com/ 
> > >      >      groups/opt_out. 
> > >      >      > 
> > >      >      > References 
> > >      >      > 
> > >      >      >    1. [21][19]https://groups.google.com/d/ 
> > >      >      msgid/pandoc-discuss/28779dd0- 
> > >      >      >    2. [22][20]http://40googlegroups.com/ 
> > >      >      >    3. [23][21]https://groups.google.com/groups/opt_out 
> > >      >      >    4. javascript:/ 
> > >      >      >    5. javascript:/ 
> > >      >      >    6. [24][22]https://groups.google.com/d/ 
> > >      >      msgid/pandoc-discuss/63edb53b-b03c-468a- 
> > >      >      >    7. [25][23]http://40googlegroups.com/ 
> > >      >      >    8. [26][24]https://groups.google.com/groups/opt_out 
> > >      >      >    9. [27][25]https://groups.google.com/d/ 
> > >      >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad% 
> > >      >      [26]40googlegroups.com 
> > >      >      >   10. [28][27]https://groups.google.com/groups/opt_out 
> > >      >      >   11. [29][28]https://groups.google.com/d/ 
> > >      >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4% 
> > >      >      [29]40googlegroups.com 
> > >      >      >   12. [30][30]https://groups.google.com/groups/opt_out 
> > >      >      >   13. [31][31]https://groups.google.com/d/ 
> > >      >      msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc% 
> > >      >      [32]40googlegroups.com 
> > >      >      >   14. [32][33]https://groups.google.com/groups/opt_out 
> > >      > 
> > >      >    -- 
> > >      >    You received this message because you are subscribed to the 
> > >      Google 
> > >      >    Groups "pandoc-discuss" group. 
> > >      >    To unsubscribe from this group and stop receiving emails 
> from 
> > >      it, send 
> > >      >    an email to [34]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >    To post to this group, send email to 
> > >      [35]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >    To view this discussion on the web visit 
> > >      >    [33][36]https://groups.google.com/d/msgid/pandoc-discuss/ 
> > >      8470226a-c385-42bd 
> > >      >    -a6ba-71037729c707%[37]40googlegroups.com. 
> > >      >    For more options, visit [34][38]https://groups.google.com/ 
> > >      groups/opt_out. 
> > >      > 
> > >      > References 
> > >      > 
> > >      >    1. [39]http://github.com/jgm/pandoc-types 
> > >      >    2. [40]http://github.com/jgm/pandoc-citeproc 
> > >      >    3. [41]http://github.com/jgm/pandoc 
> > >      >    4. [42]https://groups.google.com/d/msgid/pandoc-discuss/ 
> > >      >    5. [43]http://40googlegroups.com/ 
> > >      >    6. [44]https://groups.google.com/ 
> > >      >    7. [45]https://groups.google.com/d/msgid/pandoc-discuss/ 
> > >      >    8. [46]http://40googlegroups.com/ 
> > >      >    9. [47]https://groups.google.com/ 
> > >      >   10. [48]https://groups.google.com/d/ 
> > >      >   11. [49]http://40googlegroups.com/ 
> > >      >   12. [50]https://groups.google.com/groups/opt_out 
> > >      >   13. [51]https://groups.google.com/d/ 
> > >      >   14. [52]http://40googlegroups.com/ 
> > >      >   15. [53]https://groups.google.com/groups/opt_out 
> > >      >   16. javascript:/ 
> > >      >   17. javascript:/ 
> > >      >   18. [54]https://groups.google.com/d/ 
> > >      msgid/pandoc-discuss/233c21c4-cff5-452d 
> > >      >   19. [55]http://40googlegroups.com/ 
> > >      >   20. [56]https://groups.google.com/groups/opt_out 
> > >      >   21. [57]https://groups.google.com/d/ 
> > >      msgid/pandoc-discuss/28779dd0- 
> > >      >   22. [58]http://40googlegroups.com/ 
> > >      >   23. [59]https://groups.google.com/groups/opt_out 
> > >      >   24. [60]https://groups.google.com/d/ 
> > >      msgid/pandoc-discuss/63edb53b-b03c-468a- 
> > >      >   25. [61]http://40googlegroups.com/ 
> > >      >   26. [62]https://groups.google.com/groups/opt_out 
> > >      >   27. [63]https://groups.google.com/d/ 
> > >      msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad% 
> > >      40googlegroups.com 
> > >      >   28. [64]https://groups.google.com/groups/opt_out 
> > >      >   29. [65]https://groups.google.com/d/ 
> > >      msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4% 
> > >      40googlegroups.com 
> > >      >   30. [66]https://groups.google.com/groups/opt_out 
> > >      >   31. [67]https://groups.google.com/d/ 
> > >      msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc% 
> > >      40googlegroups.com 
> > >      >   32. [68]https://groups.google.com/groups/opt_out 
> > >      >   33. [69]https://groups.google.com/d/ 
> > >      msgid/pandoc-discuss/8470226a-c385-42bd-a6ba-71037729c707% 
> > >      40googlegroups.com 
> > >      >   34. [70]https://groups.google.com/groups/opt_out 
> > > 
> > >    -- 
> > >    You received this message because you are subscribed to the Google 
> > >    Groups "pandoc-discuss" group. 
> > >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> > >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > >    To view this discussion on the web visit 
> > >    [71]
> https://groups.google.com/d/msgid/pandoc-discuss/86ec8122-f24b-4251 
> > >    -a8f3-89953c0cf7a5%40googlegroups.com. 
> > >    For more options, visit [72]
> https://groups.google.com/groups/opt_out. 
> > > 
> > > References 
> > > 
> > >    1. http://github.com/jgm/pandoc-types 
> > >    2. http://github.com/jgm/pandoc-citeproc 
> > >    3. http://github.com/jgm/pandoc 
> > >    4. https://groups.google/ 
> > >    5. http://40googlegroups.com/ 
> > >    6. https://groups.google/ 
> > >    7. https://groups.google/ 
> > >    8. http://40googlegroups.com/ 
> > >    9. https://groups.google/ 
> > >   10. https://groups.google.com/d/ 
> > >   11. http://40googlegroups.com/ 
> > >   12. https://groups.google.com/groups/opt_out 
> > >   13. https://groups.google.com/d/ 
> > >   14. http://40googlegroups.com/ 
> > >   15. https://groups.google.com/groups/opt_out 
> > >   16. https://groups.google.com/d/msgid/pandoc-discuss/ 
> > >   17. http://40googlegroups.com/ 
> > >   18. https://groups.google.com/ 
> > >   19. https://groups.google.com/d/ 
> > >   20. http://40googlegroups.com/ 
> > >   21. https://groups.google.com/groups/opt_out 
> > >   22. https://groups.google.com/d/ 
> > >   23. http://40googlegroups.com/ 
> > >   24. https://groups.google.com/groups/opt_out 
> > >   25. https://groups.google.com/d/ 
> > >   26. http://40googlegroups.com/ 
> > >   27. https://groups.google.com/groups/opt_out 
> > >   28. https://groups.google.com/d/ 
> > >   29. http://40googlegroups.com/ 
> > >   30. https://groups.google.com/groups/opt_out 
> > >   31. https://groups.google.com/d/ 
> > >   32. http://40googlegroups.com/ 
> > >   33. https://groups.google.com/groups/opt_out 
> > >   34. javascript:/ 
> > >   35. javascript:/ 
> > >   36. 
> https://groups.google.com/d/msgid/pandoc-discuss/8470226a-c385-42bd 
> > >   37. http://40googlegroups.com/ 
> > >   38. https://groups.google.com/groups/opt_out 
> > >   39. http://github.com/jgm/pandoc-types 
> > >   40. http://github.com/jgm/pandoc-citeproc 
> > >   41. http://github.com/jgm/pandoc 
> > >   42. https://groups.google.com/d/msgid/pandoc-discuss/ 
> > >   43. http://40googlegroups.com/ 
> > >   44. https://groups.google.com/ 
> > >   45. https://groups.google.com/d/msgid/pandoc-discuss/ 
> > >   46. http://40googlegroups.com/ 
> > >   47. https://groups.google.com/ 
> > >   48. https://groups.google.com/d/ 
> > >   49. http://40googlegroups.com/ 
> > >   50. https://groups.google.com/groups/opt_out 
> > >   51. https://groups.google.com/d/ 
> > >   52. http://40googlegroups.com/ 
> > >   53. https://groups.google.com/groups/opt_out 
> > >   54. 
> https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d 
> > >   55. http://40googlegroups.com/ 
> > >   56. https://groups.google.com/groups/opt_out 
> > >   57. https://groups.google.com/d/msgid/pandoc-discuss/28779dd0- 
> > >   58. http://40googlegroups.com/ 
> > >   59. https://groups.google.com/groups/opt_out 
> > >   60. 
> https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a- 
> > >   61. http://40googlegroups.com/ 
> > >   62. https://groups.google.com/groups/opt_out 
> > >   63. 
> https://groups.google.com/d/msgid/pandoc-discuss/28779dd0-99a6-448c-8967-6d37bdba58ad%40googlegroups.com 
> > >   64. https://groups.google.com/groups/opt_out 
> > >   65. 
> https://groups.google.com/d/msgid/pandoc-discuss/63edb53b-b03c-468a-a60b-6bba4777cbf4%40googlegroups.com 
> > >   66. https://groups.google.com/groups/opt_out 
> > >   67. 
> https://groups.google.com/d/msgid/pandoc-discuss/233c21c4-cff5-452d-8269-80981156bddc%40googlegroups.com 
> > >   68. https://groups.google.com/groups/opt_out 
> > >   69. 
> https://groups.google.com/d/msgid/pandoc-discuss/8470226a-c385-42bd-a6ba-71037729c707%40googlegroups.com 
> > >   70. https://groups.google.com/groups/opt_out 
> > >   71. 
> https://groups.google.com/d/msgid/pandoc-discuss/86ec8122-f24b-4251-a8f3-89953c0cf7a5%40googlegroups.com 
> > >   72. https://groups.google.com/groups/opt_out 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "pandoc-discuss" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/20131009031248.GA51945%40Johns-MacBook-Pro.local. 
>
> > For more options, visit https://groups.google.com/groups/opt_out. 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/3892d517-a405-4c70-a9ab-0b7195d28203%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 42226 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3194 bytes --]

Date ranges in same year crash pandoc-citeproc with *some* .csl files
=====================================================================

biblio2yaml now converts biblatex dates (YYYY-MM-DD) including date ranges 
(YYYY-MM-DD/YYYY-MM-DD) very nicely. However, I've noticed an issue 
concerning date ranges within the same year, where behaviour varies, 
depending on the .csl file used, from output as expected to pandoc-citeproc 
crashing.

````
$ cat test-date-range-1.md
Foo [@item1, p. 12].

# References

---
references:
- title: Title
  id: item1
  issued:
  - day: 02
    month: 01
    year: 2003
  - day: 13
    month: 12
    year: 2003
  author:
    given:
    - Al
    family: Author
  type: article-journal
  language: en-US
...



$ pandoc -F pandoc-citeproc -t markdown-citations-markdown_in_html_blocks 
test-date-range-1.md --csl chicago-author-date.csl
pandoc-citeproc: Prelude.init: empty list
pandoc: Error running filter pandoc-citeproc



$ pandoc -F pandoc-citeproc -t markdown-citations-markdown_in_html_blocks 
test-date-range-1.md --csl chicago-note-bibliography.csl
Foo.[^1]

References
==========

Author, Al. “Title” (2–January 13, 2003).

[^1]: Author, “Title,” 12.



$ pandoc -F pandoc-citeproc -t markdown-citations-markdown_in_html_blocks 
test-date-range-1.md --csl chicago-fullnote-bibliography.csl
Foo.[^1]

References
==========

Author, Al. “Title” (January 2–December 13, 2003).

[^1]: Al Author, “Title” (January 2–December 13, 2003): 12.



$ pandoc -F pandoc-citeproc -t markdown-citations-markdown_in_html_blocks 
test-date-range-1.md --csl apa.csl
pandoc-citeproc: Prelude.init: empty list
pandoc: Error running filter pandoc-citeproc



$ pandoc -F pandoc-citeproc -t markdown-citations-markdown_in_html_blocks 
test-date-range-1.md --csl modern-language-association.csl
pandoc-citeproc: Prelude.init: empty list
pandoc: Error running filter pandoc-citeproc
````

So out of the .csl files tested here, chicago-fullnote-bibliography.csl 
works as expected, the output from chicago-note-bibliography.csl is there 
but contains errors, and chicago-author-date.csl, apa.csl, and 
modern-language-association.csl all crash the processor.

Since *some* .csl files work well, it is most probably the other .csl files 
that need fixing. Still, I'm surprised that a broken .csl file can make 
pandoc-citeproc crash.

I will report this to the CSL style maintainers but thought I'd first bring 
it up here so the behaviour of pandoc-citeproc can be checked upon.

Any thoughts?

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/b3d481a7-5937-41b5-a1a7-eb4706a57884%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3718 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3258 bytes --]

Sorting of items without an author (or editor)
==============================================

- Many styles format entries without an author to start with the title in 
the list of references. (Some styles use something like "Anon." instead of 
a missing author's name; what follows is not about these kinds of styles.) 
If a csl style encloses titles of articles, chapters, etc. in quotation 
marks, these items come out at the top of the list and not at their 
intended alphabetic position.

(Styles that do not use quotation marks around titles, such as apa.csl, do 
not show the sorting problem. With apa.csl, however, the in-text references 
come out totally wrong.)

- Also, date disambiguation (2005a, 2005b) shouldn't happen here (except 
for the unlikely case of no author *and* identical titles). 

Trying a similar example with Zotero, sorting came out as expected, and no 
date disambiguation was added.

Hence I'd suspect that at least most of these issues are caused by 
pandoc-citeproc.


$ cat test-sorting.md 

Some claim [@item1; @item2; @item3].

# References

---
references:
- author:
  - family: Doe
    given:
    - John
  id: item1
  issued:
    year: 2005
  publisher: Publisher
  publisher-place: Location
  title: First Book
  type: book
- id: item2
  issued:
    year: 2005
  publisher: Publisher
  publisher-place: Location
  title: Yellow book without an author, should appear under “Y”
  type: book
- id: item3
  issued:
    year: 2005
  container-title: The Journal
  title: Zany article without an author, should appear under “Z”
  type: article-journal
...


$ pandoc -S --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks test-sorting.md --csl 
chicago-author-date.csl
Some claim (Doe 2005; *Yellow Book Without an Author, Should Appear
Under “Y”* 2005b; “Zany Article Without an Author, Should Appear Under
“Z”” 2005a).

References
==========

“Zany Article Without an Author, Should Appear Under “Z”.” 2005a. *The
Journal*.

Doe, John. 2005. *First Book*. Location: Publisher.

*Yellow Book Without an Author, Should Appear Under “Y”*. 2005b.
Location: Publisher.


$ pandoc -S --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks test-sorting.md --csl apa.csl
Some claim (Doe, 2005; *Yellow book without an author, should appear
under “Y”*, 2005a, 2005b).

References
==========

Doe, J. (2005). *First Book*. Location: Publisher.

*Yellow book without an author, should appear under “Y”*. (2005a).
Location: Publisher.

Zany article without an author, should appear under “Z”. (2005b). *The
Journal*.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/5689c697-743a-4bb4-a97b-be60c2ded766%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3805 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 09 13 01:26 ]:
>    Sorting of items without an author (or editor)
>    ==============================================
>    - Many styles format entries without an author to start with the title
>    in the list of references. (Some styles use something like "Anon."
>    instead of a missing author's name; what follows is not about these
>    kinds of styles.) If a csl style encloses titles of articles, chapters,
>    etc. in quotation marks, these items come out at the top of the list
>    and not at their intended alphabetic position.
>    (Styles that do not use quotation marks around titles, such as apa.csl,
>    do not show the sorting problem. With apa.csl, however, the in-text
>    references come out totally wrong.)

I've fixed this by using only the portion after initial punctuation
in sorting.  I don't THINK this will have bad side effects, but it
would be good if Andrea could review the change.

>    - Also, date disambiguation (2005a, 2005b) shouldn't happen here
>    (except for the unlikely case of no author *and* identical titles).
>    Trying a similar example with Zotero, sorting came out as expected, and
>    no date disambiguation was added.

Andrea will have to help with this -- I don't understand that part of
the code yet.

Re: Decoupling citeproc and highlighting-kate from pandoc

[andrea rossato]

John MacFarlane <fiddlosopher-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:

>>    - Also, date disambiguation (2005a, 2005b) shouldn't happen here
>>    (except for the unlikely case of no author *and* identical titles).
>>    Trying a similar example with Zotero, sorting came out as expected, and
>>    no date disambiguation was added.
>
> Andrea will have to help with this -- I don't understand that part of
> the code yet.

yes, cite disambiguation is quite complicated -- well, the standard is
quite complicated too. Nick: could you please file a bug report on the
citeproc-hs issue  tracker with a minimal example(including the CSL
style)? That would be very helpful. Thanks.

Andrea

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1448 bytes --]

Bug report filed. Thank you for looking into this.

On Wednesday, October 9, 2013 8:31:19 PM UTC, arossato wrote:
>
> John MacFarlane <fiddlo...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org <javascript:>> writes: 
>
> >>    - Also, date disambiguation (2005a, 2005b) shouldn't happen here 
> >>    (except for the unlikely case of no author *and* identical titles). 
> >>    Trying a similar example with Zotero, sorting came out as expected, 
> and 
> >>    no date disambiguation was added. 
> > 
> > Andrea will have to help with this -- I don't understand that part of 
> > the code yet. 
>
> yes, cite disambiguation is quite complicated -- well, the standard is 
> quite complicated too. Nick: could you please file a bug report on the 
> citeproc-hs issue  tracker with a minimal example(including the CSL 
> style)? That would be very helpful. Thanks. 
>
> Andrea 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/4f95eb7d-cdc4-4f11-a07a-bc8fb2288d15%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2013 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

It's almost certainly a problem with formatDate in
src/Text/CSL/Eval/Date.hs

init and last should not be used in situations where there can
be an empty list.

However, it would be best if Andrea could debug this, as I'm not
a CSL expert.

+++ Nick Bart [Oct 09 13 00:07 ]:
>    Date ranges in same year crash pandoc-citeproc with *some* .csl files
>    =====================================================================
>    biblio2yaml now converts biblatex dates (YYYY-MM-DD) including date
>    ranges (YYYY-MM-DD/YYYY-MM-DD) very nicely. However, I've noticed an
>    issue concerning date ranges within the same year, where behaviour
>    varies, depending on the .csl file used, from output as expected to
>    pandoc-citeproc crashing.
>    ````
>    $ cat test-date-range-1.md
>    Foo [@item1, p. 12].
>    # References
>    ---
>    references:
>    - title: Title
>      id: item1
>      issued:
>      - day: 02
>        month: 01
>        year: 2003
>      - day: 13
>        month: 12
>        year: 2003
>      author:
>        given:
>        - Al
>        family: Author
>      type: article-journal
>      language: en-US
>    ...
>    $ pandoc -F pandoc-citeproc -t
>    markdown-citations-markdown_in_html_blocks test-date-range-1.md --csl
>    chicago-author-date.csl
>    pandoc-citeproc: Prelude.init: empty list
>    pandoc: Error running filter pandoc-citeproc
>    $ pandoc -F pandoc-citeproc -t
>    markdown-citations-markdown_in_html_blocks test-date-range-1.md --csl
>    chicago-note-bibliography.csl
>    Foo.[^1]
>    References
>    ==========
>    Author, Al. Title (2January 13, 2003).
>    [^1]: Author, Title, 12.
>    $ pandoc -F pandoc-citeproc -t
>    markdown-citations-markdown_in_html_blocks test-date-range-1.md --csl
>    chicago-fullnote-bibliography.csl
>    Foo.[^1]
>    References
>    ==========
>    Author, Al. Title (January 2December 13, 2003).
>    [^1]: Al Author, Title (January 2December 13, 2003): 12.
>    $ pandoc -F pandoc-citeproc -t
>    markdown-citations-markdown_in_html_blocks test-date-range-1.md --csl
>    apa.csl
>    pandoc-citeproc: Prelude.init: empty list
>    pandoc: Error running filter pandoc-citeproc
>    $ pandoc -F pandoc-citeproc -t
>    markdown-citations-markdown_in_html_blocks test-date-range-1.md --csl
>    modern-language-association.csl
>    pandoc-citeproc: Prelude.init: empty list
>    pandoc: Error running filter pandoc-citeproc
>    ````
>    So out of the .csl files tested here, chicago-fullnote-bibliography.csl
>    works as expected, the output from chicago-note-bibliography.csl is
>    there but contains errors, and chicago-author-date.csl, apa.csl, and
>    modern-language-association.csl all crash the processor.
>    Since *some* .csl files work well, it is most probably the other .csl
>    files that need fixing. Still, I'm surprised that a broken .csl file
>    can make pandoc-citeproc crash.
>    I will report this to the CSL style maintainers but thought I'd first
>    bring it up here so the behaviour of pandoc-citeproc can be checked
>    upon.
>    Any thoughts?
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/b3d481a7-5937-41b5-
>    a1a7-eb4706a57884%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/b3d481a7-5937-41b5-a1a7-eb4706a57884%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[andrea rossato]

John MacFarlane <fiddlosopher-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:

> It's almost certainly a problem with formatDate in
> src/Text/CSL/Eval/Date.hs
>
> init and last should not be used in situations where there can
> be an empty list.
>
> However, it would be best if Andrea could debug this, as I'm not
> a CSL expert.

I can indeed reproduce this. The fix should be easy. I'll have a look at
that.

Best,
Andrea

Re: Decoupling citeproc and highlighting-kate from pandoc

[Joseph Reagle]

On 10/08/2013 02:33 PM, Nick Bart wrote:
> did the trick. Anything else I could try?

Sounds like a purging isn't your issue, but this is the script I use to
clean/install.

#!/bin/sh

#sudo apt-get  install haskell-platform
rm ~/.pandoc ~/.cabal ~/.ghc ~/.src/{pandoc,pandoc-types,pandoc-citeproc}
cabal update
cabal install alex happy

cd .src

git clone git://github.com/jgm/pandoc-types
cd pandoc-types
cabal update
cabal install --force
cd ..

git clone git://github.com/jgm/pandoc-citeproc
cd pandoc-citeproc
cabal install --force
cd ..

git clone git://github.com/jgm/pandoc
cd pandoc
git submodule update --init
cabal install --enable-tests

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2980 bytes --]

Well, yes, if it's biblatex. It's a literal list field, and according to 
the biblatex specs, any unprotected "and" *separates* two list elements, 
it's not part of any of them (see biblatex manual, v2.7a, § 2.3.4, "Literal 
Lists")

If it's actually one publisher with an "and" in their name, the "and" needs 
to be protected by either wrapping it in braces:

   publisher = {Harcourt, Brace {and} Howe}

or by wrapping the entire name in braces:

   publisher = {{Harcourt, Brace and Howe}}

Both forms should of course be mapped to

   publisher: "Harcourt, Brace and Howe"

If it's indeed two publishers, we'd have:

   publisher = {Paternoster Press and Eerdmans}
   publisher = {Univ. of Toronto Press and Routledge {and} Kegan Paul}
   publisher = {Univ. of Toronto Press and {Routledge and Kegan Paul}}

which should come out as

   publisher: Paternoster Press; Eerdmans
   publisher: Univ. of Toronto Press; Routledge and Kegan Paul

Same for institution, organization, location, origlocation, origpublisher, 
address, and school. E.g.,

   location = {Exeter and Grand Rapids}

is mapped to

   publisher-place = {Exeter; Grand Rapids}


On Monday, October 7, 2013 4:45:16 AM UTC, fiddlosopher wrote:
>
> +++ Nick Bart [Oct 06 13 01:35 ]: 
> >    Thank you, Bruce, for clarifying. 
> >    Though I hope the CSL schema can be extended at some point - I'd like 
> >    to remind of the fact that style guides differ in their requirements 
> >    here, CMoS and APA demanding to print the first publisher and/or 
> place 
> >    only, ISO 690 and MLA wanting all of them - the sensible thing for 
> >    bibtex2pandoc, for now, would be to map all "and"s in bibtex/biblatex 
> >    "literal list" fields, i.e., address, location, origlocation, school, 
> >    institution, organization, howpublished, publisher, origpublisher - 
> >    leaving proteced "{and}"s alone, of course - to something that is 
> >    closer to the average formatting conventions, neutral with regard to 
> >    language, and does not conflict with commas in place names, either: 
> >    Hence I would recommend - for now - mapping any "and" in the fields 
> >    listed above to a semicolon. 
>
> Not so sure.  Do you really want 
>
>    publisher = {Harcourt, Brace, and Howe} 
>
> to be mapped to 
>
>    publisher: "Harcourt, Brace; Howe"? 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/31d1a9fa-e3d5-450a-af7c-89242e0bc0cd%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3733 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 761 bytes --]

One more detail:

Journal names (fields journal, journaltitle, journalsubtitle; but also 
title of @periodical) must be left as they appear, and *not* be converted 
to sentence case.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/8858a440-482b-43fe-abc7-ca41409511e9%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1029 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 06 13 23:43 ]:
>    One more detail:
>    Journal names (fields journal, journaltitle, journalsubtitle; but also
>    title of @periodical) must be left as they appear, and *not* be
>    converted to sentence case.

I suspect there might be other exceptions too.  Here's a complete list of the
fields we currently do untitlecase for.  Let me know which of these should be
removed.

issuetitle
issuesubtitle
subtitle
title
titleaddon
maintitle
booktitle
journal
journaltitle
series
mainsubtitle
booksubtitle
journalsubtitle
maintitleaddon
booktitleaddon
booktitleshort
journaltitleshort
shortjournal
series
shorttitle
eventtitle
origtitle

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2129 bytes --]

Well, it's actually only names of journals that should not be converted to 
sentence case.

The APA Manual, 6e, e.g. is very clear about this: It says for all titles, 
“Capitalize only the first word of the title and of the subtitle, if any, 
and any proper nouns; …”, with the exception of: “Periodical title: 
Journals, newsletters, magazines. Give the periodical title in full, in 
uppercase and lowercase letters. Italicize the name of the periodical. 
[Example:] Social Science Quarterly”

Thus untitlecase should not be applied to:
- journal
- journaltitle
- journalsubtitle
- shortjournal
- title, subtitle, titleaddon, and shorttitle of @periodical

On Monday, October 7, 2013 3:40:55 PM UTC, fiddlosopher wrote:
>
> +++ Nick Bart [Oct 06 13 23:43 ]: 
> >    One more detail: 
> >    Journal names (fields journal, journaltitle, journalsubtitle; but 
> also 
> >    title of @periodical) must be left as they appear, and *not* be 
> >    converted to sentence case. 
>
> I suspect there might be other exceptions too.  Here's a complete list of 
> the 
> fields we currently do untitlecase for.  Let me know which of these should 
> be 
> removed. 
>
> issuetitle 
> issuesubtitle 
> subtitle 
> title 
> titleaddon 
> maintitle 
> booktitle 
> journal 
> journaltitle 
> series 
> mainsubtitle 
> booksubtitle 
> journalsubtitle 
> maintitleaddon 
> booktitleaddon 
> booktitleshort 
> journaltitleshort 
> shortjournal 
> series 
> shorttitle 
> eventtitle 
> origtitle 
>
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/e8741a5b-2bf7-45f4-8790-6d73011ea95f%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2606 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1464 bytes --]

A few more things we'll be needing:

- bibtex/biblatex @comment and @preamble entries should be ignored, 
presently they cause trouble for biblio2yaml.

- bibtex/biblatex @string entries should be expanded. Example:

@string{AW= "Addison-Wesley"}
@book{companion,
  author        = "Goossens, Michel and Mittelbach, Franck and Samarin, 
Alexander",
  title         = "The {{\LaTeX}} {C}ompanion",
  publisher     = AW,
  year          = 1993,
} 

Note that the shorthand, AW, is *not* wrapped in braces. See 
http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf

- biblatex: @xdata, a similar mechanism for shorthands; see biblatex 
manual, v2.7a, 3.11.6 “Data Containers”

If it will take some time to implement @string and @xdata, a stopgap 
measure would be to ignore such entries: Presently at least @string, too, 
causes trouble for biblio2yaml.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1908 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 898 bytes --]

Backslash-escaped dollar signs causing trouble:

$ cat test.bib
@book{item1,
    Author = {Author, Al},
    Date = {2013},
    Hyphenation = {english},
    Title = {Title},
    Annote = {Price: \$ 34.95}
}

$ biblio2yaml test.bib
---
references:
biblio2yaml: "stdin" (line 6, column 27):
unexpected "$"
expecting letter

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/828e6e6d-708f-4722-b331-df029341bacc%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1289 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I was afraid of that.

I'm relying on an external Haskell library, 'bibtex', to parse
the bibtex file.  Looks like I might have to write my own bibtex
parser using pandoc's LaTex reader.

+++ Nick Bart [Oct 07 13 12:50 ]:
>    Backslash-escaped dollar signs causing trouble:
>    $ cat test.bib
>    @book{item1,
>        Author = {Author, Al},
>        Date = {2013},
>        Hyphenation = {english},
>        Title = {Title},
>        Annote = {Price: \$ 34.95}
>    }
>    $ biblio2yaml test.bib
>    ---
>    references:
>    biblio2yaml: "stdin" (line 6, column 27):
>    unexpected "$"
>    expecting letter
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/828e6e6d-708f-4722-
>    b331-df029341bacc%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/828e6e6d-708f-4722-b331-df029341bacc%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Fixed.

+++ Nick Bart [Oct 07 13 12:50 ]:
>    Backslash-escaped dollar signs causing trouble:
>    $ cat test.bib
>    @book{item1,
>        Author = {Author, Al},
>        Date = {2013},
>        Hyphenation = {english},
>        Title = {Title},
>        Annote = {Price: \$ 34.95}
>    }
>    $ biblio2yaml test.bib
>    ---
>    references:
>    biblio2yaml: "stdin" (line 6, column 27):
>    unexpected "$"
>    expecting letter
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/828e6e6d-708f-4722-
>    b331-df029341bacc%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/828e6e6d-708f-4722-b331-df029341bacc%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Tillmann Rendel]

Hi,

Nick Bart wrote:
> A few more things we'll be needing:
>
> - bibtex/biblatex @string entries should be expanded.

That would be great! I would also like to see support for the # operator 
for strings. For example:

   @string{Goossens = "Goossens, Michel"}
   @string{Mittelbach = "Mittelbach, Franck"}
   @string{Samarin = "Samarin, Alexander"}

   @string{AW= "Addison-Wesley"}

   @book{companion,
     author        = Goossens # "and" # Mittelbach # "and" # Samarin,
     title         = "The {{\LaTeX}} {C}ompanion",
     publisher     = AW,
     year          = 1993,
   }

Tillmann

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

New custom bibtex parser supports @string, skips @comment and @preamble.

Concatenation with # also supported.

@xdata still TODO.

+++ Nick Bart [Oct 07 13 12:46 ]:
>    A few more things we'll be needing:
>    - bibtex/biblatex @comment and @preamble entries should be ignored,
>    presently they cause trouble for biblio2yaml.
>    - bibtex/biblatex @string entries should be expanded. Example:
>    @string{AW= "Addison-Wesley"}
>    @book{companion,
>      author        = "Goossens, Michel and Mittelbach, Franck and Samarin,
>    Alexander",
>      title         = "The {{\LaTeX}} {C}ompanion",
>      publisher     = AW,
>      year          = 1993,
>    }
>    Note that the shorthand, AW, is *not* wrapped in braces. See
>    http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf
>    - biblatex: @xdata, a similar mechanism for shorthands; see biblatex
>    manual, v2.7a, 3.11.6 Data Containers
>    If it will take some time to implement @string and @xdata, a stopgap
>    measure would be to ignore such entries: Presently at least @string,
>    too, causes trouble for biblio2yaml.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-
>    aecf-bf57a71e94e1%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3652 bytes --]

Great! 
However, it seems to choke on field names it does not know:

$ cat test-book-tms.bib 
@book{item1,
    Author = {Author, Al},
    Date = {2013},
    Date-Modified = {2013-10-05 09:26:05 +0000},
    Hyphenation = {french},
    Location = {Location},
    Mainsubtitle = {Mainsubtitle},
    Maintitle = {Maintitle},
    Maintitleaddon = {Maintitleaddon},
    Number = {3},
    Publisher = {Publisher},
    Series = {Series},
    Subtitle = {Subtitle},
    Title = {Title of the Book},
    Titleaddon = {Titleaddon}}
$ biblio2yaml test-book-tms.bib 
---
references:
biblio2yaml: "stdin" (line 4, column 9):
unexpected "D"
expecting space or "}"


Context: "Date-Added", "Date-Modified", and various fields with names 
starting with "Bdsk" are added by the application BibDesk, and there are 
probably many other fields added by various other applications and/or 
users. It would probably be best if biblio2yaml ignored all fields it does 
not understand.


On Wednesday, October 9, 2013 8:35:15 PM UTC, fiddlosopher wrote:
>
> New custom bibtex parser supports @string, skips @comment and @preamble. 
>
> Concatenation with # also supported. 
>
> @xdata still TODO. 
>
> +++ Nick Bart [Oct 07 13 12:46 ]: 
> >    A few more things we'll be needing: 
> >    - bibtex/biblatex @comment and @preamble entries should be ignored, 
> >    presently they cause trouble for biblio2yaml. 
> >    - bibtex/biblatex @string entries should be expanded. Example: 
> >    @string{AW= "Addison-Wesley"} 
> >    @book{companion, 
> >      author        = "Goossens, Michel and Mittelbach, Franck and 
> Samarin, 
> >    Alexander", 
> >      title         = "The {{\LaTeX}} {C}ompanion", 
> >      publisher     = AW, 
> >      year          = 1993, 
> >    } 
> >    Note that the shorthand, AW, is *not* wrapped in braces. See 
> >    http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf 
> >    - biblatex: @xdata, a similar mechanism for shorthands; see biblatex 
> >    manual, v2.7a, 3.11.6 Data Containers 
> >    If it will take some time to implement @string and @xdata, a stopgap 
> >    measure would be to ignore such entries: Presently at least @string, 
> >    too, causes trouble for biblio2yaml. 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [1]
> https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb- 
> >    aecf-bf57a71e94e1%40googlegroups.com. 
> >    For more options, visit [2]https://groups.google.com/groups/opt_out. 
> > 
> > References 
> > 
> >    1. 
> https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%40googlegroups.com 
> >    2. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/100d2d3a-78ef-4daa-a23a-33acad3e0df3%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5580 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Well, I assumed that all field names would consist entirely of
letters.  That seems to be the case for both bibtex and biblatex
fields.  I could relax that assumption to allow hyphens, though.

+++ Nick Bart [Oct 09 13 17:00 ]:
>    Great!
>    However, it seems to choke on field names it does not know:
>    $ cat test-book-tms.bib
>    @book{item1,
>        Author = {Author, Al},
>        Date = {2013},
>        Date-Modified = {2013-10-05 09:26:05 +0000},
>        Hyphenation = {french},
>        Location = {Location},
>        Mainsubtitle = {Mainsubtitle},
>        Maintitle = {Maintitle},
>        Maintitleaddon = {Maintitleaddon},
>        Number = {3},
>        Publisher = {Publisher},
>        Series = {Series},
>        Subtitle = {Subtitle},
>        Title = {Title of the Book},
>        Titleaddon = {Titleaddon}}
>    $ biblio2yaml test-book-tms.bib
>    ---
>    references:
>    biblio2yaml: "stdin" (line 4, column 9):
>    unexpected "D"
>    expecting space or "}"
>    Context: "Date-Added", "Date-Modified", and various fields with names
>    starting with "Bdsk" are added by the application BibDesk, and there
>    are probably many other fields added by various other applications
>    and/or users. It would probably be best if biblio2yaml ignored all
>    fields it does not understand.
>    On Wednesday, October 9, 2013 8:35:15 PM UTC, fiddlosopher wrote:
> 
>      New custom bibtex parser supports @string, skips @comment and
>      @preamble.
>      Concatenation with # also supported.
>      @xdata still TODO.
>      +++ Nick Bart [Oct 07 13 12:46 ]:
>      >    A few more things we'll be needing:
>      >    - bibtex/biblatex @comment and @preamble entries should be
>      ignored,
>      >    presently they cause trouble for biblio2yaml.
>      >    - bibtex/biblatex @string entries should be expanded. Example:
>      >    @string{AW= "Addison-Wesley"}
>      >    @book{companion,
>      >      author        = "Goossens, Michel and Mittelbach, Franck and
>      Samarin,
>      >    Alexander",
>      >      title         = "The {{\LaTeX}} {C}ompanion",
>      >      publisher     = AW,
>      >      year          = 1993,
>      >    }
>      >    Note that the shorthand, AW, is *not* wrapped in braces. See
>      >    [1]http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf
>      >    - biblatex: @xdata, a similar mechanism for shorthands; see
>      biblatex
>      >    manual, v2.7a, 3.11.6 Data Containers
>      >    If it will take some time to implement @string and @xdata, a
>      stopgap
>      >    measure would be to ignore such entries: Presently at least
>      @string,
>      >    too, causes trouble for biblio2yaml.
>      >
>      >    --
>      >    You received this message because you are subscribed to the
>      Google
>      >    Groups "pandoc-discuss" group.
>      >    To unsubscribe from this group and stop receiving emails from
>      it, send
>      >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To post to this group, send email to
>      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To view this discussion on the web visit
>      >    [1][4]https://groups.google.com/d/msgid/pandoc-discuss/
>      7b95017f-3db5-40eb-
>      >    aecf-bf57a71e94e1%[5]40googlegroups.com.
>      >    For more options, visit [2][6]https://groups.google.com/
>      groups/opt_out.
>      >
>      > References
>      >
>      >    1. [7]https://groups.google.com/d/
>      msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%
>      40googlegroups.com
>      >    2. [8]https://groups.google.com/groups/opt_out
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [9]https://groups.google.com/d/msgid/pandoc-discuss/100d2d3a-78ef-4daa-
>    a23a-33acad3e0df3%40googlegroups.com.
>    For more options, visit [10]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf
>    2. javascript:/
>    3. javascript:/
>    4. https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-
>    5. http://40googlegroups.com/
>    6. https://groups.google.com/groups/opt_out
>    7. https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%40googlegroups.com
>    8. https://groups.google.com/groups/opt_out
>    9. https://groups.google.com/d/msgid/pandoc-discuss/100d2d3a-78ef-4daa-a23a-33acad3e0df3%40googlegroups.com
>   10. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've allowed hyphens and fixed a number of bugs.  Tried biblio2yaml
on a 102K real biblatex file and had no problems.  No doubt more
testing is needed, but this is getting to the useful stage, I think!

Still TODO:  proper name resolution, @xdata.

+++ John MacFarlane [Oct 09 13 17:38 ]:
> Well, I assumed that all field names would consist entirely of
> letters.  That seems to be the case for both bibtex and biblatex
> fields.  I could relax that assumption to allow hyphens, though.
> 
> +++ Nick Bart [Oct 09 13 17:00 ]:
> >    Great!
> >    However, it seems to choke on field names it does not know:
> >    $ cat test-book-tms.bib
> >    @book{item1,
> >        Author = {Author, Al},
> >        Date = {2013},
> >        Date-Modified = {2013-10-05 09:26:05 +0000},
> >        Hyphenation = {french},
> >        Location = {Location},
> >        Mainsubtitle = {Mainsubtitle},
> >        Maintitle = {Maintitle},
> >        Maintitleaddon = {Maintitleaddon},
> >        Number = {3},
> >        Publisher = {Publisher},
> >        Series = {Series},
> >        Subtitle = {Subtitle},
> >        Title = {Title of the Book},
> >        Titleaddon = {Titleaddon}}
> >    $ biblio2yaml test-book-tms.bib
> >    ---
> >    references:
> >    biblio2yaml: "stdin" (line 4, column 9):
> >    unexpected "D"
> >    expecting space or "}"
> >    Context: "Date-Added", "Date-Modified", and various fields with names
> >    starting with "Bdsk" are added by the application BibDesk, and there
> >    are probably many other fields added by various other applications
> >    and/or users. It would probably be best if biblio2yaml ignored all
> >    fields it does not understand.
> >    On Wednesday, October 9, 2013 8:35:15 PM UTC, fiddlosopher wrote:
> > 
> >      New custom bibtex parser supports @string, skips @comment and
> >      @preamble.
> >      Concatenation with # also supported.
> >      @xdata still TODO.
> >      +++ Nick Bart [Oct 07 13 12:46 ]:
> >      >    A few more things we'll be needing:
> >      >    - bibtex/biblatex @comment and @preamble entries should be
> >      ignored,
> >      >    presently they cause trouble for biblio2yaml.
> >      >    - bibtex/biblatex @string entries should be expanded. Example:
> >      >    @string{AW= "Addison-Wesley"}
> >      >    @book{companion,
> >      >      author        = "Goossens, Michel and Mittelbach, Franck and
> >      Samarin,
> >      >    Alexander",
> >      >      title         = "The {{\LaTeX}} {C}ompanion",
> >      >      publisher     = AW,
> >      >      year          = 1993,
> >      >    }
> >      >    Note that the shorthand, AW, is *not* wrapped in braces. See
> >      >    [1]http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf
> >      >    - biblatex: @xdata, a similar mechanism for shorthands; see
> >      biblatex
> >      >    manual, v2.7a, 3.11.6 Data Containers
> >      >    If it will take some time to implement @string and @xdata, a
> >      stopgap
> >      >    measure would be to ignore such entries: Presently at least
> >      @string,
> >      >    too, causes trouble for biblio2yaml.
> >      >
> >      >    --
> >      >    You received this message because you are subscribed to the
> >      Google
> >      >    Groups "pandoc-discuss" group.
> >      >    To unsubscribe from this group and stop receiving emails from
> >      it, send
> >      >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >    To post to this group, send email to
> >      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >      >    To view this discussion on the web visit
> >      >    [1][4]https://groups.google.com/d/msgid/pandoc-discuss/
> >      7b95017f-3db5-40eb-
> >      >    aecf-bf57a71e94e1%[5]40googlegroups.com.
> >      >    For more options, visit [2][6]https://groups.google.com/
> >      groups/opt_out.
> >      >
> >      > References
> >      >
> >      >    1. [7]https://groups.google.com/d/
> >      msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%
> >      40googlegroups.com
> >      >    2. [8]https://groups.google.com/groups/opt_out
> > 
> >    --
> >    You received this message because you are subscribed to the Google
> >    Groups "pandoc-discuss" group.
> >    To unsubscribe from this group and stop receiving emails from it, send
> >    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >    To view this discussion on the web visit
> >    [9]https://groups.google.com/d/msgid/pandoc-discuss/100d2d3a-78ef-4daa-
> >    a23a-33acad3e0df3%40googlegroups.com.
> >    For more options, visit [10]https://groups.google.com/groups/opt_out.
> > 
> > References
> > 
> >    1. http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf
> >    2. javascript:/
> >    3. javascript:/
> >    4. https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-
> >    5. http://40googlegroups.com/
> >    6. https://groups.google.com/groups/opt_out
> >    7. https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%40googlegroups.com
> >    8. https://groups.google.com/groups/opt_out
> >    9. https://groups.google.com/d/msgid/pandoc-discuss/100d2d3a-78ef-4daa-a23a-33acad3e0df3%40googlegroups.com
> >   10. https://groups.google.com/groups/opt_out
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131010003851.GA24278%40protagoras.phil.berkeley.edu.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ John MacFarlane [Oct 09 13 18:11 ]:
> Still TODO:  proper name resolution, @xdata.

These are now done.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 9746 bytes --]

I'd certainly say this is getting useful: Congratulations!

I tried it on a 6M real biblatex file myself and was successful, too -- in 
the sense that I could easily remove or modify the not very many constructs 
that still caused trouble.

The issues I encountered, plus a few others:

- citekeys/ids are lowercased by biblio2yaml, so none of the entries with 
citekeys containing uppercase letters are found by pandoc-citeproc 
afterwards.

- In addition to letters and hyphens, bibtex and biblatex fields commonly 
in use also contain
    - numbers (from BibDesk, e.g., Bdsk-Url-1, Bdsk-File-2)
    - underscores (export from JSTOR, e.g., Jstor_Articletype, 
Jstor_Formatteddate, Reviewedauthor_1)
    - if I find others, I'll let you know.

- "howpublished" is a "literal field", so it should be fetched with 
getField, not with getLiteralList.

- Names: The biblatex default is to cite Ludwig van Beethoven as 
“Beethoven” and alphabetize him as “Beethoven, Ludwig van”. So the default 
is to interpret the prefix, or "von" part, e.g., "von, van, of, da, de, 
della" etc. as a CSL dropping-particle, not as a non-dropping-particle, as 
biblio2yaml does now. Only if "useprefix=true" is set in the options field 
of an entry should prefixes be treated as non-dropping-particles.

- "jurisdiction" is not a biblatex field. However, since 
http://citationstyles.org/downloads/specification.html#appendix-iv-variables 
defines CSL's "jurisdiction" as "geographic scope of relevance (e.g. "US" 
for a US patent)", it would fit if a "location" field in a @patent entry 
were mapped to the CSL variable "jurisdiction", after resolving its 
biblatex keys such as "countryus".

- The biblatex "library" field could be mapped to CSL "call-number".

- In the combined CSL "note" variable, I'd suggest separating "note" and 
"addendum" by ". ".

- And then there's the issue of quotation marks that are part of titles: 
With styles wrapping article and chapter titles in (double) quotation 
marks, quotation marks that are part of titles should be switched to single 
quotation marks. (And single, if inside double, back to double.) In 
biblatex, this can be done nicely using the csquotes package, which 
supports "active characters", so you can just put smart quotes, “ and ”, 
inside your titles and let csquotes take care of all necessary changes. 
Using pandoc and CSL, that's probably difficult, in particular since it's 
style dependent, but any ideas?

On Thursday, October 10, 2013 1:11:15 AM UTC, fiddlosopher wrote:
>
> I've allowed hyphens and fixed a number of bugs.  Tried biblio2yaml 
> on a 102K real biblatex file and had no problems.  No doubt more 
> testing is needed, but this is getting to the useful stage, I think! 
>
> Still TODO:  proper name resolution, @xdata. 
>
> +++ John MacFarlane [Oct 09 13 17:38 ]: 
> > Well, I assumed that all field names would consist entirely of 
> > letters.  That seems to be the case for both bibtex and biblatex 
> > fields.  I could relax that assumption to allow hyphens, though. 
> > 
> > +++ Nick Bart [Oct 09 13 17:00 ]: 
> > >    Great! 
> > >    However, it seems to choke on field names it does not know: 
> > >    $ cat test-book-tms.bib 
> > >    @book{item1, 
> > >        Author = {Author, Al}, 
> > >        Date = {2013}, 
> > >        Date-Modified = {2013-10-05 09:26:05 +0000}, 
> > >        Hyphenation = {french}, 
> > >        Location = {Location}, 
> > >        Mainsubtitle = {Mainsubtitle}, 
> > >        Maintitle = {Maintitle}, 
> > >        Maintitleaddon = {Maintitleaddon}, 
> > >        Number = {3}, 
> > >        Publisher = {Publisher}, 
> > >        Series = {Series}, 
> > >        Subtitle = {Subtitle}, 
> > >        Title = {Title of the Book}, 
> > >        Titleaddon = {Titleaddon}} 
> > >    $ biblio2yaml test-book-tms.bib 
> > >    --- 
> > >    references: 
> > >    biblio2yaml: "stdin" (line 4, column 9): 
> > >    unexpected "D" 
> > >    expecting space or "}" 
> > >    Context: "Date-Added", "Date-Modified", and various fields with 
> names 
> > >    starting with "Bdsk" are added by the application BibDesk, and 
> there 
> > >    are probably many other fields added by various other applications 
> > >    and/or users. It would probably be best if biblio2yaml ignored all 
> > >    fields it does not understand. 
> > >    On Wednesday, October 9, 2013 8:35:15 PM UTC, fiddlosopher wrote: 
> > > 
> > >      New custom bibtex parser supports @string, skips @comment and 
> > >      @preamble. 
> > >      Concatenation with # also supported. 
> > >      @xdata still TODO. 
> > >      +++ Nick Bart [Oct 07 13 12:46 ]: 
> > >      >    A few more things we'll be needing: 
> > >      >    - bibtex/biblatex @comment and @preamble entries should be 
> > >      ignored, 
> > >      >    presently they cause trouble for biblio2yaml. 
> > >      >    - bibtex/biblatex @string entries should be expanded. 
> Example: 
> > >      >    @string{AW= "Addison-Wesley"} 
> > >      >    @book{companion, 
> > >      >      author        = "Goossens, Michel and Mittelbach, Franck 
> and 
> > >      Samarin, 
> > >      >    Alexander", 
> > >      >      title         = "The {{\LaTeX}} {C}ompanion", 
> > >      >      publisher     = AW, 
> > >      >      year          = 1993, 
> > >      >    } 
> > >      >    Note that the shorthand, AW, is *not* wrapped in braces. See 
> > >      >    [1]http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf 
> > >      >    - biblatex: @xdata, a similar mechanism for shorthands; see 
> > >      biblatex 
> > >      >    manual, v2.7a, 3.11.6 Data Containers 
> > >      >    If it will take some time to implement @string and @xdata, a 
> > >      stopgap 
> > >      >    measure would be to ignore such entries: Presently at least 
> > >      @string, 
> > >      >    too, causes trouble for biblio2yaml. 
> > >      > 
> > >      >    -- 
> > >      >    You received this message because you are subscribed to the 
> > >      Google 
> > >      >    Groups "pandoc-discuss" group. 
> > >      >    To unsubscribe from this group and stop receiving emails 
> from 
> > >      it, send 
> > >      >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >    To post to this group, send email to 
> > >      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> > >      >    To view this discussion on the web visit 
> > >      >    [1][4]https://groups.google.com/d/msgid/pandoc-discuss/ 
> > >      7b95017f-3db5-40eb- 
> > >      >    aecf-bf57a71e94e1%[5]40googlegroups.com. 
> > >      >    For more options, visit [2][6]https://groups.google.com/ 
> > >      groups/opt_out. 
> > >      > 
> > >      > References 
> > >      > 
> > >      >    1. [7]https://groups.google.com/d/ 
> > >      msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1% 
> > >      40googlegroups.com 
> > >      >    2. [8]https://groups.google.com/groups/opt_out 
> > > 
> > >    -- 
> > >    You received this message because you are subscribed to the Google 
> > >    Groups "pandoc-discuss" group. 
> > >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> > >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > >    To view this discussion on the web visit 
> > >    [9]
> https://groups.google.com/d/msgid/pandoc-discuss/100d2d3a-78ef-4daa- 
> > >    a23a-33acad3e0df3%40googlegroups.com. 
> > >    For more options, visit [10]
> https://groups.google.com/groups/opt_out. 
> > > 
> > > References 
> > > 
> > >    1. http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf 
> > >    2. javascript:/ 
> > >    3. javascript:/ 
> > >    4. 
> https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb- 
> > >    5. http://40googlegroups.com/ 
> > >    6. https://groups.google.com/groups/opt_out 
> > >    7. 
> https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%40googlegroups.com 
> > >    8. https://groups.google.com/groups/opt_out 
> > >    9. 
> https://groups.google.com/d/msgid/pandoc-discuss/100d2d3a-78ef-4daa-a23a-33acad3e0df3%40googlegroups.com 
> > >   10. https://groups.google.com/groups/opt_out 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "pandoc-discuss" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/20131010003851.GA24278%40protagoras.phil.berkeley.edu. 
>
> > For more options, visit https://groups.google.com/groups/opt_out. 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/162da1be-e24c-4b90-86ec-d10959146f42%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 14779 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've made all the changes you suggested (except for quotations; that's
a bigger issue and hard to deal with properly given how things are set
up right now).

+++ Nick Bart [Oct 10 13 04:14 ]:
>    I'd certainly say this is getting useful: Congratulations!
>    I tried it on a 6M real biblatex file myself and was successful, too --
>    in the sense that I could easily remove or modify the not very many
>    constructs that still caused trouble.
>    The issues I encountered, plus a few others:
>    - citekeys/ids are lowercased by biblio2yaml, so none of the entries
>    with citekeys containing uppercase letters are found by pandoc-citeproc
>    afterwards.
>    - In addition to letters and hyphens, bibtex and biblatex fields
>    commonly in use also contain
>        - numbers (from BibDesk, e.g., Bdsk-Url-1, Bdsk-File-2)
>        - underscores (export from JSTOR, e.g., Jstor_Articletype,
>    Jstor_Formatteddate, Reviewedauthor_1)
>        - if I find others, I'll let you know.
>    - "howpublished" is a "literal field", so it should be fetched with
>    getField, not with getLiteralList.
>    - Names: The biblatex default is to cite Ludwig van Beethoven as
>    Beethoven and alphabetize him as Beethoven, Ludwig van. So the default
>    is to interpret the prefix, or "von" part, e.g., "von, van, of, da, de,
>    della" etc. as a CSL dropping-particle, not as a non-dropping-particle,
>    as biblio2yaml does now. Only if "useprefix=true" is set in the options
>    field of an entry should prefixes be treated as non-dropping-particles.
>    - "jurisdiction" is not a biblatex field. However, since
>    http://citationstyles.org/downloads/specification.html#appendix-iv-vari
>    ables defines CSL's "jurisdiction" as "geographic scope of relevance
>    (e.g. "US" for a US patent)", it would fit if a "location" field in a
>    @patent entry were mapped to the CSL variable "jurisdiction", after
>    resolving its biblatex keys such as "countryus".
>    - The biblatex "library" field could be mapped to CSL "call-number".
>    - In the combined CSL "note" variable, I'd suggest separating "note"
>    and "addendum" by ". ".
>    - And then there's the issue of quotation marks that are part of
>    titles: With styles wrapping article and chapter titles in (double)
>    quotation marks, quotation marks that are part of titles should be
>    switched to single quotation marks. (And single, if inside double, back
>    to double.) In biblatex, this can be done nicely using the csquotes
>    package, which supports "active characters", so you can just put smart
>    quotes, and , inside your titles and let csquotes take care of all
>    necessary changes. Using pandoc and CSL, that's probably difficult, in
>    particular since it's style dependent, but any ideas?
>    On Thursday, October 10, 2013 1:11:15 AM UTC, fiddlosopher wrote:
> 
>      I've allowed hyphens and fixed a number of bugs.  Tried biblio2yaml
>      on a 102K real biblatex file and had no problems.  No doubt more
>      testing is needed, but this is getting to the useful stage, I think!
>      Still TODO:  proper name resolution, @xdata.
>      +++ John MacFarlane [Oct 09 13 17:38 ]:
>      > Well, I assumed that all field names would consist entirely of
>      > letters.  That seems to be the case for both bibtex and biblatex
>      > fields.  I could relax that assumption to allow hyphens, though.
>      >
>      > +++ Nick Bart [Oct 09 13 17:00 ]:
>      > >    Great!
>      > >    However, it seems to choke on field names it does not know:
>      > >    $ cat test-book-tms.bib
>      > >    @book{item1,
>      > >        Author = {Author, Al},
>      > >        Date = {2013},
>      > >        Date-Modified = {2013-10-05 09:26:05 +0000},
>      > >        Hyphenation = {french},
>      > >        Location = {Location},
>      > >        Mainsubtitle = {Mainsubtitle},
>      > >        Maintitle = {Maintitle},
>      > >        Maintitleaddon = {Maintitleaddon},
>      > >        Number = {3},
>      > >        Publisher = {Publisher},
>      > >        Series = {Series},
>      > >        Subtitle = {Subtitle},
>      > >        Title = {Title of the Book},
>      > >        Titleaddon = {Titleaddon}}
>      > >    $ biblio2yaml test-book-tms.bib
>      > >    ---
>      > >    references:
>      > >    biblio2yaml: "stdin" (line 4, column 9):
>      > >    unexpected "D"
>      > >    expecting space or "}"
>      > >    Context: "Date-Added", "Date-Modified", and various fields
>      with names
>      > >    starting with "Bdsk" are added by the application BibDesk,
>      and there
>      > >    are probably many other fields added by various other
>      applications
>      > >    and/or users. It would probably be best if biblio2yaml
>      ignored all
>      > >    fields it does not understand.
>      > >    On Wednesday, October 9, 2013 8:35:15 PM UTC, fiddlosopher
>      wrote:
>      > >
>      > >      New custom bibtex parser supports @string, skips @comment
>      and
>      > >      @preamble.
>      > >      Concatenation with # also supported.
>      > >      @xdata still TODO.
>      > >      +++ Nick Bart [Oct 07 13 12:46 ]:
>      > >      >    A few more things we'll be needing:
>      > >      >    - bibtex/biblatex @comment and @preamble entries
>      should be
>      > >      ignored,
>      > >      >    presently they cause trouble for biblio2yaml.
>      > >      >    - bibtex/biblatex @string entries should be expanded.
>      Example:
>      > >      >    @string{AW= "Addison-Wesley"}
>      > >      >    @book{companion,
>      > >      >      author        = "Goossens, Michel and Mittelbach,
>      Franck and
>      > >      Samarin,
>      > >      >    Alexander",
>      > >      >      title         = "The {{\LaTeX}} {C}ompanion",
>      > >      >      publisher     = AW,
>      > >      >      year          = 1993,
>      > >      >    }
>      > >      >    Note that the shorthand, AW, is *not* wrapped in
>      braces. See
>      > >      >    [1][1]http://tug.ctan.org/info/
>      bibtex/tamethebeast/ttb_en.pdf
>      > >      >    - biblatex: @xdata, a similar mechanism for
>      shorthands; see
>      > >      biblatex
>      > >      >    manual, v2.7a, 3.11.6 Data Containers
>      > >      >    If it will take some time to implement @string and
>      @xdata, a
>      > >      stopgap
>      > >      >    measure would be to ignore such entries: Presently at
>      least
>      > >      @string,
>      > >      >    too, causes trouble for biblio2yaml.
>      > >      >
>      > >      >    --
>      > >      >    You received this message because you are subscribed
>      to the
>      > >      Google
>      > >      >    Groups "pandoc-discuss" group.
>      > >      >    To unsubscribe from this group and stop receiving
>      emails from
>      > >      it, send
>      > >      >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      > >      >    To post to this group, send email to
>      > >      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      > >      >    To view this discussion on the web visit
>      > >      >    [1][4][2]https://groups.google.
>      com/d/msgid/pandoc-discuss/
>      > >      7b95017f-3db5-40eb-
>      > >      >    aecf-bf57a71e94e1%[5][3]40googlegroups.com.
>      > >      >    For more options, visit
>      [2][6][4]https://groups.google.com/
>      > >      groups/opt_out.
>      > >      >
>      > >      > References
>      > >      >
>      > >      >    1. [7][5]https://groups.google.com/d/
>      > >      msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%
>      > >      [6]40googlegroups.com
>      > >      >    2. [8][7]https://groups.google.com/groups/opt_out
>      > >
>      > >    --
>      > >    You received this message because you are subscribed to the
>      Google
>      > >    Groups "pandoc-discuss" group.
>      > >    To unsubscribe from this group and stop receiving emails from
>      it, send
>      > >    an email to [8]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      > >    To post to this group, send email to
>      [9]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      > >    To view this discussion on the web visit
>      > >    [9][10]https://groups.google.com/d/msgid/pandoc-discuss/
>      100d2d3a-78ef-4daa-
>      > >    a23a-33acad3e0df3%[11]40googlegroups.com.
>      > >    For more options, visit [10][12]https://groups.google.com/
>      groups/opt_out.
>      > >
>      > > References
>      > >
>      > >    1. [13]http://tug.ctan.org/info/
>      bibtex/tamethebeast/ttb_en.pdf
>      > >    2. javascript:/
>      > >    3. javascript:/
>      > >    4. [14]https://groups.google.com/d/
>      msgid/pandoc-discuss/7b95017f-3db5-40eb-
>      > >    5. [15]http://40googlegroups.com/
>      > >    6. [16]https://groups.google.com/groups/opt_out
>      > >    7. [17]https://groups.google.com/d/
>      msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%
>      40googlegroups.com
>      > >    8. [18]https://groups.google.com/groups/opt_out
>      > >    9. [19]https://groups.google.com/d/
>      msgid/pandoc-discuss/100d2d3a-78ef-4daa-a23a-33acad3e0df3%
>      40googlegroups.com
>      > >   10. [20]https://groups.google.com/groups/opt_out
>      >
>      > --
>      > You received this message because you are subscribed to the Google
>      Groups "pandoc-discuss" group.
>      > To unsubscribe from this group and stop receiving emails from it,
>      send an email to [21]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      > To post to this group, send email to [22]pandoc-...@googlegroups.
>      com.
>      > To view this discussion on the web visit
>      [23]https://groups.google.com/d/msgid/pandoc-discuss/
>      20131010003851.GA24278%40protagoras.phil.berkeley.edu.
>      > For more options, visit [24]https://groups.google.com/
>      groups/opt_out.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [25]https://groups.google.com/d/msgid/pandoc-discuss/162da1be-e24c-4b90
>    -86ec-d10959146f42%40googlegroups.com.
>    For more options, visit [26]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf
>    2. https://groups.google.com/d/msgid/pandoc-discuss/
>    3. http://40googlegroups.com/
>    4. https://groups.google.com/
>    5. https://groups.google.com/d/
>    6. http://40googlegroups.com/
>    7. https://groups.google.com/groups/opt_out
>    8. javascript:/
>    9. javascript:/
>   10. https://groups.google.com/d/msgid/pandoc-discuss/100d2d3a-78ef-4daa-
>   11. http://40googlegroups.com/
>   12. https://groups.google.com/groups/opt_out
>   13. http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf
>   14. https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-
>   15. http://40googlegroups.com/
>   16. https://groups.google.com/groups/opt_out
>   17. https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%40googlegroups.com
>   18. https://groups.google.com/groups/opt_out
>   19. https://groups.google.com/d/msgid/pandoc-discuss/100d2d3a-78ef-4daa-a23a-33acad3e0df3%40googlegroups.com
>   20. https://groups.google.com/groups/opt_out
>   21. javascript:/
>   22. javascript:/
>   23. https://groups.google.com/d/msgid/pandoc-discuss/20131010003851.GA24278%40protagoras.phil.berkeley.edu
>   24. https://groups.google.com/groups/opt_out
>   25. https://groups.google.com/d/msgid/pandoc-discuss/162da1be-e24c-4b90-86ec-d10959146f42%40googlegroups.com
>   26. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 16329 bytes --]

This all looks very good so far.

Three things I noticed:

- For *all* isArticle entries, the "number" field should be mapped to the 
CSL "issue", not the "number" variable, whether biblatex "issue" is present 
(and is appended to biblatex "number") or not.

(I'm not sure whether the handling of number and issue for isArticle 
entries could be improved: Concatenating the two is only a compromise 
solution to avoid that any info gets lost, but not quite what most styles 
require. In the majority of cases I've seen, issue simple contains one of 
the seasons, so in principle, "issue" could be parsed and, in case it 
contains a term for a season, mapped to the season part of CSL's "issued" 
variable. If it does not match any season, it could still be concatenated 
with biblatex's "number" as it is done now. Thoughts?)

- A regression: for isArticle entries, the biblatex series field is no 
longer appended to the CSL container-title variable.

- On quotation marks inside titles: Surprisingly, this works well if the 
title contains dumb quotes only:

$ cat test-article-quotation-marks.bib
@article{item1,
    Author = {Author, Al},
    Date = {2012},
    Hyphenation = {english},
    Journaltitle = {The Journaltitle},
    Title = {The Title with a "Quotation" inside},
    Subtitle = {and a "Quotation with "Another Quotation" inside", Too},
}

$ echo @item1 | pandoc -S --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks --bibliography 
test-article-quotation-marks.bib
Author (2012)

Author, Al. 2012. “The Title with a ‘quotation’ Inside: and a ‘quotation
with “another quotation” inside’, Too.” *The Journaltitle*.

$ echo @item1 | pandoc -S --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks --bibliography 
test-article-quotation-marks.bib --csl apa.csl
Author (2012)

Author, A. (2012). The title with a “quotation” inside: and a “quotation
with ‘another quotation’ inside”, too. *The Journaltitle*.

So, quotation marks come out flawless, for both chicago-author-date and 
apa. Only capitalization fails (for chicago).

As soon as smart double quotes are added anywhere, it no longer works, 
though ...

Since using smart quotes is so useful on the LaTeX side of things and I'd 
like to keep them in my biblatex databases, my question: Could pandoc be 
made to handle smart quotes in the same fashion, too; at least inside 
bibliographies?


On Thursday, 10 October 2013 17:00:41 UTC, fiddlosopher wrote:
>
> I've made all the changes you suggested (except for quotations; that's 
> a bigger issue and hard to deal with properly given how things are set 
> up right now). 
>
> +++ Nick Bart [Oct 10 13 04:14 ]: 
> >    I'd certainly say this is getting useful: Congratulations! 
> >    I tried it on a 6M real biblatex file myself and was successful, too 
> -- 
> >    in the sense that I could easily remove or modify the not very many 
> >    constructs that still caused trouble. 
> >    The issues I encountered, plus a few others: 
> >    - citekeys/ids are lowercased by biblio2yaml, so none of the entries 
> >    with citekeys containing uppercase letters are found by 
> pandoc-citeproc 
> >    afterwards. 
> >    - In addition to letters and hyphens, bibtex and biblatex fields 
> >    commonly in use also contain 
> >        - numbers (from BibDesk, e.g., Bdsk-Url-1, Bdsk-File-2) 
> >        - underscores (export from JSTOR, e.g., Jstor_Articletype, 
> >    Jstor_Formatteddate, Reviewedauthor_1) 
> >        - if I find others, I'll let you know. 
> >    - "howpublished" is a "literal field", so it should be fetched with 
> >    getField, not with getLiteralList. 
> >    - Names: The biblatex default is to cite Ludwig van Beethoven as 
> >    Beethoven and alphabetize him as Beethoven, Ludwig van. So the 
> default 
> >    is to interpret the prefix, or "von" part, e.g., "von, van, of, da, 
> de, 
> >    della" etc. as a CSL dropping-particle, not as a 
> non-dropping-particle, 
> >    as biblio2yaml does now. Only if "useprefix=true" is set in the 
> options 
> >    field of an entry should prefixes be treated as 
> non-dropping-particles. 
> >    - "jurisdiction" is not a biblatex field. However, since 
> >    
> http://citationstyles.org/downloads/specification.html#appendix-iv-vari 
> >    ables defines CSL's "jurisdiction" as "geographic scope of relevance 
> >    (e.g. "US" for a US patent)", it would fit if a "location" field in a 
> >    @patent entry were mapped to the CSL variable "jurisdiction", after 
> >    resolving its biblatex keys such as "countryus". 
> >    - The biblatex "library" field could be mapped to CSL "call-number". 
> >    - In the combined CSL "note" variable, I'd suggest separating "note" 
> >    and "addendum" by ". ". 
> >    - And then there's the issue of quotation marks that are part of 
> >    titles: With styles wrapping article and chapter titles in (double) 
> >    quotation marks, quotation marks that are part of titles should be 
> >    switched to single quotation marks. (And single, if inside double, 
> back 
> >    to double.) In biblatex, this can be done nicely using the csquotes 
> >    package, which supports "active characters", so you can just put 
> smart 
> >    quotes, and , inside your titles and let csquotes take care of all 
> >    necessary changes. Using pandoc and CSL, that's probably difficult, 
> in 
> >    particular since it's style dependent, but any ideas? 
> >    On Thursday, October 10, 2013 1:11:15 AM UTC, fiddlosopher wrote: 
> > 
> >      I've allowed hyphens and fixed a number of bugs.  Tried biblio2yaml 
> >      on a 102K real biblatex file and had no problems.  No doubt more 
> >      testing is needed, but this is getting to the useful stage, I 
> think! 
> >      Still TODO:  proper name resolution, @xdata. 
> >      +++ John MacFarlane [Oct 09 13 17:38 ]: 
> >      > Well, I assumed that all field names would consist entirely of 
> >      > letters.  That seems to be the case for both bibtex and biblatex 
> >      > fields.  I could relax that assumption to allow hyphens, though. 
> >      > 
> >      > +++ Nick Bart [Oct 09 13 17:00 ]: 
> >      > >    Great! 
> >      > >    However, it seems to choke on field names it does not know: 
> >      > >    $ cat test-book-tms.bib 
> >      > >    @book{item1, 
> >      > >        Author = {Author, Al}, 
> >      > >        Date = {2013}, 
> >      > >        Date-Modified = {2013-10-05 09:26:05 +0000}, 
> >      > >        Hyphenation = {french}, 
> >      > >        Location = {Location}, 
> >      > >        Mainsubtitle = {Mainsubtitle}, 
> >      > >        Maintitle = {Maintitle}, 
> >      > >        Maintitleaddon = {Maintitleaddon}, 
> >      > >        Number = {3}, 
> >      > >        Publisher = {Publisher}, 
> >      > >        Series = {Series}, 
> >      > >        Subtitle = {Subtitle}, 
> >      > >        Title = {Title of the Book}, 
> >      > >        Titleaddon = {Titleaddon}} 
> >      > >    $ biblio2yaml test-book-tms.bib 
> >      > >    --- 
> >      > >    references: 
> >      > >    biblio2yaml: "stdin" (line 4, column 9): 
> >      > >    unexpected "D" 
> >      > >    expecting space or "}" 
> >      > >    Context: "Date-Added", "Date-Modified", and various fields 
> >      with names 
> >      > >    starting with "Bdsk" are added by the application BibDesk, 
> >      and there 
> >      > >    are probably many other fields added by various other 
> >      applications 
> >      > >    and/or users. It would probably be best if biblio2yaml 
> >      ignored all 
> >      > >    fields it does not understand. 
> >      > >    On Wednesday, October 9, 2013 8:35:15 PM UTC, fiddlosopher 
> >      wrote: 
> >      > > 
> >      > >      New custom bibtex parser supports @string, skips @comment 
> >      and 
> >      > >      @preamble. 
> >      > >      Concatenation with # also supported. 
> >      > >      @xdata still TODO. 
> >      > >      +++ Nick Bart [Oct 07 13 12:46 ]: 
> >      > >      >    A few more things we'll be needing: 
> >      > >      >    - bibtex/biblatex @comment and @preamble entries 
> >      should be 
> >      > >      ignored, 
> >      > >      >    presently they cause trouble for biblio2yaml. 
> >      > >      >    - bibtex/biblatex @string entries should be expanded. 
> >      Example: 
> >      > >      >    @string{AW= "Addison-Wesley"} 
> >      > >      >    @book{companion, 
> >      > >      >      author        = "Goossens, Michel and Mittelbach, 
> >      Franck and 
> >      > >      Samarin, 
> >      > >      >    Alexander", 
> >      > >      >      title         = "The {{\LaTeX}} {C}ompanion", 
> >      > >      >      publisher     = AW, 
> >      > >      >      year          = 1993, 
> >      > >      >    } 
> >      > >      >    Note that the shorthand, AW, is *not* wrapped in 
> >      braces. See 
> >      > >      >    [1][1]http://tug.ctan.org/info/ 
> >      bibtex/tamethebeast/ttb_en.pdf 
> >      > >      >    - biblatex: @xdata, a similar mechanism for 
> >      shorthands; see 
> >      > >      biblatex 
> >      > >      >    manual, v2.7a, 3.11.6 Data Containers 
> >      > >      >    If it will take some time to implement @string and 
> >      @xdata, a 
> >      > >      stopgap 
> >      > >      >    measure would be to ignore such entries: Presently at 
> >      least 
> >      > >      @string, 
> >      > >      >    too, causes trouble for biblio2yaml. 
> >      > >      > 
> >      > >      >    -- 
> >      > >      >    You received this message because you are subscribed 
> >      to the 
> >      > >      Google 
> >      > >      >    Groups "pandoc-discuss" group. 
> >      > >      >    To unsubscribe from this group and stop receiving 
> >      emails from 
> >      > >      it, send 
> >      > >      >    an email to [2]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      > >      >    To post to this group, send email to 
> >      > >      [3]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      > >      >    To view this discussion on the web visit 
> >      > >      >    [1][4][2]https://groups.google. 
> >      com/d/msgid/pandoc-discuss/ 
> >      > >      7b95017f-3db5-40eb- 
> >      > >      >    aecf-bf57a71e94e1%[5][3]40googlegroups.com. 
> >      > >      >    For more options, visit 
> >      [2][6][4]https://groups.google.com/ 
> >      > >      groups/opt_out. 
> >      > >      > 
> >      > >      > References 
> >      > >      > 
> >      > >      >    1. [7][5]https://groups.google.com/d/ 
> >      > >      msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1% 
> >      > >      [6]40googlegroups.com 
> >      > >      >    2. [8][7]https://groups.google.com/groups/opt_out 
> >      > > 
> >      > >    -- 
> >      > >    You received this message because you are subscribed to the 
> >      Google 
> >      > >    Groups "pandoc-discuss" group. 
> >      > >    To unsubscribe from this group and stop receiving emails 
> from 
> >      it, send 
> >      > >    an email to [8]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      > >    To post to this group, send email to 
> >      [9]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      > >    To view this discussion on the web visit 
> >      > >    [9][10]https://groups.google.com/d/msgid/pandoc-discuss/ 
> >      100d2d3a-78ef-4daa- 
> >      > >    a23a-33acad3e0df3%[11]40googlegroups.com. 
> >      > >    For more options, visit [10][12]https://groups.google.com/ 
> >      groups/opt_out. 
> >      > > 
> >      > > References 
> >      > > 
> >      > >    1. [13]http://tug.ctan.org/info/ 
> >      bibtex/tamethebeast/ttb_en.pdf 
> >      > >    2. javascript:/ 
> >      > >    3. javascript:/ 
> >      > >    4. [14]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/7b95017f-3db5-40eb- 
> >      > >    5. [15]http://40googlegroups.com/ 
> >      > >    6. [16]https://groups.google.com/groups/opt_out 
> >      > >    7. [17]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1% 
> >      40googlegroups.com 
> >      > >    8. [18]https://groups.google.com/groups/opt_out 
> >      > >    9. [19]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/100d2d3a-78ef-4daa-a23a-33acad3e0df3% 
> >      40googlegroups.com 
> >      > >   10. [20]https://groups.google.com/groups/opt_out 
> >      > 
> >      > -- 
> >      > You received this message because you are subscribed to the 
> Google 
> >      Groups "pandoc-discuss" group. 
> >      > To unsubscribe from this group and stop receiving emails from it, 
> >      send an email to [21]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      > To post to this group, send email to [22]pandoc-...@googlegroups. 
> >      com. 
> >      > To view this discussion on the web visit 
> >      [23]https://groups.google.com/d/msgid/pandoc-discuss/ 
> >      20131010003851.GA24278%40protagoras.phil.berkeley.edu. 
> >      > For more options, visit [24]https://groups.google.com/ 
> >      groups/opt_out. 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [25]
> https://groups.google.com/d/msgid/pandoc-discuss/162da1be-e24c-4b90 
> >    -86ec-d10959146f42%40googlegroups.com. 
> >    For more options, visit [26]https://groups.google.com/groups/opt_out. 
>
> > 
> > References 
> > 
> >    1. http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf 
> >    2. https://groups.google.com/d/msgid/pandoc-discuss/ 
> >    3. http://40googlegroups.com/ 
> >    4. https://groups.google.com/ 
> >    5. https://groups.google.com/d/ 
> >    6. http://40googlegroups.com/ 
> >    7. https://groups.google.com/groups/opt_out 
> >    8. javascript:/ 
> >    9. javascript:/ 
> >   10. 
> https://groups.google.com/d/msgid/pandoc-discuss/100d2d3a-78ef-4daa- 
> >   11. http://40googlegroups.com/ 
> >   12. https://groups.google.com/groups/opt_out 
> >   13. http://tug.ctan.org/info/bibtex/tamethebeast/ttb_en.pdf 
> >   14. 
> https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb- 
> >   15. http://40googlegroups.com/ 
> >   16. https://groups.google.com/groups/opt_out 
> >   17. 
> https://groups.google.com/d/msgid/pandoc-discuss/7b95017f-3db5-40eb-aecf-bf57a71e94e1%40googlegroups.com 
> >   18. https://groups.google.com/groups/opt_out 
> >   19. 
> https://groups.google.com/d/msgid/pandoc-discuss/100d2d3a-78ef-4daa-a23a-33acad3e0df3%40googlegroups.com 
> >   20. https://groups.google.com/groups/opt_out 
> >   21. javascript:/ 
> >   22. javascript:/ 
> >   23. 
> https://groups.google.com/d/msgid/pandoc-discuss/20131010003851.GA24278%40protagoras.phil.berkeley.edu 
> >   24. https://groups.google.com/groups/opt_out 
> >   25. 
> https://groups.google.com/d/msgid/pandoc-discuss/162da1be-e24c-4b90-86ec-d10959146f42%40googlegroups.com 
> >   26. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/9f372a93-08e8-4565-83d0-b04caa1b8a8f%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 27307 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 11 13 02:26 ]:
> This all looks very good so far.
> 
> Three things I noticed:
> 
> - For *all* isArticle entries, the "number" field should be mapped to the 
> CSL "issue", not the "number" variable, whether biblatex "issue" is present 
> (and is appended to biblatex "number") or not.

OK, I think I've got this right now. Please test.

> (I'm not sure whether the handling of number and issue for isArticle 
> entries could be improved: Concatenating the two is only a compromise 
> solution to avoid that any info gets lost, but not quite what most styles 
> require. In the majority of cases I've seen, issue simple contains one of 
> the seasons, so in principle, "issue" could be parsed and, in case it 
> contains a term for a season, mapped to the season part of CSL's "issued" 
> variable. If it does not match any season, it could still be concatenated 
> with biblatex's "number" as it is done now. Thoughts?)

I'd prefer to keep it simple for now.

> - A regression: for isArticle entries, the biblatex series field is no 
> longer appended to the CSL container-title variable.

Fixed I think.

> - On quotation marks inside titles: Surprisingly, this works well if the 
> title contains dumb quotes only:

Looks like this is done in Text.CSL.Output.Pandoc.  I'm not sure if the
quote parsing and handling is part of the citeproc spec or not -- Andrea
could enlighten us here.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1570 bytes --]

It seems that the CSL specs do contain details about “Quote Substitution” 
and “Flipflopping” 
(<http://citationstyles.org/downloads/upgrade-notes.html#id46>) – material 
that, curiously, does not appear in 
<http://citationstyles.org/downloads/specification.html>.

Following <http://citationstyles.org/downloads/upgrade-notes.html#id46>, it 
seems that both dumb *and* smart quotes (the latter as defined by the 
open-quote,
close-quote, open-inner-quote and close-inner-quote terms in the relevant 
locale file) should be parsed by a CSL processor.

On Saturday, October 12, 2013 4:37:15 AM UTC, fiddlosopher wrote:
>
> +++ Nick Bart [Oct 11 13 02:26 ]: 
>
> > - On quotation marks inside titles: Surprisingly, this works well if the 
> > title contains dumb quotes only: 
>
> Looks like this is done in Text.CSL.Output.Pandoc.  I'm not sure if the 
> quote parsing and handling is part of the citeproc spec or not -- Andrea 
> could enlighten us here. 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/9247c44f-d639-4b7a-903b-96c02c0d4f8e%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1986 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1435 bytes --]

The section “Editor/Translator Name Collapsing” 
(<http://citationstyles.org/downloads/upgrade-notes.html#id36>) from the 
"CSL 1.0 Upgrade Notes" document might help understanding the issue of 
spurious “Edited & translated by .” strings in pandoc output (see example 
below), described earlier in this thread and at 
<https://code.google.com/p/citeproc-hs/issues/detail?id=92>.

````
$ cat test-collection-2.bib 
@collection{item1,
    Editor = {Editor, Edward},
    Hyphenation = {british},
    Location = {Location},
    Publisher = {Publisher},
    Title = {Title},
    Date = {2013}}
$ echo @item1 | pandoc  --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks --bibliography 
test-collection-2.bib 
Editor (2013)

Editor, Edward, ed. 2013. *Title*. Edited & translated by . Location:
Publisher.
````

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/fbf636ef-d74c-44a6-8be7-69513f60042f%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1828 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I'm happy to report that Andrea has fixed this bug, and I've
merged his changes into pandoc-citeproc.

+++ Nick Bart [Oct 13 13 23:39 ]:
> The section “Editor/Translator Name Collapsing” 
> (<http://citationstyles.org/downloads/upgrade-notes.html#id36>) from the 
> "CSL 1.0 Upgrade Notes" document might help understanding the issue of 
> spurious “Edited & translated by .” strings in pandoc output (see example 
> below), described earlier in this thread and at 
> <https://code.google.com/p/citeproc-hs/issues/detail?id=92>.
> 
> ````
> $ cat test-collection-2.bib 
> @collection{item1,
>     Editor = {Editor, Edward},
>     Hyphenation = {british},
>     Location = {Location},
>     Publisher = {Publisher},
>     Title = {Title},
>     Date = {2013}}
> $ echo @item1 | pandoc  --filter pandoc-citeproc -t 
> markdown-citations-markdown_in_html_blocks --bibliography 
> test-collection-2.bib 
> Editor (2013)
> 
> Editor, Edward, ed. 2013. *Title*. Edited & translated by . Location:
> Publisher.
> ````
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/fbf636ef-d74c-44a6-8be7-69513f60042f%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131015025602.GA10253%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I think I've got localized quote flipflopping working now.

+++ Nick Bart [Oct 13 13 23:23 ]:
> It seems that the CSL specs do contain details about “Quote Substitution” 
> and “Flipflopping” 
> (<http://citationstyles.org/downloads/upgrade-notes.html#id46>) – material 
> that, curiously, does not appear in 
> <http://citationstyles.org/downloads/specification.html>.
> 
> Following <http://citationstyles.org/downloads/upgrade-notes.html#id46>, it 
> seems that both dumb *and* smart quotes (the latter as defined by the 
> open-quote,
> close-quote, open-inner-quote and close-inner-quote terms in the relevant 
> locale file) should be parsed by a CSL processor.
> 
> On Saturday, October 12, 2013 4:37:15 AM UTC, fiddlosopher wrote:
> >
> > +++ Nick Bart [Oct 11 13 02:26 ]: 
> >
> > > - On quotation marks inside titles: Surprisingly, this works well if the 
> > > title contains dumb quotes only: 
> >
> > Looks like this is done in Text.CSL.Output.Pandoc.  I'm not sure if the 
> > quote parsing and handling is part of the citeproc spec or not -- Andrea 
> > could enlighten us here. 
> >
> >
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/9247c44f-d639-4b7a-903b-96c02c0d4f8e%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131117061737.GA80492%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 912 bytes --]

Great. It does work with \enquote{} in biblatex title fields. It does not 
work, however with either smart (“”) or dumb quotes. I assume there's no 
good heuristic for parsing nested dumb quotes, but I wonder whether this 
could be made to work with smart quotes, just like csquotes.sty when 
“active quotes” are defined.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/941bbcaa-276e-40ec-bae6-baa209d26d30%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1160 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I'm not sure what you mean. Perhaps you could give an example of
what doesn't work?

+++ Nick Bart [Nov 18 13 04:14 ]:
>    Great. It does work with \enquote{} in biblatex title fields. It does
>    not work, however with either smart () or dumb quotes. I assume there's
>    no good heuristic for parsing nested dumb quotes, but I wonder whether
>    this could be made to work with smart quotes, just like csquotes.sty
>    when active quotes are defined.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/941bbcaa-276e-40ec-
>    bae6-baa209d26d30%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/941bbcaa-276e-40ec-bae6-baa209d26d30%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4849 bytes --]

Sure. This is what I tried:

Note the uppercase “S” in “Doe’S”, too.

$ cat quote-enquote-flipflop.bib 

@article{article,
    Author = {Author, Ann},
    Date = {1888},
    Hyphenation = {english},
    Title = {A Reinterpretation of Doe’s {\enquote{An Analysis of 
\enquote{The Book}}}},
    Annote = {This entry uses the \verb+\enquote{}+ command from the latex 
csquotes.sty package.}}

@book{book,
    Author = {Buthor, Bob},
    Date = {1888},
    Hyphenation = {english},
    Title = {A Reinterpretation of Doe’s {\enquote{An Analysis of 
\enquote{The Book}}}},
    Annote = {This entry uses the \verb+\enquote{}+ command from the latex 
csquotes.sty package.}}

$ echo [@article] [@book] | pandoc -S -t markdown-citations --bibliography 
quote-enquote-flipflop.bib --csl chicago-author-date.csl
---
bibliography: 'quote-enquote-flipflop.bib'
csl: 'chicago-author-date.csl'
...

(Author 1888) (Buthor 1888)

<div class="references">

Author, Ann. 1888. “A Reinterpretation of Doe’S ‘An Analysis of “The
Book”’.”

Buthor, Bob. 1888. *A Reinterpretation of Doe’S “An Analysis of ‘The
Book’”*.

</div>


$ cat quote-smart-flipflop.bib 

@article{article,
    Author = {Author, Ann},
    Date = {1888},
    Hyphenation = {english},
    Title = {A Reinterpretation of Doe’s “An Analysis of “The Book””}}

@book{book,
    Author = {Buthor, Bob},
    Date = {1888},
    Hyphenation = {english},
    Title = {A Reinterpretation of Doe’s “An Analysis of “The Book””}}

$ echo [@article] [@book] | pandoc -S -t markdown-citations --bibliography 
quote-smart-flipflop.bib --csl chicago-author-date.csl
---
bibliography: 'quote-smart-flipflop.bib'
csl: 'chicago-author-date.csl'
...

(Author 1888) (Buthor 1888)

<div class="references">

Author, Ann. 1888. “A Reinterpretation of Doe’S “an Analysis of “the
Book””.”

Buthor, Bob. 1888. *A Reinterpretation of Doe’S “an Analysis of “the
Book””*.

</div>


$ cat quote-dumb-flipflop.bib 

@article{article,
    Author = {Author, Ann},
    Hyphenation = {english},
    Date = {1888},
    Title = {A Reinterpretation of Doe’s "An Analysis of "The Book""}}

@book{book,
    Author = {Buthor, Bob},
    Hyphenation = {english},
    Date = {1888},
    Title = {A Reinterpretation of Doe’s "An Analysis of "The Book""}}


$ echo [@article] [@book] | pandoc -S -t markdown-citations --bibliography 
quote-dumb-flipflop.bib --csl chicago-author-date.csl
---
bibliography: 'quote-dumb-flipflop.bib'
csl: 'chicago-author-date.csl'
...

(Author 1888) (Buthor 1888)

<div class="references">

Author, Ann. 1888. “A Reinterpretation of Doe’S "an Analysis of "the
Book"".”

Buthor, Bob. 1888. *A Reinterpretation of Doe’S "an Analysis of "the
Book""*.

</div>


On Monday, November 18, 2013 3:01:00 PM UTC, fiddlosopher wrote:
>
> I'm not sure what you mean. Perhaps you could give an example of 
> what doesn't work? 
>
> +++ Nick Bart [Nov 18 13 04:14 ]: 
> >    Great. It does work with \enquote{} in biblatex title fields. It does 
> >    not work, however with either smart () or dumb quotes. I assume 
> there's 
> >    no good heuristic for parsing nested dumb quotes, but I wonder 
> whether 
> >    this could be made to work with smart quotes, just like csquotes.sty 
> >    when active quotes are defined. 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [1]
> https://groups.google.com/d/msgid/pandoc-discuss/941bbcaa-276e-40ec- 
> >    bae6-baa209d26d30%40googlegroups.com. 
> >    For more options, visit [2]https://groups.google.com/groups/opt_out. 
> > 
> > References 
> > 
> >    1. 
> https://groups.google.com/d/msgid/pandoc-discuss/941bbcaa-276e-40ec-bae6-baa209d26d30%40googlegroups.com 
> >    2. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/4b28864c-fd8b-4215-9c6d-ba75185bf61d%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 6655 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

The "Doe'S" is a bug indeed.

The others, I'm not so sure what to do.  The way the system now works,
pandoc-citeproc uses pandoc's latex reader to parse the contents of
biblatex fields, and in particular to recognize quoted sections.

So, this will be recognized as nested quotes (and flip-flopped
appropriately):

    Title = {A Reinterpretaiton of Doe's ``An Analysis of `The Book'\,''}

but if you use " or unicode curly quotes, they will just be passed
through verbatim.  That's how the latex reader works.  (You can verify
by doing 'pandoc -f latex -t native'.)

I could change the latex reader so it tries to parse quoted sections
inside curly quotes.  That might be worth doing.  I could do the
same for dumb double quotes.  I wouldn't want to do it for dumb
single quotes, though, since a single ' already has a clear meaning
in LaTeX.

John

+++ Nick Bart [Nov 18 13 09:05 ]:
>    Sure. This is what I tried:
>    Note the uppercase S in DoeS, too.
>    $ cat quote-enquote-flipflop.bib
>    @article{article,
>        Author = {Author, Ann},
>        Date = {1888},
>        Hyphenation = {english},
>        Title = {A Reinterpretation of Does {\enquote{An Analysis of
>    \enquote{The Book}}}},
>        Annote = {This entry uses the \verb+\enquote{}+ command from the
>    latex csquotes.sty package.}}
>    @book{book,
>        Author = {Buthor, Bob},
>        Date = {1888},
>        Hyphenation = {english},
>        Title = {A Reinterpretation of Does {\enquote{An Analysis of
>    \enquote{The Book}}}},
>        Annote = {This entry uses the \verb+\enquote{}+ command from the
>    latex csquotes.sty package.}}
>    $ echo [@article] [@book] | pandoc -S -t markdown-citations
>    --bibliography quote-enquote-flipflop.bib --csl chicago-author-date.csl
>    ---
>    bibliography: 'quote-enquote-flipflop.bib'
>    csl: 'chicago-author-date.csl'
>    ...
>    (Author 1888) (Buthor 1888)
>    <div class="references">
>    Author, Ann. 1888. A Reinterpretation of DoeS An Analysis of The
>    Book.
>    Buthor, Bob. 1888. *A Reinterpretation of DoeS An Analysis of The
>    Book*.
>    </div>
>    $ cat quote-smart-flipflop.bib
>    @article{article,
>        Author = {Author, Ann},
>        Date = {1888},
>        Hyphenation = {english},
>        Title = {A Reinterpretation of Does An Analysis of The Book}}
>    @book{book,
>        Author = {Buthor, Bob},
>        Date = {1888},
>        Hyphenation = {english},
>        Title = {A Reinterpretation of Does An Analysis of The Book}}
>    $ echo [@article] [@book] | pandoc -S -t markdown-citations
>    --bibliography quote-smart-flipflop.bib --csl chicago-author-date.csl
>    ---
>    bibliography: 'quote-smart-flipflop.bib'
>    csl: 'chicago-author-date.csl'
>    ...
>    (Author 1888) (Buthor 1888)
>    <div class="references">
>    Author, Ann. 1888. A Reinterpretation of DoeS an Analysis of the
>    Book.
>    Buthor, Bob. 1888. *A Reinterpretation of DoeS an Analysis of the
>    Book*.
>    </div>
>    $ cat quote-dumb-flipflop.bib
>    @article{article,
>        Author = {Author, Ann},
>        Hyphenation = {english},
>        Date = {1888},
>        Title = {A Reinterpretation of Does "An Analysis of "The Book""}}
>    @book{book,
>        Author = {Buthor, Bob},
>        Hyphenation = {english},
>        Date = {1888},
>        Title = {A Reinterpretation of Does "An Analysis of "The Book""}}
>    $ echo [@article] [@book] | pandoc -S -t markdown-citations
>    --bibliography quote-dumb-flipflop.bib --csl chicago-author-date.csl
>    ---
>    bibliography: 'quote-dumb-flipflop.bib'
>    csl: 'chicago-author-date.csl'
>    ...
>    (Author 1888) (Buthor 1888)
>    <div class="references">
>    Author, Ann. 1888. A Reinterpretation of DoeS "an Analysis of "the
>    Book"".
>    Buthor, Bob. 1888. *A Reinterpretation of DoeS "an Analysis of "the
>    Book""*.
>    </div>
>    On Monday, November 18, 2013 3:01:00 PM UTC, fiddlosopher wrote:
> 
>      I'm not sure what you mean. Perhaps you could give an example of
>      what doesn't work?
>      +++ Nick Bart [Nov 18 13 04:14 ]:
>      >    Great. It does work with \enquote{} in biblatex title fields.
>      It does
>      >    not work, however with either smart () or dumb quotes. I assume
>      there's
>      >    no good heuristic for parsing nested dumb quotes, but I wonder
>      whether
>      >    this could be made to work with smart quotes, just like
>      csquotes.sty
>      >    when active quotes are defined.
>      >
>      >    --
>      >    You received this message because you are subscribed to the
>      Google
>      >    Groups "pandoc-discuss" group.
>      >    To unsubscribe from this group and stop receiving emails from
>      it, send
>      >    an email to [1]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To post to this group, send email to
>      [2]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To view this discussion on the web visit
>      >    [1][3]https://groups.google.com/d/msgid/pandoc-discuss/
>      941bbcaa-276e-40ec-
>      >    bae6-baa209d26d30%[4]40googlegroups.com.
>      >    For more options, visit [2][5]https://groups.google.com/
>      groups/opt_out.
>      >
>      > References
>      >
>      >    1. [6]https://groups.google.com/d/
>      msgid/pandoc-discuss/941bbcaa-276e-40ec-bae6-baa209d26d30%
>      40googlegroups.com
>      >    2. [7]https://groups.google.com/groups/opt_out
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [8]https://groups.google.com/d/msgid/pandoc-discuss/4b28864c-fd8b-4215-
>    9c6d-ba75185bf61d%40googlegroups.com.
>    For more options, visit [9]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. javascript:/
>    2. javascript:/
>    3. https://groups.google.com/d/msgid/pandoc-discuss/941bbcaa-276e-40ec-
>    4. http://40googlegroups.com/
>    5. https://groups.google.com/groups/opt_out
>    6. https://groups.google.com/d/msgid/pandoc-discuss/941bbcaa-276e-40ec-bae6-baa209d26d30%40googlegroups.com
>    7. https://groups.google.com/groups/opt_out
>    8. https://groups.google.com/d/msgid/pandoc-discuss/4b28864c-fd8b-4215-9c6d-ba75185bf61d%40googlegroups.com
>    9. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 8752 bytes --]

I see. Bit of a misunderstanding, sorry: Since I stopped using anything but 
UTF8 smart quotes, I totally forgot about `` and '', in my mental map now 
stored somewhere in the “latex prehistory” department.

So, my request would be to make UTF-8 smart quotes, “ and ”, function the 
same way as `` and '' indeed already do.

Thus

    Title = {A Reinterpretation of {Doe}’s {\enquote{An Analysis of 
\enquote{The Book}}}}

    Title = {A Reinterpretation of {Doe}’s {``An Analysis of ``The 
Book''''}}

    Title = {A Reinterpretation of {Doe}’s {“An Analysis of “The Book””}}

could all be used interchangeably.

I'm not actually interested in dumb quotes myself, but if you’d say it's 
easy to make them function like this as well, why not?

On Monday, 18 November 2013 18:22:26 UTC, fiddlosopher wrote:
>
> The "Doe'S" is a bug indeed. 
>
> The others, I'm not so sure what to do.  The way the system now works, 
> pandoc-citeproc uses pandoc's latex reader to parse the contents of 
> biblatex fields, and in particular to recognize quoted sections. 
>
> So, this will be recognized as nested quotes (and flip-flopped 
> appropriately): 
>
>     Title = {A Reinterpretaiton of Doe's ``An Analysis of `The Book'\,''} 
>
> but if you use " or unicode curly quotes, they will just be passed 
> through verbatim.  That's how the latex reader works.  (You can verify 
> by doing 'pandoc -f latex -t native'.) 
>
> I could change the latex reader so it tries to parse quoted sections 
> inside curly quotes.  That might be worth doing.  I could do the 
> same for dumb double quotes.  I wouldn't want to do it for dumb 
> single quotes, though, since a single ' already has a clear meaning 
> in LaTeX. 
>
> John 
>
> +++ Nick Bart [Nov 18 13 09:05 ]: 
> >    Sure. This is what I tried: 
> >    Note the uppercase S in DoeS, too. 
> >    $ cat quote-enquote-flipflop.bib 
> >    @article{article, 
> >        Author = {Author, Ann}, 
> >        Date = {1888}, 
> >        Hyphenation = {english}, 
> >        Title = {A Reinterpretation of Does {\enquote{An Analysis of 
> >    \enquote{The Book}}}}, 
> >        Annote = {This entry uses the \verb+\enquote{}+ command from the 
> >    latex csquotes.sty package.}} 
> >    @book{book, 
> >        Author = {Buthor, Bob}, 
> >        Date = {1888}, 
> >        Hyphenation = {english}, 
> >        Title = {A Reinterpretation of Does {\enquote{An Analysis of 
> >    \enquote{The Book}}}}, 
> >        Annote = {This entry uses the \verb+\enquote{}+ command from the 
> >    latex csquotes.sty package.}} 
> >    $ echo [@article] [@book] | pandoc -S -t markdown-citations 
> >    --bibliography quote-enquote-flipflop.bib --csl 
> chicago-author-date.csl 
> >    --- 
> >    bibliography: 'quote-enquote-flipflop.bib' 
> >    csl: 'chicago-author-date.csl' 
> >    ... 
> >    (Author 1888) (Buthor 1888) 
> >    <div class="references"> 
> >    Author, Ann. 1888. A Reinterpretation of DoeS An Analysis of The 
> >    Book. 
> >    Buthor, Bob. 1888. *A Reinterpretation of DoeS An Analysis of The 
> >    Book*. 
> >    </div> 
> >    $ cat quote-smart-flipflop.bib 
> >    @article{article, 
> >        Author = {Author, Ann}, 
> >        Date = {1888}, 
> >        Hyphenation = {english}, 
> >        Title = {A Reinterpretation of Does An Analysis of The Book}} 
> >    @book{book, 
> >        Author = {Buthor, Bob}, 
> >        Date = {1888}, 
> >        Hyphenation = {english}, 
> >        Title = {A Reinterpretation of Does An Analysis of The Book}} 
> >    $ echo [@article] [@book] | pandoc -S -t markdown-citations 
> >    --bibliography quote-smart-flipflop.bib --csl chicago-author-date.csl 
> >    --- 
> >    bibliography: 'quote-smart-flipflop.bib' 
> >    csl: 'chicago-author-date.csl' 
> >    ... 
> >    (Author 1888) (Buthor 1888) 
> >    <div class="references"> 
> >    Author, Ann. 1888. A Reinterpretation of DoeS an Analysis of the 
> >    Book. 
> >    Buthor, Bob. 1888. *A Reinterpretation of DoeS an Analysis of the 
> >    Book*. 
> >    </div> 
> >    $ cat quote-dumb-flipflop.bib 
> >    @article{article, 
> >        Author = {Author, Ann}, 
> >        Hyphenation = {english}, 
> >        Date = {1888}, 
> >        Title = {A Reinterpretation of Does "An Analysis of "The Book""}} 
> >    @book{book, 
> >        Author = {Buthor, Bob}, 
> >        Hyphenation = {english}, 
> >        Date = {1888}, 
> >        Title = {A Reinterpretation of Does "An Analysis of "The Book""}} 
> >    $ echo [@article] [@book] | pandoc -S -t markdown-citations 
> >    --bibliography quote-dumb-flipflop.bib --csl chicago-author-date.csl 
> >    --- 
> >    bibliography: 'quote-dumb-flipflop.bib' 
> >    csl: 'chicago-author-date.csl' 
> >    ... 
> >    (Author 1888) (Buthor 1888) 
> >    <div class="references"> 
> >    Author, Ann. 1888. A Reinterpretation of DoeS "an Analysis of "the 
> >    Book"". 
> >    Buthor, Bob. 1888. *A Reinterpretation of DoeS "an Analysis of "the 
> >    Book""*. 
> >    </div> 
> >    On Monday, November 18, 2013 3:01:00 PM UTC, fiddlosopher wrote: 
> > 
> >      I'm not sure what you mean. Perhaps you could give an example of 
> >      what doesn't work? 
> >      +++ Nick Bart [Nov 18 13 04:14 ]: 
> >      >    Great. It does work with \enquote{} in biblatex title fields. 
> >      It does 
> >      >    not work, however with either smart () or dumb quotes. I 
> assume 
> >      there's 
> >      >    no good heuristic for parsing nested dumb quotes, but I wonder 
> >      whether 
> >      >    this could be made to work with smart quotes, just like 
> >      csquotes.sty 
> >      >    when active quotes are defined. 
> >      > 
> >      >    -- 
> >      >    You received this message because you are subscribed to the 
> >      Google 
> >      >    Groups "pandoc-discuss" group. 
> >      >    To unsubscribe from this group and stop receiving emails from 
> >      it, send 
> >      >    an email to [1]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >    To post to this group, send email to 
> >      [2]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
> >      >    To view this discussion on the web visit 
> >      >    [1][3]https://groups.google.com/d/msgid/pandoc-discuss/ 
> >      941bbcaa-276e-40ec- 
> >      >    bae6-baa209d26d30%[4]40googlegroups.com. 
> >      >    For more options, visit [2][5]https://groups.google.com/ 
> >      groups/opt_out. 
> >      > 
> >      > References 
> >      > 
> >      >    1. [6]https://groups.google.com/d/ 
> >      msgid/pandoc-discuss/941bbcaa-276e-40ec-bae6-baa209d26d30% 
> >      40googlegroups.com 
> >      >    2. [7]https://groups.google.com/groups/opt_out 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [8]
> https://groups.google.com/d/msgid/pandoc-discuss/4b28864c-fd8b-4215- 
> >    9c6d-ba75185bf61d%40googlegroups.com. 
> >    For more options, visit [9]https://groups.google.com/groups/opt_out. 
> > 
> > References 
> > 
> >    1. javascript:/ 
> >    2. javascript:/ 
> >    3. 
> https://groups.google.com/d/msgid/pandoc-discuss/941bbcaa-276e-40ec- 
> >    4. http://40googlegroups.com/ 
> >    5. https://groups.google.com/groups/opt_out 
> >    6. 
> https://groups.google.com/d/msgid/pandoc-discuss/941bbcaa-276e-40ec-bae6-baa209d26d30%40googlegroups.com 
> >    7. https://groups.google.com/groups/opt_out 
> >    8. 
> https://groups.google.com/d/msgid/pandoc-discuss/4b28864c-fd8b-4215-9c6d-ba75185bf61d%40googlegroups.com 
> >    9. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/60857372-304b-4d5a-9868-f24516a5badb%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 13267 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've changed the latex reader so that content between dumb double-quotes
and unicode curly quotes (both double and single) will be parsed as
Quoted.  This should make all of your examples work as you expected.

I've also fixed the Doe'S bug.

+++ Nick Bart [Nov 18 13 11:21 ]:
>    I see. Bit of a misunderstanding, sorry: Since I stopped using anything
>    but UTF8 smart quotes, I totally forgot about `` and '', in my mental
>    map now stored somewhere in the latex prehistory department.
>    So, my request would be to make UTF-8 smart quotes, and , function the
>    same way as `` and '' indeed already do.
>    Thus
>        Title = {A Reinterpretation of {Doe}s {\enquote{An Analysis of
>    \enquote{The Book}}}}
>        Title = {A Reinterpretation of {Doe}s {``An Analysis of ``The
>    Book''''}}
>        Title = {A Reinterpretation of {Doe}s {An Analysis of The Book}}
>    could all be used interchangeably.
>    I'm not actually interested in dumb quotes myself, but if youd say it's
>    easy to make them function like this as well, why not?
>    On Monday, 18 November 2013 18:22:26 UTC, fiddlosopher wrote:
> 
>      The "Doe'S" is a bug indeed.
>      The others, I'm not so sure what to do.  The way the system now
>      works,
>      pandoc-citeproc uses pandoc's latex reader to parse the contents of
>      biblatex fields, and in particular to recognize quoted sections.
>      So, this will be recognized as nested quotes (and flip-flopped
>      appropriately):
>          Title = {A Reinterpretaiton of Doe's ``An Analysis of `The
>      Book'\,''}
>      but if you use " or unicode curly quotes, they will just be passed
>      through verbatim.  That's how the latex reader works.  (You can
>      verify
>      by doing 'pandoc -f latex -t native'.)
>      I could change the latex reader so it tries to parse quoted sections
>      inside curly quotes.  That might be worth doing.  I could do the
>      same for dumb double quotes.  I wouldn't want to do it for dumb
>      single quotes, though, since a single ' already has a clear meaning
>      in LaTeX.
>      John
>      +++ Nick Bart [Nov 18 13 09:05 ]:
>      >    Sure. This is what I tried:
>      >    Note the uppercase S in DoeS, too.
>      >    $ cat quote-enquote-flipflop.bib
>      >    @article{article,
>      >        Author = {Author, Ann},
>      >        Date = {1888},
>      >        Hyphenation = {english},
>      >        Title = {A Reinterpretation of Does {\enquote{An Analysis
>      of
>      >    \enquote{The Book}}}},
>      >        Annote = {This entry uses the \verb+\enquote{}+ command
>      from the
>      >    latex csquotes.sty package.}}
>      >    @book{book,
>      >        Author = {Buthor, Bob},
>      >        Date = {1888},
>      >        Hyphenation = {english},
>      >        Title = {A Reinterpretation of Does {\enquote{An Analysis
>      of
>      >    \enquote{The Book}}}},
>      >        Annote = {This entry uses the \verb+\enquote{}+ command
>      from the
>      >    latex csquotes.sty package.}}
>      >    $ echo [@article] [@book] | pandoc -S -t markdown-citations
>      >    --bibliography quote-enquote-flipflop.bib --csl
>      chicago-author-date.csl
>      >    ---
>      >    bibliography: 'quote-enquote-flipflop.bib'
>      >    csl: 'chicago-author-date.csl'
>      >    ...
>      >    (Author 1888) (Buthor 1888)
>      >    <div class="references">
>      >    Author, Ann. 1888. A Reinterpretation of DoeS An Analysis of
>      The
>      >    Book.
>      >    Buthor, Bob. 1888. *A Reinterpretation of DoeS An Analysis of
>      The
>      >    Book*.
>      >    </div>
>      >    $ cat quote-smart-flipflop.bib
>      >    @article{article,
>      >        Author = {Author, Ann},
>      >        Date = {1888},
>      >        Hyphenation = {english},
>      >        Title = {A Reinterpretation of Does An Analysis of The
>      Book}}
>      >    @book{book,
>      >        Author = {Buthor, Bob},
>      >        Date = {1888},
>      >        Hyphenation = {english},
>      >        Title = {A Reinterpretation of Does An Analysis of The
>      Book}}
>      >    $ echo [@article] [@book] | pandoc -S -t markdown-citations
>      >    --bibliography quote-smart-flipflop.bib --csl
>      chicago-author-date.csl
>      >    ---
>      >    bibliography: 'quote-smart-flipflop.bib'
>      >    csl: 'chicago-author-date.csl'
>      >    ...
>      >    (Author 1888) (Buthor 1888)
>      >    <div class="references">
>      >    Author, Ann. 1888. A Reinterpretation of DoeS an Analysis of
>      the
>      >    Book.
>      >    Buthor, Bob. 1888. *A Reinterpretation of DoeS an Analysis of
>      the
>      >    Book*.
>      >    </div>
>      >    $ cat quote-dumb-flipflop.bib
>      >    @article{article,
>      >        Author = {Author, Ann},
>      >        Hyphenation = {english},
>      >        Date = {1888},
>      >        Title = {A Reinterpretation of Does "An Analysis of "The
>      Book""}}
>      >    @book{book,
>      >        Author = {Buthor, Bob},
>      >        Hyphenation = {english},
>      >        Date = {1888},
>      >        Title = {A Reinterpretation of Does "An Analysis of "The
>      Book""}}
>      >    $ echo [@article] [@book] | pandoc -S -t markdown-citations
>      >    --bibliography quote-dumb-flipflop.bib --csl
>      chicago-author-date.csl
>      >    ---
>      >    bibliography: 'quote-dumb-flipflop.bib'
>      >    csl: 'chicago-author-date.csl'
>      >    ...
>      >    (Author 1888) (Buthor 1888)
>      >    <div class="references">
>      >    Author, Ann. 1888. A Reinterpretation of DoeS "an Analysis of
>      "the
>      >    Book"".
>      >    Buthor, Bob. 1888. *A Reinterpretation of DoeS "an Analysis of
>      "the
>      >    Book""*.
>      >    </div>
>      >    On Monday, November 18, 2013 3:01:00 PM UTC, fiddlosopher
>      wrote:
>      >
>      >      I'm not sure what you mean. Perhaps you could give an example
>      of
>      >      what doesn't work?
>      >      +++ Nick Bart [Nov 18 13 04:14 ]:
>      >      >    Great. It does work with \enquote{} in biblatex title
>      fields.
>      >      It does
>      >      >    not work, however with either smart () or dumb quotes. I
>      assume
>      >      there's
>      >      >    no good heuristic for parsing nested dumb quotes, but I
>      wonder
>      >      whether
>      >      >    this could be made to work with smart quotes, just like
>      >      csquotes.sty
>      >      >    when active quotes are defined.
>      >      >
>      >      >    --
>      >      >    You received this message because you are subscribed to
>      the
>      >      Google
>      >      >    Groups "pandoc-discuss" group.
>      >      >    To unsubscribe from this group and stop receiving emails
>      from
>      >      it, send
>      >      >    an email to [1]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >    To post to this group, send email to
>      >      [2]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >      >    To view this discussion on the web visit
>      >      >    [1][3][1]https://groups.google.
>      com/d/msgid/pandoc-discuss/
>      >      941bbcaa-276e-40ec-
>      >      >    bae6-baa209d26d30%[4][2]40googlegroups.com.
>      >      >    For more options, visit [2][5][3]https://groups.google.
>      com/
>      >      groups/opt_out.
>      >      >
>      >      > References
>      >      >
>      >      >    1. [6][4]https://groups.google.com/d/
>      >      msgid/pandoc-discuss/941bbcaa-276e-40ec-bae6-baa209d26d30%
>      >      [5]40googlegroups.com
>      >      >    2. [7][6]https://groups.google.com/groups/opt_out
>      >
>      >    --
>      >    You received this message because you are subscribed to the
>      Google
>      >    Groups "pandoc-discuss" group.
>      >    To unsubscribe from this group and stop receiving emails from
>      it, send
>      >    an email to [7]pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To post to this group, send email to
>      [8]pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>      >    To view this discussion on the web visit
>      >    [8][9]https://groups.google.com/d/msgid/pandoc-discuss/
>      4b28864c-fd8b-4215-
>      >    9c6d-ba75185bf61d%[10]40googlegroups.com.
>      >    For more options, visit [9][11]https://groups.google.com/
>      groups/opt_out.
>      >
>      > References
>      >
>      >    1. javascript:/
>      >    2. javascript:/
>      >    3. [12]https://groups.google.com/d/
>      msgid/pandoc-discuss/941bbcaa-276e-40ec-
>      >    4. [13]http://40googlegroups.com/
>      >    5. [14]https://groups.google.com/groups/opt_out
>      >    6. [15]https://groups.google.com/d/
>      msgid/pandoc-discuss/941bbcaa-276e-40ec-bae6-baa209d26d30%
>      40googlegroups.com
>      >    7. [16]https://groups.google.com/groups/opt_out
>      >    8. [17]https://groups.google.com/d/
>      msgid/pandoc-discuss/4b28864c-fd8b-4215-9c6d-ba75185bf61d%
>      40googlegroups.com
>      >    9. [18]https://groups.google.com/groups/opt_out
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [19]https://groups.google.com/d/msgid/pandoc-discuss/60857372-304b-4d5a
>    -9868-f24516a5badb%40googlegroups.com.
>    For more options, visit [20]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/
>    2. http://40googlegroups.com/
>    3. https://groups.google.com/
>    4. https://groups.google.com/d/
>    5. http://40googlegroups.com/
>    6. https://groups.google.com/groups/opt_out
>    7. javascript:/
>    8. javascript:/
>    9. https://groups.google.com/d/msgid/pandoc-discuss/4b28864c-fd8b-4215-
>   10. http://40googlegroups.com/
>   11. https://groups.google.com/groups/opt_out
>   12. https://groups.google.com/d/msgid/pandoc-discuss/941bbcaa-276e-40ec-
>   13. http://40googlegroups.com/
>   14. https://groups.google.com/groups/opt_out
>   15. https://groups.google.com/d/msgid/pandoc-discuss/941bbcaa-276e-40ec-bae6-baa209d26d30%40googlegroups.com
>   16. https://groups.google.com/groups/opt_out
>   17. https://groups.google.com/d/msgid/pandoc-discuss/4b28864c-fd8b-4215-9c6d-ba75185bf61d%40googlegroups.com
>   18. https://groups.google.com/groups/opt_out
>   19. https://groups.google.com/d/msgid/pandoc-discuss/60857372-304b-4d5a-9868-f24516a5badb%40googlegroups.com
>   20. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've fixed the bug with pluralization of vol./vols.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1455 bytes --]

Hmm, I still get "vol." instead of "vols." with "number-of-volumes: 2"

$ cat number-of-volumes.md 
@item1

---
references:
- title: Title
  number-of-volumes: 2
  id: item1
  issued:
    year: 2013
  author:
    given:
    - Al
    family: Author
  publisher: Publisher
  type: book
  publisher-place: Location
  language: en-US
...
$ pandoc -sS --filter pandoc-citeproc -t markdown-citations 
number-of-volumes.md
---
references:
- author:
    family: Author
    given:
    - Al
  id: item1
  issued:
    year: 2013
  language: 'en-US'
  number-of-volumes: 2
  publisher: Publisher
  publisher-place: Location
  title: Title
  type: book
...

Author (2013)

<div class="references">

Author, Al. 2013. *Title*. 2 vol. Location: Publisher.

</div>


On Wednesday, 20 November 2013 04:46:50 UTC, fiddlosopher wrote:
>
> I've fixed the bug with pluralization of vol./vols. 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/8ba91205-1c5d-4944-9fc6-3dd3860ccfc1%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2202 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

OK, this should be fixed now.

+++ Nick Bart [Nov 19 13 22:56 ]:
> Hmm, I still get "vol." instead of "vols." with "number-of-volumes: 2"
> 
> $ cat number-of-volumes.md 
> @item1
> 
> ---
> references:
> - title: Title
>   number-of-volumes: 2
>   id: item1
>   issued:
>     year: 2013
>   author:
>     given:
>     - Al
>     family: Author
>   publisher: Publisher
>   type: book
>   publisher-place: Location
>   language: en-US
> ...
> $ pandoc -sS --filter pandoc-citeproc -t markdown-citations 
> number-of-volumes.md
> ---
> references:
> - author:
>     family: Author
>     given:
>     - Al
>   id: item1
>   issued:
>     year: 2013
>   language: 'en-US'
>   number-of-volumes: 2
>   publisher: Publisher
>   publisher-place: Location
>   title: Title
>   type: book
> ...
> 
> Author (2013)
> 
> <div class="references">
> 
> Author, Al. 2013. *Title*. 2 vol. Location: Publisher.
> 
> </div>
> 
> 
> On Wednesday, 20 November 2013 04:46:50 UTC, fiddlosopher wrote:
> >
> > I've fixed the bug with pluralization of vol./vols. 
> >
> >
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/8ba91205-1c5d-4944-9fc6-3dd3860ccfc1%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1272 bytes --]

biblatex ver 2.8 has renamed the "hyphenation" field to "langid" (still 
accepting the alias "hyphenation" for backwards compatibility), and 
introduced a new field "langidopts". 

Example:

langid         = {english},
langidopts     = {variant=british},

The langidopts field specifies variants as defined in the polyglossia 
package, currently for english and greek only:

For greek, there's nothing to do since there is only el-GR on the CSL side 
anyway.

For english, langid = {english} with langidopts = {variant = american, us, 
or usmax} should be mapped to en-US; and  langid = {english} with 
langidopts = {variant = british, uk, australian or newzealand} should be 
mapped to en-GB.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/c1b2e9fb-5ce7-44bc-aa96-f2f531cec906%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1627 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1022 bytes --]

Is there any method for specifying, in the metadata, the default-locale I 
want pandoc-citeproc to use?

What I'm looking for is a simpler method for what can already be achieved 
by using a customized CSL style file with, e.g., "default-locale="fr-FR"" 
specified in the "<style ...>" element.

The existing pandoc variable "lang" does not seem to work here, though my 
suggestion would be to use this for setting the CSL default-locale, too.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/b701d957-b4c3-417d-86d3-4d2e1d17f59d%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1296 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1470 bytes --]

The bug with "vols." not showing up for number-of-volumes seems to be 
fixed; however, the term before "volume" comes out wrong now.

expected:
Author, Al. 2013. *Title*. Vol. 2. Location: Publisher.
actual:
Author, Al. 2013. *Title*. Vols. 2. Location: Publisher.

$ cat volume.md 
@item1

---
references:
- title: Title
  volume: 2
  id: item1
  issued:
    year: 2013
  author:
    given:
    - Al
    family: Author
  publisher: Publisher
  type: book
  publisher-place: Location
  language: en-US
...

$ pandoc -sS --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks volume.md
---
references:
- author:
    family: Author
    given:
    - Al
  id: item1
  issued:
    year: 2013
  language: 'en-US'
  publisher: Publisher
  publisher-place: Location
  title: Title
  type: book
  volume: 2
...

Author (2013)

Author, Al. 2013. *Title*. Vols. 2. Location: Publisher.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/f5538813-bf05-4f74-9037-05f7439fe61e%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2077 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1905 bytes --]

Some regression with DOIs:

pandoc currently renders DOIs like this (with chicago-author-date.csl):

[doi:http://dx.doi.org/10.1000/182](http://dx.doi.org/10.1000/182 
"10.1000/182").

or even like this (with 
modern-humanities-research-association-author-date.csl):

... 77, 1234–1239[ 
\<doi:http://dx.doi.org/10.1086/520976\>](http://dx.doi.org/10.1086/520976 
"10.1086/520976")

DOIs should, however, be rendered without the "http://dx.doi.org/" part 
being visible, and with the prefix "doi:" (or some style-specific variation 
thereof) not being clickable, i.e., they should look like they did before 
(cf., e.g., kastenholz.biblatex):

doi:[10.1000/182](http://dx.doi.org/10.1000/182 "10.1000/182").

or maybe shorter:

doi:[10.1000/182](http://dx.doi.org/10.1000/182).

Background:

CMoS, 16e, 14.18, recommends (without saying anything on clickable links):

Novak, William J. “The Myth of the ‘Weak’ American State.” *American
Historical Review* 113 (June 2008): 752–72. doi:10.1086/ahr.113.3.752.

LaTeX with biblatex & hyperref renders DOIs like this (what's displayed in 
the pdf, reproduced in markdown syntax):

doi:[10.1000/182](http://dx.doi.org/10.1000/182).

Zotero, too, renders DOIs as follows (clickable links do not seem to be 
available):

doi:10.1000/182

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/d44c4d6e-f527-4b9f-8485-245e84c33325%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2211 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I think the latest push will improve things.

+++ Nick Bart [Nov 18 13 04:28 ]:
> Some regression with DOIs:
> 
> pandoc currently renders DOIs like this (with chicago-author-date.csl):
> 
> [doi:http://dx.doi.org/10.1000/182](http://dx.doi.org/10.1000/182 
> "10.1000/182").
> 
> or even like this (with 
> modern-humanities-research-association-author-date.csl):
> 
> ... 77, 1234–1239[ 
> \<doi:http://dx.doi.org/10.1086/520976\>](http://dx.doi.org/10.1086/520976 
> "10.1086/520976")
> 
> DOIs should, however, be rendered without the "http://dx.doi.org/" part 
> being visible, and with the prefix "doi:" (or some style-specific variation 
> thereof) not being clickable, i.e., they should look like they did before 
> (cf., e.g., kastenholz.biblatex):
> 
> doi:[10.1000/182](http://dx.doi.org/10.1000/182 "10.1000/182").
> 
> or maybe shorter:
> 
> doi:[10.1000/182](http://dx.doi.org/10.1000/182).
> 
> Background:
> 
> CMoS, 16e, 14.18, recommends (without saying anything on clickable links):
> 
> Novak, William J. “The Myth of the ‘Weak’ American State.” *American
> Historical Review* 113 (June 2008): 752–72. doi:10.1086/ahr.113.3.752.
> 
> LaTeX with biblatex & hyperref renders DOIs like this (what's displayed in 
> the pdf, reproduced in markdown syntax):
> 
> doi:[10.1000/182](http://dx.doi.org/10.1000/182).
> 
> Zotero, too, renders DOIs as follows (clickable links do not seem to be 
> available):
> 
> doi:10.1000/182
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/d44c4d6e-f527-4b9f-8485-245e84c33325%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131120180217.GA86634%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 11 13 02:26 ]:
 
> $ cat test-article-quotation-marks.bib
> @article{item1,
>     Author = {Author, Al},
>     Date = {2012},
>     Hyphenation = {english},
>     Journaltitle = {The Journaltitle},
>     Title = {The Title with a "Quotation" inside},
>     Subtitle = {and a "Quotation with "Another Quotation" inside", Too},
> }
> 
> $ echo @item1 | pandoc -S --filter pandoc-citeproc -t 
> markdown-citations-markdown_in_html_blocks --bibliography 
> test-article-quotation-marks.bib
> Author (2012)
> 
> Author, Al. 2012. “The Title with a ‘quotation’ Inside: and a ‘quotation
> with “another quotation” inside’, Too.” *The Journaltitle*.
> 
> So, quotation marks come out flawless, for both chicago-author-date and 
> apa. Only capitalization fails (for chicago).

I forgot to comment on the capitalization issue.
It's a messy issue.  There are a number of problems, not just with
quoted text, but with math (e.g. $\epsilon$ will render as a capital
epsilon).

I need to work with Andrea on this one.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131012052333.GA93270%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 8160 bytes --]

A few more things I noticed:

biblatex number field
---------------------

This seems to be mapped to issue for all isArticle types now. Good.


biblatex type field
-------------------

This may contain not only keys but also free text, including latex 
formatting; should be read by getField, not getRawField, if possible.

series in isArticle
-------------------

There should be a comma, not a period between 
journal(title)/journalsubtitle and series.

biblatex @xdata entries
-----------------------

According to the biblatex manual, @xdata entries may be cascading; 
biblio2yaml apparently doesn't do that yet.

Double punctuation between title and subtitle
---------------------------------------------

Double punctuation as in "Title?: Subtitle" should be avoided: So, when 
assembling CSL title from biblatex title, subtitle, and titleaddon, "!:" or 
"!." should be replaced by "!"; and "?:" or "?." by "?". In other words, 
addColon should not add a colon if its argument string ends with "!" or 
"?". Same for addPeriod. – CMoS specifically mentions "!" and "?", no other 
punctuation; see 14.105, 14.178.

Comma before “Jr.”
------------------

biblatex and biblatex-chicago, in accordance with CMoS 6.47, default to not 
putting a comma before of "Jr.", "Sr.", etc. A comma can be forced in 
biblatex-chicago by inserting "juniorcomma=true" in the "options" field. 
Thus, biblio2yaml should output "comma-suffix: false" by default, and only 
switch to "comma-suffix: true" in case the biblatex options field contains 
"juniorcomma=true".

(The syntax proposed in 
<http://gorgias.mine.nu/repos/citeproc-hs/#name-parsing>, introducing a "!" 
to force a junior comma, cannot be recommended since this breaks 
latex/biblatex.)

Organization containing literal “and”
-------------------------------------

"Centre for This {and} That" works in biblio2yaml, "{Centre for This and 
That}" should, but does not. Example:

````
$ cat test-orga.bib
@proceedings{item1,
    Author = {Test1},
    Organization = {Centre for This {and} That and Department of That {and} 
This},
    Title = {Organization field contains \{and\} in organization names}}

@proceedings{item2,
    Author = {Test2},
    Organization = {{Centre for This and That} and {Department of That and 
This}},
    Title = {Organization field contains organization names wrapped in 
\{\}}}
````


issuetitleaddon
---------------

According to the biblatex manual, there is no field issuetitleaddon, and 
neither have I seen this in biblatex-chicago or biblatex-apa. Should be 
removed.

duplicate keys
--------------

A warning could be issued if the database(s) used contain(s) duplicate keys.

Some urls could be simplified
-----------------------------

````
$ cat test.md
@item1 @item2

---
references:
- title: Title
  id: item1
  author:
  - given:
    - Joe-Baptiste
    family: Doe
  - suffix: Jr.
    given:
    - Carl
    comma-suffix: false
    family: "Linn\xE9"
    non-dropping-particle: von
  url: http://abc.def.gh
  type: book
- title: Title
  id: item2
  issued:
    year: 2013
  author:
    given:
    - Eric
    family: Van Rompuy
  url: 
https://en.wikipedia.org/wiki/Van_%28Dutch%29#Collation_and_capitalisation
  type: article-journal
  language: en-US
...
$ pandoc -S --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks z.md
Doe and von Linné Van Rompuy (2013)

Doe, Joe-Baptiste, and Carl von Linné Jr. *Title*. <http://abc.def.gh>.

Van Rompuy, Eric. 2013. “Title.”
[https://en.wikipedia.org/wiki/Van\_%28Dutch%29\#Collation\_and\_capitalisation](https://en.wikipedia.org/wiki/Van_%28Dutch%29#Collation_and_capitalisation 
"https://en.wikipedia.org/wiki/Van_%28Dutch%29#Collation_and_capitalisation").
````

Also, see "Linné" above, is there a reason why biblio2yaml escapes *some* 
accented characters (but not all, see below)? Readability would improve if 
as few as possible were escaped.

````
$ cat test-collation-2.bib 
@collection{item1,
    Date = {2012},
    Editor = {Fernán Gómez, Fernando},
    Title = {Title}}
$ biblio2yaml test-collation-2.bib 
---
references:
- title: Title
  id: item1
  issued:
    year: 2012
  editor:
    given:
    - Fernando
    family: "Fernán Gómez"
  type: book
...
````

DOI
---

citeproc-hs has a feature, described in 
<http://gorgias.mine.nu/repos/citeproc-hs/#the-doi-variable> that if a Doi 
field starts with the prefix doi:, as in

    Doi = {doi:10.1038/171737a0},

a link is generated. Basically, this is a nice feature, but the doi: prefix 
is problematic, since it breaks biblatex (with "doi: doi:" appearing in the 
output). Would it be a problem if pandoc-citeproc simply generated such 
links for *all* doi fields by default, in as simple a form as possible?


Date disambiguation
-------------------

In the following example, no date disambiguation should come into effect:

```
$ cat test-date-range-2.bib@article{item1,
    Author = {Author, Al},
    Date = {2003/2004},
    Hyphenation = {english},
    Title = {Title},
}
@article{item2,
    Author = {Author, Al},
    Date = {2003},
    Hyphenation = {english},
    Title = {Title},
}
$ echo @item1 @item2 | pandoc  --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks --bibliography 
test-date-range-2.bib 
Author (2003–2004b) Author (2003a)

Author, Al. 2003a. “Title.”

———. 2003–2004b. “Title.”
````


Small caps
----------

In <https://github.com/jgm/pandoc/issues/931> rmzelle pointed to 
<https://www.zotero.org/support/kb/rich_text_bibliography> for the most 
recent docs on CSL inline formatting. This has `<span 
style="font-variant:small-caps;"/>` instead of `<sc>`. I don't know whether 
the "new" form has any advantages, if it doesn't, don't change it.

Related to inline formatting: Do you see any possibility, e.g., by some 
kind of inline formatting command, to *suppress* conversion to title case 
by pandoc-citeproc? For example, "Homo sapiens" should *never* be converted 
to "Homo Sapiens", but currently there seems to be no way to avoid that in 
any of the citeproc programs (except for changing the language, and thus 
suppressing title case conversion for that entry altogether). Bibtex of 
course never has such problems since strings can be explicitly protected. 
The CSL approach, by contrast, using sentence case in databases and no 
explicitly protected strings will never get 100% accuracy here, but seems 
to assume that the output can always be manually corrected after the 
processor has run. This, however, is (next to) impossible if the output 
format is, e.g., pdf, or epub. So with pandoc, it would be better if 
appropriate input together with an optimized processor could produce 
correct output requiring no additional intervention. I could imagine using 
something like the latex command `\MakeLowercase{}` here for cases in which 
conversion to title case *must* be suppressed, as in `title = {A History of 
\textit{Homo \MakeLowercase{sapiens}}},`. This would not affect latex 
output, but biblio2yaml could convert this to something like `title: "A 
history of <i>Homo <protect>sapiens</protect><i>"`, which in turn should 
keep pandoc-citeproc from converting "sapiens" to title case, even if a 
style requests this for titles in general.

Finally, on the question of a possible syntax for small caps in markdown, 
what about `^^small caps^^`?


-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/01c4f8e0-bcb6-4f43-a456-1b3194e68b01%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 9300 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 13 13 16:03 ]:
>    A few more things I noticed:
>    biblatex number field
>    ---------------------
>    This seems to be mapped to issue for all isArticle types now. Good.
>    biblatex type field
>    -------------------
>    This may contain not only keys but also free text, including latex
>    formatting; should be read by getField, not getRawField, if possible.

Done.

>    series in isArticle
>    -------------------
>    There should be a comma, not a period between
>    journal(title)/journalsubtitle and series.

Done.

>    biblatex @xdata entries
>    -----------------------
>    According to the biblatex manual, @xdata entries may be cascading;
>    biblio2yaml apparently doesn't do that yet.

Can you give me a test case?

>    Double punctuation between title and subtitle
>    ---------------------------------------------
>    Double punctuation as in "Title?: Subtitle" should be avoided: So, when
>    assembling CSL title from biblatex title, subtitle, and titleaddon,
>    "!:" or "!." should be replaced by "!"; and "?:" or "?." by "?". In
>    other words, addColon should not add a colon if its argument string
>    ends with "!" or "?". Same for addPeriod. CMoS specifically mentions
>    "!" and "?", no other punctuation; see 14.105, 14.178.

Done, I think.

Got to go now, I'll save the rest for later.

>    Comma before Jr.
>    ------------------
>    biblatex and biblatex-chicago, in accordance with CMoS 6.47, default to
>    not putting a comma before of "Jr.", "Sr.", etc. A comma can be forced
>    in biblatex-chicago by inserting "juniorcomma=true" in the "options"
>    field. Thus, biblio2yaml should output "comma-suffix: false" by
>    default, and only switch to "comma-suffix: true" in case the biblatex
>    options field contains "juniorcomma=true".
>    (The syntax proposed in
>    <http://gorgias.mine.nu/repos/citeproc-hs/#name-parsing>, introducing a
>    "!" to force a junior comma, cannot be recommended since this breaks
>    latex/biblatex.)
>    Organization containing literal and
>    -------------------------------------
>    "Centre for This {and} That" works in biblio2yaml, "{Centre for This
>    and That}" should, but does not. Example:
>    ````
>    $ cat test-orga.bib
>    @proceedings{item1,
>        Author = {Test1},
>        Organization = {Centre for This {and} That and Department of That
>    {and} This},
>        Title = {Organization field contains \{and\} in organization
>    names}}
>    @proceedings{item2,
>        Author = {Test2},
>        Organization = {{Centre for This and That} and {Department of That
>    and This}},
>        Title = {Organization field contains organization names wrapped in
>    \{\}}}
>    ````
>    issuetitleaddon
>    ---------------
>    According to the biblatex manual, there is no field issuetitleaddon,
>    and neither have I seen this in biblatex-chicago or biblatex-apa.
>    Should be removed.
>    duplicate keys
>    --------------
>    A warning could be issued if the database(s) used contain(s) duplicate
>    keys.
>    Some urls could be simplified
>    -----------------------------
>    ````
>    $ cat test.md
>    @item1 @item2
>    ---
>    references:
>    - title: Title
>      id: item1
>      author:
>      - given:
>        - Joe-Baptiste
>        family: Doe
>      - suffix: Jr.
>        given:
>        - Carl
>        comma-suffix: false
>        family: "Linn\xE9"
>        non-dropping-particle: von
>      url: http://abc.def.gh
>      type: book
>    - title: Title
>      id: item2
>      issued:
>        year: 2013
>      author:
>        given:
>        - Eric
>        family: Van Rompuy
>      url:
>    https://en.wikipedia.org/wiki/Van_%28Dutch%29#Collation_and_capitalisat
>    ion
>      type: article-journal
>      language: en-US
>    ...
>    $ pandoc -S --filter pandoc-citeproc -t
>    markdown-citations-markdown_in_html_blocks z.md
>    Doe and von Linn Van Rompuy (2013)
>    Doe, Joe-Baptiste, and Carl von Linn Jr. *Title*. <http://abc.def.gh>.
>    Van Rompuy, Eric. 2013. Title.
>    [https://en.wikipedia.org/wiki/Van\_%28Dutch%29\#Collation\_and\_capita
>    lisation](https://en.wikipedia.org/wiki/Van_%28Dutch%29#Collation_and_c
>    apitalisation
>    "https://en.wikipedia.org/wiki/Van_%28Dutch%29#Collation_and_capitalisa
>    tion").
>    ````
>    Also, see "Linn" above, is there a reason why biblio2yaml escapes
>    *some* accented characters (but not all, see below)? Readability would
>    improve if as few as possible were escaped.
>    ````
>    $ cat test-collation-2.bib
>    @collection{item1,
>        Date = {2012},
>        Editor = {Fernn Gmez, Fernando},
>        Title = {Title}}
>    $ biblio2yaml test-collation-2.bib
>    ---
>    references:
>    - title: Title
>      id: item1
>      issued:
>        year: 2012
>      editor:
>        given:
>        - Fernando
>        family: "Fernn Gmez"
>      type: book
>    ...
>    ````
>    DOI
>    ---
>    citeproc-hs has a feature, described in
>    <http://gorgias.mine.nu/repos/citeproc-hs/#the-doi-variable> that if a
>    Doi field starts with the prefix doi:, as in
>        Doi = {doi:10.1038/171737a0},
>    a link is generated. Basically, this is a nice feature, but the doi:
>    prefix is problematic, since it breaks biblatex (with "doi: doi:"
>    appearing in the output). Would it be a problem if pandoc-citeproc
>    simply generated such links for *all* doi fields by default, in as
>    simple a form as possible?
>    Date disambiguation
>    -------------------
>    In the following example, no date disambiguation should come into
>    effect:
>    ```
>    $ cat test-date-range-2.bib@article{item1,
>        Author = {Author, Al},
>        Date = {2003/2004},
>        Hyphenation = {english},
>        Title = {Title},
>    }
>    @article{item2,
>        Author = {Author, Al},
>        Date = {2003},
>        Hyphenation = {english},
>        Title = {Title},
>    }
>    $ echo @item1 @item2 | pandoc  --filter pandoc-citeproc -t
>    markdown-citations-markdown_in_html_blocks --bibliography
>    test-date-range-2.bib
>    Author (20032004b) Author (2003a)
>    Author, Al. 2003a. Title.
>    . 20032004b. Title.
>    ````
>    Small caps
>    ----------
>    In <https://github.com/jgm/pandoc/issues/931> rmzelle pointed to
>    <https://www.zotero.org/support/kb/rich_text_bibliography> for the most
>    recent docs on CSL inline formatting. This has `<span
>    style="font-variant:small-caps;"/>` instead of `<sc>`. I don't know
>    whether the "new" form has any advantages, if it doesn't, don't change
>    it.
>    Related to inline formatting: Do you see any possibility, e.g., by some
>    kind of inline formatting command, to *suppress* conversion to title
>    case by pandoc-citeproc? For example, "Homo sapiens" should *never* be
>    converted to "Homo Sapiens", but currently there seems to be no way to
>    avoid that in any of the citeproc programs (except for changing the
>    language, and thus suppressing title case conversion for that entry
>    altogether). Bibtex of course never has such problems since strings can
>    be explicitly protected. The CSL approach, by contrast, using sentence
>    case in databases and no explicitly protected strings will never get
>    100% accuracy here, but seems to assume that the output can always be
>    manually corrected after the processor has run. This, however, is (next
>    to) impossible if the output format is, e.g., pdf, or epub. So with
>    pandoc, it would be better if appropriate input together with an
>    optimized processor could produce correct output requiring no
>    additional intervention. I could imagine using something like the latex
>    command `\MakeLowercase{}` here for cases in which conversion to title
>    case *must* be suppressed, as in `title = {A History of \textit{Homo
>    \MakeLowercase{sapiens}}},`. This would not affect latex output, but
>    biblio2yaml could convert this to something like `title: "A history of
>    <i>Homo <protect>sapiens</protect><i>"`, which in turn should keep
>    pandoc-citeproc from converting "sapiens" to title case, even if a
>    style requests this for titles in general.
>    Finally, on the question of a possible syntax for small caps in
>    markdown, what about `^^small caps^^`?
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/01c4f8e0-bcb6-4f43-
>    a456-1b3194e68b01%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/01c4f8e0-bcb6-4f43-a456-1b3194e68b01%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1241 bytes --]

Sure. Adapted from the biblatex manual:

@xdata{macmillan:name,
    Publisher = {Macmillan}}

@xdata{macmillan:place,
    Location = {New York and London}}

@xdata{macmillan,
    Xdata = {macmillan:name, macmillan:place}}

@book{item1,
    Author = {Author, Ann},
    Date = {2012},
    Title = {Title},
    Xdata = {macmillan}}



On Monday, October 14, 2013 12:56:01 AM UTC, fiddlosopher wrote:
>
> +++ Nick Bart [Oct 13 13 16:03 ]: 
> >    biblatex @xdata entries 
> >    ----------------------- 
> >    According to the biblatex manual, @xdata entries may be cascading; 
> >    biblio2yaml apparently doesn't do that yet. 
>
> Can you give me a test case? 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/2a6dfa90-5eb4-42a2-b0df-ed51e1533e74%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1856 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Fixed, I think.

+++ Nick Bart [Oct 13 13 22:18 ]:
> Sure. Adapted from the biblatex manual:
> 
> @xdata{macmillan:name,
>     Publisher = {Macmillan}}
> 
> @xdata{macmillan:place,
>     Location = {New York and London}}
> 
> @xdata{macmillan,
>     Xdata = {macmillan:name, macmillan:place}}
> 
> @book{item1,
>     Author = {Author, Ann},
>     Date = {2012},
>     Title = {Title},
>     Xdata = {macmillan}}
> 
> 
> 
> On Monday, October 14, 2013 12:56:01 AM UTC, fiddlosopher wrote:
> >
> > +++ Nick Bart [Oct 13 13 16:03 ]: 
> > >    biblatex @xdata entries 
> > >    ----------------------- 
> > >    According to the biblatex manual, @xdata entries may be cascading; 
> > >    biblio2yaml apparently doesn't do that yet. 
> >
> > Can you give me a test case? 
> >
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/2a6dfa90-5eb4-42a2-b0df-ed51e1533e74%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 13 13 16:03 ]:
 
> Some urls could be simplified

This is fixed in pandoc -- if you install the latest source from github.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1847 bytes --]

Not all seems to be well with date disambiguation: Instead of the expected 

Author (2003a); Author (2003b); Buthor (2003a); Buthor (2003b)

the output in the example below is

Author (2003a); Author (2003b); Buthor (2003c); Buthor (2003d)


$ cat test-date-disambiguation.md 
@item1; @item2; @item3; @item4

---
references:
- title: Title
  id: item1
  issued:
    year: 2003
  author:
    given:
    - Al
    family: Author
  type: article-journal
  language: en-US
- title: Title two
  id: item2
  issued:
    year: 2003
  author:
    given:
    - Al
    family: Author
  type: article-journal
  language: en-US
- title: Title
  id: item3
  issued:
    year: 2003
  author:
    given:
    - Al
    family: Buthor
  type: article-journal
  language: en-US
- title: Title four
  id: item4
  issued:
    year: 2003
  author:
    given:
    - Al
    family: Buthor
  type: article-journal
  language: en-US
...
$ pandoc -S --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks test-date-disambiguation.md
Author (2003a); Author (2003b); Buthor (2003c); Buthor (2003d)

Author, Al. 2003a. “Title.”

———. 2003b. “Title Two.”

Buthor, Al. 2003c. “Title.”

———. 2003d. “Title Four.”

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a6478e98-967e-4c44-9298-263435722095%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2573 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1248 bytes --]

isArticle / number / issue:

If number is empty, no space+comma should appear in front of issue. – 
Example:

$ cat test-issue-only.bib 
@article{item1,
    Author = {Author, Al},
    Date = {2012},
    Hyphenation = {english},
    Issue = {Spring},
    Journaltitle = {Journaltitle},
    Pages = {S12-S34},
    Title = {Title},
    Volume = {5},
}
$ biblio2yaml test-issue-only.bib 
---
references:
- issue: ! ', Spring'
  page: S12-S34
  title: Title
  volume: 5
  id: item1
  issued:
    year: 2012
  author:
    given:
    - Al
    family: Author
  container-title: Journaltitle
  type: article-journal
  language: en-US
...

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/7554cc47-919e-451a-aac1-78986fdedd71%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1795 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Fixed.

+++ Nick Bart [Oct 15 13 07:12 ]:
>    isArticle / number / issue:
>    If number is empty, no space+comma should appear in front of issue.
>    Example:
>    $ cat test-issue-only.bib
>    @article{item1,
>        Author = {Author, Al},
>        Date = {2012},
>        Hyphenation = {english},
>        Issue = {Spring},
>        Journaltitle = {Journaltitle},
>        Pages = {S12-S34},
>        Title = {Title},
>        Volume = {5},
>    }
>    $ biblio2yaml test-issue-only.bib
>    ---
>    references:
>    - issue: ! ', Spring'
>      page: S12-S34
>      title: Title
>      volume: 5
>      id: item1
>      issued:
>        year: 2012
>      author:
>        given:
>        - Al
>        family: Author
>      container-title: Journaltitle
>      type: article-journal
>      language: en-US
>    ...
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/7554cc47-919e-451a-
>    aac1-78986fdedd71%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/7554cc47-919e-451a-aac1-78986fdedd71%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

This is due to some changes I brought in from Andrea's citeproc-hs.
I've reverted them for now, though I think we'll want some version
of them eventually.

+++ Nick Bart [Oct 15 13 07:06 ]:
> Not all seems to be well with date disambiguation: Instead of the expected 
> 
> Author (2003a); Author (2003b); Buthor (2003a); Buthor (2003b)
> 
> the output in the example below is
> 
> Author (2003a); Author (2003b); Buthor (2003c); Buthor (2003d)
> 
> 
> $ cat test-date-disambiguation.md 
> @item1; @item2; @item3; @item4
> 
> ---
> references:
> - title: Title
>   id: item1
>   issued:
>     year: 2003
>   author:
>     given:
>     - Al
>     family: Author
>   type: article-journal
>   language: en-US
> - title: Title two
>   id: item2
>   issued:
>     year: 2003
>   author:
>     given:
>     - Al
>     family: Author
>   type: article-journal
>   language: en-US
> - title: Title
>   id: item3
>   issued:
>     year: 2003
>   author:
>     given:
>     - Al
>     family: Buthor
>   type: article-journal
>   language: en-US
> - title: Title four
>   id: item4
>   issued:
>     year: 2003
>   author:
>     given:
>     - Al
>     family: Buthor
>   type: article-journal
>   language: en-US
> ...
> $ pandoc -S --filter pandoc-citeproc -t 
> markdown-citations-markdown_in_html_blocks test-date-disambiguation.md
> Author (2003a); Author (2003b); Buthor (2003c); Buthor (2003d)
> 
> Author, Al. 2003a. “Title.”
> 
> ———. 2003b. “Title Two.”
> 
> Buthor, Al. 2003c. “Title.”
> 
> ———. 2003d. “Title Four.”
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a6478e98-967e-4c44-9298-263435722095%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131016013237.GA37683%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 994 bytes --]

Did a fresh "git clone git://github.com/jgm/pandoc", but the long URL from 
my example did not change in the output. Did you mean anything else by 
"install the latest source from github"?

On Tuesday, October 15, 2013 6:06:38 AM UTC, fiddlosopher wrote:
>
> +++ Nick Bart [Oct 13 13 16:03 ]: 
>   
> > Some urls could be simplified 
>
> This is fixed in pandoc -- if you install the latest source from github. 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a1cb129e-221c-40bc-9431-617be8116006%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1407 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2740 bytes --]

Editor and translator: It seems that the issue with spurious "edited and 
translated by" strings when there isn't a translator at all has been 
solved. If both editor and translator exist and are identical, though, the 
output is not as expected:

item1: an editor, no translator -> as expected

item2: editor = translator -> as expected, and acceptable, though not 
ideal; apparently it's a limitation of CSL that "Editor, Edward, ed. & 
trans. 2013a. [...]" is not possible (according to a discussion on a Zotero 
forum, I can look it up if needed)

item3: editor = translator -> here the output should be "... , edited and 
translated by Edward Editor. ...", not only  "... , edited by Edward 
Editor. ..."


$ cat test-editortranslator.bib

@collection{item1,
    Date = {2013},
    Editor = {Editor, Edward},
    Location = {Location},
    Publisher = {Publisher},
    Title = {item1: Title of a Collection by an Editor, no Translator}}

@collection{item2,
    Date = {2013},
    Editor = {Editor, Edward},
    Translator = {Editor, Edward},
    Location = {Location},
    Publisher = {Publisher},
    Title = {item2: Title of a Collection by an Editor who is also 
Translator}}

@incollection{item3,
    Author = {Author, Al},
    Booktitle = {item3: Title of a Collection by an Editor who is also 
Translator},
    Date = {2013},
    Editor = {Editor, Edward},
    Location = {Location},
    Publisher = {Publisher},
    Title = {The Chapter Title},
    Translator = {Editor, Edward}}

$ echo @item1, @item2, @item3 | pandoc --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks --bibliography 
test-editortranslator.bib
Editor (2013a), Editor (2013b), Author (2013)

Author, Al. 2013. “The Chapter Title.” In *Item3: Title of a Collection
by an Editor Who Is Also Translator*, edited by Edward Editor. Location:
Publisher.

Editor, Edward, ed. 2013a. *Item1: Title of a Collection by an Editor,
No Translator*. Location: Publisher.

———, ed. 2013b. *Item2: Title of a Collection by an Editor Who Is Also
Translator*. Translated by Edward Editor. Location: Publisher.


-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/818df23f-ff11-4357-824b-7b863c75332d%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3376 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1121 bytes --]

Also,biblio2yaml does not produce *any* output from @periodical any more:

$ cat test-special-issue.bib 
@periodical{item1,
    Date = {2013},
    Editor = {Editor, Edward},
    Entrysubtype = {magazine},
    Hyphenation = {english},
    Issuetitle = {Title of the Special Issue},
    Location = {Location},
    Note = {Special Issue},
    Number = {3},
    Publisher = {Publisher},
    Series = {newseries},
    Title = {Title of the Journal},
    Volume = {7},
    Year = {2013}}
$ biblio2yaml test-special-issue.bib 
---
references:
[]
...

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/4bb3bb13-b90b-4204-90a6-78db9705262a%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1647 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I'll look into it.  Meanwhile, it occurs to me that I've relied too much
on the "ask Nick Bart" method of testing.  We should automate the
testing as much as possible.  I've added a framework for a biblio2yaml
test suite.  I'm hoping you can contribute some of the tests you're
running.

It's very simple.  Each test is just two files in the directory
tests/biblio2yaml.  The first file should be a biblatex file with a
descriptive name, like

nested-xref-resolution.bib

The second file should be the yaml you think should result,
with the same name as the biblatex file with a .yaml extension
appended:

nested-xref-resolution.bib.yaml

Doing 'cabal install --enable-tests' will run the tests automatically
for any files in this directory.  YAML files will be compared at the
semantic level, so things like ordering of fields won't matter.

It seems that you already have a lot of tests.  If you could break
them down into small, focused pieces, and send me the paired files,
I'll add them to the repository.  This should help to avoid regressions
in the future, and give us some confidence that we're getting things
right.


+++ Nick Bart [Oct 15 13 07:45 ]:
>    Also,biblio2yaml does not produce *any* output from @periodical any
>    more:
>    $ cat test-special-issue.bib
>    @periodical{item1,
>        Date = {2013},
>        Editor = {Editor, Edward},
>        Entrysubtype = {magazine},
>        Hyphenation = {english},
>        Issuetitle = {Title of the Special Issue},
>        Location = {Location},
>        Note = {Special Issue},
>        Number = {3},
>        Publisher = {Publisher},
>        Series = {newseries},
>        Title = {Title of the Journal},
>        Volume = {7},
>        Year = {2013}}
>    $ biblio2yaml test-special-issue.bib
>    ---
>    references:
>    []
>    ...
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/4bb3bb13-b90b-4204-
>    90a6-78db9705262a%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/4bb3bb13-b90b-4204-90a6-78db9705262a%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've changed the framework a bit.  Now only one file per test is
needed.  It should have extension '.biblatex' or '.bibtex', depending
on the format.  It should begin with a bibtex/biblatex section,
then have a YAML section after a line with three hyphens.
(See the example in tests/biblio2yaml.)

The nice thing is that these files count as valid bibtex/biblatex
files, since the YAML part is treated as a comment.  So you can run
biblio2yaml directly on them.

+++ John MacFarlane [Oct 15 13 11:16 ]:
> I'll look into it.  Meanwhile, it occurs to me that I've relied too much
> on the "ask Nick Bart" method of testing.  We should automate the
> testing as much as possible.  I've added a framework for a biblio2yaml
> test suite.  I'm hoping you can contribute some of the tests you're
> running.
> 
> It's very simple.  Each test is just two files in the directory
> tests/biblio2yaml.  The first file should be a biblatex file with a
> descriptive name, like
> 
> nested-xref-resolution.bib
> 
> The second file should be the yaml you think should result,
> with the same name as the biblatex file with a .yaml extension
> appended:
> 
> nested-xref-resolution.bib.yaml
> 
> Doing 'cabal install --enable-tests' will run the tests automatically
> for any files in this directory.  YAML files will be compared at the
> semantic level, so things like ordering of fields won't matter.
> 
> It seems that you already have a lot of tests.  If you could break
> them down into small, focused pieces, and send me the paired files,
> I'll add them to the repository.  This should help to avoid regressions
> in the future, and give us some confidence that we're getting things
> right.
> 
> 
> +++ Nick Bart [Oct 15 13 07:45 ]:
> >    Also,biblio2yaml does not produce *any* output from @periodical any
> >    more:
> >    $ cat test-special-issue.bib
> >    @periodical{item1,
> >        Date = {2013},
> >        Editor = {Editor, Edward},
> >        Entrysubtype = {magazine},
> >        Hyphenation = {english},
> >        Issuetitle = {Title of the Special Issue},
> >        Location = {Location},
> >        Note = {Special Issue},
> >        Number = {3},
> >        Publisher = {Publisher},
> >        Series = {newseries},
> >        Title = {Title of the Journal},
> >        Volume = {7},
> >        Year = {2013}}
> >    $ biblio2yaml test-special-issue.bib
> >    ---
> >    references:
> >    []
> >    ...
> > 
> >    --
> >    You received this message because you are subscribed to the Google
> >    Groups "pandoc-discuss" group.
> >    To unsubscribe from this group and stop receiving emails from it, send
> >    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> >    To view this discussion on the web visit
> >    [1]https://groups.google.com/d/msgid/pandoc-discuss/4bb3bb13-b90b-4204-
> >    90a6-78db9705262a%40googlegroups.com.
> >    For more options, visit [2]https://groups.google.com/groups/opt_out.
> > 
> > References
> > 
> >    1. https://groups.google.com/d/msgid/pandoc-discuss/4bb3bb13-b90b-4204-90a6-78db9705262a%40googlegroups.com
> >    2. https://groups.google.com/groups/opt_out
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131015181630.GA28168%40protagoras.phil.berkeley.edu.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 11806 bytes --]

Alright. First two instalments, including comments, and a proposal on 
title-case, below:

$ cat text-case-conversion.biblatex 
@comment{
    - bibtex and biblatex
        - expect titles in title case
        - styles use titles as is, or convert them to sentence case
        - strings wrapped {} are not converted
    - all CSL styles at <http://github.com/citation-style-language/styles> 
and <https://www.zotero.org/styles/>
        - expect titles in sentence case
        - styles use titles as is, or convert them to title case
        - except for (hardcoded) list of stop words, see 
<http://citationstyles.org/downloads/specification.html#title-case-conversion>
        - citeproc-js (MLZ only?) also recognizes a markup syntax for 
suppressing title-case changes on a range of text (see 
<https://forums.zotero.org/discussion/21991/excessive-capitalization-of-citation-titles/#Item_22>):
            - `<span class="nocase"/>lowercase</span>`
    - Proposal:
        - When converting to yaml, convert English titles to sentence case, 
            - for all strings wrapped in {} where {} is not part of a latex 
command, ...
            - ... when starting with an uppercase letter: suppress 
conversion, remove the {}
            - ... when starting with a lowercase letter ("nm", "iPod"): 
suppress conversion, replace the {} with <span class="nocase"/></span>
            - Note: Camel case ("iPod") needs to be protected in 
bibtex/biblatex anyway; the only "extension" (wrt bibtex/biblatex specs) 
we'd be introducing is wrapping lowercase-only strings in {}, something 
that is never necessary on the latex side but won't break anything there 
either.
        - citeproc-hs/pandoc-citeproc should be modified to honour this new 
syntax and suppress conversion to title case for strings wrapped in `<span 
class="nocase"/></span>`.
        - Expected output, using one of the title-case CSL styles, here 
chicago-author-date.csl:
        
            Author, Ann. 2013. “A Title, in English, with a Proper Name and 
an
            ACRONYM and a camelCase Word and Some Units, 400 nm, 3 cm, and 
a Quote,
            *Alea iacta est*.” *Journal*.
    }

@article{item1,
    Author = {Author, Ann},
    Date = {2013},
    Hyphenation = {english},
    Journaltitle = {Journal},
    Title = {A Title, in {English}, with a {Proper Name} and an {ACRONYM} 
and a {camelCase} Word and Some Units, 400~{nm}, 3~{cm}, and a Quote, 
\textit{{Alea} {iacta est}}}
}

---
references:
- title: A title, in English, with a Proper Name and an ACRONYM and a <span 
class="nocase"/>camelCase</span> word and some units, 400 <span 
class="nocase"/>nm</span>, 3 <span class="nocase"/>cm</span>, and a quote, 
<i>Alea <span class="nocase"/>iacta est</span></i>
  id: item1
  issued:
    year: 2013
  author:
    given:
    - Ann
    family: Author
  container-title: Journal
  type: article-journal
  language: en-US
...


$ cat article.biblatex 
@comment{
    - contains:
        - an article entry with just the required fields
        - an article entry with required and all optional fields
    - notes:
        - year, month to be ignored if date is present
        - journal to be ignored if journaltitle is present
        - editortype, editoratype, editorbtype, editorctype, pubstate, 
series contain keys which, unless corresponding CSL terms exist, require 
locale-specific expansion
    - limitations:
        - annotator, commentator, eid, eprint, eprintclass, eprinttype, 
issuetitle, issuesubtitle, language, origlanguage have no matching 
counterparts in CSL
        - for editor, editora, editorb, editorc (plus editortype, 
editoratype, editorbtype, editorctype) only a subset, editor and director, 
has matching counterparts in CSL
    - kludges:
        - note + addendum -> CSL note
        - number + issue -> CSL issue
        - handling of titleaddon
        - handling of (journal) series
            - done properly, this should be mapped to some CSL variable 
(version? edition? collection-number?), CSL styles would have to be adapted
            - slightly better kludge would map integer to ordinal + "ser." 
("3" -> "3rd ser."); localization keys "newseries" -> "new ser.", 
"oldseries" -> "old ser."; and print all other values as is -- but still 
wouldn't fit all styles or locales.
    }

@article{article-req,
    Author = {Author, Ann},
    Date = {2013-07-29},
    Hyphenation = {english},
    Journaltitle = {The Journaltitle},
    Title = {An Article Entry with Just the Required Fields}}

@article{article-opt,
    Addendum = {The Addendum},
    Annotator = {Annotator, A.},
    Author = {Author, Jr., Ann A.},
    Commentator = {Commentator, C.},
    Date = {2008-12-31},
    Doi = {10.1086/520976},
    Editor = {Editor, Edward},
    Editora = {Editor, A.},
    Editorb = {Editor, B.},
    Editorc = {Editor, C.},
    Eid = {eid},
    Eprint = {eprint},
    Eprintclass = {eprintclass},
    Eprinttype = {eprinttype},
    Hyphenation = {english},
    Issn = {issn},
    Issue = {issue},
    Issuesubtitle = {The Issuesubtitle},
    Issuetitle = {The Issuetitle},
    Journalsubtitle = {The Journalsubtitle},
    Journaltitle = {The Journaltitle},
    Journal = {The Journal},
    Language = {language},
    Month = {08},
    Year = {2007},
    Note = {The Note},
    Number = {number},
    Origlanguage = {origlanguage},
    Pages = {pages},
    Pubstate = {inpress},
    Series = {newseries},
    Subtitle = {The Subtitle},
    Title = {An Article Entry with the Required and All Optional Fields},
    Titleaddon = {The Titleaddon},
    Translator = {Translator, Ted},
    Url = {http://foo.bar.baz/},
    Urldate = {2013-07-29},
    Version = {version},
    Volume = {volume},
}

---
references:
- title: An article entry with just the required fields
  id: article-req
  issued:
    day: 29
    month: 07
    year: 2013
  author:
    given:
    - Ann
    family: Author
  container-title: The Journaltitle
  type: article-journal
  language: en-US
- issn: issn
  status: in press
  issue: number, issue
  page: pages
  note: The Note. The Addendum
  title: ! 'An article entry with the required and all optional fields: The 
subtitle.
    The titleaddon'
  volume: volume
  id: article-opt
  issued:
    day: 31
    month: 12
    year: 2008
  author:
    suffix: Jr.
    given:
    - Ann
    - A.
    comma-suffix: false
    family: Author
  accessed:
    day: 29
    month: 07
    year: 2013
  version: version
  container-title: ! 'The Journaltitle: The Journalsubtitle, new series'
  editor:
    given:
    - Edward
    family: Editor
  translator:
    given:
    - Ted
    family: Translator
  url: http://foo.bar.baz/
  type: article-journal
  doi: 10.1086/520976
  language: en-US
...

On Tuesday, 15 October 2013 20:09:28 UTC, fiddlosopher wrote:
>
> I've changed the framework a bit.  Now only one file per test is 
> needed.  It should have extension '.biblatex' or '.bibtex', depending 
> on the format.  It should begin with a bibtex/biblatex section, 
> then have a YAML section after a line with three hyphens. 
> (See the example in tests/biblio2yaml.) 
>
> The nice thing is that these files count as valid bibtex/biblatex 
> files, since the YAML part is treated as a comment.  So you can run 
> biblio2yaml directly on them. 
>
> +++ John MacFarlane [Oct 15 13 11:16 ]: 
> > I'll look into it.  Meanwhile, it occurs to me that I've relied too much 
> > on the "ask Nick Bart" method of testing.  We should automate the 
> > testing as much as possible.  I've added a framework for a biblio2yaml 
> > test suite.  I'm hoping you can contribute some of the tests you're 
> > running. 
> > 
> > It's very simple.  Each test is just two files in the directory 
> > tests/biblio2yaml.  The first file should be a biblatex file with a 
> > descriptive name, like 
> > 
> > nested-xref-resolution.bib 
> > 
> > The second file should be the yaml you think should result, 
> > with the same name as the biblatex file with a .yaml extension 
> > appended: 
> > 
> > nested-xref-resolution.bib.yaml 
> > 
> > Doing 'cabal install --enable-tests' will run the tests automatically 
> > for any files in this directory.  YAML files will be compared at the 
> > semantic level, so things like ordering of fields won't matter. 
> > 
> > It seems that you already have a lot of tests.  If you could break 
> > them down into small, focused pieces, and send me the paired files, 
> > I'll add them to the repository.  This should help to avoid regressions 
> > in the future, and give us some confidence that we're getting things 
> > right. 
> > 
> > 
> > +++ Nick Bart [Oct 15 13 07:45 ]: 
> > >    Also,biblio2yaml does not produce *any* output from @periodical any 
> > >    more: 
> > >    $ cat test-special-issue.bib 
> > >    @periodical{item1, 
> > >        Date = {2013}, 
> > >        Editor = {Editor, Edward}, 
> > >        Entrysubtype = {magazine}, 
> > >        Hyphenation = {english}, 
> > >        Issuetitle = {Title of the Special Issue}, 
> > >        Location = {Location}, 
> > >        Note = {Special Issue}, 
> > >        Number = {3}, 
> > >        Publisher = {Publisher}, 
> > >        Series = {newseries}, 
> > >        Title = {Title of the Journal}, 
> > >        Volume = {7}, 
> > >        Year = {2013}} 
> > >    $ biblio2yaml test-special-issue.bib 
> > >    --- 
> > >    references: 
> > >    [] 
> > >    ... 
> > > 
> > >    -- 
> > >    You received this message because you are subscribed to the Google 
> > >    Groups "pandoc-discuss" group. 
> > >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> > >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > >    To view this discussion on the web visit 
> > >    [1]
> https://groups.google.com/d/msgid/pandoc-discuss/4bb3bb13-b90b-4204- 
> > >    90a6-78db9705262a%40googlegroups.com. 
> > >    For more options, visit [2]https://groups.google.com/groups/opt_out. 
>
> > > 
> > > References 
> > > 
> > >    1. 
> https://groups.google.com/d/msgid/pandoc-discuss/4bb3bb13-b90b-4204-90a6-78db9705262a%40googlegroups.com 
> > >    2. https://groups.google.com/groups/opt_out 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "pandoc-discuss" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/20131015181630.GA28168%40protagoras.phil.berkeley.edu. 
>
> > For more options, visit https://groups.google.com/groups/opt_out. 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/3a15c700-5397-4a0c-b97d-860fec8b4146%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 17007 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2297 bytes --]

Glad you started implementing <span class="nocase"></span>. Thank you. 
However, I started finding this in authors' and publishers' names and 
journal titles, too (cluttering the yaml file, and turning strings into a 
shade of green in pdf output); but I think this is only ever required for 
title variables where citeproc in combination with a title-casing CLS style 
actually carries out title-case conversion. Thus, I think it should be 
strictly limited to titles in English (with the exception of 
journal/journaltitle/journalsubtitle, and title/subtitle/titleaddon of 
@periodical, where it is unnecessary, too.)

$ cat test-case-conversion-2.bib 
@article{item2,
    Author = {{al-Haj}, Ahmed},
    Date = {2012-03-18},
    Hyphenation = {english},
    Journaltitle = {XYZ {eLibrary}},
    Title = {Title with {Upper} {lower} {caMel}}}

@book{item1,
    Author = {Author, Ann},
    Date = {2013},
    Hyphenation = {english},
    Publisher = {Holt, Rinehart {and} Winston},
    Title = {7 {nm} (nanometre), or 7 {Nm} ({Newton} metre)}}
mytests $ biblio2yaml test-case-conversion-2.bib 
---
references:
- title: Title with Upper <span class="nocase">lower</span> <span 
class="nocase">caMel</span>
  id: item2
  issued:
    day: 18
    month: 03
    year: 2012
  author:
    given:
    - Ahmed
    family: <span class="nocase">al-Haj</span>
  container-title: XYZ <span class="nocase">eLibrary</span>
  type: article-journal
  language: en-US
- title: 7 <span class="nocase">nm</span> (nanometre), or 7 Nm (Newton 
metre)
  id: item1
  issued:
    year: 2013
  author:
    given:
    - Ann
    family: Author
  publisher: Holt, Rinehart <span class="nocase">and</span> Winston
  type: book
  language: en-US
...

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/1e21d06f-63e8-4ac8-83e9-ad03ee7b812b%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3128 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Okay, I've made a change that should have this effect.

+++ Nick Bart [Oct 16 13 23:31 ]:
> Glad you started implementing <span class="nocase"></span>. Thank you. 
> However, I started finding this in authors' and publishers' names and 
> journal titles, too (cluttering the yaml file, and turning strings into a 
> shade of green in pdf output); but I think this is only ever required for 
> title variables where citeproc in combination with a title-casing CLS style 
> actually carries out title-case conversion. Thus, I think it should be 
> strictly limited to titles in English (with the exception of 
> journal/journaltitle/journalsubtitle, and title/subtitle/titleaddon of 
> @periodical, where it is unnecessary, too.)
> 
> $ cat test-case-conversion-2.bib 
> @article{item2,
>     Author = {{al-Haj}, Ahmed},
>     Date = {2012-03-18},
>     Hyphenation = {english},
>     Journaltitle = {XYZ {eLibrary}},
>     Title = {Title with {Upper} {lower} {caMel}}}
> 
> @book{item1,
>     Author = {Author, Ann},
>     Date = {2013},
>     Hyphenation = {english},
>     Publisher = {Holt, Rinehart {and} Winston},
>     Title = {7 {nm} (nanometre), or 7 {Nm} ({Newton} metre)}}
> mytests $ biblio2yaml test-case-conversion-2.bib 
> ---
> references:
> - title: Title with Upper <span class="nocase">lower</span> <span 
> class="nocase">caMel</span>
>   id: item2
>   issued:
>     day: 18
>     month: 03
>     year: 2012
>   author:
>     given:
>     - Ahmed
>     family: <span class="nocase">al-Haj</span>
>   container-title: XYZ <span class="nocase">eLibrary</span>
>   type: article-journal
>   language: en-US
> - title: 7 <span class="nocase">nm</span> (nanometre), or 7 Nm (Newton 
> metre)
>   id: item1
>   issued:
>     year: 2013
>   author:
>     given:
>     - Ann
>     family: Author
>   publisher: Holt, Rinehart <span class="nocase">and</span> Winston
>   type: book
>   language: en-US
> ...
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/1e21d06f-63e8-4ac8-83e9-ad03ee7b812b%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1050 bytes --]

Hmmm, right now (almost) all title words are lowercased:

$ biblio2yaml 
../pandoc-citeproc/tests/biblio2yaml/test-case-conversion.biblatex 
---
references:
- title: A title, in english, with a proper name and an acronym and a 
camelcase word
    and some units, 400 nm, 3 cm, and a quote, <i>Alea iacta est</i>
  id: item1
  issued:
    year: 2013
  author:
    given:
    - Ann
    family: Author
  container-title: Journal
  type: article-journal
  language: en-US
...

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/01963ff0-a123-4dd3-bcd6-32a150e6828e%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1484 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 5880 bytes --]

And here's a test for dates:

$ cat dates.biblatex 

@comment{
    - Dates
    - Not included in tests:
        - malformed dates and date ranges
        - literal dates ("13th century"; not supported by biblatex)
        - seasons (would have to come from parsing the issue field)
        - uncertain dates (use "doubtfuldate" from biblatex-apa?)
        - dates with single-digit day or month (not supported by biblatex)
        - negative dates (not supported by biblatex)
    - Note: biblatex supports years < 1000 but only if padded with leading 
zeros
    - TODO:
        - either biblio2yaml or, probably better, citeproc should strip 
leading zeros
         from days and months (CSL can add leading zeros to its output, but 
not remove
         them when they are in its input data)
}

@article{year-month-old,
    Author = {Author, Al},
    Journal = {Journal},
    Month = aug,
    Title = {Year and Month, bibtex style, supported by biblatex for 
backwards compatibility},
    Year = {1999}}

@article{year-month-new,
    Author = {Author, Al},
    Journal = {Journal},
    Month = {08},
    Title = {Year and Month, biblatex style; note that biblatex does not 
have a "day" field},
    Year = {1999}}

@article{dates,
    Author = {Author, Al},
    Date = {2012-12-13},
    Eventdate = {2011-10-03},
    Journal = {Journal},
    Month = may,
    Origdate = {1888-10-01},
    Title = {Dates, default biblatex style; year, month to be ignored if 
date exists},
    Urldate = {1999-05-23},
    Year = {9999}}

@article{date-ranges-different-years,
    Author = {Author, Al},
    Date = {1999-10-14/2010-01-23},
    Eventdate = {1999-12/2000-01},
    Journal = {Journal},
    Origdate = {1888-01-02/1913-12-13},
    Title = {Date ranges; different years},
    Urldate = {2012-10-12/2013-01-31}}

@article{date-ranges-same-year,
    Author = {Author, Al},
    Date = {1999-10-14/1999-10-15},
    Eventdate = {1999-10/1999-11},
    Journal = {Journal},
    Origdate = {1888-01-02/1888-12-13},
    Title = {Date ranges; same year},
    Urldate = {2012-10-31/2012-11-01}}

@article{date-ranges-open,
    Author = {Author, Al},
    Date = {1999-10-14/},
    Eventdate = {1999-10/},
    Journal = {Journal},
    Origdate = {1888-01-02/},
    Title = {Date ranges, open-ended},
    Urldate = {2012-10-31/}}

@article{dates-very-old,
    Author = {Author, Al},
    Date = {0712-12-13},
    Eventdate = {0311-10-03},
    Journal = {Journal},
    Month = may,
    Origdate = {0088-10-01},
    Title = {Dates, year less than 1000},
    Urldate = {0999-12-14}}

---
references:
- title: Year and month, bibtex style, supported by biblatex for backwards 
compatibility
  id: year-month-old
  issued:
    month: 08
    year: 1999
  author:
    given:
    - Al
    family: Author
  container-title: Journal
  type: article-journal
- title: Year and month, biblatex style; note that biblatex does not have a 
"day"
    field
  id: year-month-new
  issued:
    month: 08
    year: 1999
  author:
    given:
    - Al
    family: Author
  container-title: Journal
  type: article-journal
- original-date:
    day: 01
    month: 10
    year: 1888
  title: Dates, default biblatex style; year, month to be ignored if date 
exists
  event-date:
    day: 03
    month: 10
    year: 2011
  id: dates
  issued:
    day: 13
    month: 12
    year: 2012
  author:
    given:
    - Al
    family: Author
  accessed:
    day: 23
    month: 05
    year: 1999
  container-title: Journal
  type: article-journal
- original-date:
  - day: 02
    month: 01
    year: 1888
  - day: 13
    month: 12
    year: 1913
  title: Date ranges; different years
  event-date:
  - month: 12
    year: 1999
  - month: 01
    year: 2000
  id: date-ranges-different-years
  issued:
  - day: 14
    month: 10
    year: 1999
  - day: 23
    month: 01
    year: 2010
  author:
    given:
    - Al
    family: Author
  accessed:
  - day: 12
    month: 10
    year: 2012
  - day: 31
    month: 01
    year: 2013
  container-title: Journal
  type: article-journal
- original-date:
  - day: 02
    month: 01
    year: 1888
  - day: 13
    month: 12
    year: 1888
  title: Date ranges; same year
  event-date:
  - month: 10
    year: 1999
  - month: 11
    year: 1999
  id: date-ranges-same-year
  issued:
  - day: 14
    month: 10
    year: 1999
  - day: 15
    month: 10
    year: 1999
  author:
    given:
    - Al
    family: Author
  accessed:
  - day: 31
    month: 10
    year: 2012
  - day: 01
    month: 11
    year: 2012
  container-title: Journal
  type: article-journal
- original-date:
  - day: 02
    month: 01
    year: 1888
  - {}
  title: Date ranges, open-ended
  event-date:
  - month: 10
    year: 1999
  - {}
  id: date-ranges-open
  issued:
  - day: 14
    month: 10
    year: 1999
  - {}
  author:
    given:
    - Al
    family: Author
  accessed:
  - day: 31
    month: 10
    year: 2012
  - {}
  container-title: Journal
  type: article-journal
- original-date:
    day: 01
    month: 10
    year: 0088
  title: Dates, year less than 1000
  event-date:
    day: 03
    month: 10
    year: 0311
  id: dates-very-old
  issued:
    day: 13
    month: 12
    year: 0712
  author:
    given:
    - Al
    family: Author
  accessed:
    day: 14
    month: 12
    year: 0999
  container-title: Journal
  type: article-journal
...

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/38197416-012d-4e4e-b74b-f1714901fe22%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 9597 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Here's what I see:

% dist/build/biblio2yaml/biblio2yaml tests/biblio2yaml/test-case-conversion.biblatex
---
references:
- title: A title, in English, with a Proper Name and an ACRONYM and a <span class="nocase">camelCase</span>
    word and some units, 400 <span class="nocase">nm</span>, 3 <span class="nocase">cm</span>,
    and a quote, <i>Alea iacta est</i>
  id: item1
  issued:
    year: 2013
  author:
    given:
    - Ann
    family: Author
  container-title: Journal
  type: article-journal
  language: en-US
...

I'm not sure what is happening with your setup.  Try running the
built version, dist/build/biblio2yaml/biblio2yaml.  Perhaps it didn't
install (this would happen if you did cabal install --enable-tests,
since some of the tests fail and it won't install when they do).

If that doesn't work, you might try getting the latest pandoc-types
and pandoc. Then,

    cd pandoc-types
    cabal install --force
    cd ../pandoc
    cabal install --force
    cd ../pandoc-citeproc
    cabal install


+++ Nick Bart [Oct 17 13 23:59 ]:
> Hmmm, right now (almost) all title words are lowercased:
> 
> $ biblio2yaml 
> ../pandoc-citeproc/tests/biblio2yaml/test-case-conversion.biblatex 
> ---
> references:
> - title: A title, in english, with a proper name and an acronym and a 
> camelcase word
>     and some units, 400 nm, 3 cm, and a quote, <i>Alea iacta est</i>
>   id: item1
>   issued:
>     year: 2013
>   author:
>     given:
>     - Ann
>     family: Author
>   container-title: Journal
>   type: article-journal
>   language: en-US
> ...
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/01963ff0-a123-4dd3-bcd6-32a150e6828e%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1.1: Type: text/plain, Size: 3190 bytes --]

Thank you, you're probably right: After a cabal install --force in all of 
the subdirs everything's back to normal.

The attached tests.zip contains a few more tests, most of them include 
comments.

On Friday, October 18, 2013 3:10:10 PM UTC, fiddlosopher wrote:
>
> Here's what I see: 
>
> % dist/build/biblio2yaml/biblio2yaml 
> tests/biblio2yaml/test-case-conversion.biblatex 
> --- 
> references: 
> - title: A title, in English, with a Proper Name and an ACRONYM and a 
> <span class="nocase">camelCase</span> 
>     word and some units, 400 <span class="nocase">nm</span>, 3 <span 
> class="nocase">cm</span>, 
>     and a quote, <i>Alea iacta est</i> 
>   id: item1 
>   issued: 
>     year: 2013 
>   author: 
>     given: 
>     - Ann 
>     family: Author 
>   container-title: Journal 
>   type: article-journal 
>   language: en-US 
> ... 
>
> I'm not sure what is happening with your setup.  Try running the 
> built version, dist/build/biblio2yaml/biblio2yaml.  Perhaps it didn't 
> install (this would happen if you did cabal install --enable-tests, 
> since some of the tests fail and it won't install when they do). 
>
> If that doesn't work, you might try getting the latest pandoc-types 
> and pandoc. Then, 
>
>     cd pandoc-types 
>     cabal install --force 
>     cd ../pandoc 
>     cabal install --force 
>     cd ../pandoc-citeproc 
>     cabal install 
>
>
> +++ Nick Bart [Oct 17 13 23:59 ]: 
> > Hmmm, right now (almost) all title words are lowercased: 
> > 
> > $ biblio2yaml 
> > ../pandoc-citeproc/tests/biblio2yaml/test-case-conversion.biblatex 
> > --- 
> > references: 
> > - title: A title, in english, with a proper name and an acronym and a 
> > camelcase word 
> >     and some units, 400 nm, 3 cm, and a quote, <i>Alea iacta est</i> 
> >   id: item1 
> >   issued: 
> >     year: 2013 
> >   author: 
> >     given: 
> >     - Ann 
> >     family: Author 
> >   container-title: Journal 
> >   type: article-journal 
> >   language: en-US 
> > ... 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "pandoc-discuss" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/01963ff0-a123-4dd3-bcd6-32a150e6828e%40googlegroups.com. 
>
> > For more options, visit https://groups.google.com/groups/opt_out. 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/afd215e6-4482-4ce4-b6a4-2184a4c160f7%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #1.2: Type: text/html, Size: 4441 bytes --]

[-- Attachment #2: tests.zip --]
[-- Type: application/zip, Size: 20676 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1.1: Type: text/plain, Size: 2949 bytes --]

A bit more on the date disambiguation issue:

Formatting this file:

$ cat test-no-author.md 
@item1, @item2, @item3, @item4, @item5, @item6 

---
references:
- title: Title A
  id: item1
  issued:
    year: 2012
  container-title: Magazine
  type: article-magazine
- title: Title b
  id: item2
  issued:
    year: 2012
  container-title: Magazine
  type: article-magazine
- title: Title c
  id: item3
  issued:
    year: 2012
  container-title: Magazine
  type: article-magazine
- title: Title d
  id: item4
  issued:
    year: 2012
  container-title: Magazine
  type: article-magazine
- title: Title e
  id: item5
  issued:
    year: 2012
  container-title: Newspaper
  type: article-magazine
- title: Title f
  id: item6
  issued:
    year: 2012
  container-title: Newspaper
  type: article-magazine
...


with chicago-author-date.csl gives

$ pandoc --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks test-no-author.md
(“Title A” 2012a), (“Title B” 2012b), (“Title C” 2012c), (“Title D”
2012d), (“Title E” 2012e), (“Title F” 2012f),

“Title A.” 2012a. *Magazine*.

“Title B.” 2012b. *Magazine*.

“Title C.” 2012c. *Magazine*.

“Title D.” 2012d. *Magazine*.

“Title E.” 2012e. *Newspaper*.

“Title F.” 2012f. *Newspaper*.

Here, *no* date disambiguation would be correct since the titles are all 
different.


Using the attached chicago-author-date-TEST-20131018.csl, which adds 
support for origdate, status, and for substituting the magazine/newspaper 
name for author (to correctly format “unsigned newspaper articles” where 
“the name of the newspaper stands in place of the author”; see CMoS, 16e, 
14.207), the output is:

$ pandoc --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks test-no-author.md --csl 
chicago-author-date-TEST-20131018.csl
 (*Magazine* 2012a), (*Magazine* 2012b), (*Magazine* 2012c), (*Magazine*
2012d), (*Newspaper* 2012e), (*Newspaper* 2012f)

*Magazine*. 2012a. “Title A.”

*Magazine*. 2012b. “Title B.”

*Magazine*. 2012c. “Title C.”

*Magazine*. 2012d. “Title D.”

*Newspaper*. 2012e. “Title E.”

*Newspaper*. 2012f. “Title F.”


Here, *Magazine* 2012a to *Magazine* 2012d is correct, but *Newspaper* 
2012e, *Newspaper* 2012f should of course be *Newspaper* 2012a, *Newspaper* 
2012b.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/259dcbfd-1596-4955-b2e6-f9b1532ed8a3%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #1.2: Type: text/html, Size: 3576 bytes --]

[-- Attachment #2: chicago-author-date-TEST-20131018.csl --]
[-- Type: application/octet-stream, Size: 17168 bytes --]

<?xml version="1.0" encoding="utf-8"?>
<style xmlns="http://purl.org/net/xbiblio/csl" class="in-text" version="1.0" demote-non-dropping-particle="sort-only">
  <info>
    <title>Chicago Manual of Style 16th edition (author-date) TEST-20131018</title>
    <id>chicago-author-date-TEST-20131018</id>
    <link href="http://www.chicagomanualofstyle.org/tools_citationguide.html" rel="documentation"/>
    <author>
      <name>Julian Onions</name>
      <email>julian.onions-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org</email>
    </author>
    <contributor>
      <name>Sebastian Karcher</name>
    </contributor>
    <contributor>
      <name>Richard Karnesky</name>
      <email>karnesky+zotero-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org</email>
      <uri>http://arc.nucapt.northwestern.edu/Richard_Karnesky</uri>
    </contributor>
    <category citation-format="author-date"/>
    <category field="generic-base"/>
    <summary>The author-date variant of the Chicago style</summary>
    <updated>2013-04-24T02:44:57+00:00</updated>
    <rights license="http://creativecommons.org/licenses/by-sa/3.0/">This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 License</rights>
  </info>
  <locale>
    <terms>
      <term name="editor" form="verb-short">ed.</term>
      <term name="container-author" form="verb">by</term>
      <term name="translator" form="verb-short">trans.</term>
      <term name="editortranslator" form="verb">
        <single>edited and translated by</single>
        <multiple>edited and translated by</multiple>
      </term>
      <term name="translator" form="short">trans.</term>
    </terms>
  </locale>

  <macro name="secondary-contributors">
    <choose>
      <if type="chapter paper-conference" match="none">
        <group>
          <names variable="editor translator" delimiter=". ">
            <label form="verb" text-case="capitalize-first" suffix=" " plural="never"/>
            <name and="text" delimiter=", "/>
          </names>
        </group>
      </if>
    </choose>
  </macro>

  <macro name="container-contributors">
    <choose>
      <if type="chapter paper-conference" match="any">
        <group prefix=", " delimiter=", ">
          <names variable="container-author editor" delimiter=", ">
            <label form="verb" suffix=" " plural="never"/>
            <name and="text" delimiter=", "/>
          </names>
        </group>
      </if>
    </choose>
  </macro>
  <macro name="editor">
    <names variable="editor">
      <name name-as-sort-order="first" and="text" sort-separator=", " delimiter=", " delimiter-precedes-last="always"/>
      <label form="short" prefix=", "/>
    </names>
  </macro>
  <macro name="translator">
    <names variable="translator">
      <name name-as-sort-order="first" and="text" sort-separator=", " delimiter=", " delimiter-precedes-last="always"/>
      <label form="short" prefix=", " plural="never"/>
    </names>
  </macro>
  <macro name="recipient">
    <choose>
      <if type="personal_communication">
        <choose>
          <if variable="genre">
            <text variable="genre" text-case="capitalize-first"/>
          </if>
          <else>
            <text term="letter" text-case="capitalize-first"/>
          </else>
        </choose>
      </if>
    </choose>
    <names variable="recipient" delimiter=", ">
      <label form="verb" prefix=" " text-case="lowercase" suffix=" "/>
      <name and="text" delimiter=", "/>
    </names>
  </macro>
  <macro name="contributors">
    <group delimiter=". ">
      <names variable="author">
        <name and="text" name-as-sort-order="first" sort-separator=", " delimiter=", " delimiter-precedes-last="always"/>
        <label form="short" plural="never" prefix=", "/>
        <substitute>
          <names variable="editor"/>
          <names variable="translator"/>
          <choose>
            <if type="article-magazine article-newspaper" match="any">
              <text macro="container-title"/>
            </if>
          </choose>
          <text macro="title"/>
        </substitute>
      </names>
      <text macro="recipient"/>
    </group>
  </macro>

  <macro name="contributors-short">
    <names variable="author">
      <name form="short" and="text" delimiter=", " initialize-with=". "/>
      <substitute>
        <names variable="editor"/>
        <names variable="translator"/>
          <!-- NEW -->
          <choose>
            <if type="article-magazine article-newspaper" match="any">
              <text macro="container-title"/>
            </if>
          </choose>
        <text macro="title"/>
      </substitute>
    </names>
  </macro>

  <macro name="interviewer">
    <names variable="interviewer" delimiter=", ">
      <label form="verb" prefix=" " text-case="capitalize-first" suffix=" "/>
      <name and="text" delimiter=", "/>
    </names>
  </macro>
  <macro name="archive">
    <group delimiter=". ">
      <text variable="archive_location" text-case="capitalize-first"/>
      <text variable="archive"/>
      <text variable="archive-place"/>
    </group>
  </macro>
  <macro name="access">
    <group delimiter=". ">
      <choose>
        <if type="graphic report" match="any">
          <text macro="archive"/>
        </if>
        <else-if type="article-magazine article-newspaper bill book chapter graphic legal_case legislation motion_picture paper-conference report song thesis" match="none">
          <text macro="archive"/>
        </else-if>
      </choose>
      <text variable="DOI" prefix="doi:"/>
      <choose>
        <if variable="DOI issued" match="none">
          <choose>
            <if variable="URL accessed" match="all">
              <group delimiter=" ">
                <text term="accessed" text-case="capitalize-first"/>
                <date variable="accessed" delimiter=" ">
                  <date-part name="month"/>
                  <date-part name="day"/>
                </date>
              </group>
            </if>
          </choose>
        </if>
        <else-if type="webpage">
          <date variable="issued" delimiter=" ">
            <date-part name="month"/>
            <date-part name="day"/>
          </date>
        </else-if>
      </choose>
      <choose>
        <if type="legal_case" match="none">
          <text variable="URL"/>
        </if>
      </choose>
    </group>
  </macro>
  <macro name="title">
    <choose>
      <if variable="title" match="none">
        <choose>
          <if type="personal_communication" match="none">
            <text variable="genre" text-case="capitalize-first"/>
          </if>
        </choose>
      </if>
      <else-if type="bill book graphic legal_case legislation motion_picture song" match="any">
        <text variable="title" text-case="title" font-style="italic"/>
      </else-if>
      <else>
        <text variable="title" text-case="title" quotes="true"/>
      </else>
    </choose>
  </macro>
  <macro name="edition">
    <choose>
      <if type="bill book graphic legal_case legislation motion_picture report song" match="any">
        <choose>
          <if is-numeric="edition">
            <group delimiter=" " prefix=". ">
              <number variable="edition" form="ordinal"/>
              <text term="edition" form="short" strip-periods="true"/>
            </group>
          </if>
          <else>
            <text variable="edition" prefix=". "/>
          </else>
        </choose>
      </if>
      <else-if type="chapter  paper-conference" match="any">
        <choose>
          <if is-numeric="edition">
            <group delimiter=" " prefix=", ">
              <number variable="edition" form="ordinal"/>
              <text term="edition" form="short"/>
            </group>
          </if>
          <else>
            <text variable="edition" prefix=", "/>
          </else>
        </choose>
      </else-if>
    </choose>
  </macro>
  <macro name="locators">
    <choose>
      <if type="article-journal">
        <text variable="volume" prefix=" "/>
        <text variable="issue" prefix=" (" suffix=")"/>
      </if>
      <else-if type="legal_case">
        <text variable="volume" prefix=", "/>
        <text variable="container-title" prefix=" "/>
        <text variable="page" prefix=" "/>
      </else-if>
      <else-if type="bill book graphic legal_case legislation motion_picture report song" match="any">
        <group prefix=". " delimiter=". ">
          <group>
            <text term="volume" form="short" text-case="capitalize-first" suffix=" "/>
            <number variable="volume" form="numeric"/>
          </group>
          <group>
            <number variable="number-of-volumes" form="numeric"/>
            <text term="volume" form="short" prefix=" " plural="true"/>
          </group>
        </group>
      </else-if>
      <else-if type="chapter paper-conference" match="any">
        <choose>
          <if variable="page" match="none">
            <group prefix=". ">
              <text term="volume" form="short" text-case="capitalize-first" suffix=" "/>
              <number variable="volume" form="numeric"/>
            </group>
          </if>
        </choose>
      </else-if>
    </choose>
  </macro>
  <macro name="locators-chapter">
    <choose>
      <if type="chapter paper-conference" match="any">
        <choose>
          <if variable="page">
            <group prefix=", ">
              <text variable="volume" suffix=":"/>
              <text variable="page"/>
            </group>
          </if>
        </choose>
      </if>
    </choose>
  </macro>
  <macro name="locators-article">
    <choose>
      <if type="article-newspaper">
        <group prefix=", " delimiter=", ">
          <group>
            <text variable="edition" suffix=" "/>
            <text term="edition" prefix=" "/>
          </group>
          <group>
            <text term="section" form="short" suffix=" "/>
            <text variable="section"/>
          </group>
        </group>
      </if>
      <else-if type="article-journal">
        <text variable="page" prefix=": "/>
      </else-if>
    </choose>
  </macro>
  <macro name="point-locators">
    <choose>
      <if variable="locator">
        <choose>
          <if locator="page" match="none">
            <choose>
              <if type="bill book graphic legal_case legislation motion_picture report song" match="any">
                <choose>
                  <if variable="volume">
                    <group>
                      <text term="volume" form="short" suffix=" "/>
                      <number variable="volume" form="numeric"/>
                      <label variable="locator" form="short" prefix=", " suffix=" "/>
                    </group>
                  </if>
                  <else>
                    <label variable="locator" form="short" suffix=" "/>
                  </else>
                </choose>
              </if>
              <else>
                <label variable="locator" form="short" suffix=" "/>
              </else>
            </choose>
          </if>
          <else-if type="bill book graphic legal_case legislation motion_picture report song" match="any">
            <number variable="volume" form="numeric" suffix=":"/>
          </else-if>
        </choose>
        <text variable="locator"/>
      </if>
    </choose>
  </macro>
  <macro name="container-prefix">
    <text term="in" text-case="capitalize-first"/>
  </macro>

  <macro name="container-title">
    <choose>
      <if type="chapter paper-conference" match="any">
        <text macro="container-prefix" suffix=" "/>
      </if>
    </choose>
    <choose>
      <if type="legal_case" match="none">
        <text variable="container-title" text-case="title" font-style="italic"/>
      </if>
    </choose>
  </macro>

  <macro name="publisher">
    <group delimiter=": ">
      <text variable="publisher-place"/>
      <text variable="publisher"/>
    </group>
  </macro>
  <macro name="date-citation">
      <choose>
        <if variable="original-date">
          <date variable="original-date" prefix=" ["  suffix="]">
          <date-part name="year"/>
        </date>
        </if>
      </choose>
    <choose>
      <if variable="issued">
        <date variable="issued" prefix=" ">
          <date-part name="year"/>
        </date>
      </if>
      <else-if variable="status">
        <text variable="status" prefix=", "/>
      </else-if>
      <else-if variable="accessed">
        <date variable="accessed" prefix=" ">
          <date-part name="year"/>
        </date>
      </else-if>
      </choose>
  </macro>
  <macro name="date-bibliography">
      <choose>
        <if variable="original-date">
          <date variable="original-date" prefix=" ("  suffix=") ">
          <date-part name="year"/>
        </date>
        </if>
      </choose>
    <choose>
      <if variable="issued">
        <date variable="issued">
          <date-part name="year"/>
        </date>
      </if>
      <else-if variable="status">
        <text variable="status" text-case="capitalize-first"/>
      </else-if>
      <else-if variable="accessed">
        <date variable="accessed">
          <date-part name="year"/>
        </date>
      </else-if>
      </choose>
  </macro>
  <macro name="day-month">
    <date variable="issued">
      <date-part name="month"/>
      <date-part name="day" prefix=" " form="numeric" />
    </date>
  </macro>
  <macro name="collection-title">
    <text variable="collection-title" text-case="title"/>
    <text variable="collection-number" prefix=" "/>
  </macro>
  <macro name="event">
    <group>
      <text term="presented at" suffix=" "/>
      <text variable="event"/>
    </group>
  </macro>
  <macro name="description">
    <choose>
      <if type="interview">
        <group delimiter=". ">
          <text macro="interviewer"/>
          <text variable="medium" text-case="capitalize-first"/>
        </group>
      </if>
      <else>
        <text variable="medium" text-case="capitalize-first" prefix=". "/>
      </else>
    </choose>
    <choose>
      <if variable="title" match="none"/>
      <else-if type="thesis personal_communication" match="any"/>
      <else>
        <group delimiter=" " prefix=". ">
          <text variable="genre" text-case="capitalize-first"/>
          <choose>
            <if type="report patent" match="any">
              <text variable="number"/>
            </if>
          </choose>
        </group>
      </else>
    </choose>
    <!--This is for computer programs only. Localization new to 1.0.1, so may be missing in many locales-->
    <group delimiter=" " prefix=" (" suffix=")">
      <text term="version"/>
      <text variable="version"/>
    </group>
  </macro>
  <macro name="issue">
    <choose>
      <if type="article-journal">
        <text macro="day-month" prefix=" (" suffix=")"/>
      </if>
      <else-if type="legal_case">
        <text variable="authority" prefix=". "/>
      </else-if>
      <else-if type="patent">
        <text variable="jurisdiction" prefix=". "/>
      </else-if>
      <else-if type="speech">
        <group prefix=" " delimiter=", ">
          <text macro="event"/>
          <text macro="day-month"/>
          <text variable="event-place"/>
        </group>
      </else-if>
      <else-if type="article-newspaper article-magazine personal_communication" match="any">
        <text macro="day-month" prefix=", "/>
      </else-if>
      <else>
        <group prefix=". " delimiter=", ">
          <choose>
            <if type="thesis">
              <text variable="genre" text-case="capitalize-first"/>
            </if>
          </choose>
          <text macro="publisher"/>
        </group>
      </else>
    </choose>
  </macro>
  <citation et-al-min="4" et-al-use-first="1" disambiguate-add-year-suffix="true" disambiguate-add-names="true" disambiguate-add-givenname="true" givenname-disambiguation-rule="primary-name">
    <layout prefix="(" suffix=")" delimiter="; ">
      <group delimiter=", ">
        <group delimiter="">
          <text macro="contributors-short"/>
          <text macro="date-citation"/>
        </group>
        <text macro="point-locators"/>
      </group>
    </layout>
  </citation>
  <bibliography hanging-indent="true" et-al-min="11" et-al-use-first="7" subsequent-author-substitute="&#8212;&#8212;&#8212;" entry-spacing="0">
    <sort>
      <key macro="contributors"/>
      <key variable="issued"/>
    </sort>
    <layout suffix=".">
      <group delimiter=". ">
        <text macro="contributors"/>
        <text macro="date-bibliography"  text-case="capitalize-first"/>
        <text macro="title"/>
      </group>
      <text macro="description"/>
      <text macro="secondary-contributors" prefix=". "/>
      <text macro="container-title" prefix=". "/>
      <text macro="container-contributors"/>
      <text macro="edition"/>
      <text macro="locators-chapter"/>
      <text macro="locators"/>
      <text macro="collection-title" prefix=". "/>
      <text macro="issue"/>
      <text macro="locators-article"/>
      <text macro="access" prefix=". "/>
    </layout>
  </bibliography>
</style>

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 3207 bytes --]


I believe this is fixed by the latest commit.  However,
more testing in this area would be useful.

+++ Nick Bart [Oct 18 13 09:19 ]:
> A bit more on the date disambiguation issue:
> 
> Formatting this file:
> 
> $ cat test-no-author.md 
> @item1, @item2, @item3, @item4, @item5, @item6 
> 
> ---
> references:
> - title: Title A
>   id: item1
>   issued:
>     year: 2012
>   container-title: Magazine
>   type: article-magazine
> - title: Title b
>   id: item2
>   issued:
>     year: 2012
>   container-title: Magazine
>   type: article-magazine
> - title: Title c
>   id: item3
>   issued:
>     year: 2012
>   container-title: Magazine
>   type: article-magazine
> - title: Title d
>   id: item4
>   issued:
>     year: 2012
>   container-title: Magazine
>   type: article-magazine
> - title: Title e
>   id: item5
>   issued:
>     year: 2012
>   container-title: Newspaper
>   type: article-magazine
> - title: Title f
>   id: item6
>   issued:
>     year: 2012
>   container-title: Newspaper
>   type: article-magazine
> ...
> 
> 
> with chicago-author-date.csl gives
> 
> $ pandoc --filter pandoc-citeproc -t 
> markdown-citations-markdown_in_html_blocks test-no-author.md
> (“Title A” 2012a), (“Title B” 2012b), (“Title C” 2012c), (“Title D”
> 2012d), (“Title E” 2012e), (“Title F” 2012f),
> 
> “Title A.” 2012a. *Magazine*.
> 
> “Title B.” 2012b. *Magazine*.
> 
> “Title C.” 2012c. *Magazine*.
> 
> “Title D.” 2012d. *Magazine*.
> 
> “Title E.” 2012e. *Newspaper*.
> 
> “Title F.” 2012f. *Newspaper*.
> 
> Here, *no* date disambiguation would be correct since the titles are all 
> different.
> 
> 
> Using the attached chicago-author-date-TEST-20131018.csl, which adds 
> support for origdate, status, and for substituting the magazine/newspaper 
> name for author (to correctly format “unsigned newspaper articles” where 
> “the name of the newspaper stands in place of the author”; see CMoS, 16e, 
> 14.207), the output is:
> 
> $ pandoc --filter pandoc-citeproc -t 
> markdown-citations-markdown_in_html_blocks test-no-author.md --csl 
> chicago-author-date-TEST-20131018.csl
>  (*Magazine* 2012a), (*Magazine* 2012b), (*Magazine* 2012c), (*Magazine*
> 2012d), (*Newspaper* 2012e), (*Newspaper* 2012f)
> 
> *Magazine*. 2012a. “Title A.”
> 
> *Magazine*. 2012b. “Title B.”
> 
> *Magazine*. 2012c. “Title C.”
> 
> *Magazine*. 2012d. “Title D.”
> 
> *Newspaper*. 2012e. “Title E.”
> 
> *Newspaper*. 2012f. “Title F.”
> 
> 
> Here, *Magazine* 2012a to *Magazine* 2012d is correct, but *Newspaper* 
> 2012e, *Newspaper* 2012f should of course be *Newspaper* 2012a, *Newspaper* 
> 2012b.
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe@googlegroups.com.
> To post to this group, send email to pandoc-discuss@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/259dcbfd-1596-4955-b2e6-f9b1532ed8a3%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I believe this is now fixed.

+++ Nick Bart [Oct 18 13 09:19 ]:
>    A bit more on the date disambiguation issue:
>    Formatting this file:
>    $ cat test-no-author.md
>    @item1, @item2, @item3, @item4, @item5, @item6
>    ---
>    references:
>    - title: Title A
>      id: item1
>      issued:
>        year: 2012
>      container-title: Magazine
>      type: article-magazine
>    - title: Title b
>      id: item2
>      issued:
>        year: 2012
>      container-title: Magazine
>      type: article-magazine
>    - title: Title c
>      id: item3
>      issued:
>        year: 2012
>      container-title: Magazine
>      type: article-magazine
>    - title: Title d
>      id: item4
>      issued:
>        year: 2012
>      container-title: Magazine
>      type: article-magazine
>    - title: Title e
>      id: item5
>      issued:
>        year: 2012
>      container-title: Newspaper
>      type: article-magazine
>    - title: Title f
>      id: item6
>      issued:
>        year: 2012
>      container-title: Newspaper
>      type: article-magazine
>    ...
>    with chicago-author-date.csl gives
>    $ pandoc --filter pandoc-citeproc -t
>    markdown-citations-markdown_in_html_blocks test-no-author.md
>    (Title A 2012a), (Title B 2012b), (Title C 2012c), (Title D
>    2012d), (Title E 2012e), (Title F 2012f),
>    Title A. 2012a. *Magazine*.
>    Title B. 2012b. *Magazine*.
>    Title C. 2012c. *Magazine*.
>    Title D. 2012d. *Magazine*.
>    Title E. 2012e. *Newspaper*.
>    Title F. 2012f. *Newspaper*.
>    Here, *no* date disambiguation would be correct since the titles are
>    all different.
>    Using the attached chicago-author-date-TEST-20131018.csl, which adds
>    support for origdate, status, and for substituting the
>    magazine/newspaper name for author (to correctly format unsigned
>    newspaper articles where the name of the newspaper stands in place of
>    the author; see CMoS, 16e, 14.207), the output is:
>    $ pandoc --filter pandoc-citeproc -t
>    markdown-citations-markdown_in_html_blocks test-no-author.md --csl
>    chicago-author-date-TEST-20131018.csl
>     (*Magazine* 2012a), (*Magazine* 2012b), (*Magazine* 2012c),
>    (*Magazine*
>    2012d), (*Newspaper* 2012e), (*Newspaper* 2012f)
>    *Magazine*. 2012a. Title A.
>    *Magazine*. 2012b. Title B.
>    *Magazine*. 2012c. Title C.
>    *Magazine*. 2012d. Title D.
>    *Newspaper*. 2012e. Title E.
>    *Newspaper*. 2012f. Title F.
>    Here, *Magazine* 2012a to *Magazine* 2012d is correct, but *Newspaper*
>    2012e, *Newspaper* 2012f should of course be *Newspaper* 2012a,
>    *Newspaper* 2012b.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/259dcbfd-1596-4955-
>    b2e6-f9b1532ed8a3%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/259dcbfd-1596-4955-b2e6-f9b1532ed8a3%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've added all the tests you sent in your zip file.
Very useful.  I think I corrected two mistakes in the test output.
Let me know if I'm off base.  I've fixed several of the
issues the tests revealed.

diff --git a/tests/biblio2yaml/patent.biblatex b/tests/biblio2yaml/patent.biblatex
index 42b1d1a..9144df3 100644
--- a/tests/biblio2yaml/patent.biblatex
+++ b/tests/biblio2yaml/patent.biblatex
@@ -156,7 +156,7 @@ references:
-  title: High-Speed digital-to-RF converter
+  title: High-speed digital-to-RF converter

commit 45392c988c8f533a20a0cd931655878b1f24b55e
Author: John MacFarlane <jgm-TVLZxgkOlNX2fBVCVOL8/A@public.gmane.org>
Date:   Fri Oct 18 17:22:44 2013 -0700

    biblio2yaml: Added missing 'span class="nocase"' in test case.

diff --git a/tests/biblio2yaml/incollection.biblatex b/tests/biblio2yaml/incollection.biblatex
index c9a8bc3..9bf23ac 100644
--- a/tests/biblio2yaml/incollection.biblatex
+++ b/tests/biblio2yaml/incollection.biblatex
@@ -110,8 +110,7 @@ references:
-  title: The limitations of human knowledge according to Al-Farabi, ibn Bajja, and
+  title: 'The limitations of human knowledge according to Al-Farabi, <span class="nocase">ibn Bajja</span>, and

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

All of these tests pass now...!

+++ John MacFarlane [Oct 18 13 17:41 ]:
> I've added all the tests you sent in your zip file.
> Very useful.  I think I corrected two mistakes in the test output.
> Let me know if I'm off base.  I've fixed several of the
> issues the tests revealed.
> 
> diff --git a/tests/biblio2yaml/patent.biblatex b/tests/biblio2yaml/patent.biblatex
> index 42b1d1a..9144df3 100644
> --- a/tests/biblio2yaml/patent.biblatex
> +++ b/tests/biblio2yaml/patent.biblatex
> @@ -156,7 +156,7 @@ references:
> -  title: High-Speed digital-to-RF converter
> +  title: High-speed digital-to-RF converter
> 
> commit 45392c988c8f533a20a0cd931655878b1f24b55e
> Author: John MacFarlane <jgm-TVLZxgkOlNX2fBVCVOL8/A@public.gmane.org>
> Date:   Fri Oct 18 17:22:44 2013 -0700
> 
>     biblio2yaml: Added missing 'span class="nocase"' in test case.
> 
> diff --git a/tests/biblio2yaml/incollection.biblatex b/tests/biblio2yaml/incollection.biblatex
> index c9a8bc3..9bf23ac 100644
> --- a/tests/biblio2yaml/incollection.biblatex
> +++ b/tests/biblio2yaml/incollection.biblatex
> @@ -110,8 +110,7 @@ references:
> -  title: The limitations of human knowledge according to Al-Farabi, ibn Bajja, and
> +  title: 'The limitations of human knowledge according to Al-Farabi, <span class="nocase">ibn Bajja</span>, and
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131019004115.GA18998%40protagoras.phil.berkeley.edu.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1.1: Type: text/plain, Size: 4971 bytes --]

Brilliant!

And your corrections of two mistakes in the test output are spot on.

On title vs. booktitle, however:

I'm not quite sure whether I understand the Bibtex.hs code and your 
comments in the commits correctly, but when biblio2yaml is dealing with 
title/booktitle/maintitle, it should never be *comparing* strings (though 
citeproc should compare author and bookauthor, see below).

- book, collection, proceedings, reference can have a title, and 
(optionally) a maintitle. These should be mapped to CSL title and 
container-title. If such entries also have a booktitle, this would only be 
used by the bibtex program to obtain a booktitle for an @inbook, 
@incollection, etc. entry cross-referencing a @book, @collection, etc. 
entry, and
should never be mapped to any CSL variable for a @book etc. entry itself. 
As far as I see, Bibtex.hs uses the biblatex inheritance mechanisms for 
both biblatex and bibtex, which yields correct results for bibtex entries, 
too, so in fact the booktitle field *of book, collection, proceedings, 
reference entries* is never used.

The booktitle field has to be used, by contrast, for inbook, incollection, 
inproceedings, inreference entries without crossref, where it is mapped to 
container-title if no maintitle exists; else maintitle is mapped to 
container-title, and booktitle to volume-title.

BTW, crossref and xref are two different fields in biblatex. crossref seems 
to work well in biblio2yaml, while xref is not about data inheritance but 
about bookkeeping how many child entries share a parent entry, and printing 
the parent entry separately if some threshold is reached. I don't think CSL 
can do anything like this, so most likely biblio2yaml can safely ignore the 
xref field for now.

In contrast, citeproc should compare author and bookauthor. They could be 
different, as in

Scholar, S. 2000. “Introduction.” In: *Collected Essays*, by W. Writer ...

or identical, in which case we'd want

Writer, W. 2000. “Essay One.” In: *Collected Essays.* ...

instead of

Writer, W. 2000. “Essay One.” In: *Collected Essays*, by W. Writer ...

A few additional biblio2yaml tests for title/booktitle are attached.

On Saturday, October 19, 2013 5:24:34 AM UTC, fiddlosopher wrote:
>
> All of these tests pass now...! 
>
> +++ John MacFarlane [Oct 18 13 17:41 ]: 
> > I've added all the tests you sent in your zip file. 
> > Very useful.  I think I corrected two mistakes in the test output. 
> > Let me know if I'm off base.  I've fixed several of the 
> > issues the tests revealed. 
> > 
> > diff --git a/tests/biblio2yaml/patent.biblatex 
> b/tests/biblio2yaml/patent.biblatex 
> > index 42b1d1a..9144df3 100644 
> > --- a/tests/biblio2yaml/patent.biblatex 
> > +++ b/tests/biblio2yaml/patent.biblatex 
> > @@ -156,7 +156,7 @@ references: 
> > -  title: High-Speed digital-to-RF converter 
> > +  title: High-speed digital-to-RF converter 
> > 
> > commit 45392c988c8f533a20a0cd931655878b1f24b55e 
> > Author: John MacFarlane <j...-TVLZxgkOlNX2fBVCVOL8/A@public.gmane.org <javascript:>> 
> > Date:   Fri Oct 18 17:22:44 2013 -0700 
> > 
> >     biblio2yaml: Added missing 'span class="nocase"' in test case. 
> > 
> > diff --git a/tests/biblio2yaml/incollection.biblatex 
> b/tests/biblio2yaml/incollection.biblatex 
> > index c9a8bc3..9bf23ac 100644 
> > --- a/tests/biblio2yaml/incollection.biblatex 
> > +++ b/tests/biblio2yaml/incollection.biblatex 
> > @@ -110,8 +110,7 @@ references: 
> > -  title: The limitations of human knowledge according to Al-Farabi, ibn 
> Bajja, and 
> > +  title: 'The limitations of human knowledge according to Al-Farabi, 
> <span class="nocase">ibn Bajja</span>, and 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "pandoc-discuss" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/20131019004115.GA18998%40protagoras.phil.berkeley.edu. 
>
> > For more options, visit https://groups.google.com/groups/opt_out. 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/bcd8674d-af74-47ed-a088-7268152a4b6c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #1.2: Type: text/html, Size: 5968 bytes --]

[-- Attachment #2: test2.zip --]
[-- Type: application/zip, Size: 1870 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1485 bytes --]

A period right after the title is missing in the following example, most 
likely because of some interaction between <span class="nocase"> and <i>:

$ echo @item1 | pandoc --filter pandoc-citeproc -t 
markdown-citations-markdown_in_html_blocks --bibliography 
../pandoc-citeproc/tests/biblio2yaml/test-case-conversion.biblatex 
Author (2013)

Author, Ann. 2013. “A Title, in English, with a Proper Name and an
ACRONYM and a <span class="nocase">camelCase</span> Word and Some Units,
400 <span class="nocase">nm</span>, 3 <span class="nocase">cm</span>,
and a Quote, *Alea <span class="nocase">iacta est</span>*” *Journal*.


Also, I'd say that <span class="nocase"></span> elements in all types of 
pandoc output (except for metadata "references" sections) are unnecessary 
and should be removed: Case doesn't need to be protected here since 
citeproc won't ever see this as input.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/0a54dd61-a375-4bfa-8b80-8b1709ddec34%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1824 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've added some code to strip out those span tags.

I don't know about the missing period...probably we need Andrea
to hunt that one down.

+++ Nick Bart [Oct 19 13 07:19 ]:
> A period right after the title is missing in the following example, most 
> likely because of some interaction between <span class="nocase"> and <i>:
> 
> $ echo @item1 | pandoc --filter pandoc-citeproc -t 
> markdown-citations-markdown_in_html_blocks --bibliography 
> ../pandoc-citeproc/tests/biblio2yaml/test-case-conversion.biblatex 
> Author (2013)
> 
> Author, Ann. 2013. “A Title, in English, with a Proper Name and an
> ACRONYM and a <span class="nocase">camelCase</span> Word and Some Units,
> 400 <span class="nocase">nm</span>, 3 <span class="nocase">cm</span>,
> and a Quote, *Alea <span class="nocase">iacta est</span>*” *Journal*.
> 
> 
> Also, I'd say that <span class="nocase"></span> elements in all types of 
> pandoc output (except for metadata "references" sections) are unnecessary 
> and should be removed: Case doesn't need to be protected here since 
> citeproc won't ever see this as input.
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/0a54dd61-a375-4bfa-8b80-8b1709ddec34%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131019232356.GE84789%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

This is not due to the interaction between <span> and <i>; I can
reproduce it without the <i>:

@article{item1,
    Author = {Author, Ann},
    Date = {2013},
    Hyphenation = {english},
    Journaltitle = {Journal},
    Title = {alea {est}}
}

Oddly, if you change this to

@article{item1,
    Author = {Author, Ann},
    Date = {2013},
    Hyphenation = {english},
    Journaltitle = {Journal},
    Title = {{est}}
}

then the period appears.  That should be a big clue.


+++ Nick Bart [Oct 19 13 07:19 ]:
> A period right after the title is missing in the following example, most 
> likely because of some interaction between <span class="nocase"> and <i>:
> 
> $ echo @item1 | pandoc --filter pandoc-citeproc -t 
> markdown-citations-markdown_in_html_blocks --bibliography 
> ../pandoc-citeproc/tests/biblio2yaml/test-case-conversion.biblatex 
> Author (2013)
> 
> Author, Ann. 2013. “A Title, in English, with a Proper Name and an
> ACRONYM and a <span class="nocase">camelCase</span> Word and Some Units,
> 400 <span class="nocase">nm</span>, 3 <span class="nocase">cm</span>,
> and a Quote, *Alea <span class="nocase">iacta est</span>*” *Journal*.
> 
> 
> Also, I'd say that <span class="nocase"></span> elements in all types of 
> pandoc output (except for metadata "references" sections) are unnecessary 
> and should be removed: Case doesn't need to be protected here since 
> citeproc won't ever see this as input.
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/0a54dd61-a375-4bfa-8b80-8b1709ddec34%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131022013305.GA71166%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Aha -- I found the problem and fixed it.  The period is back.
It comes, oddly, inside the emphasis; fixing that would require
more changes to the way citeproc-hs does things.

+++ John MacFarlane [Oct 21 13 18:33 ]:
> This is not due to the interaction between <span> and <i>; I can
> reproduce it without the <i>:
> 
> @article{item1,
>     Author = {Author, Ann},
>     Date = {2013},
>     Hyphenation = {english},
>     Journaltitle = {Journal},
>     Title = {alea {est}}
> }
> 
> Oddly, if you change this to
> 
> @article{item1,
>     Author = {Author, Ann},
>     Date = {2013},
>     Hyphenation = {english},
>     Journaltitle = {Journal},
>     Title = {{est}}
> }
> 
> then the period appears.  That should be a big clue.
> 
> 
> +++ Nick Bart [Oct 19 13 07:19 ]:
> > A period right after the title is missing in the following example, most 
> > likely because of some interaction between <span class="nocase"> and <i>:
> > 
> > $ echo @item1 | pandoc --filter pandoc-citeproc -t 
> > markdown-citations-markdown_in_html_blocks --bibliography 
> > ../pandoc-citeproc/tests/biblio2yaml/test-case-conversion.biblatex 
> > Author (2013)
> > 
> > Author, Ann. 2013. “A Title, in English, with a Proper Name and an
> > ACRONYM and a <span class="nocase">camelCase</span> Word and Some Units,
> > 400 <span class="nocase">nm</span>, 3 <span class="nocase">cm</span>,
> > and a Quote, *Alea <span class="nocase">iacta est</span>*” *Journal*.
> > 
> > 
> > Also, I'd say that <span class="nocase"></span> elements in all types of 
> > pandoc output (except for metadata "references" sections) are unnecessary 
> > and should be removed: Case doesn't need to be protected here since 
> > citeproc won't ever see this as input.
> > 
> > -- 
> > You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> > To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> > To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> > To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/0a54dd61-a375-4bfa-8b80-8b1709ddec34%40googlegroups.com.
> > For more options, visit https://groups.google.com/groups/opt_out.
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131022013305.GA71166%40Johns-MacBook-Pro.local.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131022051454.GA80364%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 5738 bytes --]

Good to find that the improved pandoc-citeproc sees a public release, and 
it does indeed start producing much better results than before when 
bibutils was used.

One of the biggest remaining issues right now seems to be date 
disambiguation: With the style file chicago-author-date.csl, pandoc 
produces wrong disambiguation letters, and with the the style files apa.csl 
none at all (see examples below). Both .csl files do work well with Zotero, 
so it's most likely not a style file issue.

Maybe the following gives a hint:

chicago-author-date.csl contains the line

<citation et-al-min="4" et-al-use-first="1" 
disambiguate-add-year-suffix="true" disambiguate-add-names="true" 
disambiguate-add-givenname="true" 
givenname-disambiguation-rule="primary-name">

If you remove 'disambiguate-add-names="true"', the behaviour, at least for 
the example given below, is as expected again.

Note that disambiguation not only has to deal with same author(s)/same 
year, but also, if there is no author, with whatever replaces the 
author(s), either same title/same year, or, in the case of 
magazines/newspapers, same journal/same year.

For title-year in-text citations, CSL title-short, if available, should be 
used; see CMoS, 16e, 15.32: “Text citations may refer to a short form of 
the title but must include the first word (other than an initial article).”

For testing same journal/same year cases, my modified 
chicago-author-date-TEST-20131018.csl, attached to an earlier post in this 
thread, needs to be used.

Andrea, if you are reading this, could you give us a hint as to what you 
think could be done about it?


pandoc-citeproc also has a few issues concerning capitalization: In some 
cases it does too much when converting to title case:

two (of four) -> Two (Of Four) (expected: Two (of Four))

... and in some cases not enough:

big/small -> Big/small (expected: Big/Small)
post-mortem -> Post-mortem (expected: Post-Mortem)



Tests:

$ cat date-disambiguation-author.bib 

@article{item1,
    Author = {Author, Al},
    Date = {2003},
    Hyphenation = {english},
    Title = {Title One}}

@article{item2,
    Author = {Author, Al},
    Date = {2003},
    Hyphenation = {english},
    Title = {Title Two}}

@article{item3,
    Author = {Buthor, Al},
    Date = {2003},
    Hyphenation = {english},
    Title = {Title Three}}

@article{item4,
    Author = {Buthor, Al},
    Date = {2003},
    Hyphenation = {english},
    Title = {Title Four}}

$ echo [@item1] [@item2] [@item3] [@item4] | pandoc -S -t 
markdown-citations-markdown_in_html_blocks --bibliography 
date-disambiguation-author.bib --csl chicago-author-date.csl
(Author 2003a) (Author 2003b) (Buthor 2003c) (Buthor 2003d)

Author, Al. 2003a. “Title One.”

———. 2003b. “Title Two.”

Buthor, Al. 2003c. “Title Three.”

———. 2003d. “Title Four.”


$ echo [@item1] [@item2] [@item3] [@item4] | pandoc -S -t 
markdown-citations-markdown_in_html_blocks --bibliography 
date-disambiguation-author.bib --csl apa.csl
(Author, 2003) (Author, 2003) (Buthor, 2003) (Buthor, 2003)

Author, A. (2003). Title one.

Author, A. (2003). Title two.

Buthor, A. (2003). Title three.

Buthor, A. (2003). Title four.



$ cat date-disambiguation-title.bib 


@article{item1,
    Date = {2003},
    Hyphenation = {english},
    Title = {Title One}}

@article{item2,
    Date = {2003},
    Hyphenation = {english},
    Title = {Title One}}

@book{item3,
    Date = {2003},
    Hyphenation = {english},
    Title = {Title Two}}

@book{item4,
    Date = {2003},
    Hyphenation = {english},
    Title = {Title Two}}

$ echo [@item1] [@item2] [@item3] [@item4] | pandoc -S -t 
markdown-citations-markdown_in_html_blocks --bibliography 
date-disambiguation-title.bib --csl chicago-author-date.csl
(“Title One” 2003a) (“Title One” 2003b) (*Title Two* 2003c) (*Title Two*
2003d)

“Title One.” 2003a.

“Title One.” 2003b.

*Title Two*. 2003c.

*Title Two*. 2003d.



$ cat date-disambiguation-magazine.bib 
@article{item1,
    Date = {2003},
    Entrysubtype = {magazine},
    Hyphenation = {english},
    Journal = {Magazine},
    Title = {Title One}}

@article{item2,
    Date = {2003},
    Entrysubtype = {magazine},
    Hyphenation = {english},
    Journal = {Magazine},
    Title = {Title Two}}

@article{item3,
    Date = {2003},
    Entrysubtype = {magazine},
    Hyphenation = {english},
    Journal = {Newspaper},
    Title = {Title Three}}

@article{item4,
    Date = {2003},
    Entrysubtype = {magazine},
    Hyphenation = {english},
    Journal = {Newspaper},
    Title = {Title Four}}

$ echo [@item1] [@item2] [@item3] [@item4] | pandoc -S -t 
markdown-citations-markdown_in_html_blocks --bibliography 
date-disambiguation-magazine.bib --csl chicago-author-date-TEST-20131018.csl
(*Magazine* 2003a) (*Magazine* 2003b) (*Newspaper* 2003c) (*Newspaper*
2003d)

*Magazine*. 2003a. “Title One.”

*Magazine*. 2003b. “Title Two.”

*Newspaper*. 2003c. “Title Three.”

*Newspaper*. 2003d. “Title Four.”

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/94195ec7-4de6-45b9-930e-30523ff04477%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 7001 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4766 bytes --]

I’ve started checking out the biblatex-chicago examples 
(http://mirrors.ctan.org/macros/latex/contrib/biblatex-contrib/biblatex-chicago/doc/examples/dates-test.bib; 
there's also a notes-test.bib I will have a look at later). A few things I 
noticed here are the following:

- "~" is used in author and editor fields, e.g., "Author = {Author, 
Margaret~M.}," or "Editor = {Ross, W.~D.},"

I don't think there's a reason not to treat these as spaces, and to output

  author:
    given:
    - Margaret
    - M.

instead of 

  author:
    given:
    - ! "Margaret\_M."

- "\ " is used in various fields, e.g., "Publisher = {Oxford Univ.\ Press 
and Sandpiper Books},"

In latex, the "\ " does *not* stand for a no-break space, but indicates 
that a single space is to be used, e.g., after an abbreviation dot. If 
vanilla latex sees just ". ", it assumes it's the end of a sentence and 
outputs two spaces. Since pandoc (quite sensibly, see CMoS) does not add 
extra space after sentences, there is, again, no reason not to treat latex 
"\ " like a normal space char. Or use the "punctuation space" char, U+2008. 
(Also, to follow CMoS recommendations, "\frenchspacing" could be added to 
the preamble of default.latex to suppress printing two spaces after a 
period.)

BTW, I realize that "\_" in yaml does work, at least pandoc output seems 
correct, but I find the notation of "\_" for protected space in yaml mildly 
irritating. If it's treated like a normal space by pandoc anyway, as it 
seems to be, why not use a normal space in yaml, too? And if "\_" is 
intended to represent a genuine no-break space, is this the only 
alternative? I for one would prefer using Unicode wherever possible, in 
this case a no-break space, U+00A0, would do nicely.

- Interestingly, while the biblatex date field is restricted to (variations 
of) yyyy-mm-dd, the year field can contain pretty much anything, and 
biblatex-chicago uses this option, so it would be good if pandoc-citeproc 
could map everything it finds in a year field that is not a number, after 
expanding it as necessary, to yaml as a literal string to "issued:  other:" 

biblatex-chicago examples include:

    Year = {1860s}
    Year = {\mkbibbrackets{1957?}}
    Year = {[1920?]}
    Year = {\bibstring{nodate}}}

These could be mapped to CSL "issued" as follows:

  issued:
    other: 1860s

  issued:
    other: [1957?]

  issued:
    other: [1920?]

  issued:
    other: no date

The following is a special case:

    Year = {\autocap{f}orthcoming}
    Year = {forthcoming}

should be mapped to CSL "status" (!):

   status: forthcoming

(\autocap{} can be disregarded; CSL can do its own capitalization here, if 
needed.)

The biblatex-chicago examples also include things such as "year = 
{1951--63}," and "{1951\---63},", but I would regard these as malformed 
(should be "date = {1951/1963}" etc.).


The biblatex-chicago example databases also include a number of 
\bibstring{} commands, and it would be nice if pandoc-citeproc could expand 
these. I have added the  English equivalents (usually the short ones from 
/usr/local/texlive/2013/texmf-dist/tex/latex/biblatex/lbx/english.lbx; 
pseudonym and revisededition from 
/usr/local/texlive/2013/texmf-dist/tex/latex/biblatex-chicago/cms-american.lbx):

\bibstring{and}             -> and
\bibstring{by}              -> by
\bibstring{edition}         -> ed.
\bibstring{introduction}    -> intro.
\bibstring{jourser}         -> ser.
\bibstring{newseries}       -> new ser.
\bibstring{nodate}          -> no date
\bibstring{number}          -> no.
\bibstring{pseudonym}       -> pseud.
\bibstring{reviewof}        -> review of
\bibstring{revisededition}  -> rev. ed.

I’d suggest using the long form of "no date" since this is a CSL "term" and 
can be processed (e.g., shortened) by CSL only if it appears exactly in 
this form.

There's also one command in the example databases that could be translated:

\partedit                   -> ed. by

(defined in 
/usr/local/texlive/2013/texmf-dist/tex/latex/biblatex-chicago/chicago-authordate.cbx)

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/551cc4f2-546c-4386-8d61-e8ea8d3220fc%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5929 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Haven't had a chance to look into this yet.  I've added a new
optional test suite, which runs the citeproc test suite.

To use it, first compile pandoc-citeproc with the 'test_citeproc'
flag:

cabal install --enable-tests -ftest_citeproc

Then,

test-citeproc

will run the suite.  The first time you run it, it will clone
the test suite repository (you need hg installed) and run a
program that converts the tests to machine-readable form (you
need python installed for this).

Score so far is 525 passed, 226 failed, 6 skipped, so there's work to be
done.

You can specify an argument or arguments to test-citeproc if you
just want to look at certain tests.  e.g.

test-citeproc affix prefix suffix

will give you all the tests with any of these three words in the
filename.

The nice thing about the test suite is that it will make it easy to
see if I've broken something once I start messing with things like
date disambiguation.  (Interestingly, there isn't now a test that gets
at the problem Nick noted.)

+++ Nick Bart [Oct 22 13 01:15 ]:
> Good to find that the improved pandoc-citeproc sees a public release, and 
> it does indeed start producing much better results than before when 
> bibutils was used.
> 
> One of the biggest remaining issues right now seems to be date 
> disambiguation...

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've made some large-scale changes in the pandoc-citeproc code,
including the part that overlaps citeproc-hs.  Let me just explain
the changes and what motivated them.

First, I removed dependency on the json library. It seems silly to
depend on two different json libraries, and we were already using aeson.
I defined FromJSON instances for the various citeproc types, doing so
in a really permissive way so that even the somewhat chaotic JSON in
the citeproc test suite will work.  (Numbers are sometimes used where
strings will also go; 0 and 1 are sometimes used for booleans; fields
that can be lists can also have the value 'false', etc. Ah, dynamic
languages!)

I then moved these instance definitions to the modules where the
relevant types are defined, to avoid orphan instances.  This required
me to merge Text.CSL.Parser into Text.CSL.Style, in order to avoid a
circular module dependency.  (The FromJSON instance for Style needs to
be able to parse the XML for the CSL style.)  So, no more
Text.CSL.Parser; the parsing functions are now in Text.CSL.Style.

Doing this allowed me to remove Text.CSL.Input.Json entirely; one can
now just use 'Data.Aeson.decode' to read JSON into citeproc data
structures.

I also removed dependency on utf8-string; pandoc supplies functions that
do the conversions we need.

In the process of doing this, I cleaned up some duplication and moved a
bunch of general-purpose code to a new (unexported) module,
Text.CSL.Util.

So, I think all of this is an improvement.  I hope Andrea agrees.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2845 bytes --]

All this seems extremely useful: Thank you.

Another suggestion for Bibtex.hs:

In case a biblatex entry contains a hyphenation field with a language term 
not listed in Bibtex.hs, its output (in yaml) should either copy the 
content of the hyphenation field into "language: ", or output something 
like "language: unknown".

I stumbled across this in biblatex-examples.bib which contains entries in 
Latin. For Latin in particular, I’d suggest adding the following line to 
Bibtex.hs:

toLocale "latin"    = "la"



On Friday, 25 October 2013 05:37:20 UTC, fiddlosopher wrote:
>
> I've made some large-scale changes in the pandoc-citeproc code, 
> including the part that overlaps citeproc-hs.  Let me just explain 
> the changes and what motivated them. 
>
> First, I removed dependency on the json library. It seems silly to 
> depend on two different json libraries, and we were already using aeson. 
> I defined FromJSON instances for the various citeproc types, doing so 
> in a really permissive way so that even the somewhat chaotic JSON in 
> the citeproc test suite will work.  (Numbers are sometimes used where 
> strings will also go; 0 and 1 are sometimes used for booleans; fields 
> that can be lists can also have the value 'false', etc. Ah, dynamic 
> languages!) 
>
> I then moved these instance definitions to the modules where the 
> relevant types are defined, to avoid orphan instances.  This required 
> me to merge Text.CSL.Parser into Text.CSL.Style, in order to avoid a 
> circular module dependency.  (The FromJSON instance for Style needs to 
> be able to parse the XML for the CSL style.)  So, no more 
> Text.CSL.Parser; the parsing functions are now in Text.CSL.Style. 
>
> Doing this allowed me to remove Text.CSL.Input.Json entirely; one can 
> now just use 'Data.Aeson.decode' to read JSON into citeproc data 
> structures. 
>
> I also removed dependency on utf8-string; pandoc supplies functions that 
> do the conversions we need. 
>
> In the process of doing this, I cleaned up some duplication and moved a 
> bunch of general-purpose code to a new (unexported) module, 
> Text.CSL.Util. 
>
> So, I think all of this is an improvement.  I hope Andrea agrees. 
>
>
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/39973df3-822d-4f25-96b9-466a7ad669e9%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3333 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've made this change.

+++ Nick Bart [Oct 25 13 03:33 ]:
> All this seems extremely useful: Thank you.
> 
> Another suggestion for Bibtex.hs:
> 
> In case a biblatex entry contains a hyphenation field with a language term 
> not listed in Bibtex.hs, its output (in yaml) should either copy the 
> content of the hyphenation field into "language: ", or output something 
> like "language: unknown".
> 
> I stumbled across this in biblatex-examples.bib which contains entries in 
> Latin. For Latin in particular, I’d suggest adding the following line to 
> Bibtex.hs:
> 
> toLocale "latin"    = "la"
> 

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131025171704.GD41641%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2122 bytes --]

Great. I wouldn't map forthcoming to *both* status and issued, though; just 
to status is perfectly ok. (When "forthcoming" is in the biblatex pubstate 
field, we don't map it to both status and issued either; that fact that can 
appear in a year field at all is just an idiosyncrasy of biblatex-chicago, 
probably from before biblatex had a pubstate field.)

Not too many csl styles can handle "status" yet, but try 
antarctic-science.csl, or my chicago-author-date-TEST-20131018.csl.

One of our next steps will have to be to encourage CSL style authors to 
cater for those valid CSL variables that have not been implemented in most 
style files so far (usually because no frontend was actually using them) 
but that can be easily accessed now via pandoc-citeproc / biblio2yaml.


On Friday, 25 October 2013 17:17:04 UTC, fiddlosopher wrote:
>
> I've made this change. 
>
> +++ Nick Bart [Oct 25 13 03:33 ]: 
> > All this seems extremely useful: Thank you. 
> > 
> > Another suggestion for Bibtex.hs: 
> > 
> > In case a biblatex entry contains a hyphenation field with a language 
> term 
> > not listed in Bibtex.hs, its output (in yaml) should either copy the 
> > content of the hyphenation field into "language: ", or output something 
> > like "language: unknown". 
> > 
> > I stumbled across this in biblatex-examples.bib which contains entries 
> in 
> > Latin. For Latin in particular, I’d suggest adding the following line to 
> > Bibtex.hs: 
> > 
> > toLocale "latin"    = "la" 
> > 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/5bd1f091-3114-4e1a-b477-20f1ed8c0777%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2573 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1.1: Type: text/plain, Size: 760 bytes --]

… and here are a few more tests for biblio2yaml: The entries of vanilla 
biblatex’s "biblatex-example.bib", reformatted for testing, with a few 
notes (grep for "NOTES:").

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a88bfdf5-f8a2-4cde-b4d1-a7bcbfc54810%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #1.2: Type: text/html, Size: 1012 bytes --]

[-- Attachment #2: biblio2yaml-tests-from-biblatex-examples-bib.zip --]
[-- Type: application/zip, Size: 82572 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1078 bytes --]

One more detail on years:

Any value in a biblatex year field that could sensibly be understood as a 
negative year, "-200", "--333", or "–753" (unlike the first, the last two 
would look good in latex, too) should be mapped to "issued: year:" instead 
of "issued: other:", with a *single* hyphen in front of the number.

Done this way, citeproc correctly formats a value of "-753" as "753BC"; 
with a negative year mapped to "other", citeproc gets confused and prints 
"-753753AD."

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/ec945e8b-2e50-4716-aefe-d7a7ccf792e0%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1337 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Done.

+++ Nick Bart [Oct 25 13 15:12 ]:
> One more detail on years:
> 
> Any value in a biblatex year field that could sensibly be understood as a 
> negative year, "-200", "--333", or "–753" (unlike the first, the last two 
> would look good in latex, too) should be mapped to "issued: year:" instead 
> of "issued: other:", with a *single* hyphen in front of the number.
> 
> Done this way, citeproc correctly formats a value of "-753" as "753BC"; 
> with a negative year mapped to "other", citeproc gets confused and prints 
> "-753753AD."
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/ec945e8b-2e50-4716-aefe-d7a7ccf792e0%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131026031252.GA54079%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Excellent, I've added these.  Quite a few currently fail.

+++ Nick Bart [Oct 25 13 14:38 ]:
> … and here are a few more tests for biblio2yaml: The entries of vanilla 
> biblatex’s "biblatex-example.bib", reformatted for testing, with a few 
> notes (grep for "NOTES:").
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a88bfdf5-f8a2-4cde-b4d1-a7bcbfc54810%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.


-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131026031305.GB54079%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

The latest code passes all but 3 of the 118 tests.  Note:  you must
first compile the latest pandoc from github (which fixes a bug in markdown
rendering of spans with attributes).

Two of the failures are of this type:

--- expected    2013-10-27 20:24:54.000000000 -0700
+++ actual      2013-10-27 20:24:54.000000000 -0700
@@ -18,7 +18,7 @@
         }
       ],
       "container-title": "Journal of Computational Chemistry",
-      "doi": "10.1002/(SICI)1096-987X(199803)19:4\u003c377::AID-JCC1\u003e3.0.CO;2-P",
+      "doi": "doi:10.1002/(SICI)1096-987X(199803)19:4\u003c377::AID-JCC1\u003e3.0.CO;2-P",
       "id": "sigfridsson",
       "issue": 4,
       "issued": {

The difference here is that biblio2yaml puts the 'doi:' prefix in front
of the DOI.  I did this on your earlier request, since citeproc-hs code
only linkifies the DOI if that prefix is there.

I could change citeproc to linkify all DOIs.  Or I could change biblio2yaml
so that it doesn't put the 'doi:' prefix there when converting from biblatex.
Or we could change the test cases so they expect a 'doi:' prefix.  Which
is best?

John

+++ Nick Bart [Oct 25 13 14:38 ]:
> … and here are a few more tests for biblio2yaml: The entries of vanilla 
> biblatex’s "biblatex-example.bib", reformatted for testing, with a few 
> notes (grep for "NOTES:").
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a88bfdf5-f8a2-4cde-b4d1-a7bcbfc54810%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.


-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131028033052.GA12640%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4171 bytes --]

Well, I still think pandoc-citeproc should generate links for *all* DOI 
fields by default (that's what I suggested earlier).

I feel that linkifying just some DOIs but not all does not make sense and 
will only confuse readers. Hence, I don't think we need a mechanism for 
that, and I don't like the "doi:" prefix in particular since it's not part 
of the DOI name spec and also, if used in a biblatex database, breaks the 
latex output.

A different question would be whether users should be able to specify 
whether they want all DOIs linkified or none. In latex, all urls are 
clickable by default whereas DOIs become clickable only when the hyperref 
package is loaded. My feeling is that clickable URLs and DOIs are perfectly 
harmless, so pandoc would not need to offer such an option, but maybe there 
are use cases that would call for documents stripped of any clickable 
links: If anyone is aware of such cases, please let us know.

Yet another question is an item-specific one: whether to include URL or DOI 
information *at all* for a given item. Here, biblatex has url=true/false 
and doi=true/false in the options field, with both defaulting to true. If 
set to false, pandoc-citeproc could exclude the respective field from being 
mapped to CSL. (However, as in standard biblatex, the url field should 
always be included in online entries, regardless of the state of this 
option.)

On Monday, 28 October 2013 03:30:52 UTC, fiddlosopher wrote:
>
> The latest code passes all but 3 of the 118 tests.  Note:  you must 
> first compile the latest pandoc from github (which fixes a bug in markdown 
> rendering of spans with attributes). 
>
> Two of the failures are of this type: 
>
> --- expected    2013-10-27 20:24:54.000000000 -0700 
> +++ actual      2013-10-27 20:24:54.000000000 -0700 
> @@ -18,7 +18,7 @@ 
>          } 
>        ], 
>        "container-title": "Journal of Computational Chemistry", 
> -      "doi": "10.1002/(SICI)1096-987X(199803)19:4\u003c377::AID-JCC1\
> u003e3.0.CO;2-P", 
> +      "doi": "doi:10.1002/(SICI)1096-987X(199803)19:4\u003c377::AID-JCC1\
> u003e3.0.CO;2-P", 
>        "id": "sigfridsson", 
>        "issue": 4, 
>        "issued": { 
>
> The difference here is that biblio2yaml puts the 'doi:' prefix in front 
> of the DOI.  I did this on your earlier request, since citeproc-hs code 
> only linkifies the DOI if that prefix is there. 
>
> I could change citeproc to linkify all DOIs.  Or I could change 
> biblio2yaml 
> so that it doesn't put the 'doi:' prefix there when converting from 
> biblatex. 
> Or we could change the test cases so they expect a 'doi:' prefix.  Which 
> is best? 
>
> John 
>
> +++ Nick Bart [Oct 25 13 14:38 ]: 
> > … and here are a few more tests for biblio2yaml: The entries of vanilla 
> > biblatex’s "biblatex-example.bib", reformatted for testing, with a few 
> > notes (grep for "NOTES:"). 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "pandoc-discuss" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/a88bfdf5-f8a2-4cde-b4d1-a7bcbfc54810%40googlegroups.com. 
>
> > For more options, visit https://groups.google.com/groups/opt_out. 
>
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/78a76dee-7c31-4323-8d66-21de3212d267%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5231 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4504 bytes --]

To sum it up: I would change biblio2yaml so that it doesn't put the 'doi:' 
prefix there when converting from biblatex, *and* I would change citeproc 
to linkify all DOIs. 

On Monday, 28 October 2013 12:43:27 UTC, Nick Bart wrote:
>
> Well, I still think pandoc-citeproc should generate links for *all* DOI 
> fields by default (that's what I suggested earlier).
>
> I feel that linkifying just some DOIs but not all does not make sense and 
> will only confuse readers. Hence, I don't think we need a mechanism for 
> that, and I don't like the "doi:" prefix in particular since it's not part 
> of the DOI name spec and also, if used in a biblatex database, breaks the 
> latex output.
>
> A different question would be whether users should be able to specify 
> whether they want all DOIs linkified or none. In latex, all urls are 
> clickable by default whereas DOIs become clickable only when the hyperref 
> package is loaded. My feeling is that clickable URLs and DOIs are perfectly 
> harmless, so pandoc would not need to offer such an option, but maybe there 
> are use cases that would call for documents stripped of any clickable 
> links: If anyone is aware of such cases, please let us know.
>
> Yet another question is an item-specific one: whether to include URL or 
> DOI information *at all* for a given item. Here, biblatex has 
> url=true/false and doi=true/false in the options field, with both 
> defaulting to true. If set to false, pandoc-citeproc could exclude the 
> respective field from being mapped to CSL. (However, as in standard 
> biblatex, the url field should always be included in online entries, 
> regardless of the state of this option.)
>
> On Monday, 28 October 2013 03:30:52 UTC, fiddlosopher wrote:
>>
>> The latest code passes all but 3 of the 118 tests.  Note:  you must 
>> first compile the latest pandoc from github (which fixes a bug in 
>> markdown 
>> rendering of spans with attributes). 
>>
>> Two of the failures are of this type: 
>>
>> --- expected    2013-10-27 20:24:54.000000000 -0700 
>> +++ actual      2013-10-27 20:24:54.000000000 -0700 
>> @@ -18,7 +18,7 @@ 
>>          } 
>>        ], 
>>        "container-title": "Journal of Computational Chemistry", 
>> -      "doi": "10.1002/(SICI)1096-987X(199803)19:4\u003c377::AID-JCC1\
>> u003e3.0.CO;2-P", 
>> +      "doi": "doi:10.1002/(SICI)1096-987X(199803)19:4\u003c377::AID-JCC1\
>> u003e3.0.CO;2-P", 
>>        "id": "sigfridsson", 
>>        "issue": 4, 
>>        "issued": { 
>>
>> The difference here is that biblio2yaml puts the 'doi:' prefix in front 
>> of the DOI.  I did this on your earlier request, since citeproc-hs code 
>> only linkifies the DOI if that prefix is there. 
>>
>> I could change citeproc to linkify all DOIs.  Or I could change 
>> biblio2yaml 
>> so that it doesn't put the 'doi:' prefix there when converting from 
>> biblatex. 
>> Or we could change the test cases so they expect a 'doi:' prefix.  Which 
>> is best? 
>>
>> John 
>>
>> +++ Nick Bart [Oct 25 13 14:38 ]: 
>> > … and here are a few more tests for biblio2yaml: The entries of vanilla 
>> > biblatex’s "biblatex-example.bib", reformatted for testing, with a few 
>> > notes (grep for "NOTES:"). 
>> > 
>> > -- 
>> > You received this message because you are subscribed to the Google 
>> Groups "pandoc-discuss" group. 
>> > To unsubscribe from this group and stop receiving emails from it, send 
>> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>> > To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/pandoc-discuss/a88bfdf5-f8a2-4cde-b4d1-a7bcbfc54810%40googlegroups.com. 
>>
>> > For more options, visit https://groups.google.com/groups/opt_out. 
>>
>>
>>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/7b646c3e-be17-44f0-b56a-d49519e63b6c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5488 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Matthias Hüning]

[-- Attachment #1: Type: text/plain, Size: 4773 bytes --]

What do you mean by 'linkify'? Putting http://dx.doi.org/... in front of 
the DOI?

- Matthias


Am Montag, 28. Oktober 2013 13:54:50 UTC+1 schrieb Nick Bart:
>
> To sum it up: I would change biblio2yaml so that it doesn't put the 'doi:' 
> prefix there when converting from biblatex, *and* I would change citeproc 
> to linkify all DOIs. 
>
> On Monday, 28 October 2013 12:43:27 UTC, Nick Bart wrote:
>>
>> Well, I still think pandoc-citeproc should generate links for *all* DOI 
>> fields by default (that's what I suggested earlier).
>>
>> I feel that linkifying just some DOIs but not all does not make sense and 
>> will only confuse readers. Hence, I don't think we need a mechanism for 
>> that, and I don't like the "doi:" prefix in particular since it's not part 
>> of the DOI name spec and also, if used in a biblatex database, breaks the 
>> latex output.
>>
>> A different question would be whether users should be able to specify 
>> whether they want all DOIs linkified or none. In latex, all urls are 
>> clickable by default whereas DOIs become clickable only when the hyperref 
>> package is loaded. My feeling is that clickable URLs and DOIs are perfectly 
>> harmless, so pandoc would not need to offer such an option, but maybe there 
>> are use cases that would call for documents stripped of any clickable 
>> links: If anyone is aware of such cases, please let us know.
>>
>> Yet another question is an item-specific one: whether to include URL or 
>> DOI information *at all* for a given item. Here, biblatex has 
>> url=true/false and doi=true/false in the options field, with both 
>> defaulting to true. If set to false, pandoc-citeproc could exclude the 
>> respective field from being mapped to CSL. (However, as in standard 
>> biblatex, the url field should always be included in online entries, 
>> regardless of the state of this option.)
>>
>> On Monday, 28 October 2013 03:30:52 UTC, fiddlosopher wrote:
>>>
>>> The latest code passes all but 3 of the 118 tests.  Note:  you must 
>>> first compile the latest pandoc from github (which fixes a bug in 
>>> markdown 
>>> rendering of spans with attributes). 
>>>
>>> Two of the failures are of this type: 
>>>
>>> --- expected    2013-10-27 20:24:54.000000000 -0700 
>>> +++ actual      2013-10-27 20:24:54.000000000 -0700 
>>> @@ -18,7 +18,7 @@ 
>>>          } 
>>>        ], 
>>>        "container-title": "Journal of Computational Chemistry", 
>>> -      "doi": "10.1002/(SICI)1096-987X(199803)19:4\u003c377::AID-JCC1\
>>> u003e3.0.CO;2-P", 
>>> +      "doi": 
>>> "doi:10.1002/(SICI)1096-987X(199803)19:4\u003c377::AID-JCC1\u003e3.0.CO;2-P", 
>>>
>>>        "id": "sigfridsson", 
>>>        "issue": 4, 
>>>        "issued": { 
>>>
>>> The difference here is that biblio2yaml puts the 'doi:' prefix in front 
>>> of the DOI.  I did this on your earlier request, since citeproc-hs code 
>>> only linkifies the DOI if that prefix is there. 
>>>
>>> I could change citeproc to linkify all DOIs.  Or I could change 
>>> biblio2yaml 
>>> so that it doesn't put the 'doi:' prefix there when converting from 
>>> biblatex. 
>>> Or we could change the test cases so they expect a 'doi:' prefix.  Which 
>>> is best? 
>>>
>>> John 
>>>
>>> +++ Nick Bart [Oct 25 13 14:38 ]: 
>>> > … and here are a few more tests for biblio2yaml: The entries of 
>>> vanilla 
>>> > biblatex’s "biblatex-example.bib", reformatted for testing, with a few 
>>> > notes (grep for "NOTES:"). 
>>> > 
>>> > -- 
>>> > You received this message because you are subscribed to the Google 
>>> Groups "pandoc-discuss" group. 
>>> > To unsubscribe from this group and stop receiving emails from it, send 
>>> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>>> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>>> > To view this discussion on the web visit 
>>> https://groups.google.com/d/msgid/pandoc-discuss/a88bfdf5-f8a2-4cde-b4d1-a7bcbfc54810%40googlegroups.com. 
>>>
>>> > For more options, visit https://groups.google.com/groups/opt_out. 
>>>
>>>
>>>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/7efb5264-92d3-4970-8b52-88efb285ec69%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5805 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 5111 bytes --]

It means rendering a DOI as a clickable link, in markdown output:

doi:[10.1063/1.2172593](http://dx.doi.org/10.1063/1.2172593 
"10.1063/1.2172593")


On Monday, 28 October 2013 13:27:20 UTC, Matthias Hüning wrote:
>
> What do you mean by 'linkify'? Putting http://dx.doi.org/... in front of 
> the DOI?
>
> - Matthias
>
>
> Am Montag, 28. Oktober 2013 13:54:50 UTC+1 schrieb Nick Bart:
>>
>> To sum it up: I would change biblio2yaml so that it doesn't put the 
>> 'doi:' prefix there when converting from biblatex, *and* I would change 
>> citeproc to linkify all DOIs. 
>>
>> On Monday, 28 October 2013 12:43:27 UTC, Nick Bart wrote:
>>>
>>> Well, I still think pandoc-citeproc should generate links for *all* DOI 
>>> fields by default (that's what I suggested earlier).
>>>
>>> I feel that linkifying just some DOIs but not all does not make sense 
>>> and will only confuse readers. Hence, I don't think we need a mechanism for 
>>> that, and I don't like the "doi:" prefix in particular since it's not part 
>>> of the DOI name spec and also, if used in a biblatex database, breaks the 
>>> latex output.
>>>
>>> A different question would be whether users should be able to specify 
>>> whether they want all DOIs linkified or none. In latex, all urls are 
>>> clickable by default whereas DOIs become clickable only when the hyperref 
>>> package is loaded. My feeling is that clickable URLs and DOIs are perfectly 
>>> harmless, so pandoc would not need to offer such an option, but maybe there 
>>> are use cases that would call for documents stripped of any clickable 
>>> links: If anyone is aware of such cases, please let us know.
>>>
>>> Yet another question is an item-specific one: whether to include URL or 
>>> DOI information *at all* for a given item. Here, biblatex has 
>>> url=true/false and doi=true/false in the options field, with both 
>>> defaulting to true. If set to false, pandoc-citeproc could exclude the 
>>> respective field from being mapped to CSL. (However, as in standard 
>>> biblatex, the url field should always be included in online entries, 
>>> regardless of the state of this option.)
>>>
>>> On Monday, 28 October 2013 03:30:52 UTC, fiddlosopher wrote:
>>>>
>>>> The latest code passes all but 3 of the 118 tests.  Note:  you must 
>>>> first compile the latest pandoc from github (which fixes a bug in 
>>>> markdown 
>>>> rendering of spans with attributes). 
>>>>
>>>> Two of the failures are of this type: 
>>>>
>>>> --- expected    2013-10-27 20:24:54.000000000 -0700 
>>>> +++ actual      2013-10-27 20:24:54.000000000 -0700 
>>>> @@ -18,7 +18,7 @@ 
>>>>          } 
>>>>        ], 
>>>>        "container-title": "Journal of Computational Chemistry", 
>>>> -      "doi": "10.1002/(SICI)1096-987X(199803)19:4\u003c377::AID-JCC1\
>>>> u003e3.0.CO;2-P", 
>>>> +      "doi": 
>>>> "doi:10.1002/(SICI)1096-987X(199803)19:4\u003c377::AID-JCC1\u003e3.0.CO;2-P", 
>>>>
>>>>        "id": "sigfridsson", 
>>>>        "issue": 4, 
>>>>        "issued": { 
>>>>
>>>> The difference here is that biblio2yaml puts the 'doi:' prefix in front 
>>>> of the DOI.  I did this on your earlier request, since citeproc-hs code 
>>>> only linkifies the DOI if that prefix is there. 
>>>>
>>>> I could change citeproc to linkify all DOIs.  Or I could change 
>>>> biblio2yaml 
>>>> so that it doesn't put the 'doi:' prefix there when converting from 
>>>> biblatex. 
>>>> Or we could change the test cases so they expect a 'doi:' prefix. 
>>>>  Which 
>>>> is best? 
>>>>
>>>> John 
>>>>
>>>> +++ Nick Bart [Oct 25 13 14:38 ]: 
>>>> > … and here are a few more tests for biblio2yaml: The entries of 
>>>> vanilla 
>>>> > biblatex’s "biblatex-example.bib", reformatted for testing, with a 
>>>> few 
>>>> > notes (grep for "NOTES:"). 
>>>> > 
>>>> > -- 
>>>> > You received this message because you are subscribed to the Google 
>>>> Groups "pandoc-discuss" group. 
>>>> > To unsubscribe from this group and stop receiving emails from it, 
>>>> send an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>>>> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org 
>>>> > To view this discussion on the web visit 
>>>> https://groups.google.com/d/msgid/pandoc-discuss/a88bfdf5-f8a2-4cde-b4d1-a7bcbfc54810%40googlegroups.com. 
>>>>
>>>> > For more options, visit https://groups.google.com/groups/opt_out. 
>>>>
>>>>
>>>>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/05de5fac-df12-454a-a771-98a96278ecbe%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 6229 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Matthias Hüning]

[-- Attachment #1: Type: text/plain, Size: 837 bytes --]



Am Montag, 28. Oktober 2013 16:02:36 UTC+1 schrieb Nick Bart:
>
> It means rendering a DOI as a clickable link, in markdown output:
>
> doi:[10.1063/1.2172593](http://dx.doi.org/10.1063/1.2172593"10.1063/1.2172593")
>
>
Yes, that's perfect.

Best, Matthias

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/7b3d0e8e-fdf0-4e7c-b4c2-0d2e93543352%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1340 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1530 bytes --]

Another small issue: Leading zeroes are removed from days, as expected, 
when using an inline bibliography (first example) but they are *not* 
removed when using an external .bib file (second example):

$ cat date-leading-zeroes-month-day.md 
@item1

---
references:
- title: Title
  id: item1
  issued:
    day: 02
    month: 01
    year: 1999
  author:
    family: Author
  type: article-journal
...

$ pandoc -S --filter pandoc-citeproc -t markdown-citations 
date-leading-zeroes-month-day.md
Author (1999)

<div class="references">

Author. 1999. “Title” (January 2).

</div>


$ cat date-leading-zeroes-month-day.bib 

@article{item1,
    Author = {Author},
    Date = {1999-01-02},
    Title = {Title},
}

$ echo [@item1] | pandoc -S -t markdown-citations --bibliography 
date-leading-zeroes-month-day.bib
(Author 1999)

<div class="references">

Author. 1999. “Title” (January 02).

</div>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/e4205d49-e559-428c-99e1-a090bfd9b00e%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2017 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

This should now be fixed.

+++ Nick Bart [Oct 28 13 07:59 ]:
>    Another small issue: Leading zeroes are removed from days, as expected,
>    when using an inline bibliography (first example) but they are *not*
>    removed when using an external .bib file (second example):
>    $ cat date-leading-zeroes-month-day.md
>    @item1
>    ---
>    references:
>    - title: Title
>      id: item1
>      issued:
>        day: 02
>        month: 01
>        year: 1999
>      author:
>        family: Author
>      type: article-journal
>    ...
>    $ pandoc -S --filter pandoc-citeproc -t markdown-citations
>    date-leading-zeroes-month-day.md
>    Author (1999)
>    <div class="references">
>    Author. 1999. Title (January 2).
>    </div>
>    $ cat date-leading-zeroes-month-day.bib
>    @article{item1,
>        Author = {Author},
>        Date = {1999-01-02},
>        Title = {Title},
>    }
>    $ echo [@item1] | pandoc -S -t markdown-citations --bibliography
>    date-leading-zeroes-month-day.bib
>    (Author 1999)
>    <div class="references">
>    Author. 1999. Title (January 02).
>    </div>
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/e4205d49-e559-428c-
>    99e1-a090bfd9b00e%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/e4205d49-e559-428c-99e1-a090bfd9b00e%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've done this.  Now only one test fails:  the one with "series."
Handling this the way you expect would require quite a bit of
localization.  The word "series" can be handled in the present
framework, as we already resolve bibstrings (though so far only
English localization is present).  Ordinalization of the number is
trickier.  citeproc does it, so the information must be in the
citeproc locale files.  Really I need to think about how to do
localization more uniformly for both citeproc and the bibtex conversion.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2775 bytes --]

Right, I know this is tricky. Earlier on, I thought that for article etc. 
the series field would ideally be mapped to some CSL variable (either by 
introducing a new one or by hijacking an existing one not used for article 
etc. yet, such as collection-number, edition, version, …). However, since I 
realized that CSL simply does not support testing for variable content, it 
has become clear that essentially any kind of localization of variable 
content has to take place in biblio2yaml.

I’d expect that CSL eventually will not be able to get around introducing 
some kind of testing for variable content. In particular, I don't quite see 
how CSL can properly localize certain of the already existing variables, 
e.g., "status", without doing so. The terms "in press" and "forthcoming" 
are defined for various languages, but I fail to see how these can be used 
without testing for the content of the status variable, e.g., IF language = 
fr-FR AND content of status = "in press" THEN print "sous presse".

Still, that's the situation for now. However, there's one idea to shift 
some of the burden away from biblio2yaml: If we could convince the CSL 
community to introduce or repurpose some variable for journal series, we 
could let biblio2yaml localize "oldseries" and "newseries" as it does 
already, and pass on everything else as is to this new variable, letting 
CSL handle numerical values in a similar way as it currently handles 
edition. Do you think this would be easier than letting biblio2yaml do all 
of the work?

On Monday, October 28, 2013 4:03:27 PM UTC, fiddlosopher wrote:
>
> I've done this.  Now only one test fails:  the one with "series." 
> Handling this the way you expect would require quite a bit of 
> localization.  The word "series" can be handled in the present 
> framework, as we already resolve bibstrings (though so far only 
> English localization is present).  Ordinalization of the number is 
> trickier.  citeproc does it, so the information must be in the 
> citeproc locale files.  Really I need to think about how to do 
> localization more uniformly for both citeproc and the bibtex conversion. 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/5c690247-2054-4198-87b3-92521dd489bb%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3165 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 28 13 12:00 ]:
> Right, I know this is tricky. Earlier on, I thought that for article etc. 
> the series field would ideally be mapped to some CSL variable (either by 
> introducing a new one or by hijacking an existing one not used for article 
> etc. yet, such as collection-number, edition, version, …). However, since I 
> realized that CSL simply does not support testing for variable content, it 
> has become clear that essentially any kind of localization of variable 
> content has to take place in biblio2yaml.
> 
> I’d expect that CSL eventually will not be able to get around introducing 
> some kind of testing for variable content. In particular, I don't quite see 
> how CSL can properly localize certain of the already existing variables, 
> e.g., "status", without doing so. The terms "in press" and "forthcoming" 
> are defined for various languages, but I fail to see how these can be used 
> without testing for the content of the status variable, e.g., IF language = 
> fr-FR AND content of status = "in press" THEN print "sous presse".
> 
> Still, that's the situation for now. However, there's one idea to shift 
> some of the burden away from biblio2yaml: If we could convince the CSL 
> community to introduce or repurpose some variable for journal series, we 
> could let biblio2yaml localize "oldseries" and "newseries" as it does 
> already, and pass on everything else as is to this new variable, letting 
> CSL handle numerical values in a similar way as it currently handles 
> edition. Do you think this would be easier than letting biblio2yaml do all 
> of the work?

It seems a better approach to me.  I'd prefer not to duplicate
citeproc's existing ordinalization functionality, for example.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131029043626.GB23351%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 28 13 12:00 ]:
> Right, I know this is tricky. Earlier on, I thought that for article etc. 
> the series field would ideally be mapped to some CSL variable (either by 
> introducing a new one or by hijacking an existing one not used for article 
> etc. yet, such as collection-number, edition, version, …). However, since I 
> realized that CSL simply does not support testing for variable content, it 
> has become clear that essentially any kind of localization of variable 
> content has to take place in biblio2yaml.
> 
> I’d expect that CSL eventually will not be able to get around introducing 
> some kind of testing for variable content. In particular, I don't quite see 
> how CSL can properly localize certain of the already existing variables, 
> e.g., "status", without doing so. The terms "in press" and "forthcoming" 
> are defined for various languages, but I fail to see how these can be used 
> without testing for the content of the status variable, e.g., IF language = 
> fr-FR AND content of status = "in press" THEN print "sous presse".
> 
> Still, that's the situation for now. However, there's one idea to shift 
> some of the burden away from biblio2yaml: If we could convince the CSL 
> community to introduce or repurpose some variable for journal series, we 
> could let biblio2yaml localize "oldseries" and "newseries" as it does 
> already, and pass on everything else as is to this new variable, letting 
> CSL handle numerical values in a similar way as it currently handles 
> edition. Do you think this would be easier than letting biblio2yaml do all 
> of the work?

While you investigate this, I think I'll disable the current behavior
of adding 'series' to the title, and adjust that test case.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131029043910.GC23351%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 28 13 12:00 ]:
> Right, I know this is tricky. Earlier on, I thought that for article etc. 
> the series field would ideally be mapped to some CSL variable (either by 
> introducing a new one or by hijacking an existing one not used for article 
> etc. yet, such as collection-number, edition, version, …). However, since I 
> realized that CSL simply does not support testing for variable content, it 
> has become clear that essentially any kind of localization of variable 
> content has to take place in biblio2yaml.
> 
> I’d expect that CSL eventually will not be able to get around introducing 
> some kind of testing for variable content. In particular, I don't quite see 
> how CSL can properly localize certain of the already existing variables, 
> e.g., "status", without doing so. The terms "in press" and "forthcoming" 
> are defined for various languages, but I fail to see how these can be used 
> without testing for the content of the status variable, e.g., IF language = 
> fr-FR AND content of status = "in press" THEN print "sous presse".
> 
> Still, that's the situation for now. However, there's one idea to shift 
> some of the burden away from biblio2yaml: If we could convince the CSL 
> community to introduce or repurpose some variable for journal series, we 
> could let biblio2yaml localize "oldseries" and "newseries" as it does 
> already, and pass on everything else as is to this new variable, letting 
> CSL handle numerical values in a similar way as it currently handles 
> edition. Do you think this would be easier than letting biblio2yaml do all 
> of the work?

Hm, now I'm not so sure.  I see that many of your examples have a purely
textual 'series' which is well treated the way we do it now.

Still, it would be worth consulting with citeproc people on how best
to handle this issue.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131029044704.GD23351%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2931 bytes --]

Just to avoid misunderstandings: The issue at hand is only about the 
biblatex "series" field when used in article, periodical, or suppperiodical 
entries. For all other entry types, there's nothing wrong with the handling 
of the series field, and thus nothing should be changed there.

On Tuesday, 29 October 2013 04:47:04 UTC, fiddlosopher wrote:
>
> +++ Nick Bart [Oct 28 13 12:00 ]: 
> > Right, I know this is tricky. Earlier on, I thought that for article 
> etc. 
> > the series field would ideally be mapped to some CSL variable (either by 
> > introducing a new one or by hijacking an existing one not used for 
> article 
> > etc. yet, such as collection-number, edition, version, …). However, 
> since I 
> > realized that CSL simply does not support testing for variable content, 
> it 
> > has become clear that essentially any kind of localization of variable 
> > content has to take place in biblio2yaml. 
> > 
> > I’d expect that CSL eventually will not be able to get around 
> introducing 
> > some kind of testing for variable content. In particular, I don't quite 
> see 
> > how CSL can properly localize certain of the already existing variables, 
> > e.g., "status", without doing so. The terms "in press" and "forthcoming" 
> > are defined for various languages, but I fail to see how these can be 
> used 
> > without testing for the content of the status variable, e.g., IF 
> language = 
> > fr-FR AND content of status = "in press" THEN print "sous presse". 
> > 
> > Still, that's the situation for now. However, there's one idea to shift 
> > some of the burden away from biblio2yaml: If we could convince the CSL 
> > community to introduce or repurpose some variable for journal series, we 
> > could let biblio2yaml localize "oldseries" and "newseries" as it does 
> > already, and pass on everything else as is to this new variable, letting 
> > CSL handle numerical values in a similar way as it currently handles 
> > edition. Do you think this would be easier than letting biblio2yaml do 
> all 
> > of the work? 
>
> Hm, now I'm not so sure.  I see that many of your examples have a purely 
> textual 'series' which is well treated the way we do it now. 
>
> Still, it would be worth consulting with citeproc people on how best 
> to handle this issue. 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/25f97fa9-4ec7-4414-a13a-1d69e84e5fde%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3405 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1232 bytes --]

I noticed that there are style files that include instructions to convert 
journal titles ("container-title") to title-case, too. biblio2yaml should 
therefore protect lowercase or camelCase strings wrapped in curly braces in 
the biblatex fields journal, journaltitle, journalsubtitle (and title, 
subtitle fields of periodical entries) by enclosing them in <span 
class="nocase"></span>, too.

It would also be nice if biblio2yaml could map the latex command 
\textnormal{} to <span style="font-style:normal;"></span>: This would 
permit correct formatting of expressions such as "new ser." if one chooses 
to attach these manually to a journal title.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/663baeef-dcc3-4fc4-af80-f36e0a8a509c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1519 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I've made these changes.

+++ Nick Bart [Oct 29 13 07:41 ]:
> I noticed that there are style files that include instructions to convert 
> journal titles ("container-title") to title-case, too. biblio2yaml should 
> therefore protect lowercase or camelCase strings wrapped in curly braces in 
> the biblatex fields journal, journaltitle, journalsubtitle (and title, 
> subtitle fields of periodical entries) by enclosing them in <span 
> class="nocase"></span>, too.
> 
> It would also be nice if biblio2yaml could map the latex command 
> \textnormal{} to <span style="font-style:normal;"></span>: This would 
> permit correct formatting of expressions such as "new ser." if one chooses 
> to attach these manually to a journal title.
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/663baeef-dcc3-4fc4-af80-f36e0a8a509c%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4154 bytes --]

Great. The output from biblio2yaml looks good, except that I think there 
should be only one backslash inside verbatim spans such as 
`{\textnormal{}}` (see example below).

The next step would be to let citeproc format the <span></span> elements 
appropriately, and to fix the protection of braced lowercase strings: 
Currently, neither \textnormal{{new. ser.}} nor {\textnormal{new. ser.}} 
works as I would expect it (for both, see example below).

$ cat textnormal-textsc.bib
@article{item1,
    Author = {Author, Ann},
    Date = {2007},
    volume = {77},
    pages = {131-139},
    Journal = {Journal, {\textnormal{new ser.}},},
    Title = {Title -- series info inside \verb+{\textnormal{}}+}}
}

@article{item2,
    Author = {Author, Ann},
    Date = {2007},
    volume = {77},
    pages = {131-139},
    Journal = {Journal, \textsc{{new ser.},}},
    Title = {Title -- series info inside \verb+\textsc{{}}+}}
}


$ biblio2yaml textnormal-textsc.bib
---
references:
- page: 131-139
  title: "Title – series info inside `{\\textnormal{}}`"
  volume: 77
  id: item1
  issued:
    year: 2007
  author:
    given:
    - Ann
    family: Author
  container-title: Journal, <span style="font-style:normal;">new 
ser.</span>,
  type: article-journal
- page: 131-139
  title: "Title – series info inside `\\textsc{{}}`"
  volume: 77
  id: item2
  issued:
    year: 2007
  author:
    given:
    - Ann
    family: Author
  container-title: Journal, <span style="font-variant:small-caps;">new 
ser.,</span>
  type: article-journal
...


$ echo [@item1] [@item2] | pandoc -S -t markdown-citations --bibliography 
textnormal-textsc.bib
(Author 2007a) (Author 2007b)

<div class="references">

Author, Ann. 2007a. “Title – Series Info Inside \`{\\textnormal{}}\`.”
*Journal, \<span Style="font-style:normal;"\>new Ser.\</span\>,* 77:
131–139.

———. 2007b. “Title – Series Info Inside \`\\textsc{{}}\`.” *Journal,
\<span Style="font-variant:small-caps;"\>new Ser.,\</span\>* 77:
131–139.

</div>


On Wednesday, 30 October 2013 06:42:01 UTC, fiddlosopher wrote:
>
> I've made these changes. 
>
> +++ Nick Bart [Oct 29 13 07:41 ]: 
> > I noticed that there are style files that include instructions to 
> convert 
> > journal titles ("container-title") to title-case, too. biblio2yaml 
> should 
> > therefore protect lowercase or camelCase strings wrapped in curly braces 
> in 
> > the biblatex fields journal, journaltitle, journalsubtitle (and title, 
> > subtitle fields of periodical entries) by enclosing them in <span 
> > class="nocase"></span>, too. 
> > 
> > It would also be nice if biblio2yaml could map the latex command 
> > \textnormal{} to <span style="font-style:normal;"></span>: This would 
> > permit correct formatting of expressions such as "new ser." if one 
> chooses 
> > to attach these manually to a journal title. 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "pandoc-discuss" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/663baeef-dcc3-4fc4-af80-f36e0a8a509c%40googlegroups.com. 
>
> > For more options, visit https://groups.google.com/groups/opt_out. 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/55a6616f-948d-49eb-bbd5-5de5ef48344a%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5500 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 30 13 01:13 ]:
> Great. The output from biblio2yaml looks good, except that I think there 
> should be only one backslash inside verbatim spans such as 
> `{\textnormal{}}` (see example below).

That's proper YAML escaping for a single backslash.

> The next step would be to let citeproc format the <span></span> elements 
> appropriately, and to fix the protection of braced lowercase strings: 
> Currently, neither \textnormal{{new. ser.}} nor {\textnormal{new. ser.}} 
> works as I would expect it (for both, see example below).

citeproc doesn't know about this yet.  In fact, this aspect of things
is a terrible mess right now.  Text.CSL.Input.Bibtex is parsing
everything into a pandoc structure, then writing this to a string
(which goes in the Reference data structure); Text.CSL.Eval.Output
and Text.CSL.Output.Pandoc then parse this string back into a pandoc
structure.  Currently this isn't very reliable, and it's needlessly
complex.  Hence my desire to cut out the intermediate string
representation, and change everything so that the Reference
and Cite structures have Pandoc fields instead of strings. Unfortunately
this is a pretty big change and will take quite a lot of work.  Andrea
said he was going to work on it; I don't know if he has made progress.

John

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ John MacFarlane [Oct 30 13 09:34 ]:
> +++ Nick Bart [Oct 30 13 01:13 ]:
> > Great. The output from biblio2yaml looks good, except that I think there 
> > should be only one backslash inside verbatim spans such as 
> > `{\textnormal{}}` (see example below).
> 
> That's proper YAML escaping for a single backslash.
> 
> > The next step would be to let citeproc format the <span></span> elements 
> > appropriately, and to fix the protection of braced lowercase strings: 
> > Currently, neither \textnormal{{new. ser.}} nor {\textnormal{new. ser.}} 
> > works as I would expect it (for both, see example below).
> 
> citeproc doesn't know about this yet.  In fact, this aspect of things
> is a terrible mess right now.  Text.CSL.Input.Bibtex is parsing
> everything into a pandoc structure, then writing this to a string
> (which goes in the Reference data structure); Text.CSL.Eval.Output
> and Text.CSL.Output.Pandoc then parse this string back into a pandoc
> structure.  Currently this isn't very reliable, and it's needlessly
> complex.  Hence my desire to cut out the intermediate string
> representation, and change everything so that the Reference
> and Cite structures have Pandoc fields instead of strings. Unfortunately
> this is a pretty big change and will take quite a lot of work.  Andrea
> said he was going to work on it; I don't know if he has made progress.

I've done a bit of preliminary cleaning in Text.CSL.Output.Pandoc.

I think the project could be done in several steps:

1.  FormattedOutput already contains FPan to hold a Pandoc [Inline].
So the first step could be to change the routines that produce
FormattedOutput so that they only use FPan.

2.  The Plain output module would then use writePlain from pandoc.

3.  At this point, FormattedOutput could be removed and replaced with
pandoc [Inline] (or it could be made a synonym).

4.  Output could probably be simplified at this point.  E.g. I think
we could get rid of OUrl and just use OPan.

5.  Now the big change:  use [Inline] instead of String in Reference
and Affix.  This would require lots of other changes, but the compiler
would tell us where they were needed.  A lot of code could be
removed, here, because we wouldn't be reparsing strings for formatting.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3375 bytes --]

I see that a lot of work is being put into pandoc-citeproc, and most likely 
not everything is finished yet. Just wanted to mention that biblio2yaml 
from the latest version seems to choke on escaped chars and double 
underscores in URLs, and removes single underscores from URLs (try 
biblio2yaml pandoc-citeproc/tests/biblio2yaml/markey.biblatex).

On Wednesday, 30 October 2013 17:57:02 UTC, fiddlosopher wrote:
>
> +++ John MacFarlane [Oct 30 13 09:34 ]: 
> > +++ Nick Bart [Oct 30 13 01:13 ]: 
> > > Great. The output from biblio2yaml looks good, except that I think 
> there 
> > > should be only one backslash inside verbatim spans such as 
> > > `{\textnormal{}}` (see example below). 
> > 
> > That's proper YAML escaping for a single backslash. 
> > 
> > > The next step would be to let citeproc format the <span></span> 
> elements 
> > > appropriately, and to fix the protection of braced lowercase strings: 
> > > Currently, neither \textnormal{{new. ser.}} nor {\textnormal{new. 
> ser.}} 
> > > works as I would expect it (for both, see example below). 
> > 
> > citeproc doesn't know about this yet.  In fact, this aspect of things 
> > is a terrible mess right now.  Text.CSL.Input.Bibtex is parsing 
> > everything into a pandoc structure, then writing this to a string 
> > (which goes in the Reference data structure); Text.CSL.Eval.Output 
> > and Text.CSL.Output.Pandoc then parse this string back into a pandoc 
> > structure.  Currently this isn't very reliable, and it's needlessly 
> > complex.  Hence my desire to cut out the intermediate string 
> > representation, and change everything so that the Reference 
> > and Cite structures have Pandoc fields instead of strings. Unfortunately 
> > this is a pretty big change and will take quite a lot of work.  Andrea 
> > said he was going to work on it; I don't know if he has made progress. 
>
> I've done a bit of preliminary cleaning in Text.CSL.Output.Pandoc. 
>
> I think the project could be done in several steps: 
>
> 1.  FormattedOutput already contains FPan to hold a Pandoc [Inline]. 
> So the first step could be to change the routines that produce 
> FormattedOutput so that they only use FPan. 
>
> 2.  The Plain output module would then use writePlain from pandoc. 
>
> 3.  At this point, FormattedOutput could be removed and replaced with 
> pandoc [Inline] (or it could be made a synonym). 
>
> 4.  Output could probably be simplified at this point.  E.g. I think 
> we could get rid of OUrl and just use OPan. 
>
> 5.  Now the big change:  use [Inline] instead of String in Reference 
> and Affix.  This would require lots of other changes, but the compiler 
> would tell us where they were needed.  A lot of code could be 
> removed, here, because we wouldn't be reparsing strings for formatting. 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a0649979-54f8-4647-99ca-c98bec0d69cc%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4042 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Yes, I'm in the middle of a very large-scale change (making
pandoc-citeproc use [Inline] internally instead of String to store
most bibliographic fields, so that formatting can be reliably
preserved and we avoid an inefficient parsing step).

I'm aware that not everything is working properly right now, but
it should be fixed in time.

+++ Nick Bart [Nov 04 13 00:19 ]:
> I see that a lot of work is being put into pandoc-citeproc, and most likely 
> not everything is finished yet. Just wanted to mention that biblio2yaml 
> from the latest version seems to choke on escaped chars and double 
> underscores in URLs, and removes single underscores from URLs (try 
> biblio2yaml pandoc-citeproc/tests/biblio2yaml/markey.biblatex).
> 
> On Wednesday, 30 October 2013 17:57:02 UTC, fiddlosopher wrote:
> >
> > +++ John MacFarlane [Oct 30 13 09:34 ]: 
> > > +++ Nick Bart [Oct 30 13 01:13 ]: 
> > > > Great. The output from biblio2yaml looks good, except that I think 
> > there 
> > > > should be only one backslash inside verbatim spans such as 
> > > > `{\textnormal{}}` (see example below). 
> > > 
> > > That's proper YAML escaping for a single backslash. 
> > > 
> > > > The next step would be to let citeproc format the <span></span> 
> > elements 
> > > > appropriately, and to fix the protection of braced lowercase strings: 
> > > > Currently, neither \textnormal{{new. ser.}} nor {\textnormal{new. 
> > ser.}} 
> > > > works as I would expect it (for both, see example below). 
> > > 
> > > citeproc doesn't know about this yet.  In fact, this aspect of things 
> > > is a terrible mess right now.  Text.CSL.Input.Bibtex is parsing 
> > > everything into a pandoc structure, then writing this to a string 
> > > (which goes in the Reference data structure); Text.CSL.Eval.Output 
> > > and Text.CSL.Output.Pandoc then parse this string back into a pandoc 
> > > structure.  Currently this isn't very reliable, and it's needlessly 
> > > complex.  Hence my desire to cut out the intermediate string 
> > > representation, and change everything so that the Reference 
> > > and Cite structures have Pandoc fields instead of strings. Unfortunately 
> > > this is a pretty big change and will take quite a lot of work.  Andrea 
> > > said he was going to work on it; I don't know if he has made progress. 
> >
> > I've done a bit of preliminary cleaning in Text.CSL.Output.Pandoc. 
> >
> > I think the project could be done in several steps: 
> >
> > 1.  FormattedOutput already contains FPan to hold a Pandoc [Inline]. 
> > So the first step could be to change the routines that produce 
> > FormattedOutput so that they only use FPan. 
> >
> > 2.  The Plain output module would then use writePlain from pandoc. 
> >
> > 3.  At this point, FormattedOutput could be removed and replaced with 
> > pandoc [Inline] (or it could be made a synonym). 
> >
> > 4.  Output could probably be simplified at this point.  E.g. I think 
> > we could get rid of OUrl and just use OPan. 
> >
> > 5.  Now the big change:  use [Inline] instead of String in Reference 
> > and Affix.  This would require lots of other changes, but the compiler 
> > would tell us where they were needed.  A lot of code could be 
> > removed, here, because we wouldn't be reparsing strings for formatting. 
> >
> >
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a0649979-54f8-4647-99ca-c98bec0d69cc%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2430 bytes --]

Re: url=false: In online entries, this option should be ignored. See 
attached test.

$ cat options-url-false-doi-false.biblatex

@comment{

Entry contains url and doi fields; these should be discarded since the 
options field specifies url=false, doi=false.

Exception: As in standard biblatex, in online entries url should never be 
discarded, even if options contains url=false.

}

@article{item1,
    Author = {Author, Andy},
    Date = {2012},
    Doi = {1234/5678.90},
    Journal = {Journal},
    Options = {url=false, doi=false},
    Title = {Title, Any Entry Type Except online},
    Url = {http://foo.bar}
}

@online{item2,
    Author = {Author, Andy},
    Date = {2012},
    Doi = {1234/5678.90},
    Journal = {Journal},
    Options = {url=false, doi=false},
    Title = {Title, Entry Type online},
    Url = {http://foo.bar}
}

---
references:
- title: Title, any entry type except online
  id: item1
  issued:
    year: 2012
  author:
    given:
    - Andy
    family: Author
  container-title: Journal
  type: article-journal
- title: Title, entry type online
  id: item2
  issued:
    year: 2012
  author:
    given:
    - Andy
    family: Author
  container-title: Journal
  type: webpage
  url: http://foo.bar
...




On Monday, October 28, 2013 4:03:27 PM UTC, fiddlosopher wrote:
>
> I've done this.  Now only one test fails:  the one with "series." 
> Handling this the way you expect would require quite a bit of 
> localization.  The word "series" can be handled in the present 
> framework, as we already resolve bibstrings (though so far only 
> English localization is present).  Ordinalization of the number is 
> trickier.  citeproc does it, so the information must be in the 
> citeproc locale files.  Really I need to think about how to do 
> localization more uniformly for both citeproc and the bibtex conversion. 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/0b566c5f-d991-43f8-bc65-52c843bef62c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3427 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Why would somebody bother to put 'Options = {url=false}' in an
online entry, if they just wanted the option to be ignored?

+++ Nick Bart [Oct 28 13 13:53 ]:
>    Re: url=false: In online entries, this option should be ignored. See
>    attached test.
>    $ cat options-url-false-doi-false.biblatex
>    @comment{
>    Entry contains url and doi fields; these should be discarded since the
>    options field specifies url=false, doi=false.
>    Exception: As in standard biblatex, in online entries url should never
>    be discarded, even if options contains url=false.
>    }
>    @article{item1,
>        Author = {Author, Andy},
>        Date = {2012},
>        Doi = {1234/5678.90},
>        Journal = {Journal},
>        Options = {url=false, doi=false},
>        Title = {Title, Any Entry Type Except online},
>        Url = {http://foo.bar}
>    }
>    @online{item2,
>        Author = {Author, Andy},
>        Date = {2012},
>        Doi = {1234/5678.90},
>        Journal = {Journal},
>        Options = {url=false, doi=false},
>        Title = {Title, Entry Type online},
>        Url = {http://foo.bar}
>    }
>    ---
>    references:
>    - title: Title, any entry type except online
>      id: item1
>      issued:
>        year: 2012
>      author:
>        given:
>        - Andy
>        family: Author
>      container-title: Journal
>      type: article-journal
>    - title: Title, entry type online
>      id: item2
>      issued:
>        year: 2012
>      author:
>        given:
>        - Andy
>        family: Author
>      container-title: Journal
>      type: webpage
>      url: http://foo.bar
>    ...
>    On Monday, October 28, 2013 4:03:27 PM UTC, fiddlosopher wrote:
> 
>      I've done this.  Now only one test fails:  the one with "series."
>      Handling this the way you expect would require quite a bit of
>      localization.  The word "series" can be handled in the present
>      framework, as we already resolve bibstrings (though so far only
>      English localization is present).  Ordinalization of the number is
>      trickier.  citeproc does it, so the information must be in the
>      citeproc locale files.  Really I need to think about how to do
>      localization more uniformly for both citeproc and the bibtex
>      conversion.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/0b566c5f-d991-43f8-
>    bc65-52c843bef62c%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/0b566c5f-d991-43f8-bc65-52c843bef62c%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4291 bytes --]

Good point. It's what the biblatex & biblatex-chicago manuals say, useful 
when you set url=false as a per-document option (which is not available in 
pandoc), but as a safety net for single entries most probably overkill. 
Keep the test, if you wish, but throw out the online entry.

On Monday, October 28, 2013 9:16:35 PM UTC, fiddlosopher wrote:
>
> Why would somebody bother to put 'Options = {url=false}' in an 
> online entry, if they just wanted the option to be ignored? 
>
> +++ Nick Bart [Oct 28 13 13:53 ]: 
> >    Re: url=false: In online entries, this option should be ignored. See 
> >    attached test. 
> >    $ cat options-url-false-doi-false.biblatex 
> >    @comment{ 
> >    Entry contains url and doi fields; these should be discarded since 
> the 
> >    options field specifies url=false, doi=false. 
> >    Exception: As in standard biblatex, in online entries url should 
> never 
> >    be discarded, even if options contains url=false. 
> >    } 
> >    @article{item1, 
> >        Author = {Author, Andy}, 
> >        Date = {2012}, 
> >        Doi = {1234/5678.90}, 
> >        Journal = {Journal}, 
> >        Options = {url=false, doi=false}, 
> >        Title = {Title, Any Entry Type Except online}, 
> >        Url = {http://foo.bar} 
> >    } 
> >    @online{item2, 
> >        Author = {Author, Andy}, 
> >        Date = {2012}, 
> >        Doi = {1234/5678.90}, 
> >        Journal = {Journal}, 
> >        Options = {url=false, doi=false}, 
> >        Title = {Title, Entry Type online}, 
> >        Url = {http://foo.bar} 
> >    } 
> >    --- 
> >    references: 
> >    - title: Title, any entry type except online 
> >      id: item1 
> >      issued: 
> >        year: 2012 
> >      author: 
> >        given: 
> >        - Andy 
> >        family: Author 
> >      container-title: Journal 
> >      type: article-journal 
> >    - title: Title, entry type online 
> >      id: item2 
> >      issued: 
> >        year: 2012 
> >      author: 
> >        given: 
> >        - Andy 
> >        family: Author 
> >      container-title: Journal 
> >      type: webpage 
> >      url: http://foo.bar 
> >    ... 
> >    On Monday, October 28, 2013 4:03:27 PM UTC, fiddlosopher wrote: 
> > 
> >      I've done this.  Now only one test fails:  the one with "series." 
> >      Handling this the way you expect would require quite a bit of 
> >      localization.  The word "series" can be handled in the present 
> >      framework, as we already resolve bibstrings (though so far only 
> >      English localization is present).  Ordinalization of the number is 
> >      trickier.  citeproc does it, so the information must be in the 
> >      citeproc locale files.  Really I need to think about how to do 
> >      localization more uniformly for both citeproc and the bibtex 
> >      conversion. 
> > 
> >    -- 
> >    You received this message because you are subscribed to the Google 
> >    Groups "pandoc-discuss" group. 
> >    To unsubscribe from this group and stop receiving emails from it, 
> send 
> >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> >    To view this discussion on the web visit 
> >    [1]
> https://groups.google.com/d/msgid/pandoc-discuss/0b566c5f-d991-43f8- 
> >    bc65-52c843bef62c%40googlegroups.com. 
> >    For more options, visit [2]https://groups.google.com/groups/opt_out. 
> > 
> > References 
> > 
> >    1. 
> https://groups.google.com/d/msgid/pandoc-discuss/0b566c5f-d991-43f8-bc65-52c843bef62c%40googlegroups.com 
> >    2. https://groups.google.com/groups/opt_out 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/572bb2df-907e-4bb0-ac0e-31fd0a038f5f%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 6857 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Well, if that's what the manual says, it's not a big deal for me
to add the extra check.

+++ Nick Bart [Oct 28 13 14:34 ]:
> Good point. It's what the biblatex & biblatex-chicago manuals say, useful 
> when you set url=false as a per-document option (which is not available in 
> pandoc), but as a safety net for single entries most probably overkill. 
> Keep the test, if you wish, but throw out the online entry.
> 
> On Monday, October 28, 2013 9:16:35 PM UTC, fiddlosopher wrote:
> >
> > Why would somebody bother to put 'Options = {url=false}' in an 
> > online entry, if they just wanted the option to be ignored? 
> >
> > +++ Nick Bart [Oct 28 13 13:53 ]: 
> > >    Re: url=false: In online entries, this option should be ignored. See 
> > >    attached test. 
> > >    $ cat options-url-false-doi-false.biblatex 
> > >    @comment{ 
> > >    Entry contains url and doi fields; these should be discarded since 
> > the 
> > >    options field specifies url=false, doi=false. 
> > >    Exception: As in standard biblatex, in online entries url should 
> > never 
> > >    be discarded, even if options contains url=false. 
> > >    } 
> > >    @article{item1, 
> > >        Author = {Author, Andy}, 
> > >        Date = {2012}, 
> > >        Doi = {1234/5678.90}, 
> > >        Journal = {Journal}, 
> > >        Options = {url=false, doi=false}, 
> > >        Title = {Title, Any Entry Type Except online}, 
> > >        Url = {http://foo.bar} 
> > >    } 
> > >    @online{item2, 
> > >        Author = {Author, Andy}, 
> > >        Date = {2012}, 
> > >        Doi = {1234/5678.90}, 
> > >        Journal = {Journal}, 
> > >        Options = {url=false, doi=false}, 
> > >        Title = {Title, Entry Type online}, 
> > >        Url = {http://foo.bar} 
> > >    } 
> > >    --- 
> > >    references: 
> > >    - title: Title, any entry type except online 
> > >      id: item1 
> > >      issued: 
> > >        year: 2012 
> > >      author: 
> > >        given: 
> > >        - Andy 
> > >        family: Author 
> > >      container-title: Journal 
> > >      type: article-journal 
> > >    - title: Title, entry type online 
> > >      id: item2 
> > >      issued: 
> > >        year: 2012 
> > >      author: 
> > >        given: 
> > >        - Andy 
> > >        family: Author 
> > >      container-title: Journal 
> > >      type: webpage 
> > >      url: http://foo.bar 
> > >    ... 
> > >    On Monday, October 28, 2013 4:03:27 PM UTC, fiddlosopher wrote: 
> > > 
> > >      I've done this.  Now only one test fails:  the one with "series." 
> > >      Handling this the way you expect would require quite a bit of 
> > >      localization.  The word "series" can be handled in the present 
> > >      framework, as we already resolve bibstrings (though so far only 
> > >      English localization is present).  Ordinalization of the number is 
> > >      trickier.  citeproc does it, so the information must be in the 
> > >      citeproc locale files.  Really I need to think about how to do 
> > >      localization more uniformly for both citeproc and the bibtex 
> > >      conversion. 
> > > 
> > >    -- 
> > >    You received this message because you are subscribed to the Google 
> > >    Groups "pandoc-discuss" group. 
> > >    To unsubscribe from this group and stop receiving emails from it, 
> > send 
> > >    an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > >    To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
> >
> > >    To view this discussion on the web visit 
> > >    [1]
> > https://groups.google.com/d/msgid/pandoc-discuss/0b566c5f-d991-43f8- 
> > >    bc65-52c843bef62c%40googlegroups.com. 
> > >    For more options, visit [2]https://groups.google.com/groups/opt_out. 
> > > 
> > > References 
> > > 
> > >    1. 
> > https://groups.google.com/d/msgid/pandoc-discuss/0b566c5f-d991-43f8-bc65-52c843bef62c%40googlegroups.com 
> > >    2. https://groups.google.com/groups/opt_out 
> >
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/572bb2df-907e-4bb0-ac0e-31fd0a038f5f%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 25 13 11:39 ]:
> Great. I wouldn't map forthcoming to *both* status and issued, though; just 
> to status is perfectly ok. (When "forthcoming" is in the biblatex pubstate 
> field, we don't map it to both status and issued either; that fact that can 
> appear in a year field at all is just an idiosyncrasy of biblatex-chicago, 
> probably from before biblatex had a pubstate field.)
> 
> Not too many csl styles can handle "status" yet, but try 
> antarctic-science.csl, or my chicago-author-date-TEST-20131018.csl.
> 
> One of our next steps will have to be to encourage CSL style authors to 
> cater for those valid CSL variables that have not been implemented in most 
> style files so far (usually because no frontend was actually using them) 
> but that can be easily accessed now via pandoc-citeproc / biblio2yaml.

Does it do any harm to include "forthcoming" in "issued/other" as well?

Wouldn't it be a good idea to do so, as long as many CSL styles
don't pay attention to the "status" field?

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3330 bytes --]

"forthcoming" in both "status" and "issued/other": As a stopgap, maybe; as 
a more permanent solution, no.

- Formatting requirements for date vs. publication status are different: 
Chicago author-date, e.g., wants "(Doe 2007)" but "(Doe, forthcoming)". In 
these cases, this proposal would break styles that try CSL "issued" 
*before* CSL "status".

- This proposal would also break styles that use both date and publication 
status. “Cite it right”, University of Limerick, 
<http://www.ul.ie/~library/pdf/citeitright.pdf>, e.g., requires:

    > Stetter, S. and Nathanson, R., eds. (2007, forthcoming) A Region 
under Stress: EU-Israeli Relations and Wider Middle East Politics (Working 
Title), Tel Aviv and Berlin: Fredrich-Ebert-Stiftung.

    > … (Stetter and Nathanson 2007, forthcoming) …

Something else I just realized to its full extent is that in CSL testing 
for variable *content* is not supported. Thus, proper localization of 
"forthcoming", "in press", etc., is impossible in CSL: Though terms such as 
"forthcoming" are defined in the CSL locale files, they cannot be accessed 
in any useful way: E.g., we could test for the existence of "status", but 
most likely we wouldn't want to print "forthcoming" for *every* item where 
status is non-empty.

This means that whenever we think about localization of variable content, 
we have to keep in mind that most of it will have to be done by 
biblio2yaml. (Unless, that is, the CSL community could be convinced to 
extend CSL to allow testing for variable content …)


On Saturday, 26 October 2013 03:14:41 UTC, fiddlosopher wrote:
>
> +++ Nick Bart [Oct 25 13 11:39 ]: 
> > Great. I wouldn't map forthcoming to *both* status and issued, though; 
> just 
> > to status is perfectly ok. (When "forthcoming" is in the biblatex 
> pubstate 
> > field, we don't map it to both status and issued either; that fact that 
> can 
> > appear in a year field at all is just an idiosyncrasy of 
> biblatex-chicago, 
> > probably from before biblatex had a pubstate field.) 
> > 
> > Not too many csl styles can handle "status" yet, but try 
> > antarctic-science.csl, or my chicago-author-date-TEST-20131018.csl. 
> > 
> > One of our next steps will have to be to encourage CSL style authors to 
> > cater for those valid CSL variables that have not been implemented in 
> most 
> > style files so far (usually because no frontend was actually using them) 
> > but that can be easily accessed now via pandoc-citeproc / biblio2yaml. 
>
> Does it do any harm to include "forthcoming" in "issued/other" as well? 
>
> Wouldn't it be a good idea to do so, as long as many CSL styles 
> don't pay attention to the "status" field? 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/9dc5392b-5152-46cb-b76c-0ddd89dfdd9a%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3804 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1354 bytes --]

Another small case of excessive punctuation: '?”.' instead of '?”' at the 
end of (foot)notes, a case apparently not included in the citeproc test 
suite. Example:

$ cat punctuation-after-note-with-question-mark.md 
Foo.[@item1]

---
csl: chicago-note-bibliography.csl
references:
- title: Title ending with a question mark?
  id: item1
  issued:
    year: 1999
  author:
    given:
    - Al
    family: Author
  container-title: Journal
  type: article-journal
...
$  pandoc -S --filter pandoc-citeproc -t markdown-citations 
punctuation-after-note-with-question-mark.md
Foo.[^1]

<div class="references">

Author, Al. “Title Ending with a Question Mark?” *Journal* (1999).

</div>

[^1]: Author, “Title Ending with a Question Mark?”.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/0b13117f-caee-48d4-8323-50ded709c103%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1766 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3999 bytes --]

A suggestion (wrapped in a biblio2yaml test file) for unpublished 
conference contributions:

$ cat unpublished-speech.biblatex
@comment{

Unpublished papers, posters, presentations, speeches, and other 
contributions given at conferences or meetings do not have a dedicated 
entry type in biblatex.

CSL has the "speech" type, which is used specifically for such unpublished 
conference contributions, as seen by the fact that all "speech" entries 
containing a nonempty "event" variable are formatted using the string 
"presented at the [event]" (or some localization of it).

In biblatex, one could imagine using the “inproceedings” entry type, 
leaving “publisher” and “location” empty but adding type (e.g., = 
{Poster}), eventtitle, venue, and maybe eventdate (though date alone will 
usually be sufficient here), and letting processors infer from this 
configuration that it’s an *unpublished* conference contribution. One could 
also use the “unpublished” entry type and infer that it's a conference 
contribution from the existence of the fields eventtitle, venue, and maybe 
eventdate. In both cases one could facilitate this by adding an appropriate 
entrysubtype field.

This, however, is *not* how this is usually done in biblatex (and bibtex).

Typically, an "unpublished" entry is used in the following way (example 
from biblatex-apa-test-references.bib):

% (APA 7.04 Example 37)
% Unpublished conference papers are UNPUBLISHED type
[at]UNPUBLISHED{7.04:37,
    AUTHOR = {S. Liu},
    TITLE = {Defending Against Business Crises with the Help of Intelligent 
Agent Based Early Warning Solutions},
    DATE = {2005-05},
    LOCATION = {Miami, FL},
    NOTE = {Paper presented at the Seventh International Conference on 
Enterprise Information Systems},
    ABSTRACTURL = {http://www.iceis.org/iceis2005/abstracts_2005.htm}
}

My suggestion – that would not require any changes on the biblatex side but 
would still be fairly reliable and render correct output in CSL – is to try 
to parse this common biblatex format, i.e., search for the string “XXX 
presented at the YYY” in “note” and “howpublished” fields of “unpublished” 
entries (XXX might be empty), and if this is found, set the CSL type to 
“speech”, map XXX to CSL “genre”, and YYY to CSL “event”, and map 
“location” to CSL “event-place”. 

For French, Spanish, etc., one could search for "présenté à", "presentado 
en", etc. (see the definitions of the term "presented at" in the various 
CSL locale files).


pandoc using the yaml entry below yields the expected output:

Author (2012)

Author, Ann. 2012. “Title of the Paper, Poster, Presentation, or
Speech.” Poster presented at the Conference of Something, December 30,
Paris, France. <http://foo.bar>.

}

@unpublished{item1,
    author = {Author, Ann},
    title = {Title of the Paper, Poster, Presentation, or Speech},
    howpublished = {Poster presented at the Conference of Something},
    location = {Paris, France},
    date = {2012-12-30},
    url = {http://foo.bar},
}

---
references:
- title: Title of the paper, poster, presentation, or speech
  id: item1
  issued:
    day: 30
    month: 12
    year: 2012
  author:
    given:
    - Ann
    family: Author
  genre: Poster
  event: Conference of Something
  event-place: Paris, France
  url: http://foo.bar
  type: speech
...

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/94074c52-b2ff-473b-86a4-ba1ca49128dc%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4659 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Nick Bart [Oct 19 13 04:45 ]:
> Brilliant!
> 
> And your corrections of two mistakes in the test output are spot on.
> 
> On title vs. booktitle, however:
> 
> I'm not quite sure whether I understand the Bibtex.hs code and your 
> comments in the commits correctly, but when biblio2yaml is dealing with 
> title/booktitle/maintitle, it should never be *comparing* strings (though 
> citeproc should compare author and bookauthor, see below).

That was a hack.  I think I've implemented the correct logic, now.
One test fails:

[biblio2yaml/incollection-2.biblatex] FAILED (-expected, +actual)
    1 {
    2   "references": [
    3     {
    4       "annote": "A cross-referenced article from a collection. This is an incollection entry with a crossref field. Note the subtitle and indextitle fields",
    5       "author": {
    6         "family": "Westfahl",
    7         "given": [
    8           "Gary"
    9         ]
   10       },
   11       "container-title": "Space and beyond: The frontier theme in science fiction",
   12       "editor": {
   13         "family": "Westfahl",
   14         "given": [
   15           "Gary"
   16         ]
   17       },
   18       "id": "westfahl:space",
   19       "issued": {
   20         "year": 2000
   21       },
   22       "language": "en-US",
   23       "page": "55-65",
   24       "publisher": "Greenwood",
   25       "publisher-place": "Westport, Conn.; London",
   26       "title": "The true frontier: Confronting and avoiding the realities of space in American science fiction films",
-  27       "type": "chapter"
+  27       "type": "chapter",
+  28       "volume-title": "Space and beyond: The frontier theme in science fiction"
   29     },

The only difference is the presence of "volume-title".  You say:

> The booktitle field has to be used, by contrast, for inbook, incollection, 
> inproceedings, inreference entries without crossref, where it is mapped to 
> container-title if no maintitle exists; else maintitle is mapped to 
> container-title, and booktitle to volume-title.

Here we have an entry with both 'maintitle' (remapped from 'title' in
crossref) and 'booktitle', so the program has dutifully mapped one to
container-title and the other to volume-title.  (In this case they are
the same.)

Where have I gone wrong?

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 3901 bytes --]

The problem is in the remapping (biblatex: "inheritance") rules.

Bibtex.hs seems to remap all  book, collection, proceedings, reference, 
mvbook, mvcollection, mvproceedings, mvreference titles to maintitles

       "title"          -> ["maintitle"]
       "subtitle"       -> ["mainsubtitle"]
       "titleaddon"     -> ["maintitleaddon"]

This is only correct for mvbook, mvcollection, mvproceedings, mvreference.

For book, collection, proceedings, reference, we need:

       "title"          -> ["booktitle"]
       "subtitle"       -> ["booksubtitle"]
       "titleaddon"     -> ["booktitleaddon"]

See biblatex manual, v2.7a, Appendix B, p. 240–242

On Saturday, October 19, 2013 10:31:12 PM UTC, fiddlosopher wrote:
>
> +++ Nick Bart [Oct 19 13 04:45 ]: 
> > Brilliant! 
> > 
> > And your corrections of two mistakes in the test output are spot on. 
> > 
> > On title vs. booktitle, however: 
> > 
> > I'm not quite sure whether I understand the Bibtex.hs code and your 
> > comments in the commits correctly, but when biblio2yaml is dealing with 
> > title/booktitle/maintitle, it should never be *comparing* strings 
> (though 
> > citeproc should compare author and bookauthor, see below). 
>
> That was a hack.  I think I've implemented the correct logic, now. 
> One test fails: 
>
> [biblio2yaml/incollection-2.biblatex] FAILED (-expected, +actual) 
>     1 { 
>     2   "references": [ 
>     3     { 
>     4       "annote": "A cross-referenced article from a collection. This 
> is an incollection entry with a crossref field. Note the subtitle and 
> indextitle fields", 
>     5       "author": { 
>     6         "family": "Westfahl", 
>     7         "given": [ 
>     8           "Gary" 
>     9         ] 
>    10       }, 
>    11       "container-title": "Space and beyond: The frontier theme in 
> science fiction", 
>    12       "editor": { 
>    13         "family": "Westfahl", 
>    14         "given": [ 
>    15           "Gary" 
>    16         ] 
>    17       }, 
>    18       "id": "westfahl:space", 
>    19       "issued": { 
>    20         "year": 2000 
>    21       }, 
>    22       "language": "en-US", 
>    23       "page": "55-65", 
>    24       "publisher": "Greenwood", 
>    25       "publisher-place": "Westport, Conn.; London", 
>    26       "title": "The true frontier: Confronting and avoiding the 
> realities of space in American science fiction films", 
> -  27       "type": "chapter" 
> +  27       "type": "chapter", 
> +  28       "volume-title": "Space and beyond: The frontier theme in 
> science fiction" 
>    29     }, 
>
> The only difference is the presence of "volume-title".  You say: 
>
> > The booktitle field has to be used, by contrast, for inbook, 
> incollection, 
> > inproceedings, inreference entries without crossref, where it is mapped 
> to 
> > container-title if no maintitle exists; else maintitle is mapped to 
> > container-title, and booktitle to volume-title. 
>
> Here we have an entry with both 'maintitle' (remapped from 'title' in 
> crossref) and 'booktitle', so the program has dutifully mapped one to 
> container-title and the other to volume-title.  (In this case they are 
> the same.) 
>
> Where have I gone wrong? 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/40fcf73d-5a6d-45dd-9cad-a79ee246b9e2%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5572 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4063 bytes --]

Two more test files below.

Note that an author field of book and mvbook needs to be inherited to both 
author and bookauthor of the child, and that a child should inherit fields 
from its *immediate* ancestor (in the example, the inbook entry should 
inherit "year: 1975" from its parent, not "year: 1970 [to] year: 1980" from 
its grandparent).


$ cat crossref-inbook-mvbook.biblatex
@comment{

crossref, directly from inbook to mvbook

}

@inbook{inbook-1,
    Crossref = {mvbook-1},
    Title = {Macbeth [title field of inbook-1]},
    Date = {1975},
    Volume = {3},
    Chapter = {7},
    Pages = {100-200},
}

@mvbook{mvbook-1,
    Author = {Shakespeare},
    Date = {1970/1980},
    Title = {Collected Works [title field of mvbook-1]},
    Location = {Location},
    Publisher = {Publisher},
    Volumes = {4},
}

---
references:

- type: chapter
id: inbook-1
title: Macbeth [title field of inbook-1]
container-title: Collected works [title field of mvbook-1]
author:
family: Shakespeare
container-author:
family: Shakespeare
issued:
year: 1975
volume: 3
chapter-number: 7
page: 100-200
number-of-volumes: 4
publisher: Publisher
publisher-place: Location

- type: book
id: mvbook-1
title: Collected works [title field of mvbook-1]
author:
family: Shakespeare
issued:
- year: 1970
- year: 1980
number-of-volumes: 4
publisher: Publisher
publisher-place: Location
...

$ cat crossref-nested.biblatex 
@comment{

Nested crossreferences (see biber manual v 1.7)

}

@bookinbook{bookinbook-1,
    Crossref = {book-1},
    Title = {Macbeth [title field of bookinbook-1]},
    Chapter = {7},
    Pages = {100-200},
}

@inbook{inbook-1,
    Crossref = {book-1},
    Title = {Macbeth [title field of inbook-1]},
    Chapter = {7},
    Pages = {100-200},
}

@book{book-1,
    Crossref = {mvbook-1},
    Date = {1975},
    Title = {Tragedies [title field of book-1]},
    Volume = {3}
}

@mvbook{mvbook-1,
    Author = {Shakespeare},
    Date = {1970/1980},
    Title = {Collected Works [title field of mvbook-1]},
    Location = {Location},
    Publisher = {Publisher},
    Volumes = {4}
}

---
references:

- chapter-number: 7
  page: 100-200
  title: Macbeth [title field of bookinbook-1]
  volume-title: Tragedies [title field of book-1]
  container-title: Collected works [title field of mvbook-1]
  volume: 3
  id: bookinbook-1
  issued:
  - year: 1975
  author:
    family: Shakespeare
  container-author:
    family: Shakespeare
  number-of-volumes: 4
  publisher: Publisher
  type: book
  publisher-place: Location

- title: Macbeth [title field of inbook-1]
  volume-title: Tragedies [title field of book-1]
  container-title: Collected works [title field of mvbook-1]
  author:
    family: Shakespeare
  container-author:
    family: Shakespeare
  volume: 3
  chapter-number: 7
  page: 100-200
  id: inbook-1
  issued:
  - year: 1975
  number-of-volumes: 4
  publisher: Publisher
  type: chapter
  publisher-place: Location

- title: Tragedies [title field of book-1]
  container-title: Collected works [title field of mvbook-1]
  author:
    family: Shakespeare
  container-author:
    family: Shakespeare
  volume: 3
  id: book-1
  issued:
    year: 1975
  number-of-volumes: 4
  publisher: Publisher
  type: book
  publisher-place: Location

- title: Collected works [title field of mvbook-1]
  author:
    family: Shakespeare
  id: mvbook-1
  issued:
  - year: 1970
  - year: 1980
  number-of-volumes: 4
  publisher: Publisher
  type: book
  publisher-place: Location
...

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/797dd21b-5a51-48aa-900e-12eb5cd59d79%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 5627 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

After a few fixes, I still get one failure:

[biblio2yaml/crossref-nested.biblatex] FAILED
--- expected    2013-10-20 22:41:15.000000000 -0700
+++ actual      2013-10-20 22:41:15.000000000 -0700
@@ -48,9 +48,6 @@
       "author": {
         "family": "Shakespeare"
       },
-      "container-author": {
-        "family": "Shakespeare"
-      },
       "container-title": "Collected works [title field of mvbook-1]",
       "id": "book-1",
       "issued": {

I think the test case may be wrong.  The BibTeX manual tells us to
map author to both author and bookauthor for

source = mvbook or book
target = inbook, bookinbook, or suppbook

In this case source is mvbook and target is book.  So am I right
that we should not get the container-author that is expected by
your test?


+++ Nick Bart [Oct 20 13 04:41 ]:
> Two more test files below.
> 
> Note that an author field of book and mvbook needs to be inherited to both 
> author and bookauthor of the child, and that a child should inherit fields 
> from its *immediate* ancestor (in the example, the inbook entry should 
> inherit "year: 1975" from its parent, not "year: 1970 [to] year: 1980" from 
> its grandparent).
> 
> 
> $ cat crossref-inbook-mvbook.biblatex
> @comment{
> 
> crossref, directly from inbook to mvbook
> 
> }
> 
> @inbook{inbook-1,
>     Crossref = {mvbook-1},
>     Title = {Macbeth [title field of inbook-1]},
>     Date = {1975},
>     Volume = {3},
>     Chapter = {7},
>     Pages = {100-200},
> }
> 
> @mvbook{mvbook-1,
>     Author = {Shakespeare},
>     Date = {1970/1980},
>     Title = {Collected Works [title field of mvbook-1]},
>     Location = {Location},
>     Publisher = {Publisher},
>     Volumes = {4},
> }
> 
> ---
> references:
> 
> - type: chapter
> id: inbook-1
> title: Macbeth [title field of inbook-1]
> container-title: Collected works [title field of mvbook-1]
> author:
> family: Shakespeare
> container-author:
> family: Shakespeare
> issued:
> year: 1975
> volume: 3
> chapter-number: 7
> page: 100-200
> number-of-volumes: 4
> publisher: Publisher
> publisher-place: Location
> 
> - type: book
> id: mvbook-1
> title: Collected works [title field of mvbook-1]
> author:
> family: Shakespeare
> issued:
> - year: 1970
> - year: 1980
> number-of-volumes: 4
> publisher: Publisher
> publisher-place: Location
> ...
> 
> $ cat crossref-nested.biblatex 
> @comment{
> 
> Nested crossreferences (see biber manual v 1.7)
> 
> }
> 
> @bookinbook{bookinbook-1,
>     Crossref = {book-1},
>     Title = {Macbeth [title field of bookinbook-1]},
>     Chapter = {7},
>     Pages = {100-200},
> }
> 
> @inbook{inbook-1,
>     Crossref = {book-1},
>     Title = {Macbeth [title field of inbook-1]},
>     Chapter = {7},
>     Pages = {100-200},
> }
> 
> @book{book-1,
>     Crossref = {mvbook-1},
>     Date = {1975},
>     Title = {Tragedies [title field of book-1]},
>     Volume = {3}
> }
> 
> @mvbook{mvbook-1,
>     Author = {Shakespeare},
>     Date = {1970/1980},
>     Title = {Collected Works [title field of mvbook-1]},
>     Location = {Location},
>     Publisher = {Publisher},
>     Volumes = {4}
> }
> 
> ---
> references:
> 
> - chapter-number: 7
>   page: 100-200
>   title: Macbeth [title field of bookinbook-1]
>   volume-title: Tragedies [title field of book-1]
>   container-title: Collected works [title field of mvbook-1]
>   volume: 3
>   id: bookinbook-1
>   issued:
>   - year: 1975
>   author:
>     family: Shakespeare
>   container-author:
>     family: Shakespeare
>   number-of-volumes: 4
>   publisher: Publisher
>   type: book
>   publisher-place: Location
> 
> - title: Macbeth [title field of inbook-1]
>   volume-title: Tragedies [title field of book-1]
>   container-title: Collected works [title field of mvbook-1]
>   author:
>     family: Shakespeare
>   container-author:
>     family: Shakespeare
>   volume: 3
>   chapter-number: 7
>   page: 100-200
>   id: inbook-1
>   issued:
>   - year: 1975
>   number-of-volumes: 4
>   publisher: Publisher
>   type: chapter
>   publisher-place: Location
> 
> - title: Tragedies [title field of book-1]
>   container-title: Collected works [title field of mvbook-1]
>   author:
>     family: Shakespeare
>   container-author:
>     family: Shakespeare
>   volume: 3
>   id: book-1
>   issued:
>     year: 1975
>   number-of-volumes: 4
>   publisher: Publisher
>   type: book
>   publisher-place: Location
> 
> - title: Collected works [title field of mvbook-1]
>   author:
>     family: Shakespeare
>   id: mvbook-1
>   issued:
>   - year: 1970
>   - year: 1980
>   number-of-volumes: 4
>   publisher: Publisher
>   type: book
>   publisher-place: Location
> ...
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/797dd21b-5a51-48aa-900e-12eb5cd59d79%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 7130 bytes --]

You're right, this is not part of the biblatex default rules. I added it 
because it seems to make sense within the CSL logic; imagine an example 
like 

- title: Tragedies by Shakespeare [title field of book-1]
  container-title: Collected works of Shakespeare and Marlowe [title field 
of mvbook-1]
  author:
    - family: Shakespeare
  container-author:
    - family: Shakespeare
    - family: Marlowe
  volume: 3

something you can represent in CSL, but not in biblatex (where there is no 
mainauthor field). I do admit that it's a bit of a corner case, so I leave 
it up to you whether you want to deal with it or just stick to the biblatex 
defaults.



On Monday, October 21, 2013 5:50:49 AM UTC, fiddlosopher wrote:
>
> After a few fixes, I still get one failure: 
>
> [biblio2yaml/crossref-nested.biblatex] FAILED 
> --- expected    2013-10-20 22:41:15.000000000 -0700 
> +++ actual      2013-10-20 22:41:15.000000000 -0700 
> @@ -48,9 +48,6 @@ 
>        "author": { 
>          "family": "Shakespeare" 
>        }, 
> -      "container-author": { 
> -        "family": "Shakespeare" 
> -      }, 
>        "container-title": "Collected works [title field of mvbook-1]", 
>        "id": "book-1", 
>        "issued": { 
>
> I think the test case may be wrong.  The BibTeX manual tells us to 
> map author to both author and bookauthor for 
>
> source = mvbook or book 
> target = inbook, bookinbook, or suppbook 
>
> In this case source is mvbook and target is book.  So am I right 
> that we should not get the container-author that is expected by 
> your test? 
>
>
> +++ Nick Bart [Oct 20 13 04:41 ]: 
> > Two more test files below. 
> > 
> > Note that an author field of book and mvbook needs to be inherited to 
> both 
> > author and bookauthor of the child, and that a child should inherit 
> fields 
> > from its *immediate* ancestor (in the example, the inbook entry should 
> > inherit "year: 1975" from its parent, not "year: 1970 [to] year: 1980" 
> from 
> > its grandparent). 
> > 
> > 
> > $ cat crossref-inbook-mvbook.biblatex 
> > @comment{ 
> > 
> > crossref, directly from inbook to mvbook 
> > 
> > } 
> > 
> > @inbook{inbook-1, 
> >     Crossref = {mvbook-1}, 
> >     Title = {Macbeth [title field of inbook-1]}, 
> >     Date = {1975}, 
> >     Volume = {3}, 
> >     Chapter = {7}, 
> >     Pages = {100-200}, 
> > } 
> > 
> > @mvbook{mvbook-1, 
> >     Author = {Shakespeare}, 
> >     Date = {1970/1980}, 
> >     Title = {Collected Works [title field of mvbook-1]}, 
> >     Location = {Location}, 
> >     Publisher = {Publisher}, 
> >     Volumes = {4}, 
> > } 
> > 
> > --- 
> > references: 
> > 
> > - type: chapter 
> > id: inbook-1 
> > title: Macbeth [title field of inbook-1] 
> > container-title: Collected works [title field of mvbook-1] 
> > author: 
> > family: Shakespeare 
> > container-author: 
> > family: Shakespeare 
> > issued: 
> > year: 1975 
> > volume: 3 
> > chapter-number: 7 
> > page: 100-200 
> > number-of-volumes: 4 
> > publisher: Publisher 
> > publisher-place: Location 
> > 
> > - type: book 
> > id: mvbook-1 
> > title: Collected works [title field of mvbook-1] 
> > author: 
> > family: Shakespeare 
> > issued: 
> > - year: 1970 
> > - year: 1980 
> > number-of-volumes: 4 
> > publisher: Publisher 
> > publisher-place: Location 
> > ... 
> > 
> > $ cat crossref-nested.biblatex 
> > @comment{ 
> > 
> > Nested crossreferences (see biber manual v 1.7) 
> > 
> > } 
> > 
> > @bookinbook{bookinbook-1, 
> >     Crossref = {book-1}, 
> >     Title = {Macbeth [title field of bookinbook-1]}, 
> >     Chapter = {7}, 
> >     Pages = {100-200}, 
> > } 
> > 
> > @inbook{inbook-1, 
> >     Crossref = {book-1}, 
> >     Title = {Macbeth [title field of inbook-1]}, 
> >     Chapter = {7}, 
> >     Pages = {100-200}, 
> > } 
> > 
> > @book{book-1, 
> >     Crossref = {mvbook-1}, 
> >     Date = {1975}, 
> >     Title = {Tragedies [title field of book-1]}, 
> >     Volume = {3} 
> > } 
> > 
> > @mvbook{mvbook-1, 
> >     Author = {Shakespeare}, 
> >     Date = {1970/1980}, 
> >     Title = {Collected Works [title field of mvbook-1]}, 
> >     Location = {Location}, 
> >     Publisher = {Publisher}, 
> >     Volumes = {4} 
> > } 
> > 
> > --- 
> > references: 
> > 
> > - chapter-number: 7 
> >   page: 100-200 
> >   title: Macbeth [title field of bookinbook-1] 
> >   volume-title: Tragedies [title field of book-1] 
> >   container-title: Collected works [title field of mvbook-1] 
> >   volume: 3 
> >   id: bookinbook-1 
> >   issued: 
> >   - year: 1975 
> >   author: 
> >     family: Shakespeare 
> >   container-author: 
> >     family: Shakespeare 
> >   number-of-volumes: 4 
> >   publisher: Publisher 
> >   type: book 
> >   publisher-place: Location 
> > 
> > - title: Macbeth [title field of inbook-1] 
> >   volume-title: Tragedies [title field of book-1] 
> >   container-title: Collected works [title field of mvbook-1] 
> >   author: 
> >     family: Shakespeare 
> >   container-author: 
> >     family: Shakespeare 
> >   volume: 3 
> >   chapter-number: 7 
> >   page: 100-200 
> >   id: inbook-1 
> >   issued: 
> >   - year: 1975 
> >   number-of-volumes: 4 
> >   publisher: Publisher 
> >   type: chapter 
> >   publisher-place: Location 
> > 
> > - title: Tragedies [title field of book-1] 
> >   container-title: Collected works [title field of mvbook-1] 
> >   author: 
> >     family: Shakespeare 
> >   container-author: 
> >     family: Shakespeare 
> >   volume: 3 
> >   id: book-1 
> >   issued: 
> >     year: 1975 
> >   number-of-volumes: 4 
> >   publisher: Publisher 
> >   type: book 
> >   publisher-place: Location 
> > 
> > - title: Collected works [title field of mvbook-1] 
> >   author: 
> >     family: Shakespeare 
> >   id: mvbook-1 
> >   issued: 
> >   - year: 1970 
> >   - year: 1980 
> >   number-of-volumes: 4 
> >   publisher: Publisher 
> >   type: book 
> >   publisher-place: Location 
> > ... 
> > 
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "pandoc-discuss" group. 
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
>
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/797dd21b-5a51-48aa-900e-12eb5cd59d79%40googlegroups.com. 
>
> > For more options, visit https://groups.google.com/groups/opt_out. 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a6d828b7-11a1-4cbd-bf9a-0da86af74cf2%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 9735 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I agree that it makes sense -- if you get container-title, you should
get container-author too.  I don't understand why biblatex doesn't do
it, but I've made the change for now.  (All tests pass!)

+++ Nick Bart [Oct 21 13 02:30 ]:
> You're right, this is not part of the biblatex default rules. I added it 
> because it seems to make sense within the CSL logic; imagine an example 
> like 
> 
> - title: Tragedies by Shakespeare [title field of book-1]
>   container-title: Collected works of Shakespeare and Marlowe [title field 
> of mvbook-1]
>   author:
>     - family: Shakespeare
>   container-author:
>     - family: Shakespeare
>     - family: Marlowe
>   volume: 3
> 
> something you can represent in CSL, but not in biblatex (where there is no 
> mainauthor field). I do admit that it's a bit of a corner case, so I leave 
> it up to you whether you want to deal with it or just stick to the biblatex 
> defaults.
> 
> 
> 
> On Monday, October 21, 2013 5:50:49 AM UTC, fiddlosopher wrote:
> >
> > After a few fixes, I still get one failure: 
> >
> > [biblio2yaml/crossref-nested.biblatex] FAILED 
> > --- expected    2013-10-20 22:41:15.000000000 -0700 
> > +++ actual      2013-10-20 22:41:15.000000000 -0700 
> > @@ -48,9 +48,6 @@ 
> >        "author": { 
> >          "family": "Shakespeare" 
> >        }, 
> > -      "container-author": { 
> > -        "family": "Shakespeare" 
> > -      }, 
> >        "container-title": "Collected works [title field of mvbook-1]", 
> >        "id": "book-1", 
> >        "issued": { 
> >
> > I think the test case may be wrong.  The BibTeX manual tells us to 
> > map author to both author and bookauthor for 
> >
> > source = mvbook or book 
> > target = inbook, bookinbook, or suppbook 
> >
> > In this case source is mvbook and target is book.  So am I right 
> > that we should not get the container-author that is expected by 
> > your test? 
> >
> >
> > +++ Nick Bart [Oct 20 13 04:41 ]: 
> > > Two more test files below. 
> > > 
> > > Note that an author field of book and mvbook needs to be inherited to 
> > both 
> > > author and bookauthor of the child, and that a child should inherit 
> > fields 
> > > from its *immediate* ancestor (in the example, the inbook entry should 
> > > inherit "year: 1975" from its parent, not "year: 1970 [to] year: 1980" 
> > from 
> > > its grandparent). 
> > > 
> > > 
> > > $ cat crossref-inbook-mvbook.biblatex 
> > > @comment{ 
> > > 
> > > crossref, directly from inbook to mvbook 
> > > 
> > > } 
> > > 
> > > @inbook{inbook-1, 
> > >     Crossref = {mvbook-1}, 
> > >     Title = {Macbeth [title field of inbook-1]}, 
> > >     Date = {1975}, 
> > >     Volume = {3}, 
> > >     Chapter = {7}, 
> > >     Pages = {100-200}, 
> > > } 
> > > 
> > > @mvbook{mvbook-1, 
> > >     Author = {Shakespeare}, 
> > >     Date = {1970/1980}, 
> > >     Title = {Collected Works [title field of mvbook-1]}, 
> > >     Location = {Location}, 
> > >     Publisher = {Publisher}, 
> > >     Volumes = {4}, 
> > > } 
> > > 
> > > --- 
> > > references: 
> > > 
> > > - type: chapter 
> > > id: inbook-1 
> > > title: Macbeth [title field of inbook-1] 
> > > container-title: Collected works [title field of mvbook-1] 
> > > author: 
> > > family: Shakespeare 
> > > container-author: 
> > > family: Shakespeare 
> > > issued: 
> > > year: 1975 
> > > volume: 3 
> > > chapter-number: 7 
> > > page: 100-200 
> > > number-of-volumes: 4 
> > > publisher: Publisher 
> > > publisher-place: Location 
> > > 
> > > - type: book 
> > > id: mvbook-1 
> > > title: Collected works [title field of mvbook-1] 
> > > author: 
> > > family: Shakespeare 
> > > issued: 
> > > - year: 1970 
> > > - year: 1980 
> > > number-of-volumes: 4 
> > > publisher: Publisher 
> > > publisher-place: Location 
> > > ... 
> > > 
> > > $ cat crossref-nested.biblatex 
> > > @comment{ 
> > > 
> > > Nested crossreferences (see biber manual v 1.7) 
> > > 
> > > } 
> > > 
> > > @bookinbook{bookinbook-1, 
> > >     Crossref = {book-1}, 
> > >     Title = {Macbeth [title field of bookinbook-1]}, 
> > >     Chapter = {7}, 
> > >     Pages = {100-200}, 
> > > } 
> > > 
> > > @inbook{inbook-1, 
> > >     Crossref = {book-1}, 
> > >     Title = {Macbeth [title field of inbook-1]}, 
> > >     Chapter = {7}, 
> > >     Pages = {100-200}, 
> > > } 
> > > 
> > > @book{book-1, 
> > >     Crossref = {mvbook-1}, 
> > >     Date = {1975}, 
> > >     Title = {Tragedies [title field of book-1]}, 
> > >     Volume = {3} 
> > > } 
> > > 
> > > @mvbook{mvbook-1, 
> > >     Author = {Shakespeare}, 
> > >     Date = {1970/1980}, 
> > >     Title = {Collected Works [title field of mvbook-1]}, 
> > >     Location = {Location}, 
> > >     Publisher = {Publisher}, 
> > >     Volumes = {4} 
> > > } 
> > > 
> > > --- 
> > > references: 
> > > 
> > > - chapter-number: 7 
> > >   page: 100-200 
> > >   title: Macbeth [title field of bookinbook-1] 
> > >   volume-title: Tragedies [title field of book-1] 
> > >   container-title: Collected works [title field of mvbook-1] 
> > >   volume: 3 
> > >   id: bookinbook-1 
> > >   issued: 
> > >   - year: 1975 
> > >   author: 
> > >     family: Shakespeare 
> > >   container-author: 
> > >     family: Shakespeare 
> > >   number-of-volumes: 4 
> > >   publisher: Publisher 
> > >   type: book 
> > >   publisher-place: Location 
> > > 
> > > - title: Macbeth [title field of inbook-1] 
> > >   volume-title: Tragedies [title field of book-1] 
> > >   container-title: Collected works [title field of mvbook-1] 
> > >   author: 
> > >     family: Shakespeare 
> > >   container-author: 
> > >     family: Shakespeare 
> > >   volume: 3 
> > >   chapter-number: 7 
> > >   page: 100-200 
> > >   id: inbook-1 
> > >   issued: 
> > >   - year: 1975 
> > >   number-of-volumes: 4 
> > >   publisher: Publisher 
> > >   type: chapter 
> > >   publisher-place: Location 
> > > 
> > > - title: Tragedies [title field of book-1] 
> > >   container-title: Collected works [title field of mvbook-1] 
> > >   author: 
> > >     family: Shakespeare 
> > >   container-author: 
> > >     family: Shakespeare 
> > >   volume: 3 
> > >   id: book-1 
> > >   issued: 
> > >     year: 1975 
> > >   number-of-volumes: 4 
> > >   publisher: Publisher 
> > >   type: book 
> > >   publisher-place: Location 
> > > 
> > > - title: Collected works [title field of mvbook-1] 
> > >   author: 
> > >     family: Shakespeare 
> > >   id: mvbook-1 
> > >   issued: 
> > >   - year: 1970 
> > >   - year: 1980 
> > >   number-of-volumes: 4 
> > >   publisher: Publisher 
> > >   type: book 
> > >   publisher-place: Location 
> > > ... 
> > > 
> > > -- 
> > > You received this message because you are subscribed to the Google 
> > Groups "pandoc-discuss" group. 
> > > To unsubscribe from this group and stop receiving emails from it, send 
> > an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org <javascript:>. 
> > > To post to this group, send email to pandoc-...-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org<javascript:>. 
> >
> > > To view this discussion on the web visit 
> > https://groups.google.com/d/msgid/pandoc-discuss/797dd21b-5a51-48aa-900e-12eb5cd59d79%40googlegroups.com. 
> >
> > > For more options, visit https://groups.google.com/groups/opt_out. 
> >
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a6d828b7-11a1-4cbd-bf9a-0da86af74cf2%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Excellent.  I think I was using an older version of the manual
when I wrote that code.  I've brought it up to date, and now all
tests pass.

+++ Nick Bart [Oct 19 13 23:10 ]:
> The problem is in the remapping (biblatex: "inheritance") rules.
> 
> Bibtex.hs seems to remap all  book, collection, proceedings, reference, 
> mvbook, mvcollection, mvproceedings, mvreference titles to maintitles
> 
>        "title"          -> ["maintitle"]
>        "subtitle"       -> ["mainsubtitle"]
>        "titleaddon"     -> ["maintitleaddon"]
> 
> This is only correct for mvbook, mvcollection, mvproceedings, mvreference.
> 
> For book, collection, proceedings, reference, we need:
> 
>        "title"          -> ["booktitle"]
>        "subtitle"       -> ["booksubtitle"]
>        "titleaddon"     -> ["booktitleaddon"]
> 
> See biblatex manual, v2.7a, Appendix B, p. 240–242
> 
> On Saturday, October 19, 2013 10:31:12 PM UTC, fiddlosopher wrote:
> >
> > +++ Nick Bart [Oct 19 13 04:45 ]: 
> > > Brilliant! 
> > > 
> > > And your corrections of two mistakes in the test output are spot on. 
> > > 
> > > On title vs. booktitle, however: 
> > > 
> > > I'm not quite sure whether I understand the Bibtex.hs code and your 
> > > comments in the commits correctly, but when biblio2yaml is dealing with 
> > > title/booktitle/maintitle, it should never be *comparing* strings 
> > (though 
> > > citeproc should compare author and bookauthor, see below). 
> >
> > That was a hack.  I think I've implemented the correct logic, now. 
> > One test fails: 
> >
> > [biblio2yaml/incollection-2.biblatex] FAILED (-expected, +actual) 
> >     1 { 
> >     2   "references": [ 
> >     3     { 
> >     4       "annote": "A cross-referenced article from a collection. This 
> > is an incollection entry with a crossref field. Note the subtitle and 
> > indextitle fields", 
> >     5       "author": { 
> >     6         "family": "Westfahl", 
> >     7         "given": [ 
> >     8           "Gary" 
> >     9         ] 
> >    10       }, 
> >    11       "container-title": "Space and beyond: The frontier theme in 
> > science fiction", 
> >    12       "editor": { 
> >    13         "family": "Westfahl", 
> >    14         "given": [ 
> >    15           "Gary" 
> >    16         ] 
> >    17       }, 
> >    18       "id": "westfahl:space", 
> >    19       "issued": { 
> >    20         "year": 2000 
> >    21       }, 
> >    22       "language": "en-US", 
> >    23       "page": "55-65", 
> >    24       "publisher": "Greenwood", 
> >    25       "publisher-place": "Westport, Conn.; London", 
> >    26       "title": "The true frontier: Confronting and avoiding the 
> > realities of space in American science fiction films", 
> > -  27       "type": "chapter" 
> > +  27       "type": "chapter", 
> > +  28       "volume-title": "Space and beyond: The frontier theme in 
> > science fiction" 
> >    29     }, 
> >
> > The only difference is the presence of "volume-title".  You say: 
> >
> > > The booktitle field has to be used, by contrast, for inbook, 
> > incollection, 
> > > inproceedings, inreference entries without crossref, where it is mapped 
> > to 
> > > container-title if no maintitle exists; else maintitle is mapped to 
> > > container-title, and booktitle to volume-title. 
> >
> > Here we have an entry with both 'maintitle' (remapped from 'title' in 
> > crossref) and 'booktitle', so the program has dutifully mapped one to 
> > container-title and the other to volume-title.  (In this case they are 
> > the same.) 
> >
> > Where have I gone wrong? 
> >
> >
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/40fcf73d-5a6d-45dd-9cad-a79ee246b9e2%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/20131021023606.GA46387%40Johns-MacBook-Pro.local.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Fixed.

+++ Nick Bart [Oct 15 13 07:45 ]:
>    Also,biblio2yaml does not produce *any* output from @periodical any
>    more:
>    $ cat test-special-issue.bib
>    @periodical{item1,
>        Date = {2013},
>        Editor = {Editor, Edward},
>        Entrysubtype = {magazine},
>        Hyphenation = {english},
>        Issuetitle = {Title of the Special Issue},
>        Location = {Location},
>        Note = {Special Issue},
>        Number = {3},
>        Publisher = {Publisher},
>        Series = {newseries},
>        Title = {Title of the Journal},
>        Volume = {7},
>        Year = {2013}}
>    $ biblio2yaml test-special-issue.bib
>    ---
>    references:
>    []
>    ...
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/4bb3bb13-b90b-4204-
>    90a6-78db9705262a%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/4bb3bb13-b90b-4204-90a6-78db9705262a%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

You'll need to reinstall pandoc-citeproc too, so it uses the
latest pandoc library.

+++ Nick Bart [Oct 15 13 07:16 ]:
> Did a fresh "git clone git://github.com/jgm/pandoc", but the long URL from 
> my example did not change in the output. Did you mean anything else by 
> "install the latest source from github"?
> 
> On Tuesday, October 15, 2013 6:06:38 AM UTC, fiddlosopher wrote:
> >
> > +++ Nick Bart [Oct 13 13 16:03 ]: 
> >   
> > > Some urls could be simplified 
> >
> > This is fixed in pandoc -- if you install the latest source from github. 
> >
> >
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a1cb129e-221c-40bc-9431-617be8116006%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

To be explicit:

Go to the pandoc source directory you cloned, then:

git submodule update --init
cabal install  --force

Then go to pandoc-citeproc source directory and

cabal install

+++ Nick Bart [Oct 15 13 07:16 ]:
> Did a fresh "git clone git://github.com/jgm/pandoc", but the long URL from 
> my example did not change in the output. Did you mean anything else by 
> "install the latest source from github"?
> 
> On Tuesday, October 15, 2013 6:06:38 AM UTC, fiddlosopher wrote:
> >
> > +++ Nick Bart [Oct 13 13 16:03 ]: 
> >   
> > > Some urls could be simplified 
> >
> > This is fixed in pandoc -- if you install the latest source from github. 
> >
> >
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a1cb129e-221c-40bc-9431-617be8116006%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1387 bytes --]

Couldn't pandoc choose localized bibliography terms based on the value of 
its own "lang" variable, e.g. --variable=lang:french?

The necessary locale files exist: 
https://github.com/citation-style-language/locales

On Sunday, 29 September 2013 14:49:59 UTC, fiddlosopher wrote:
>
> +++ Nick Bart [Sep 29 13 04:39 ]: 
> >    On a related note, how is per-document localization done in 
> >    pandoc(-citeproc)? 
>
> I'm going to look into this more, but I think it just takes the 
> locale from the environment, and uses it to find appropriate 
> terms for things like "editor".  (These can be overridden, I 
> believe, in the CSL files.)  In any case, the per-entry 'language' 
> field wouldn't be used for this; rather, we'd want the language 
> for the document containing the bibliography. 
>
> John 
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/4e1cb117-16ba-4526-b956-f5700b833198%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1847 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 4003 bytes --]

Below is another patch with a few fixes: number, issue, series field, and 
article with entrysubtype=magazine.

The following code, based on your suggestions did not compile, however:

  toLocale :: String -> String
  toLocale "english"    = "en-EN"
  toLocale "american"   = "en-US"
  toLocale "british"    = "en-UK"
  toLocale "canadian"   = "en-CA"
  toLocale "australian" = "en-AU"
  toLocale "newzealand" = "en-NZ"
  toLocale "USenglish"  = "en-US"
  toLocale "UKenglish"  = "en-UK"
  opt $ getRawField "hyphenation" >>= setRawField "language" . toLocale

Error message:

$  ghc --make bibtex2pandoc
[1 of 1] Compiling Main             ( bibtex2pandoc.hs, bibtex2pandoc.o )

bibtex2pandoc.hs:283:25: parse error on input `='

Any ideas?


--- ../pandoc-citeproc/bibtex2pandoc.hs    2013-09-29 08:56:30.000000000 
+0100
+++ bibtex2pandoc.hs    2013-09-29 21:53:25.000000000 +0100
@@ -271,6 +271,26 @@
        "unpublished"     -> setType "manuscript"
        "www"             -> setType "webpage"
        _                 -> setType "no-type"
+
+  opt $ do
+    val <- getRawField "entrysubtype"
+    if  et == "article" && val == "magazine" then
+          setType "article-magazine"
+    else
+          return ()
+
+  toLocale :: String -> String
+  toLocale "english"    = "en-EN"
+  toLocale "american"   = "en-US"
+  toLocale "british"    = "en-UK"
+  toLocale "canadian"   = "en-CA"
+  toLocale "australian" = "en-AU"
+  toLocale "newzealand" = "en-NZ"
+  toLocale "USenglish"  = "en-US"
+  toLocale "UKenglish"  = "en-UK"
+  opt $ getRawField "hyphenation" >>= setRawField "language" . toLocale
+
+
   opt $ getRawField "type" >>= setRawField "genre"
   hyphenation <- getRawField "hyphenation" <|> return "english"
   let processTitle = if (map toLower hyphenation) `elem`
@@ -303,10 +323,15 @@
                           then "volume-title"
                           else "container-title") addPeriod . processTitle
   opt $ getField "shorttitle" >>= setField "title-short" . processTitle
-  opt $ getField "series" >>= setField "collection-title" . processTitle
   opt $ getField "pages" >>= setField "page"
   opt $ getField "volume" >>= setField "volume"
-  opt $ getField "number" >>= setField "number"
+  opt $ getField "number" >>=
+             setField (if et `elem` ["report"]
+                       then "number"
+                       else if et `elem` ["article", "periodical"]
+                       then "issue"
+                       else "collection-number")
+  opt $ getField "issue" >>= appendField "issue" addComma
   opt $ getField "chapter" >>= setField "chapter-number"
   opt $ getField "edition" >>= setField "edition"
   opt $ getField "note" >>= setField "note"
@@ -352,11 +377,16 @@
   opt $ getField "urldate" >>= setField "accessed"
   opt $ getField "version" >>= setField "version"
   opt $ getField "volumes" >>= setField "number-of-volumes"
-  opt $ getRawField "hyphenation" >>= setRawField "language"
+  opt $ getField "series" >>= if et `elem` ["article","periodical"]
+                                        then appendField "container-title" 
addComma
+                                        else setField "collection-title" . 
processTitle
 
 addColon :: [Inline] -> [Inline]
 addColon xs = [Str ":",Space] ++ xs
 
+addComma :: [Inline] -> [Inline]
+addComma xs = [Str ",",Space] ++ xs
+
 addPeriod :: [Inline] -> [Inline]
 addPeriod xs = [Str ".",Space] ++ xs
 

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/87563807-c8d8-47ff-a774-bc13342f76a2%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 6746 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

Sorry - I should have said that the toLocale function needs to go
at the outer level -- i.e. it should not be inside the
itemToMetaValue function.
Also you should have a catch-all at the end

    toLocale _ = ""

or it will raise an error for other values.

+++ Nick Bart [Sep 29 13 13:57 ]:
>    Below is another patch with a few fixes: number, issue, series field,
>    and article with entrysubtype=magazine.
>    The following code, based on your suggestions did not compile, however:
>      toLocale :: String -> String
>      toLocale "english"    = "en-EN"
>      toLocale "american"   = "en-US"
>      toLocale "british"    = "en-UK"
>      toLocale "canadian"   = "en-CA"
>      toLocale "australian" = "en-AU"
>      toLocale "newzealand" = "en-NZ"
>      toLocale "USenglish"  = "en-US"
>      toLocale "UKenglish"  = "en-UK"
>      opt $ getRawField "hyphenation" >>= setRawField "language" . toLocale
>    Error message:
>    $  ghc --make bibtex2pandoc
>    [1 of 1] Compiling Main             ( bibtex2pandoc.hs, bibtex2pandoc.o
>    )
>    bibtex2pandoc.hs:283:25: parse error on input `='
>    Any ideas?
>    --- ../pandoc-citeproc/bibtex2pandoc.hs    2013-09-29
>    08:56:30.000000000 +0100
>    +++ bibtex2pandoc.hs    2013-09-29 21:53:25.000000000 +0100
>    @@ -271,6 +271,26 @@
>            "unpublished"     -> setType "manuscript"
>            "www"             -> setType "webpage"
>            _                 -> setType "no-type"
>    +
>    +  opt $ do
>    +    val <- getRawField "entrysubtype"
>    +    if  et == "article" && val == "magazine" then
>    +          setType "article-magazine"
>    +    else
>    +          return ()
>    +
>    +  toLocale :: String -> String
>    +  toLocale "english"    = "en-EN"
>    +  toLocale "american"   = "en-US"
>    +  toLocale "british"    = "en-UK"
>    +  toLocale "canadian"   = "en-CA"
>    +  toLocale "australian" = "en-AU"
>    +  toLocale "newzealand" = "en-NZ"
>    +  toLocale "USenglish"  = "en-US"
>    +  toLocale "UKenglish"  = "en-UK"
>    +  opt $ getRawField "hyphenation" >>= setRawField "language" .
>    toLocale
>    +
>    +
>       opt $ getRawField "type" >>= setRawField "genre"
>       hyphenation <- getRawField "hyphenation" <|> return "english"
>       let processTitle = if (map toLower hyphenation) `elem`
>    @@ -303,10 +323,15 @@
>                               then "volume-title"
>                               else "container-title") addPeriod .
>    processTitle
>       opt $ getField "shorttitle" >>= setField "title-short" .
>    processTitle
>    -  opt $ getField "series" >>= setField "collection-title" .
>    processTitle
>       opt $ getField "pages" >>= setField "page"
>       opt $ getField "volume" >>= setField "volume"
>    -  opt $ getField "number" >>= setField "number"
>    +  opt $ getField "number" >>=
>    +             setField (if et `elem` ["report"]
>    +                       then "number"
>    +                       else if et `elem` ["article", "periodical"]
>    +                       then "issue"
>    +                       else "collection-number")
>    +  opt $ getField "issue" >>= appendField "issue" addComma
>       opt $ getField "chapter" >>= setField "chapter-number"
>       opt $ getField "edition" >>= setField "edition"
>       opt $ getField "note" >>= setField "note"
>    @@ -352,11 +377,16 @@
>       opt $ getField "urldate" >>= setField "accessed"
>       opt $ getField "version" >>= setField "version"
>       opt $ getField "volumes" >>= setField "number-of-volumes"
>    -  opt $ getRawField "hyphenation" >>= setRawField "language"
>    +  opt $ getField "series" >>= if et `elem` ["article","periodical"]
>    +                                        then appendField
>    "container-title" addComma
>    +                                        else setField
>    "collection-title" . processTitle
> 
>     addColon :: [Inline] -> [Inline]
>     addColon xs = [Str ":",Space] ++ xs
> 
>    +addComma :: [Inline] -> [Inline]
>    +addComma xs = [Str ",",Space] ++ xs
>    +
>     addPeriod :: [Inline] -> [Inline]
>     addPeriod xs = [Str ".",Space] ++ xs
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/87563807-c8d8-47ff-
>    a774-bc13342f76a2%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/87563807-c8d8-47ff-a774-bc13342f76a2%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

I'll wait for a revised patch that compiles before applying any of this.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Rintze Zelle]

[-- Attachment #1: Type: text/plain, Size: 1070 bytes --]

On Saturday, September 28, 2013 9:41:30 PM UTC-4, fiddlosopher wrote:

> I guess they are Babel language names.
>

In CSL we use 2 or 4-letter IETF language tags ( 
http://en.wikipedia.org/wiki/IETF_language_tag ). For the languages for 
which we have CSL locale files, I maintain a list of language codes with 
their human readable names: 
https://github.com/citation-style-language/locales/blob/master/locales.json . 
Note that the code for British English is "en-GB", not "en-UK".

Rintze

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/ebfe0ce3-e011-4716-8669-a28d7813ba50%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1675 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Paulo Ney de Souza]

[-- Attachment #1: Type: text/plain, Size: 1748 bytes --]

  Rintze,

There is one locale file that is puzzling me a bit. It is the one named 
"locales-ar-AR.xml". The code is supposed to be the combination of a 
language and a country. Language "ar" is Arabic and country "AR" is 
Argentina, which makes for an odd locale!

The file is indeed in Arabic, so most likely it is a labeling error and 
the correct could be ar-AE , for Arabic in United Arab Emirates (R and E 
are close on the keyboard) .. or may of the other countries speaking 
Arabic,...

Do you guys keep track of who did these translations ? How reliable they 
are and if anyone has complained about specific entries ?

Paulo Ney


On 9/29/2013 3:58 PM, Rintze Zelle wrote:
> On Saturday, September 28, 2013 9:41:30 PM UTC-4, fiddlosopher wrote:
>
>     I guess they are Babel language names.
>
>
> In CSL we use 2 or 4-letter IETF language tags ( 
> http://en.wikipedia.org/wiki/IETF_language_tag ). For the languages 
> for which we have CSL locale files, I maintain a list of language 
> codes with their human readable names: 
> https://github.com/citation-style-language/locales/blob/master/locales.json . 
> Note that the code for British English is "en-GB", not "en-UK".
>
> Rintze

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/524CD904.4000405%40gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 3051 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 2872 bytes --]

Good you spotted this, Paulo.

As far as I can tell, "locales-ar-AR.xml" contains Modern Standard Arabic 
(MSA) terms. Since MSA is used in so many different countries, is probably 
best to just use the language code "ar" and the file name "locales-ar.xml" 
(as it is already done with basque, "eu").

Also, a few things are missing in "locales-ar-AR.xml": The terms "available 
at", and "director" and "illustrator" (including derived forms) stand 
untranslated.

Given that "available at" is most likely required a lot, and "director", 
though not used by, e.g., Zotero when I last checked, can be used via 
bibtex2pandoc now, it would be nice if this could be fixed.

At least locales-de-CH.xml, locales-it-IT.xml, locales-pt-BR.xml contain 
some untranslated English terms, too.

Finally, I think it would be a good idea to include, in the header of 
locale files, information on what exact language they contain, plus, if 
available, contact details of authors or maintainers.


On Thursday, 3 October 2013 02:40:04 UTC, Paulo Ney de Souza wrote:
>
>  Rintze,
>
> There is one locale file that is puzzling me a bit. It is the one named 
> "locales-ar-AR.xml". The code is supposed to be the combination of a 
> language and a country. Language "ar" is Arabic and country "AR" is 
> Argentina, which makes for an odd locale!
>
> The file is indeed in Arabic, so most likely it is a labeling error and 
> the correct could be ar-AE , for Arabic in United Arab Emirates (R and E 
> are close on the keyboard) .. or may of the other countries speaking 
> Arabic,...
>
> Do you guys keep track of who did these translations ? How reliable they 
> are and if anyone has complained about specific entries ? 
>
> Paulo Ney
>
>
> On 9/29/2013 3:58 PM, Rintze Zelle wrote: 
>
> On Saturday, September 28, 2013 9:41:30 PM UTC-4, fiddlosopher wrote:
>
>> I guess they are Babel language names.
>>
>
>  In CSL we use 2 or 4-letter IETF language tags ( 
> http://en.wikipedia.org/wiki/IETF_language_tag ). For the languages for 
> which we have CSL locale files, I maintain a list of language codes with 
> their human readable names: 
> https://github.com/citation-style-language/locales/blob/master/locales.json . 
> Note that the code for British English is "en-GB", not "en-UK".
>
>  Rintze
>  
>  

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/3f290ad0-44ff-4559-a614-a0add2b5089b%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 4063 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Paulo Ney de Souza]

[-- Attachment #1: Type: text/plain, Size: 4785 bytes --]

  Hi Nick, Rintze,

The idea of using the "locales-ar.xml" is grand, it complies to 
standards and allows one to have a a file that loads its definitions and 
then add something local, like for example (locale-ar-EG, locale-ar-DZ, 
...) ... but just like the locales-eu already defined - it begs an 
explanation of how it will be used because the chances of a particular 
user to have a computer set to this locale is next to nill!

I have parsed all the CSL locale files for a project at BibLaTeX - they 
have some 75 terms which are plain-common in between the two projects 
and in derived ways a lot more, so I wrote them out to a a db to be able 
to compare. Some of them look like high quality translations, but some 
of them seem like output of Google translator! It is only inside a db 
that one will be able to compare and possibly sort some of the repeated, 
missing and low-quality entries.

It would be nice if they could use the same set of files, but TeX parses 
XML very poorly and would slow down things tremendously. One of the 
consequences of this project is that I have a single command now that is 
able to generate all the locale files from a database. In the DB one can:

    * compare many translations for the same term
    * eliminate duplicates as in pt-PT and pt-BR having a common 
locale-pt.xml underneath
    * mark tokens to be studied or translated
    * summarize missing terms, locales or languages

If you guys want to use this later for CSL, please feel welcome. We will 
use it at BibLaTeX to drive the expansion of language/locale support and 
study each term a bit more carefully.

Paulo Ney



On 10/3/2013 12:04 AM, Nick Bart wrote:
> Good you spotted this, Paulo.
>
> As far as I can tell, "locales-ar-AR.xml" contains Modern Standard 
> Arabic (MSA) terms. Since MSA is used in so many different countries, 
> is probably best to just use the language code "ar" and the file name 
> "locales-ar.xml" (as it is already done with basque, "eu").
>
> Also, a few things are missing in "locales-ar-AR.xml": The terms 
> "available at", and "director" and "illustrator" (including derived 
> forms) stand untranslated.
>
> Given that "available at" is most likely required a lot, and 
> "director", though not used by, e.g., Zotero when I last checked, can 
> be used via bibtex2pandoc now, it would be nice if this could be fixed.
>
> At least locales-de-CH.xml, locales-it-IT.xml, locales-pt-BR.xml 
> contain some untranslated English terms, too.
>
> Finally, I think it would be a good idea to include, in the header of 
> locale files, information on what exact language they contain, plus, 
> if available, contact details of authors or maintainers.
>
>
> On Thursday, 3 October 2013 02:40:04 UTC, Paulo Ney de Souza wrote:
>
>     Rintze,
>
>     There is one locale file that is puzzling me a bit. It is the one
>     named "locales-ar-AR.xml". The code is supposed to be the
>     combination of a language and a country. Language "ar" is Arabic
>     and country "AR" is Argentina, which makes for an odd locale!
>
>     The file is indeed in Arabic, so most likely it is a labeling
>     error and the correct could be ar-AE , for Arabic in United Arab
>     Emirates (R and E are close on the keyboard) .. or may of the
>     other countries speaking Arabic,...
>
>     Do you guys keep track of who did these translations ? How
>     reliable they are and if anyone has complained about specific
>     entries ?
>
>     Paulo Ney
>
>
>     On 9/29/2013 3:58 PM, Rintze Zelle wrote:
>>     On Saturday, September 28, 2013 9:41:30 PM UTC-4, fiddlosopher wrote:
>>
>>         I guess they are Babel language names.
>>
>>
>>     In CSL we use 2 or 4-letter IETF language tags (
>>     http://en.wikipedia.org/wiki/IETF_language_tag
>>     <http://en.wikipedia.org/wiki/IETF_language_tag> ). For the
>>     languages for which we have CSL locale files, I maintain a list
>>     of language codes with their human readable names:
>>     https://github.com/citation-style-language/locales/blob/master/locales.json
>>     <https://github.com/citation-style-language/locales/blob/master/locales.json> .
>>     Note that the code for British English is "en-GB", not "en-UK".
>>
>>     Rintze
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/524D5E27.1070706%40gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 7130 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Rintze Zelle]

On Thu, Oct 3, 2013 at 8:08 AM, Paulo Ney de Souza <pauloney-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> If you guys want to use this later for CSL, please feel welcome. We will use
> it at BibLaTeX to drive the expansion of language/locale support and study
> each term a bit more carefully.

Cool! It's good if there is some cross-pollination between the
BibLaTeX and CSL worlds :). I personally (foolishly?) always shied
away from the BibTeX world in my work on CSL as I found it a bit
intimidating and hard to navigate, but clearly there are lessons to be
learned both ways.

> On 10/3/2013 12:04 AM, Nick Bart wrote:
>> As far as I can tell, "locales-ar-AR.xml" contains Modern Standard Arabic
>> (MSA) terms. Since MSA is used in so many different countries, is probably
>> best to just use the language code "ar" and the file name "locales-ar.xml"
>> (as it is already done with basque, "eu").

Agreed. I prefer to include country codes when possible, but since MSA
doesn't have a single originating country, "locales-ar.xml" makes
sense. And the "AR" is clearly wrong. I nowadays often check language
tags at http://rishida.net/utils/subtags/ , but "ar-AR" must have
slipped by.

>> Also, a few things are missing in "locales-ar-AR.xml": The terms "available
>> at", and "director" and "illustrator" (including derived forms) stand
>> untranslated.

With every CSL update that affected the locale files, I prepopulated
new terms with the translations from en-US. See e.g.
https://github.com/citation-style-language/locales/commit/0374188ca0784c1d5cde3d1a2655897b9f4c8292
for the CSL 1.0.1 update. So sometimes these new terms aren't
translated yet.

>> Given that "available at" is most likely required a lot, and "director",
>> though not used by, e.g., Zotero when I last checked, can be used via
>> bibtex2pandoc now, it would be nice if this could be fixed.

The most recent (and only?) contributor to ar-AR translations is Ahmed
Rizk (from Egypt). You could ping him in the following pull request:
https://github.com/citation-style-language/locales/pull/17

>> Finally, I think it would be a good idea to include, in the header of locale
>> files, information on what exact language they contain, plus, if available,
>> contact details of authors or maintainers.

CSL 1.0.1 introduced the ability to specify translators. See
http://citationstyles.org/downloads/release-notes-csl101.html#locale-metadata
. The only place where we currently link language tags to language
names is https://github.com/citation-style-language/locales/blob/master/locales.json
. Is that enough?

Rintze

P.S. This discussion about CSL locale files probably belongs more on
CSL's xbiblio mailing list, or at least in a separate pandoc-discuss
thread. The "Decoupling citeproc and highlighting-kate from pandoc"
thread already has over a 100 posts!

>> On Thursday, 3 October 2013 02:40:04 UTC, Paulo Ney de Souza wrote:
>>>
>>> Rintze,
>>>
>>> There is one locale file that is puzzling me a bit. It is the one named
>>> "locales-ar-AR.xml". The code is supposed to be the combination of a
>>> language and a country. Language "ar" is Arabic and country "AR" is
>>> Argentina, which makes for an odd locale!
>>>
>>> The file is indeed in Arabic, so most likely it is a labeling error and
>>> the correct could be ar-AE , for Arabic in United Arab Emirates (R and E are
>>> close on the keyboard) .. or may of the other countries speaking Arabic,...
>>>
>>> Do you guys keep track of who did these translations ? How reliable they
>>> are and if anyone has complained about specific entries ?
>>>
>>> Paulo Ney>

CSL locale files

[Paulo Ney de Souza]

  On 10/3/2013 6:51 AM, Rintze Zelle wrote:
> Cool! It's good if there is some cross-pollination between the
> BibLaTeX and CSL worlds :).

There should be! The intersection (in this area of i18n) is huge.

> I personally (foolishly?) always shied
> away from the BibTeX world in my work on CSL as I found it a bit
> intimidating and hard to navigate, but clearly there are lessons to be
> learned both ways.

The BibLaTeX people are also literally staying away from BibTeX too!  :)

That is in fact the whole reason for the existence of BibLaTeX.

> Agreed. I prefer to include country codes when possible, but since MSA
> doesn't have a single originating country, "locales-ar.xml" makes
> sense. And the "AR" is clearly wrong. I nowadays often check language
> tags at http://rishida.net/utils/subtags/ , but "ar-AR" must have
> slipped by.

I am wondering how is that supposed to work at the user level, since 
while it is possible that some user will have locale set to "eu" (for 
separatist reasons) - it is quite unlikely that someone will have it set 
to "ar" alone.
Will it pick up all locales set to ar-* and load the file "locales-ar.xml".


> With every CSL update that affected the locale files, I prepopulated
> new terms with the translations from en-US. See e.g.
> https://github.com/citation-style-language/locales/commit/0374188ca0784c1d5cde3d1a2655897b9f4c8292
> for the CSL 1.0.1 update. So sometimes these new terms aren't
> translated yet.

At the end of this piece I'll propose lots of missing token-translation, 
a few changes and some new languages/locale to Rintze.


> The most recent (and only?) contributor to ar-AR translations is Ahmed
> Rizk (from Egypt). You could ping him in the following pull request:
> https://github.com/citation-style-language/locales/pull/17
>
>>> Finally, I think it would be a good idea to include, in the header of locale
>>> files, information on what exact language they contain, plus, if available,
>>> contact details of authors or maintainers.

I second Nick's request to have the name of the author IN the file, I 
know I can always go through the entire repository and check who did 
what, but that is pretty hard for most people.

IN fact, I go further and say that for that file to use the CC license, 
that is currently using - it is required that the name of the author be 
present on the file. I know it is legalese - but so is the license. 
Attributions in the work have been required in CC licenses since 2004.

> CSL 1.0.1 introduced the ability to specify translators. See
> http://citationstyles.org/downloads/release-notes-csl101.html#locale-metadata
> . The only place where we currently link language tags to language
> names is https://github.com/citation-style-language/locales/blob/master/locales.json
> . Is that enough?
>
> Rintze

It would be nice if we could place the ISO language code as one of the 
columns in that file, in this way it would make easier for people to 
layout the data in their own db.

Paulo Ney

Re: CSL locale files

[Rintze Zelle]

[-- Attachment #1: Type: text/plain, Size: 4592 bytes --]

On Thursday, October 3, 2013 12:57:09 PM UTC-4, Paulo Ney de Souza wrote:

> > Agreed. I prefer to include country codes when possible, but since MSA 
> > doesn't have a single originating country, "locales-ar.xml" makes 
> > sense. And the "AR" is clearly wrong. I nowadays often check language 
> > tags at http://rishida.net/utils/subtags/ , but "ar-AR" must have 
> > slipped by. 
>
> I am wondering how is that supposed to work at the user level, since 
> while it is possible that some user will have locale set to "eu" (for 
> separatist reasons) - it is quite unlikely that someone will have it set 
> to "ar" alone. 
> Will it pick up all locales set to ar-* and load the file "locales-ar.xml".
>

There are some pointers on this at 
http://citationstyles.org/downloads/specification.html#locale-fallback . 
Note that locale codes can be set at various levels in CSL: there is the 
user supplied locale, the style locale, the locale files, and the (zero or 
more) locale elements within the styles. IIRC, in Zotero, the user supplied 
locale doesn't override the style locale. The CSL specification doesn't 
currently discuss explicitly how "eu" and "ar" locale files should work. I 
think they should function as "primary dialects" (in the jargon of the CSL 
specification). If the user supplied locale is country-specific (e.g. 
"ar-EG" for Egyptian Arabic), and "ar-EG" and "ar" locale files are 
available, "ar-EG" should probably be used with "ar" as fallback. If the 
user supplied locale is "ar", the "ar-EG" locale file should probably be 
skipped in favor of the "ar" locale file.

I don't think such complicated cases have actually come up yet, though.
 

> > With every CSL update that affected the locale files, I prepopulated 
> > new terms with the translations from en-US. See e.g. 
> > 
> https://github.com/citation-style-language/locales/commit/0374188ca0784c1d5cde3d1a2655897b9f4c8292 
> > for the CSL 1.0.1 update. So sometimes these new terms aren't 
> > translated yet. 
>
> At the end of this piece I'll propose lots of missing token-translation, 
> a few changes and some new languages/locale to Rintze. 
>

Can you do that at 
https://lists.sourceforge.net/lists/listinfo/xbiblio-devel ? That seems the 
best avenue.
 

> >>> Finally, I think it would be a good idea to include, in the header of 
> locale 
> >>> files, information on what exact language they contain, plus, if 
> available, 
> >>> contact details of authors or maintainers. 
>
> I second Nick's request to have the name of the author IN the file, I 
> know I can always go through the entire repository and check who did 
> what, but that is pretty hard for most people. 
>
> IN fact, I go further and say that for that file to use the CC license, 
> that is currently using - it is required that the name of the author be 
> present on the file. I know it is legalese - but so is the license. 
> Attributions in the work have been required in CC licenses since 2004.
>

Translators are free to include themselves in the locale file metadata, and 
I mention the possibility in our documentation: 
http://citation-style-language.readthedocs.org/en/latest/translating-locale-files.html#locale-file-metadata

I don't see it as my job to request that every translator does so, though, 
or to track down past contributors and include them as translators. There 
is always the git commit trail.
 

> > CSL 1.0.1 introduced the ability to specify translators. See 
> > 
> http://citationstyles.org/downloads/release-notes-csl101.html#locale-metadata 
> > . The only place where we currently link language tags to language 
> > names is 
> https://github.com/citation-style-language/locales/blob/master/locales.json 
> > . Is that enough? 
> > 
> > Rintze 
>
> It would be nice if we could place the ISO language code as one of the 
> columns in that file, in this way it would make easier for people to 
> layout the data in their own db.


Don't the JSON file and the locale files already list the ISO codes? 

Rintze

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/9d91d79f-8fe1-4eb6-b0d9-31312756c6d7%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 6474 bytes --]

Re: CSL locale files

[Paulo Ney de Souza]

[-- Attachment #1: Type: text/plain, Size: 5795 bytes --]

  Rintze,

Thanks! My understanding of the whole process is getting a bit clear now.

On the last issue, the code for the language in the json file, I should 
have said the "3-letter" ISO code. The 2-letter code has been deprecated 
now, because it is not enough to discern the 8K languages out there and  
not even the Wikipedia languages - Cebuano, Minangkabau and Waray are 
among the top 50 Wikipedia languages that do not have 2-letter codes 
assigned to them - so anyone working with a DB of languages is more 
likely to have it indexed by the 3-letter codes.

I have the list in my DB if you want.

Paulo Ney



On 10/3/2013 10:55 AM, Rintze Zelle wrote:
> On Thursday, October 3, 2013 12:57:09 PM UTC-4, Paulo Ney de Souza wrote:
>
>     > Agreed. I prefer to include country codes when possible, but
>     since MSA
>     > doesn't have a single originating country, "locales-ar.xml" makes
>     > sense. And the "AR" is clearly wrong. I nowadays often check
>     language
>     > tags at http://rishida.net/utils/subtags/
>     <http://rishida.net/utils/subtags/> , but "ar-AR" must have
>     > slipped by.
>
>     I am wondering how is that supposed to work at the user level, since
>     while it is possible that some user will have locale set to "eu" (for
>     separatist reasons) - it is quite unlikely that someone will have
>     it set
>     to "ar" alone.
>     Will it pick up all locales set to ar-* and load the file
>     "locales-ar.xml".
>
>
> There are some pointers on this at 
> http://citationstyles.org/downloads/specification.html#locale-fallback . 
> Note that locale codes can be set at various levels in CSL: there is 
> the user supplied locale, the style locale, the locale files, and the 
> (zero or more) locale elements within the styles. IIRC, in Zotero, the 
> user supplied locale doesn't override the style locale. The CSL 
> specification doesn't currently discuss explicitly how "eu" and "ar" 
> locale files should work. I think they should function as "primary 
> dialects" (in the jargon of the CSL specification). If the user 
> supplied locale is country-specific (e.g. "ar-EG" for Egyptian 
> Arabic), and "ar-EG" and "ar" locale files are available, "ar-EG" 
> should probably be used with "ar" as fallback. If the user supplied 
> locale is "ar", the "ar-EG" locale file should probably be skipped in 
> favor of the "ar" locale file.
>
> I don't think such complicated cases have actually come up yet, though.
>
>     > With every CSL update that affected the locale files, I
>     prepopulated
>     > new terms with the translations from en-US. See e.g.
>     >
>     https://github.com/citation-style-language/locales/commit/0374188ca0784c1d5cde3d1a2655897b9f4c8292
>     <https://github.com/citation-style-language/locales/commit/0374188ca0784c1d5cde3d1a2655897b9f4c8292>
>
>     > for the CSL 1.0.1 update. So sometimes these new terms aren't
>     > translated yet.
>
>     At the end of this piece I'll propose lots of missing
>     token-translation,
>     a few changes and some new languages/locale to Rintze.
>
>
> Can you do that at 
> https://lists.sourceforge.net/lists/listinfo/xbiblio-devel ? That 
> seems the best avenue.
>
>     >>> Finally, I think it would be a good idea to include, in the
>     header of locale
>     >>> files, information on what exact language they contain, plus,
>     if available,
>     >>> contact details of authors or maintainers.
>
>     I second Nick's request to have the name of the author IN the file, I
>     know I can always go through the entire repository and check who did
>     what, but that is pretty hard for most people.
>
>     IN fact, I go further and say that for that file to use the CC
>     license,
>     that is currently using - it is required that the name of the
>     author be
>     present on the file. I know it is legalese - but so is the license.
>     Attributions in the work have been required in CC licenses since 2004.
>
>
> Translators are free to include themselves in the locale file 
> metadata, and I mention the possibility in our documentation: 
> http://citation-style-language.readthedocs.org/en/latest/translating-locale-files.html#locale-file-metadata
>
> I don't see it as my job to request that every translator does so, 
> though, or to track down past contributors and include them as 
> translators. There is always the git commit trail.
>
>     > CSL 1.0.1 introduced the ability to specify translators. See
>     >
>     http://citationstyles.org/downloads/release-notes-csl101.html#locale-metadata
>     <http://citationstyles.org/downloads/release-notes-csl101.html#locale-metadata>
>
>     > . The only place where we currently link language tags to language
>     > names is
>     https://github.com/citation-style-language/locales/blob/master/locales.json
>     <https://github.com/citation-style-language/locales/blob/master/locales.json>
>
>     > . Is that enough?
>     >
>     > Rintze
>
>     It would be nice if we could place the ISO language code as one of
>     the
>     columns in that file, in this way it would make easier for people to
>     layout the data in their own db.
>
>
> Don't the JSON file and the locale files already list the ISO codes?
>
> Rintze

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/524DB89B.1090401%40gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 9440 bytes --]

Re: CSL locale files

[Rintze Zelle]

On Thu, Oct 3, 2013 at 2:34 PM, Paulo Ney de Souza <pauloney-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> On the last issue, the code for the language in the json file, I should have
> said the "3-letter" ISO code. The 2-letter code has been deprecated now,
> because it is not enough to discern the 8K languages out there and  not even
> the Wikipedia languages - Cebuano, Minangkabau and Waray are among the top
> 50 Wikipedia languages that do not have 2-letter codes assigned to them - so
> anyone working with a DB of languages is more likely to have it indexed by
> the 3-letter codes.

I think we settled for 2-letter codes for CSL back in the day since
that matched the locales as provided by Firefox, and hence, Zotero.
But I can include the 3-letter variants in the JSON file. Just not
sure about the cleanest way to do that in JSON.

Rintze

Re: CSL locale files

[BP Jonsson]

2013-10-03 18:57, Paulo Ney de Souza skrev:
> I am wondering how is that supposed to work at the user level,
> since while it is possible that some user will have locale set to
> "eu" (for separatist reasons) - it is quite unlikely that someone
> will have it set to "ar" alone.
> Will it pick up all locales set to ar-* and load the file
> "locales-ar.xml".

The supposition that country and language are always coupled, and
that the preferences of all speakers of the same language in the
same contry, or for individual speakers at all times, are the same
is only the most obvious way the locale concept is broken. See
<http://www.i18nguy.com/locales/iuc22-locales.html> and make sure
to follow these links at the bottom:
<http://www.i18nguy.com/locales/index.html>
<http://www.i18nguy.com/locales/Locales.pdf>

For example four different styles of quotation, three of them in
two variants making a total of seven, have been used in Swedish
scholarly works during the last 150 years or so, and when citing
older works one is faced with the choice to preserve the original
style or 'updating' it. The problem is that the decision may
often be reversed/changed during the editing process, and the
poor editor (me) must be prepared for that, and I have to jump
backwards through hoops to get things working in pandoc/LaTeX.
It's *not* a matter of simply globally replacing some characters,
since the meaning of double quotes is different in different
styles -- it is outer quotes in the modern style but was inner
quotes in some of the older styles -- and usually the modern style
and perhaps several of the older ones co-occur in the same pages.
Add to that that linguists usually cite meanings in single quotes
regardless of the style they use for actual quotations!

The seven different styles in case you wonder, and both the top
ones are still in use, and even the third is sometimes seen,
though not listed in the style guides!

     ”outer ’and inner’ quotes”

     »outer ’and inner’ quotes»

     »outer ’and inner’ quotes«

     «outer ’and inner’ quotes»

     »outer ”and inner” quotes»

     »outer ”and inner” quotes«

     «outer ”and inner” quotes»

Imagine three or four of them in a document plus some citations
in German, French and English!  I deal with that regularly.
Even if pandoc should come to support locales or variables
for deciding how to render its DoubleQuote and SingleQuote
elements that wouldn't quite cut it for me!

/bpj

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/524DB94F.3080309%40gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1711 bytes --]

I feel bibtex2pandoc should split literal lists into their components, and 
pandoc-citeproc should be modified to accept arrays here.

I think this becomes clear where considering this: Whereas for one 
publisher with a literal "and" in its name, such as "Holt, Rinehart {and} 
Winston" the "and" must always be retained in the output as is, for two 
different publishers such as "Univ. of Toronto Press and Routledge" the 
output will have to be modified according to the various styles' rules 
(could become "Univ. of Toronto Press; Routledge", or just the first 
publisher "Univ. of Toronto Press") and/or language (in French, "Univ. of 
Toronto Press et Routledge"). All of this won't work if bibtex2pandoc 
doesn't split literal lists first.

Still we'd better double check the CSL specs and how other citeprocs handle 
this.

On Saturday, September 28, 2013 7:34:09 PM UTC, fiddlosopher wrote:
>
>
> The literal {and} will not cause breaking of author lists or literal 
> lists. 
>
> I'll leave the multiple publishers question open for now.  Either 
> bibtex2pandoc or pandoc-citeproc should be changed, not sure which. 
>

 

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/7ddad469-30dd-47b3-a18c-8fad8c5d25c4%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2143 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1924 bytes --]



On Saturday, September 28, 2013 7:16:16 AM UTC, fiddlosopher wrote:
>
> ...
> >    Yes, but this is about, e.g., pubstate={inpress}, where the "inpress" 
> >    should be transformed to 'in press'. 
>
> So how does it work? Is it just a list of strings that are to be 
> expanded macro-like before further processing? 
>

Yes, basically, but they are used in biblatex for localization at the same 
time; see biblatex manual (v2.7a): "4.9.2 Localization Keys". 
If CSL has corresponding "terms" (see 
http://citationstyles.org/downloads/specification.html#miscellaneous), I 
assume these should be used, and CSL can take care of localization.
If not – in the case of "status" the only "terms" available seem to be 
"forthcoming" and "in press" – I guess the only workaround for now is to 
use (natural) English language terms, in the case of pubstate/status: 
inpreparation -> "in preparation";  prepublished -> "pre-published"; 
"submitted" of course as is.

Other keys that appear in database fields include type fields of @thesis, 
@report, @misc, and others; e.g., mathesis, phdthesis, candthesis, 
techreport, resreport, software, datacd, audiocd.

All these would look decidedly odd in the output if copied verbatim, so I 
feel the effort of having bibtex2pandoc somehow translate these is worth 
it. 

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a784b5d3-6731-4b36-86e0-bed7db57babc%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2352 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[Joseph Reagle]

On 08/25/2013 03:15 PM, John MacFarlane wrote:
> Pandoc currently includes built-in support for citation processing
> (using citeproc-hs) and syntax highlighting (using highlighting-kate).

I'm in a bit of a pickle, I need something after 1.11.1 for my slides,
but haven't adapted to the new citation processing. Is there a coherent
version of pandoc and its dependencies that supports reveal.js natively
but is before the citation switch?

> and specify the bibliography and csl file in the document's metadata.

Is the '---' preceding a YAML section in your examples necessary to
designate a metadata block?

Could you attach a minimal example and the exact command line?

Re: Decoupling citeproc and highlighting-kate from pandoc

[Joseph Reagle]

[-- Attachment #1: Type: text/plain, Size: 1048 bytes --]

On 09/09/2013 04:42 PM, Joseph Reagle wrote:
> Could you attach a minimal example and the exact command line?

For instance, with the attached, I do know the filter is kicking in
(because it throws a file not found error if I don't specify the '.csl'
extension) but I get no bibliography...? (I'm using the
chicago-author-date.csl from git tests).

~~~~

urd:~/data/2web/reagle.org/joseph/2013/test$ pandoc -s --filter
pandoc-citeproc small.md
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
  <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
  <meta http-equiv="Content-Style-Type" content="text/css" />
  <meta name="generator" content="pandoc" />
  <title>Test</title>
  <style type="text/css">code{white-space: pre;}</style>
</head>
<body>
<div id="header">
<h1 class="title">Test</h1>
</div>
<p>Some claim [fenner12a].</p>
<div class="references">
<h1>References</h1>
</div>
</body>
</html>

~~~~

Re: Decoupling citeproc and highlighting-kate from pandoc

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1603 bytes --]

This file, test1.md

~~~~
% A Test

# Main

Some claim [@item1].

# References

---
references:
- author:
  - family: Doe
    given:
    - John
  id: item1
  issued:
    year: 2005
  publisher: Cambridge University Press
  publisher-place: Cambridge
  title: First Book
  type: book
...
~~~~

when converted with
pandoc -s --filter pandoc-citeproc -t html test1.md
or
pandoc -s --filter pandoc-citeproc -t latex test1.md
produces the expected output.

Using
pandoc -s --filter pandoc-citeproc -t plain test1.md
produces no references:

~~~~
A Test

Main
====

Some claim (Doe 2005).
~~~~

And 
pandoc -s --filter pandoc-citeproc -t markdown test1.md
produces

~~~~
---
references:
- author:
  - family: Doe
    given:
    - John
  id: item1
  issued:
    year: 2005
  publisher: Cambridge University Press
  publisher-place: Cambridge
  title: First Book
  type: book
title: A Test
...

Main
====

Some claim [@item1].
~~~~

... but is there any way to obtain a markdown-formatted file in which the 
references are resolved?

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/38cf4434-b15b-4ad1-9423-869182709bf3%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2245 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

You can do

pandoc -t markdown-citations

if you want references to be resolved in markdown output.

As for plain, it may be a bug.


+++ Nick Bart [Sep 12 13 02:09 ]:
>    This file, test1.md
>    ~~~~
>    % A Test
>    # Main
>    Some claim [@item1].
>    # References
>    ---
>    references:
>    - author:
>      - family: Doe
>        given:
>        - John
>      id: item1
>      issued:
>        year: 2005
>      publisher: Cambridge University Press
>      publisher-place: Cambridge
>      title: First Book
>      type: book
>    ...
>    ~~~~
>    when converted with
>    pandoc -s --filter pandoc-citeproc -t html test1.md
>    or
>    pandoc -s --filter pandoc-citeproc -t latex test1.md
>    produces the expected output.
>    Using
>    pandoc -s --filter pandoc-citeproc -t plain test1.md
>    produces no references:
>    ~~~~
>    A Test
>    Main
>    ====
>    Some claim (Doe 2005).
>    ~~~~
>    And
>    pandoc -s --filter pandoc-citeproc -t markdown test1.md
>    produces
>    ~~~~
>    ---
>    references:
>    - author:
>      - family: Doe
>        given:
>        - John
>      id: item1
>      issued:
>        year: 2005
>      publisher: Cambridge University Press
>      publisher-place: Cambridge
>      title: First Book
>      type: book
>    title: A Test
>    ...
>    Main
>    ====
>    Some claim [@item1].
>    ~~~~
>    ... but is there any way to obtain a markdown-formatted file in which
>    the references are resolved?
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/38cf4434-b15b-4ad1-
>    9423-869182709bf3%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/38cf4434-b15b-4ad1-9423-869182709bf3%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

plain output is fixed now.

+++ Nick Bart [Sep 12 13 02:09 ]:
>    This file, test1.md
>    ~~~~
>    % A Test
>    # Main
>    Some claim [@item1].
>    # References
>    ---
>    references:
>    - author:
>      - family: Doe
>        given:
>        - John
>      id: item1
>      issued:
>        year: 2005
>      publisher: Cambridge University Press
>      publisher-place: Cambridge
>      title: First Book
>      type: book
>    ...
>    ~~~~
>    when converted with
>    pandoc -s --filter pandoc-citeproc -t html test1.md
>    or
>    pandoc -s --filter pandoc-citeproc -t latex test1.md
>    produces the expected output.
>    Using
>    pandoc -s --filter pandoc-citeproc -t plain test1.md
>    produces no references:
>    ~~~~
>    A Test
>    Main
>    ====
>    Some claim (Doe 2005).
>    ~~~~
>    And
>    pandoc -s --filter pandoc-citeproc -t markdown test1.md
>    produces
>    ~~~~
>    ---
>    references:
>    - author:
>      - family: Doe
>        given:
>        - John
>      id: item1
>      issued:
>        year: 2005
>      publisher: Cambridge University Press
>      publisher-place: Cambridge
>      title: First Book
>      type: book
>    title: A Test
>    ...
>    Main
>    ====
>    Some claim [@item1].
>    ~~~~
>    ... but is there any way to obtain a markdown-formatted file in which
>    the references are resolved?
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/38cf4434-b15b-4ad1-
>    9423-869182709bf3%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/38cf4434-b15b-4ad1-9423-869182709bf3%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

bib-yaml on the command line (Was: Decoupling citeproc and highlighting-kate from pandoc)

[Joseph Reagle]

On 08/25/2013 03:15 PM, John MacFarlane wrote:
> Of course, you can also store the YAML bibliography in a separate
> file, and provide it to pandoc on the command line with the main
> file.

My bibliography tool is now generating remedial yaml, and biblio2yaml
converts to yaml, but how is it specified/provided on the commandline?
While json files work, the following doesn't:

`--bibliography=/home/reagle/joseph/readings.yaml`

Re: bib-yaml on the command line (Was: Decoupling citeproc and highlighting-kate from pandoc)

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1108 bytes --]

Another patch for bibtex2pandoc, to be found here: 
https://gist.github.com/anonymous/6811257

- added more biblatex entry types
- regrouped code for field mappings in loose categories: author/editor, 
titles, dates, publisher/location, etc.
- added (probably overly complicated and un-haskell-ish) hack for biblatex 
date and urldate; no support for date ranges yet
- fixed titles of @periodical
- various other small fixes

... and one new issue:
- pandoc-citeproc:
    - entries whose cite keys contain "--" are not recognized

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/a54e9afd-db7e-4ac1-91a4-4fb0f5e02b67%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1419 bytes --]

Re: bib-yaml on the command line (Was: Decoupling citeproc and highlighting-kate from pandoc)

[John MacFarlane]

>    Another patch for bibtex2pandoc, to be found here:
>    https://gist.github.com/anonymous/6811257

Merged!

The date hack is not good, because it will crash if the list of
date components doesn't contain all three elements.  But I can fix
that later.

>    ... and one new issue:
>    - pandoc-citeproc:
>        - entries whose cite keys contain "--" are not recognized

Example to reproduce?

Re: bib-yaml on the command line (Was: Decoupling citeproc and highlighting-kate from pandoc)

[Nick Bart]

[-- Attachment #1: Type: text/plain, Size: 1819 bytes --]

$ cat citekey-two-dashes.md 
Foo [@item--1].

Bar [@item-2].

# References

---
references:
- author:
  - family: Doe
    given:
    - John
  id: item--1
  issued:
    year: 2005
  title: First Book
  type: book
- author:
  - family: Doe
    given:
    - John
  id: item-2
  issued:
    year: 2005
  title: Second Book
  type: book
...


$ pandoc -t markdown-citations -F pandoc-citeproc citekey-two-dashes.md 
pandoc-citeproc: reference item not found
Foo (<span
class="citeproc-not-found"data-reference-id="item">**???**</span>).

Bar (Doe 2005).

<div class="references">

References
==========

Doe, John. 2005. *Second Book*.

</div>

Remove the second dash, and this works as expected.

Same effect with two underscores, "__", BTW.

On Friday, October 4, 2013 3:13:38 AM UTC, fiddlosopher wrote:
>
> >    Another patch for bibtex2pandoc, to be found here: 
> >    https://gist.github.com/anonymous/6811257 
>
> Merged! 
>
> The date hack is not good, because it will crash if the list of 
> date components doesn't contain all three elements.  But I can fix 
> that later. 
>
> >    ... and one new issue: 
> >    - pandoc-citeproc: 
> >        - entries whose cite keys contain "--" are not recognized 
>
> Example to reproduce? 
>
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/70a255af-aae0-4fae-90f9-9c6d95dae746%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 2715 bytes --]

Re: bib-yaml on the command line (Was: Decoupling citeproc and highlighting-kate from pandoc)

[John MacFarlane]

Pandoc's citation parser only allows one internal punctuation character
in a row for citation references.  I'm reluctant to relax this, for the
following reason:

    @item1---and to a certain extent @item2---are guilty of this
    fallacy.

Here you certainly don't want the key to be parsed as 'item1---and'.
Same goes for '--'.

+++ Nick Bart [Oct 03 13 22:31 ]:
>    $ cat citekey-two-dashes.md
>    Foo [@item--1].
>    Bar [@item-2].
>    # References
>    ---
>    references:
>    - author:
>      - family: Doe
>        given:
>        - John
>      id: item--1
>      issued:
>        year: 2005
>      title: First Book
>      type: book
>    - author:
>      - family: Doe
>        given:
>        - John
>      id: item-2
>      issued:
>        year: 2005
>      title: Second Book
>      type: book
>    ...
>    $ pandoc -t markdown-citations -F pandoc-citeproc citekey-two-dashes.md
>    pandoc-citeproc: reference item not found
>    Foo (<span
>    class="citeproc-not-found"data-reference-id="item">**???**</span>).
>    Bar (Doe 2005).
>    <div class="references">
>    References
>    ==========
>    Doe, John. 2005. *Second Book*.
>    </div>
>    Remove the second dash, and this works as expected.
>    Same effect with two underscores, "__", BTW.
>    On Friday, October 4, 2013 3:13:38 AM UTC, fiddlosopher wrote:
> 
>      >    Another patch for bibtex2pandoc, to be found here:
>      >    [1]https://gist.github.com/anonymous/6811257
>      Merged!
>      The date hack is not good, because it will crash if the list of
>      date components doesn't contain all three elements.  But I can fix
>      that later.
>      >    ... and one new issue:
>      >    - pandoc-citeproc:
>      >        - entries whose cite keys contain "--" are not recognized
>      Example to reproduce?
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [2]https://groups.google.com/d/msgid/pandoc-discuss/70a255af-aae0-4fae-
>    90f9-9c6d95dae746%40googlegroups.com.
>    For more options, visit [3]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://gist.github.com/anonymous/6811257
>    2. https://groups.google.com/d/msgid/pandoc-discuss/70a255af-aae0-4fae-90f9-9c6d95dae746%40googlegroups.com
>    3. https://groups.google.com/groups/opt_out

Re: bib-yaml on the command line (Was: Decoupling citeproc and highlighting-kate from pandoc)

[Tillmann Rendel]

Hi,

John MacFarlane wrote:
> Pandoc's citation parser only allows one internal punctuation character
> in a row for citation references.  I'm reluctant to relax this, for the
> following reason:
>
>      @item1---and to a certain extent @item2---are guilty of this
>      fallacy.
>
> Here you certainly don't want the key to be parsed as 'item1---and'.
> Same goes for '--'.

Could this depend on whether there is a "item1---and" reference key 
available in the bibliography or not? That is, the parsing logic would 
be something like:

   referenceKeyOrExample :: Parser Inline
   referenceKeyOrExample = try $ do
     char '@'
     key <- parseKeyPermissive
     if key `elem` allReferenceKeys
       then return (Reference key)
       else if notTerriblyUnlikely key
         then return (Example key)
         else pzero

If I remember correctly, parsing already depends on the set of available 
reference keys, so this might be possible, too.

   Tillmann

Re: bib-yaml on the command line (Was: Decoupling citeproc and highlighting-kate from pandoc)

[John MacFarlane]

You can either embed the YAML section into the markdown file itself,
or specify it as a separate input file.

    pandoc input.txt readings.yaml -F pandoc-citeproc

+++ Joseph Reagle [Oct 03 13 10:17 ]:
> On 08/25/2013 03:15 PM, John MacFarlane wrote:
> > Of course, you can also store the YAML bibliography in a separate
> > file, and provide it to pandoc on the command line with the main
> > file.
> 
> My bibliography tool is now generating remedial yaml, and biblio2yaml
> converts to yaml, but how is it specified/provided on the commandline?
> While json files work, the following doesn't:
> 
> `--bibliography=/home/reagle/joseph/readings.yaml`
> 
> -- 
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/524D7C6A.7070704%40reagle.org.
> For more options, visit https://groups.google.com/groups/opt_out.

Re: bib-yaml on the command line (Was: Decoupling citeproc and highlighting-kate from pandoc)

[Joseph Reagle]

pandoc (1.12.1) is now able to take my (very large) bibliography
(converted with biblio2yaml) without throwing a parsing error.

However, it is still reporting errors (e.g., "pandoc-citeproc: reference
danieltosh not found") when the `@` character appears in the title. Does
this need to be escaped?

~~~~

- url: https://twitter.com/louisck/status/222849861224108035
  author:
    family: C.K.
    given:
    - Louis
  id: CK2012dys
  accessed:
    day: 28
    year: 2013
    month: 08
  issued:
    day: 10
    year: 2012
    month: 07
  title: ! '@danieltosh your show makes ...'
  type: webpage
  publisher: Twitter

~~~~

Re: bib-yaml on the command line (Was: Decoupling citeproc and highlighting-kate from pandoc)

[Joseph Reagle]

Hello John, I'm returning to this citeproc/yaml stuff after a few months.

I'm now generating valid yaml (according to pyyaml at least) but pandoc
(1.12.3.3) is choking on the large bibliography (sent separately). I did
some bisecting but couldn't suss it out. Fortunately, in practice I
don't throw the whole bib at pandoc because of speed issues (same with
the bibtex) and subset the yaml first based on the keys in the document,
but I still find that parsing error odd. The error I'm getting is:
"pandoc-citeproc: Error parsing references: Could not read `PostWebblog'"

On 10/03/2013 11:45 AM, John MacFarlane wrote:
> You can either embed the YAML section into the markdown file itself,
> or specify it as a separate input file.
> 
>     pandoc input.txt readings.yaml -F pandoc-citeproc

Also, as I transition from bibtex to yaml, it'd be handy if there were
parallel syntaxes.  Any chance  you could support
`--bibliography=readings.yaml`

Re: Decoupling citeproc and highlighting-kate from pandoc

[andrea rossato]

John MacFarlane <fiddlosopher-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:

> I'm pretty excited about these changes, which I would like to include
> in the 1.12 release.  The release has been held up waiting for Andrea
> to release citeproc-hs.  I haven't heard from Andrea in a long time, so
> I now have a contingency plan.  I will simply include the code from
> citeproc-hs, including a couple of bug fixes, in the pandoc-citations
> package.  When Andrea comes back, as I hope he does, I'll remove this
> and depend once more on citeproc-hs.  But at least this will let us
> go ahead with the release.

I'm deeply sorry I've disappeared for the last 2/3 months but I had a
long series of personal troubles that kept me away from coding and left
me with no time to even read my mail (I could came back to work full
time two weeks ago but I still have to catch up with all the stuff I
left behind me starting from July).

I've just released citeproc-hs-0.3.9 with quite a few bug fixes but I
understand, by just having a brief look at this thread, that a lot has
been done and probably this release is already outdated. I did not even
try to compile pandoc-1.12 (I'm not using cabal-install and the process
of manually updating all the needed packages will take quite some time).
So I do not yet know what I have to do to update my code to work with
the new pandoc infrastructure.

Could someone please send me a (private) message to summarize what's new
and what is expected on my side? I'll try to catch up as soon as
possible.

With reference to the MODS parser: this was meant to be a quick hack in
order to use bibutils -- which in turn is far from being perfect -- and
let people stick with their favorite bibliographic database format.
Still I had the hope that the presence of the CSL JSON data model --
which is natively read by citeproc-hs to instantiate every possible CSL
variable -- together with the possibility to combine databases in
different format gave the users enough flexibility to us pandoc/citeproc
for producing well formatted references and bibliographies out of the
box and without post-processing the pandoc output -- this is working for
me. This is the reason why my priority, given the limited resources I
can invest in this project, is the fixing of the CSL implementation. If
I understand it correctly you are working on a native Haskell bib(la)tex
parses. If this is correct I believe that the CSL JSON should be the
target.

Best,
Andrea

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ andrea rossato [Oct 03 13 19:28 ]:
> John MacFarlane <fiddlosopher-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:
> 
> > I'm pretty excited about these changes, which I would like to include
> > in the 1.12 release.  The release has been held up waiting for Andrea
> > to release citeproc-hs.  I haven't heard from Andrea in a long time, so
> > I now have a contingency plan.  I will simply include the code from
> > citeproc-hs, including a couple of bug fixes, in the pandoc-citations
> > package.  When Andrea comes back, as I hope he does, I'll remove this
> > and depend once more on citeproc-hs.  But at least this will let us
> > go ahead with the release.
> 
> I'm deeply sorry I've disappeared for the last 2/3 months but I had a
> long series of personal troubles that kept me away from coding and left
> me with no time to even read my mail (I could came back to work full
> time two weeks ago but I still have to catch up with all the stuff I
> left behind me starting from July).
> 
> I've just released citeproc-hs-0.3.9 with quite a few bug fixes but I
> understand, by just having a brief look at this thread, that a lot has
> been done and probably this release is already outdated. I did not even
> try to compile pandoc-1.12 (I'm not using cabal-install and the process
> of manually updating all the needed packages will take quite some time).
> So I do not yet know what I have to do to update my code to work with
> the new pandoc infrastructure.
> 
> Could someone please send me a (private) message to summarize what's new
> and what is expected on my side? I'll try to catch up as soon as
> possible.
> 
> With reference to the MODS parser: this was meant to be a quick hack in
> order to use bibutils -- which in turn is far from being perfect -- and
> let people stick with their favorite bibliographic database format.
> Still I had the hope that the presence of the CSL JSON data model --
> which is natively read by citeproc-hs to instantiate every possible CSL
> variable -- together with the possibility to combine databases in
> different format gave the users enough flexibility to us pandoc/citeproc
> for producing well formatted references and bibliographies out of the
> box and without post-processing the pandoc output -- this is working for
> me. This is the reason why my priority, given the limited resources I
> can invest in this project, is the fixing of the CSL implementation. If
> I understand it correctly you are working on a native Haskell bib(la)tex
> parses. If this is correct I believe that the CSL JSON should be the
> target.

I've provided ToJSON and FromJSON (aeson) instances for Reference
and its associated types in Text.CSL.Reference.  That allows one
to convert between JSON (or equivalent YAML) and [Reference].
There are probably still issues with these instances, and the
Reference structure may also need some work.  But it seems to me
that if we target [Reference], we can get the JSON for free...

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

You want

[@fenner12a]

not

[fenner12a]

+++ Joseph Reagle [Sep 09 13 16:54 ]:
> On 09/09/2013 04:42 PM, Joseph Reagle wrote:
> > Could you attach a minimal example and the exact command line?
> 
> For instance, with the attached, I do know the filter is kicking in
> (because it throws a file not found error if I don't specify the '.csl'
> extension) but I get no bibliography...? (I'm using the
> chicago-author-date.csl from git tests).
> 
> ~~~~
> 
> urd:~/data/2web/reagle.org/joseph/2013/test$ pandoc -s --filter
> pandoc-citeproc small.md
> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
> "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
> <html xmlns="http://www.w3.org/1999/xhtml">
> <head>
>   <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
>   <meta http-equiv="Content-Style-Type" content="text/css" />
>   <meta name="generator" content="pandoc" />
>   <title>Test</title>
>   <style type="text/css">code{white-space: pre;}</style>
> </head>
> <body>
> <div id="header">
> <h1 class="title">Test</h1>
> </div>
> <p>Some claim [fenner12a].</p>
> <div class="references">
> <h1>References</h1>
> </div>
> </body>
> </html>
> 
> ~~~~
> 

> % Test
> 
> Some claim [fenner12a]. 
> 
> # References
> 
> --- 
> csl: chicago-author-date.csl
> references: 
> - id: fenner12a 
>   title: One-click science marketing 
>   author: 
>    family: Fenner 
>    given: [Martin] 
>   container-title: Nature Materials 
>   volume: 11 
>   url: 'http://dx.doi.org/10.1038/nmat3283' 
>   doi: '10.1038/nmat3283' 
>   issue: 4 
>   publisher: Nature Publishing Group 
>   page: 261-263 
>   type: article-journal 
>   issued: 
>     year: 2012 
>     month: 3 
> ---

----- End forwarded message -----

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ Joseph Reagle [Sep 09 13 16:42 ]:
> On 08/25/2013 03:15 PM, John MacFarlane wrote:
> > Pandoc currently includes built-in support for citation processing
> > (using citeproc-hs) and syntax highlighting (using highlighting-kate).
> 
> I'm in a bit of a pickle, I need something after 1.11.1 for my slides,
> but haven't adapted to the new citation processing. Is there a coherent
> version of pandoc and its dependencies that supports reveal.js natively
> but is before the citation switch?

Yes.  Try the version before the citation switch.  You'll need to
install citeproc-hs from darcs.

> > and specify the bibliography and csl file in the document's metadata.
> 
> Is the '---' preceding a YAML section in your examples necessary to
> designate a metadata block?

Yes.

> Could you attach a minimal example and the exact command line?

Note, in the latest versions, you can use --bibliography and --csl
as before to specify these things, if you don't want to put them in
the document's metadata.   So your old workflow should work fine if
you just add '--filter pandoc-citeproc' to your command line, and
make sure you've installed pandoc-citeproc.

Re: Decoupling citeproc and highlighting-kate from pandoc

[David Sanson]

[-- Attachment #1: Type: text/plain, Size: 1350 bytes --]


>
> Note, in the latest versions, you can use --bibliography and --csl
> as before to specify these things, if you don't want to put them in
> the document's metadata.   So your old workflow should work fine if
> you just add '--filter pandoc-citeproc' to your command line, and
> make sure you've installed pandoc-citeproc.
>

Is there a use case (present or future) that requires `--bibliography` but 
not `--filter pandoc-citeproc`? If not, perhaps specifying `--bibliography` 
on the command line (but not in the metadata) should imply `--filter 
pandoc-citeproc`?

This would make for a shorter command line for people using the 
`--bibliography` option. It would also provide backward compatibility for 
scripts/makefiles/etc. that use the `--bibliography` option.

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/3456a162-43cb-42aa-b2de-6a0d3793e877%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[-- Attachment #2: Type: text/html, Size: 1781 bytes --]

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ David Sanson [Sep 10 13 07:06 ]:
>      Note, in the latest versions, you can use --bibliography and --csl
>      as before to specify these things, if you don't want to put them in
>      the document's metadata.   So your old workflow should work fine if
>      you just add '--filter pandoc-citeproc' to your command line, and
>      make sure you've installed pandoc-citeproc.
> 
>    Is there a use case (present or future) that requires `--bibliography`
>    but not `--filter pandoc-citeproc`? If not, perhaps specifying
>    `--bibliography` on the command line (but not in the metadata) should
>    imply `--filter pandoc-citeproc`?

I tried that, briefly.  The problem is that if the user ALSO specifies
--filter pandoc-citeproc on the command line, it will run the filter
twice (since --filter arguments can be repeated).  That isn't desirable.
But maybe there's another way of implementing the idea that wouldn't
have this effect.

>    This would make for a shorter command line for people using the
>    `--bibliography` option. It would also provide backward compatibility
>    for scripts/makefiles/etc. that use the `--bibliography` option.
> 
>    --
>    You received this message because you are subscribed to the Google
>    Groups "pandoc-discuss" group.
>    To unsubscribe from this group and stop receiving emails from it, send
>    an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To post to this group, send email to pandoc-discuss-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
>    To view this discussion on the web visit
>    [1]https://groups.google.com/d/msgid/pandoc-discuss/3456a162-43cb-42aa-
>    b2de-6a0d3793e877%40googlegroups.com.
>    For more options, visit [2]https://groups.google.com/groups/opt_out.
> 
> References
> 
>    1. https://groups.google.com/d/msgid/pandoc-discuss/3456a162-43cb-42aa-b2de-6a0d3793e877%40googlegroups.com
>    2. https://groups.google.com/groups/opt_out

Re: Decoupling citeproc and highlighting-kate from pandoc

[John MacFarlane]

+++ David Sanson [Sep 10 13 07:06 ]:
>      Note, in the latest versions, you can use --bibliography and --csl
>      as before to specify these things, if you don't want to put them in
>      the document's metadata.   So your old workflow should work fine if
>      you just add '--filter pandoc-citeproc' to your command line, and
>      make sure you've installed pandoc-citeproc.
> 
>    Is there a use case (present or future) that requires `--bibliography`
>    but not `--filter pandoc-citeproc`? If not, perhaps specifying
>    `--bibliography` on the command line (but not in the metadata) should
>    imply `--filter pandoc-citeproc`?
>    This would make for a shorter command line for people using the
>    `--bibliography` option. It would also provide backward compatibility
>    for scripts/makefiles/etc. that use the `--bibliography` option.

I've made it work this way, so everything should be nicely backward
compatible.

Re: Decoupling citeproc and highlighting-kate from pandoc

[Joseph Reagle]

On 09/09/2013 05:09 PM, John MacFarlane wrote:
> Note, in the latest versions, you can use --bibliography and --csl
> as before to specify these things, if you don't want to put them in
> the document's metadata.   So your old workflow should work fine if
> you just add '--filter pandoc-citeproc' to your command line, and
> make sure you've installed pandoc-citeproc.
I just pulled the latest and it looks like it'll work without  '--filter
pandoc-citeproc'?

However, it's not quite "as before" as "--csl" requires one to use the
'.csl' extension? (I omitted that in the past.)