issue with beta from 2019.11.29 21:47 in Windows

issue with beta from 2019.11.29 21:47 in Windows

[Pablo Rodriguez]

Dear list,

I have the following sample:

    \def\ThisOption{ab}
    \def\ThatOption{ábc}
    \starttext
    \executesystemcommand{contextjit --purgeall
--arguments="OptionThis={\ThisOption},OptionThat={\ThatOption}" second.tex}

    \contextversion
    \stoptext

The contents of second.tex read:

    \starttext
    \enablemode[\env{OptionThis}]
    \enablemode[\env{OptionThat}]
    This: \doifmodeelse{ab}{enabled}{disabled}.\par
    That: \doifmodeelse{ábc}{enabled}{disabled}.
    \stoptext

I use --arguments to pass modes to documents compiled via
\executesystemcommand.

Everything worked fine. This morning I updated ConTeXt at work (with
Win7) and modes with non-ASCII chars aren’t recognized.

Could anyone confirm the issue I’m describing in Windows?

Is there any ConTeXt command (or Lua function) that translates non-ASCII
chars to their ASCII values?

Many thanks for your help,

Pablo
--
http://www.ousia.tk
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Wolfgang Schuster]

Pablo Rodriguez schrieb am 02.12.2019 um 16:18:
> Dear list,
>
> I have the following sample:
>
>      \def\ThisOption{ab}
>      \def\ThatOption{ábc}
>      \starttext
>      \executesystemcommand{contextjit --purgeall
> --arguments="OptionThis={\ThisOption},OptionThat={\ThatOption}" second.tex}

Limited alternative (no rerun when second.pdf exists and by default the 
resulting PDF is loaded as image).

\typesetfile[second.tex][--arguments="OptionThis={\ThisOption},OptionThat={\ThatOption}"][object=no]

>      \contextversion
>      \stoptext
>
> The contents of second.tex read:
>
>      \starttext
>      \enablemode[\env{OptionThis}]
>      \enablemode[\env{OptionThat}]
>      This: \doifmodeelse{ab}{enabled}{disabled}.\par
>      That: \doifmodeelse{ábc}{enabled}{disabled}.
>      \stoptext
>
> I use --arguments to pass modes to documents compiled via
> \executesystemcommand.
>
> Everything worked fine. This morning I updated ConTeXt at work (with
> Win7) and modes with non-ASCII chars aren’t recognized.
>
> Could anyone confirm the issue I’m describing in Windows?

I get the same results with MkIV but LMTX works.

> Is there any ConTeXt command (or Lua function) that translates non-ASCII
> chars to their ASCII values?

Lua: characters.shaped(...)

Wolfgang

___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Hans Hagen]

On 12/2/2019 4:18 PM, Pablo Rodriguez wrote:
> Dear list,
> 
> I have the following sample:
> 
>      \def\ThisOption{ab}
>      \def\ThatOption{ábc}
>      \starttext
>      \executesystemcommand{contextjit --purgeall
> --arguments="OptionThis={\ThisOption},OptionThat={\ThatOption}" second.tex}
> 
>      \contextversion
>      \stoptext
> 
> The contents of second.tex read:
> 
>      \starttext
>      \enablemode[\env{OptionThis}]
>      \enablemode[\env{OptionThat}]
>      This: \doifmodeelse{ab}{enabled}{disabled}.\par
>      That: \doifmodeelse{ábc}{enabled}{disabled}.
>      \stoptext
> 
> I use --arguments to pass modes to documents compiled via
> \executesystemcommand.
> 
> Everything worked fine. This morning I updated ConTeXt at work (with
> Win7) and modes with non-ASCII chars aren’t recognized.
> 
> Could anyone confirm the issue I’m describing in Windows?
> 
> Is there any ConTeXt command (or Lua function) that translates non-ASCII
> chars to their ASCII values?
define ASCII .. there are no accented characters in ASCII

windows uses code pages so it all depends on what trickles down to the 
command processor; here you pass utf that then gets interpreted 
depending on your environment

as you use luajittex it depends on what the windows binary does (i 
remember reading that something was changed in the native windows 
binaries because it was needed/decided at the latex end) ... one reason 
more to never use non-ascii for critital stuff (lots of chicken egg 
issues there, kind of guaranteed fix this, break that)

anyway, in luametatex with lmtx we're (hopefully) code page neutral (as 
far as i could test; all utf8 and windows utf16) and we're not going to 
touch the default luatex internals like that

Hans

(windows 7 is kind of outdated so you can't expect someone to check that 
setup out)

-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Pablo Rodriguez]

On 12/2/19 6:01 PM, Wolfgang Schuster wrote:
> Pablo Rodriguez schrieb am 02.12.2019 um 16:18:
>>      \starttext
>>      \executesystemcommand{contextjit --purgeall
>> --arguments="OptionThis={\ThisOption},OptionThat={\ThatOption}" second.tex}
>
> Limited alternative (no rerun when second.pdf exists and by default the
> resulting PDF is loaded as image).
>
> \typesetfile[second.tex][--arguments="OptionThis={\ThisOption},OptionThat={\ThatOption}"][object=no]

Many thanks for your reply, Wolfgang.

This is an interesting way to avoid the issue.

>> Could anyone confirm the issue I’m describing in Windows?
>
> I get the same results with MkIV but LMTX works.

I’d love to switch to LMTX, but this isn’t an option for me. It has an
issue with some fonts (I already reported).

In this sample \ss and \cg fonts cannot be used, even invoking them by
file name:

    \definefontfamily[mainface][rm][TeX Gyre Heros]
    \definefontfamily[mainface][mm][TeX Gyre Termes Math]
    \definefontfamily[mainface][ss][Gill Sans MT][tf=file:GIL_____.TTF]
    \definefontfamily[mainface][tt][Cousine]
    \definefontfamily[mainface][cg][Arial Narrow][tf=file:ARIALN.TTF]
    \setupbodyfont[mainface]
    \startbuffer
    \ConTeXt\ is awesome!\par
    \stopbuffer
    \starttext
    \startTEXpage[offset=1em]
    \getbuffer
    \ss\getbuffer
    \tt\getbuffer
    \cg\getbuffer
    \stopTEXpage
    \stoptext

https://mailman.ntg.nl/pipermail/ntg-context/2019/094953.html
https://mailman.ntg.nl/pipermail/ntg-context/2019/095353.html

>> Is there any ConTeXt command (or Lua function) that translates non-ASCII
>> chars to their ASCII values?
>
> Lua: characters.shaped(...)

Many thanks for your help,

Pablo
--
http://www.ousia.tk
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Pablo Rodriguez]

On 12/2/19 6:05 PM, Hans Hagen wrote:
> [...]
> anyway, in luametatex with lmtx we're (hopefully) code page neutral (as
> far as i could test; all utf8 and windows utf16) and we're not going to
> touch the default luatex internals like that

Many thanks for your reply, Hans.

As written before, I would love to use LMTX, but using some Gill Sans
and Arial Narrow is mandatory for us.

For some strange reason (as reported in my previous message), LMTX isn’t
able to deal with them.

> (windows 7 is kind of outdated so you can't expect someone to check that
> setup out)

I only use Windows at work. All I wanted to know was whether Windows was
affected by this issue.

Many thanks for your help,

Pablo
--
http://www.ousia.tk
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Wolfgang Schuster]

Pablo Rodriguez schrieb am 02.12.2019 um 18:46:
>>> Could anyone confirm the issue I’m describing in Windows?
>> I get the same results with MkIV but LMTX works.
> I’d love to switch to LMTX, but this isn’t an option for me. It has an
> issue with some fonts (I already reported).
>
> In this sample \ss and \cg fonts cannot be used, even invoking them by
> file name:
>
>      \definefontfamily[mainface][rm][TeX Gyre Heros]
>      \definefontfamily[mainface][mm][TeX Gyre Termes Math]
>      \definefontfamily[mainface][ss][Gill Sans MT][tf=file:GIL_____.TTF]
>      \definefontfamily[mainface][tt][Cousine]
>      \definefontfamily[mainface][cg][Arial Narrow][tf=file:ARIALN.TTF]
>      \setupbodyfont[mainface]
>      \startbuffer
>      \ConTeXt\ is awesome!\par
>      \stopbuffer
>      \starttext
>      \startTEXpage[offset=1em]
>      \getbuffer
>      \ss\getbuffer
>      \tt\getbuffer
>      \cg\getbuffer
>      \stopTEXpage
>      \stoptext

LMTX finds the fonts and loads them but nothing appears in the final PDF.

\nopdfcompression

\starttext
\definedfont[file:arialn.ttf*default]Arial Narrow
\stoptext


A alternative to Arial Narrow is Arial Nova which has condensed variant 
and comes as part of the pan-european font pack in windows 10.

\startmkivmode

     \definefontfamily [arial-narrow] [ss] [Arial Narrow]

     \definefontfamily [gill-sans] [ss] [Gill Sans MT]

\stopmkivmode

\startlmtxmode

     \definefontfamily [arial-narrow] [ss] [Arial Nova]
         [tf=style:condensed,
          it=style:condenseditalic,
          bf=style:condensedbold,
          bi=style:condensedbolditalic]

     \definefontfamily [gill-sans] [ss] [Gill Sans Nova]

\stoplmtxmode

\starttext

regular \italic{italic} \bold{bold} \bolditalic{bolditalic}

\switchtobodyfont[arial-narrow]

regular \italic{italic} \bold{bold} \bolditalic{bolditalic}

\switchtobodyfont[gill-sans]

regular \italic{italic} \bold{bold} \bolditalic{bolditalic}

\stoptext

Wolfgang

___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Hans Hagen]

On 12/2/2019 6:51 PM, Pablo Rodriguez wrote:
> On 12/2/19 6:05 PM, Hans Hagen wrote:
>> [...]
>> anyway, in luametatex with lmtx we're (hopefully) code page neutral (as
>> far as i could test; all utf8 and windows utf16) and we're not going to
>> touch the default luatex internals like that
> 
> Many thanks for your reply, Hans.
> 
> As written before, I would love to use LMTX, but using some Gill Sans
> and Arial Narrow is mandatory for us.
> 
> For some strange reason (as reported in my previous message), LMTX isn’t
> able to deal with them.

arial looks ok here ... (2014 version windows) so maybe you should look 
for a new version than yours

>> (windows 7 is kind of outdated so you can't expect someone to check that
>> setup out)
> 
> I only use Windows at work. All I wanted to know was whether Windows was
> affected by this issue.
> 
> Many thanks for your help,
> 
> Pablo
> --
> http://www.ousia.tk
> ___________________________________________________________________________________
> If your question is of interest to others as well, please add an entry to the Wiki!
> 
> maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
> webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
> archive  : https://bitbucket.org/phg/context-mirror/commits/
> wiki     : http://contextgarden.net
> ___________________________________________________________________________________
> 


-- 

-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Pablo Rodriguez]

On 12/2/19 9:45 PM, Wolfgang Schuster wrote:
> [...]
> LMTX finds the fonts and loads them but nothing appears in the final PDF.
>
> \nopdfcompression
>
> \starttext
> \definedfont[file:arialn.ttf*default]Arial Narrow
> \stoptext

Many thanks for your reply, Wolfgang.

I’m afraid there is no font there.

Compiling the source above with MkIV generates a PDF document with the
object 1 0 obj, which contains a /Font dictionary including two /Type0
font dictionaries.

Compiling the source above with LMTX the resulting PDF document contains
a /Page dictionary (6 0 obj) with /Font dictionary within. This
dictionary refers to an existing /Type0 font dictionary (1 0 obj) and a
missing "2 0 obj".

I don’t know why "2 0 obj" isn’t generated, but this should be the
missing font.

> An alternative to Arial Narrow is Arial Nova which has condensed variant
> and comes as part of the pan-european font pack in windows 10.

I’m afraid a new font is not an option for me at work.

Many thanks for your help,

Pablo
--
http://www.ousia.tk
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Hans Hagen]

On 12/3/2019 8:37 PM, Pablo Rodriguez wrote:
> On 12/2/19 9:45 PM, Wolfgang Schuster wrote:
>> [...]
>> LMTX finds the fonts and loads them but nothing appears in the final PDF.
>>
>> \nopdfcompression
>>
>> \starttext
>> \definedfont[file:arialn.ttf*default]Arial Narrow
>> \stoptext
> 
> Many thanks for your reply, Wolfgang.
> 
> I’m afraid there is no font there.
> 
> Compiling the source above with MkIV generates a PDF document with the
> object 1 0 obj, which contains a /Font dictionary including two /Type0
> font dictionaries.
> 
> Compiling the source above with LMTX the resulting PDF document contains
> a /Page dictionary (6 0 obj) with /Font dictionary within. This
> dictionary refers to an existing /Type0 font dictionary (1 0 obj) and a
> missing "2 0 obj".
> 
> I don’t know why "2 0 obj" isn’t generated, but this should be the
> missing font.

I don't have that font. The version I have here works ok.

>> An alternative to Arial Narrow is Arial Nova which has condensed variant
>> and comes as part of the pan-european font pack in windows 10.
> 
> I’m afraid a new font is not an option for me at work.
But there are better (unicode) versions available on windows, why nmot 
use those then (as wolfgang pointed out).

Hans


-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Pablo Rodriguez]

On 12/3/19 10:14 PM, Hans Hagen wrote:
> On 12/3/2019 8:37 PM, Pablo Rodriguez wrote:
>> [...]
>> I’m afraid a new font is not an option for me at work.
> But there are better (unicode) versions available on windows, why nmot
> use those then (as wolfgang pointed out).

Many thanks for your reply, Hans.

I would use the Unicode versions at home (if I had a Windows license),
but at work it is problematic even to ask for free fonts.

My company may be to strict with these policies, but I don’t think it is
a total exception in this matter.

I hope we might switch to Windows 10 in the no so near future 😃.

Many thanks for your help,

Pablo
--
http://www.ousia.tk
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Pablo Rodriguez]

On 12/2/19 6:01 PM, Wolfgang Schuster wrote:
> [...]
> Limited alternative (no rerun when second.pdf exists and by default the
> resulting PDF is loaded as image).
>
> \typesetfile[second.tex][--arguments="OptionThis={\ThisOption},OptionThat={\ThatOption}"][object=no]

Hi Wolfgang,

is there any way to add more options to the typeset file?

I need to setup "--result", "--purgeall" (and maybe another one [I
cannot remember, I’m not in front of the computer at work]).

Many thanks for your help,

Pablo
--
http://www.ousia.tk
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Hans Hagen]

On 12/4/2019 10:19 PM, Pablo Rodriguez wrote:
> On 12/3/19 10:14 PM, Hans Hagen wrote:
>> On 12/3/2019 8:37 PM, Pablo Rodriguez wrote:
>>> [...]
>>> I’m afraid a new font is not an option for me at work.
>> But there are better (unicode) versions available on windows, why nmot
>> use those then (as wolfgang pointed out).
> 
> Many thanks for your reply, Hans.
> 
> I would use the Unicode versions at home (if I had a Windows license),
> but at work it is problematic even to ask for free fonts.
> 
> My company may be to strict with these policies, but I don’t think it is
> a total exception in this matter.
> 
> I hope we might switch to Windows 10 in the no so near future 😃.
- if you use tex, use fonts that are put in the tex tree, the only 
guarantee you have for continuity
- i checked a few versions of that font on an old windows xp vm and it 
works ok ... so does your company ever update the machines?

Hans

-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Pablo Rodriguez]

On 12/5/19 9:23 AM, Hans Hagen wrote:
> On 12/4/2019 10:19 PM, Pablo Rodriguez wrote:
>> [...]
>> I hope we might switch to Windows 10 in the no so near future 😃.
> - if you use tex, use fonts that are put in the tex tree, the only
> guarantee you have for continuity.
> > - i checked a few versions of that font on an old windows xp vm and it
> works ok ... so does your company ever update the machines?

Hi Hans,

I wonder whether my issue with the fonts might be caused by my ConTeXt
distribution (such as the one with expansion, reported some ).

Windows is updated with security patches (or whichever the name might be).

If I try to explain to the IT people that a font is misbehaving and an
application (I don’t even try to explain what ConTeXt actually is [they
won’t understand]) cannot recognize it, they will reply that the
application is wrong. Because of that, I’m not even trying to explain
what is wrong with the font at all.

In my previous company, I remember a guy (which had as slightly idea of
what TeX was [he used LaTeX for his PhD decades ago]) with a strong
background in computer science. He was impressed that I “developed” a
system with no coding knowledge, but he objected that ConTeXt wasn’t
standard software (the standard for him was OpenOffice.org). I replied
that the standard was in the output PDF documents (which were PDF/A-3a).
(I didn’t mention that working with OOo was a real pain and that trying
to write conditionals with document merging was extremely annoying for me.)

Many thanks for your help and your new betas,

Pablo
--
http://www.ousia.tk
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Arthur Reutenauer]

On Thu, Dec 05, 2019 at 08:31:45PM +0100, Pablo Rodriguez wrote:
>                                 He was impressed that I “developed” a
> system with no coding knowledge, but he objected that ConTeXt wasn’t
> standard software (the standard for him was OpenOffice.org). I replied
> that the standard was in the output PDF documents (which were PDF/A-3a).

  And what did he have to say to that?

> (I didn’t mention that working with OOo was a real pain and that trying
> to write conditionals with document merging was extremely annoying for me.)

  You should have.

	Best,

		Arthur
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Pablo Rodriguez]

On 12/6/19 10:01 AM, Arthur Reutenauer wrote:
> On Thu, Dec 05, 2019 at 08:31:45PM +0100, Pablo Rodriguez wrote:
>>                                 He was impressed that I “developed” a
>> system with no coding knowledge, but he objected that ConTeXt wasn’t
>> standard software (the standard for him was OpenOffice.org). I replied
>> that the standard was in the output PDF documents (which were PDF/A-3a).
>
>   And what did he have to say to that?

He argued that if any other person had to improve the system, (s)he
would have to learn everything.

While I cannot disagree with that, I’m afraid that this would happen, no
matter which software I deployed (library, tool, or anything else).

I must admit that altough I cannot code, GUIs distract me when working
with text. But this would be my limitation.

>> (I didn’t mention that working with OOo was a real pain and that trying
>> to write conditionals with document merging was extremely annoying for me.)
>
>   You should have.

I know, but he was one of my bosses. And I remember I was totally
shocked when he explained to me that the standard document format for
any word processor was OpenOffice.org.

I cannot recall the accurate details from his explanation, but he seemed
to think that even Microsoft Word was at fault for not implementing the
Open Document Text format (.odt) as its native format.

The reasoning was so bizarre and shocking to me that I understood that
it was better to me simply to decline the discussion. Also other factual
inaccuracies about the work done in other departments made me clear that
it was better to avoid the conversation.

OOo was the cause of many compatibility issues in that company, because
they simply uninstalled Microsoft Office and installed OpenOffice.org
(no previous warning) in one evening.

The installation was so poorly performed that they forgot to assign
Microsoft Office extensions to OOo programs in Windows. AOnly this minor
incident was a huge problem for the vast majority of users.

It was clear to me what I knew decades ago: free software isn’t programs
for free. I think they still have to deal with issues in OOo.

I only talked no more than five times with that guy. But if he was the
evangelist of free software in that company, I’m afraid I totally agree
with the people working there that hated OOo.

Pablo
--
http://www.ousia.tk
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Hans Hagen]

On 12/7/2019 2:40 PM, Pablo Rodriguez wrote:

> I know, but he was one of my bosses. And I remember I was totally
> shocked when he explained to me that the standard document format for
> any word processor was OpenOffice.org.

Anyone claiming that something is a standard (esp in computer science) 
is unaware of history. I bet that our ancestors of thousands of years 
ago also considered themselves modern, with standards etc. But sometimes 
I think that many folks today think of themselves as being on top of the 
human (intelligence, progress, morale, etc) pyramid compared to whoever 
came before. History proved them wrong. (Similar are claims of this or 
that being better (software, operating systems, etc) while in the end 
much converges to the same.)

Anyway, it's a waste of time and energy discussing with those folks.

And we as texies should also be honest: how many of the acclaimed 
'happy' tex users are really 'happy' with their system and are those 
'millions' really (unforced) users who couldn't as well be using word or 
google docs or ... given what and how they deal with documents. And 
let's not add the quality argument because a coupel of weeks ago I 
noticed how bad tex output didn't look any better in a display of some 
80 summaries at some meeting (the, oh, look how bad that table looks 
experience).

(I'm sure Arthur, when reading this, can comment as we sort of had this 
at a theme of a talk!)

(And, yes I consider myself a happy tex user, but I also admit that I 
don't have to write much. And yes, it's a specific kind of user and usage.)

> I cannot recall the accurate details from his explanation, but he seemed
> to think that even Microsoft Word was at fault for not implementing the
> Open Document Text format (.odt) as its native format.

I suppose he read the specs of both formats in detail (in print of course).

> The reasoning was so bizarre and shocking to me that I understood that
> it was better to me simply to decline the discussion. Also other factual
> inaccuracies about the work done in other departments made me clear that
> it was better to avoid the conversation.

Indeed. Waste of time. Just think of this: you could kind of check his 
claims, so how about all the other claims someone makes ... stuff you 
know little about ... how valid are those claims then.

> OOo was the cause of many compatibility issues in that company, because
> they simply uninstalled Microsoft Office and installed OpenOffice.org
> (no previous warning) in one evening.

Well, as long as they're happy ... in most cases no one cares how output 
looks, nor cares about long term storage and exchange of data. Going 
belly up means 'delete all data and thrash the machines'.

> The installation was so poorly performed that they forgot to assign
> Microsoft Office extensions to OOo programs in Windows. AOnly this minor
> incident was a huge problem for the vast majority of users.

And then they entered denial state.

> It was clear to me what I knew decades ago: free software isn’t programs
> for free. I think they still have to deal with issues in OOo.

Although, not all free software comes for free. I'm not that sure of 
online tex services are cheaper than bulk microsoft licenses.

> I only talked no more than five times with that guy. But if he was the
> evangelist of free software in that company, I’m afraid I totally agree
> with the people working there that hated OOo.

It's all about honnesty isn't it? And about people spending time and 
energy, That doesn't always goes well with commercial objectives. And 
there's always the knowledge issue. And expecially when open source and 
such starts looking like a religion (one without a long history of 
dealing with itself and communicating properly) it gets even trickier.

Hans


-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Pablo Rodriguez]

On 12/7/19 4:32 PM, Hans Hagen wrote:
> On 12/7/2019 2:40 PM, Pablo Rodriguez wrote:
>
>> I know, but he was one of my bosses. And I remember I was totally
>> shocked when he explained to me that the standard document format for
>> any word processor was OpenOffice.org.
>
> Anyone claiming that something is a standard (esp in computer science)
> is unaware of history. I bet that our ancestors of thousands of years
> ago also considered themselves modern, with standards etc. But sometimes
> I think that many folks today think of themselves as being on top of the
> human (intelligence, progress, morale, etc) pyramid compared to whoever
> came before. History proved them wrong. (Similar are claims of this or
> that being better (software, operating systems, etc) while in the end
> much converges to the same.)

The main problem with his explanation of the standard was that he mixed
both tools and formats. I’m not sure any tool is a standard, the file
formrat for some kind of data is relevant to be standarized.

> Anyway, it's a waste of time and energy discussing with those folks.
>
> And we as texies should also be honest: how many of the acclaimed
> 'happy' tex users are really 'happy' with their system and are those
> 'millions' really (unforced) users who couldn't as well be using word or
> google docs or ... given what and how they deal with documents. And
> let's not add the quality argument because a coupel of weeks ago I
> noticed how bad tex output didn't look any better in a display of some
> 80 summaries at some meeting (the, oh, look how bad that table looks
> experience).

I must admit I’m having my worst time since I started using TeX (decades
ago), because some documents I generate have simply unreadable parts and
they are simply and plainly wrong (even considering the lowest
standards). It makes me suspect that one of approaches to basic
functionality could be flawed.

But I don’t doubt ConTeXt is the most useful piece of software I use on
a daily basis (followed by git and pandoc [probably in that order]),
because it makes me achieve things I couldn’t do using other tools.

I’m extremely confortable using it (although I report all bugs I hit).
And I think it is worth the effort to investigate further to contribute
to the fix to the problem described above (even if it drives me crazy).

> (And, yes I consider myself a happy tex user, but I also admit that I
> don't have to write much. And yes, it's a specific kind of user and usage.)

Well, Hans, this specific kind of user and usage is called LuaTeX,
LuaMetaTeX and ConTeXt development.

>> I cannot recall the accurate details from his explanation, but he seemed
>> to think that even Microsoft Word was at fault for not implementing the
>> Open Document Text format (.odt) as its native format.
>
> I suppose he read the specs of both formats in detail (in print of course).

His reasoning was flawed from the start. You cannot complain an egg is
not an elephant.

Well, Microsoft Word format may be propietary (it is actually a
standard), but there is nothing wrong in being different from another
text document format.

XML may have its flaws, but it is perfectly fine that it isn’t
PostScript. (Otherwise, what the reasoning demands is the removal of XML
as format.)

>> The reasoning was so bizarre and shocking to me that I understood that
>> it was better to me simply to decline the discussion. Also other factual
>> inaccuracies about the work done in other departments made me clear that
>> it was better to avoid the conversation.
>
> Indeed. Waste of time. Just think of this: you could kind of check his
> claims, so how about all the other claims someone makes ... stuff you
> know little about ... how valid are those claims then.

I perfectly recall one factual statement he did about one detail from
another department. I told him that I worked two years ago there and I
thought that was different. After leaving his office, I went to the
department and checked the detail. He was wrong and I thought that I
should avoid him (just in case he wanted to discuss general questions).

>> The installation was so poorly performed that they forgot to assign
>> Microsoft Office extensions to OOo programs in Windows. AOnly this minor
>> incident was a huge problem for the vast majority of users.
>
> And then they entered denial state.

They didn’t even need that. They didn’t experienced the problems, since
the users didn’t complain to them (installation was remote).

>> It was clear to me what I knew decades ago: free software isn’t programs
>> for free. I think they still have to deal with issues in OOo.
>
> Although, not all free software comes for free. I'm not that sure of
> online tex services are cheaper than bulk microsoft licenses.

I know and the problem with that migration was they wanted to avoid
spending in the license fees.

>> I only talked no more than five times with that guy. But if he was the
>> evangelist of free software in that company, I’m afraid I totally agree
>> with the people working there that hated OOo.
>
> It's all about honnesty isn't it? And about people spending time and
> energy, That doesn't always goes well with commercial objectives. And
> there's always the knowledge issue. And expecially when open source and
> such starts looking like a religion (one without a long history of
> dealing with itself and communicating properly) it gets even trickier.

I don’t have any problem with OOo myself, but I understand that people
thought that OOo was a cheap and bad alternative to Microsoft Office.

The problem with the way they used OOo is that you shouldn’t expect that
bugs disappear by magic. At least, start learning how to report the bugs
(to report them next).

As ideology, I think free software is most problematic. I agree that
free software may be a good contribution to society, but it is no moral
problem at all. (Probably we should start questioning why we apply
copyright law to trade secrets.)

Pablo
--
http://www.ousia.tk
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Hans Hagen]

On 12/7/2019 11:49 PM, Pablo Rodriguez wrote:

> free software may be a good contribution to society, but it is no moral
> problem at all. (Probably we should start questioning why we apply
> copyright law to trade secrets.)
What always puzzles me is that companies that otherwise depend on open 
source have no problems with quite rigourous copyright claims, even go 
for lawsuits, and don't care about violating privacy. And I also have 
not much sympathy for companies that start using free software, boast 
about it, kind of use the community and then cash out, quit support, put 
the usefull stuff in expensive variants only and in the end leave users 
in the cold. I have no problem with commercial activities but they 
should be open and can not be hidden.

Btw, it's one reason why open source / free sw can fail in cases: 
support can be more expensive than for commercial stuff, although often 
tex is dirt cheap, if only because much can be achieved with little 
coding and once it works, it works forever. One problem in your case is 
that managers probably don't make a proper cost-benefit analysis. Maybe 
we should make some templates for that some day.

I wonder how much backslash  - pun intended - there is because of all 
that pseudo open source. Makes a nice topic for late-night discussions 
at a next ctx meeting.

Hans


-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Hans Hagen]

On 12/3/2019 8:45 PM, Pablo Rodriguez wrote:

> I’m afraid I cannot get this working (I get a nil value error).
> 
>      \starttext
>       \startluacode
>          print(lua.getcodepage())
>       \stopluacode
>      \stoptext
a recent addition so your binary doesn't have it i guess

Hans

-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Pablo Rodriguez]

On 12/3/19 9:17 AM, Hans Hagen wrote:
> On 12/3/2019 12:07 AM, Akira Kakuto wrote:
>> [...]
>> Thus your example does not work unless a system call GetACP()
>> returns CP_UTF8 or 65001, sorry for the inconvenience.
> in luatex one can check this from lua with
>
> print(lua.getcodepage())

Many thanks for your reply, Hans.

I’m afraid I cannot get this working (I get a nil value error).

    \starttext
     \startluacode
        print(lua.getcodepage())
     \stopluacode
    \stoptext

Which is the right way to check it?

Many thanks for your help,

Pablo
--
http://www.ousia.tk
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: issue with beta from 2019.11.29 21:47 in Windows

[Hans Hagen]

On 12/3/2019 12:07 AM, Akira Kakuto wrote:
> Dear Pablo,
> 
>> \executesystemcommand{contextjit --purgeall
>> --arguments="OptionThis={\ThisOption},OptionThat={\ThatOption}" 
>> second.tex}
> 
> In my new lua[jit]tex binaries, encoding in a command line
> is assumed to be the default code page of a system which depends
> on a country or a machine. The code page of arguments in the
> command line is changed into utf-8 in the initial stage by using the 
> information
> of the default code page.
> Thus your example does not work unless a system call GetACP()
> returns CP_UTF8 or 65001, sorry for the inconvenience.
in luatex one can check this from lua with

print(lua.getcodepage())

Hans

-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://context.aanhet.net
archive  : https://bitbucket.org/phg/context-mirror/commits/
wiki     : http://contextgarden.net
___________________________________________________________________________________