Re: Syntax errors in YAML blocks

[William Lupton <wlupton-QSt+ys/nuMyEUIsrzH9SikB+6BGkLq7r@public.gmane.org>]

[-- Attachment #1: Type: text/plain, Size: 10748 bytes --]

Thanks. https://github.com/jgm/pandoc/issues/7400

On Mon, 21 Jun 2021 at 23:01, John MacFarlane <jgm-TVLZxgkOlNX2fBVCVOL8/A@public.gmane.org> wrote:

>
> Yes, this definitely used to work.  Maybe due to the change to
> HsYAML? I will look into it, but an issue on the tracker will
> help ensure it doesn't get forgotten.
>
> BPJ <bpj-J3H7GcXPSITLoDKTGw+V6w@public.gmane.org> writes:
>
> > It used to work.
> >
> > Den mån 21 juni 2021 15:44William Lupton <wlupton@broadband-forum.org>
> > skrev:
> >
> >> All,
> >>
> >> I don't know how I missed this before, but I have noticed that this
> isn't
> >> treated as a valid YAML block (at least not with pandoc 2.11.4).
> >>
> >> *% *cat comment.md
> >>
> >> ---
> >>
> >> # comment
> >>
> >> ...
> >>
> >> *% *pandoc comment.md
> >>
> >> <hr />
> >>
> >> <h1 id="comment">comment</h1>
> >>
> >> <p>…</p>
> >>
> >> Inserting some YAML works:
> >>
> >> *% *cat comment-plus-variable.md
> >>
> >> ---
> >>
> >> # comment
> >>
> >> variable: 42
> >>
> >> ...
> >>
> >> *% *pandoc comment-plus-variable.md
> >>
> >>
> >>
> >> I realise that this is an abuse, and that I could have used an HTML
> >> comment (for example), but I believe that this is valid YAML, so should
> it
> >> have worked?
> >>
> >> *% *yq eval comment.md
> >>
> >>
> >>
> >> # comment
> >>
> >>
> >>
> >> *% *yq eval comment-plus-variable.md
> >>
> >> # comment
> >>
> >> variable: 42
> >>
> >> Apologies if this behaviour has changed in the latest pandoc...
> >>
> >> Thanks,
> >> William
> >>
> >> On Thu, 3 Jun 2021 at 09:10, William Lupton <
> wlupton-QSt+ys/nuMyEUIsrzH9SikB+6BGkLq7r@public.gmane.org>
> >> wrote:
> >>
> >>> Thanks!
> >>>
> >>> On Thu, 3 Jun 2021 at 05:49, John MacFarlane <jgm-TVLZxgkOlNX2fBVCVOL8/A@public.gmane.org> wrote:
> >>>
> >>>>
> >>>> These are good suggestions.  I've revised the manual accordingly.
> >>>>
> >>>> BPJ <melroch-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:
> >>>>
> >>>> > Perhaps it should be pointed out that the body of YAML block scalars
> >>>> must
> >>>> > be indented relative to the line with the `|` or `>`. Not doing so
> is
> >>>> > indeed a YAML syntax error, so Pandoc does the right thing when not
> >>>> > considering it valid YAML, however unfortunate it is that Pandoc may
> >>>> think
> >>>> > it is something else.
> >>>> >
> >>>> > The overview of YAML syntax on Wikipedia is quite good. Perhaps the
> >>>> manual
> >>>> > could link to (some specific version of) it.
> >>>> >
> >>>> > https://en.m.wikipedia.org/wiki/YAML#Syntax
> >>>> >
> >>>> > Den ons 2 juni 2021 21:21John MacFarlane <jgm-TVLZxgkOlNX2fBVCVOL8/A@public.gmane.org> skrev:
> >>>> >
> >>>> >>
> >>>> >> The problem is that basically anything is valid Markdown,
> >>>> >> it's just a matter of how it is interpreted.  It does seem
> >>>> >> to me that in the second case we might want to at least emit
> >>>> >> a warning -- if not downright fail.  It's just so unlikely
> >>>> >> that someone would write a block beginning with `---`
> >>>> >> and ending with `...` and intend it to be something other
> >>>> >> than YAML metadata.  If you'd like, you could submit an
> >>>> >> issue for this.  I'd rather not suggest avoiding complex
> >>>> >> YAML blocks -- sometimes they're fine, if the syntax is
> >>>> >> correct!
> >>>> >>
> >>>> >> William Lupton <wlupton-QSt+ys/nuMyEUIsrzH9SikB+6BGkLq7r@public.gmane.org> writes:
> >>>> >>
> >>>> >> > Thanks. Yes, I guess I realised it had something to do with
> >>>> ambiguous
> >>>> >> > grammars, but I hadn't really considered that my input was a
> valid
> >>>> table
> >>>> >> > (mea culpa).
> >>>> >> >
> >>>> >> > So I thought aha! I'll change the terminating "---" to "..." (in
> >>>> the hope
> >>>> >> > of an error), but this of course gives the output shown below.
> >>>> >> >
> >>>> >> > I wonder whether there could be a warning in the manual (or maybe
> >>>> there
> >>>> >> is
> >>>> >> > and I've missed it?) and/or perhaps a suggestion that it might
> be a
> >>>> good
> >>>> >> > idea to avoid complex YAML metadata blocks (in favour of metadata
> >>>> files)?
> >>>> >> > I'd be happy to create an issue and/or a MANUAL.md pull request.
> >>>> >> >
> >>>> >> > % cat bad.md
> >>>> >> > ---
> >>>> >> > history:
> >>>> >> > - number: Release 1.2
> >>>> >> >   changes: |
> >>>> >> >   - TBD
> >>>> >> > ...
> >>>> >> >
> >>>> >> > % pandoc bad.md
> >>>> >> > <hr />
> >>>> >> > <p>history: - number: Release 1.2 changes: | - TBD …</p>
> >>>> >> >
> >>>> >> >
> >>>> >> > On Wed, 2 Jun 2021 at 17:47, John MacFarlane <jgm-TVLZxgkOlNX2fBVCVOL8/A@public.gmane.org>
> >>>> wrote:
> >>>> >> >
> >>>> >> >>
> >>>> >> >> Well, this actually IS a valid "simple table".  So how is
> >>>> >> >> pandoc to know that you intended it as a YAML metadata block?
> >>>> >> >> Pandoc tries first to parse things as YAML blocks, and then as
> >>>> >> >> tables, if either works, the parse succeeds with no error.
> >>>> >> >>
> >>>> >> >> William Lupton <wlupton-QSt+ys/nuMyEUIsrzH9SikB+6BGkLq7r@public.gmane.org> writes:
> >>>> >> >>
> >>>> >> >> > Should I expect syntax errors in YAML blocks to be detected
> and
> >>>> >> reported?
> >>>> >> >> >
> >>>> >> >> > This has a syntax error in the YAML block:
> >>>> >> >> >
> >>>> >> >> > % cat bad.md
> >>>> >> >> > ---
> >>>> >> >> > history:
> >>>> >> >> > - number: Release 1.2
> >>>> >> >> >   changes: |
> >>>> >> >> >   - TBD
> >>>> >> >> > ---
> >>>> >> >> >
> >>>> >> >> > This fixes it:
> >>>> >> >> >
> >>>> >> >> > % diff bad.md good.md
> >>>> >> >> > 5c5
> >>>> >> >> > <   - TBD
> >>>> >> >> > ---
> >>>> >> >> >>     - TBD
> >>>> >> >> >
> >>>> >> >> > The latest pandoc:
> >>>> >> >> >
> >>>> >> >> > % ~/Downloads/pandoc-2.14.0.1/bin/pandoc -v
> >>>> >> >> > pandoc 2.14.0.1
> >>>> >> >> > ...
> >>>> >> >> >
> >>>> >> >> > Pandoc thinks the bad YAML is a table:
> >>>> >> >> >
> >>>> >> >> > % ~/Downloads/pandoc-2.14.0.1/bin/pandoc bad.md
> >>>> >> >> > <table>
> >>>> >> >> > <tbody>
> >>>> >> >> > <tr class="odd">
> >>>> >> >> > <td>history:</td>
> >>>> >> >> > </tr>
> >>>> >> >> > <tr class="even">
> >>>> >> >> > <td>- number: Release 1.2</td>
> >>>> >> >> > </tr>
> >>>> >> >> > <tr class="odd">
> >>>> >> >> > <td>changes: |</td>
> >>>> >> >> > </tr>
> >>>> >> >> > <tr class="even">
> >>>> >> >> > <td>- TBD</td>
> >>>> >> >> > </tr>
> >>>> >> >> > </tbody>
> >>>> >> >> > </table>
> >>>> >> >> >
> >>>> >> >> > But it's OK with the good YAML:
> >>>> >> >> >
> >>>> >> >> > % ~/Downloads/pandoc-2.14.0.1/bin/pandoc good.md
> >>>> >> >> >
> >>>> >> >> > Parsing the file as YAML reports the error:
> >>>> >> >> >
> >>>> >> >> > % ~/Downloads/pandoc-2.14.0.1/bin/pandoc --metadata-file
> bad.md
> >>>> >> >> > Error at "bad.md" (line 5, column 0):
> >>>> >> >> > Unexpected ' '
> >>>> >> >> >
> >>>> >> >> > % ~/Downloads/pandoc-2.14.0.1/bin/pandoc --metadata-file
> good.md
> >>>> >> >> > ^C
> >>>> >> >> >
> >>>> >> >> > --
> >>>> >> >> > You received this message because you are 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 view this discussion on the web visit
> >>>> >> >>
> >>>> >>
> >>>>
> https://groups.google.com/d/msgid/pandoc-discuss/CAEe_xxjF3Un2J-xg5UOs7uReZ5cO27TATA%2BfDY%2BAtghAsVhShA%40mail.gmail.com
> >>>> >> >> .
> >>>> >> >>
> >>>> >>
> >>>> >> --
> >>>> >> You received this message because you are 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 view this discussion on the web visit
> >>>> >>
> >>>>
> https://groups.google.com/d/msgid/pandoc-discuss/m2a6o8ul91.fsf%40johnmacfarlane.net
> >>>> >> .
> >>>> >>
> >>>> >
> >>>> > --
> >>>> > You received this message because you are 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 view this discussion on the web visit
> >>>>
> https://groups.google.com/d/msgid/pandoc-discuss/CADAJKhC2g4z_H%2BK%2BVtGoVd1T2nmYeMuRrnUquhDPpoYfuxSwWw%40mail.gmail.com
> >>>> .
> >>>>
> >>> --
> >> You received this message because you are 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 view this discussion on the web visit
> >>
> https://groups.google.com/d/msgid/pandoc-discuss/CAEe_xxj-bHpfTLDhz5QRz51x-zcs%2BVW5dgWbmRTRaNL%2Bx6Y31Q%40mail.gmail.com
> >> <
> https://groups.google.com/d/msgid/pandoc-discuss/CAEe_xxj-bHpfTLDhz5QRz51x-zcs%2BVW5dgWbmRTRaNL%2Bx6Y31Q%40mail.gmail.com?utm_medium=email&utm_source=footer
> >
> >> .
> >>
> >
> > --
> > You received this message because you are 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 view this discussion on the web visit
> https://groups.google.com/d/msgid/pandoc-discuss/CADAJKhAr3Q_FR8FaubSDUxj692uSwAhKgdVV8mDRw5eU7HmO0g%40mail.gmail.com
> .
>
> --
> You received this message because you are 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 view this discussion on the web visit
> https://groups.google.com/d/msgid/pandoc-discuss/m27dimg9nv.fsf%40Johns-Air.hsd1.ca.comcast.net
> .
>

-- 
You received this message because you are 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 view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CAEe_xxh-geLJmwn4cuYftHAVzA6UZKq9wUkE7s5hPvsUw88%3DEw%40mail.gmail.com.

[-- Attachment #2: Type: text/html, Size: 18992 bytes --]