Re: [pretext-announce] [CHANGE] side-by-side in the schema

[Jeremy Sylvestre]

On Wed, 26 Aug 2026 at 15:13, 'Rob Beezer' via PreTeXt announcements <pretext-...@googlegroups.com> wrote:

So a one-panel #sidebyside has been deprecated.  Two panels or more are
required.  No changes yet to existing code, you will simply get no-rush warnings.

How do I get subfigures in a figure with vertical flow?

(Necessary if the subfigures are too wide to actually sit side-by-side.)

I would have authored this as

figure

  caption

  sbsgroup

    sidebyside

      figure

        caption

        image

    sidebyside

      figure

        caption

        image

to produce output

[image]

(a) Caption for first subfigure.

[image]

(b) Caption for second subfigure.

Figure 1. Overall caption.

This is now illegal? Or is there another way to achieve this?

[Rob Beezer]

On 8/27/26 14:18, Jeremy Sylvestre wrote:
> This is now illegal?

Yes.

> Or is there another way to achieve this?

I can't think of a way. I'd move to two consecutive (wide) #figure, and use
their #caption to make any connection between them.

Rob

[Jeremy Sylvestre]

So vertical flow of subfigures is sometimes permitted (Figure 26.42 of the sample article), and sometimes not.

--
You received this message because you are subscribed to the Google Groups "PreTeXt development" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pretext-dev...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pretext-dev/MTAwMDAwLmpBVnhkVUg.1787868034%40pnsh.

[Rob Beezer]

Those are #sbsgroup, which are a collection of #sidebyside. Each #sidebyside
must have at least two panels.

Another minor change, that went in with this one, is that a #sbsgroup must now
contain at least two #sidebyside.

> vertical flow of subfigures

If that is what you are calling it, then maybe so. The changes speak for
themselves.

Rob

On 8/27/26 15:07, Jeremy Sylvestre wrote:
> So vertical flow of subfigures is sometimes permitted (Figure 26.42 <https://
> pretextbook.org/examples/sample-article/html/section-side-by-
> side.html#subsection-sbsgroup-10> of the sample article), and sometimes not.

>
> On Thu, 27 Aug 2026 at 16:00, 'Rob Beezer' via PreTeXt development <pretext-
> d...@googlegroups.com <mailto:prete...@googlegroups.com>> wrote:
>
> On 8/27/26 14:18, Jeremy Sylvestre wrote:
> > This is now illegal?
>
> Yes.
>
> > Or is there another way to achieve this?
> I can't think of a way.  I'd move to two consecutive (wide) #figure, and use
> their #caption to make any connection between them.
>
> Rob
>
> --
> You received this message because you are subscribed to the Google Groups
> "PreTeXt development" group.
> To unsubscribe from this group and stop receiving emails from it, send an

> email to pretext-dev...@googlegroups.com <mailto:pretext-
> dev%2Bunsu...@googlegroups.com>.

> To view this discussion visit https://groups.google.com/d/msgid/pretext-dev/

> MTAwMDAwLmpBVnhkVUg.1787868034%40pnsh <https://groups.google.com/d/msgid/
> pretext-dev/MTAwMDAwLmpBVnhkVUg.1787868034%40pnsh>.

>
> --
> You received this message because you are subscribed to the Google Groups
> "PreTeXt development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email

> to pretext-dev...@googlegroups.com <mailto:pretext-
> dev+uns...@googlegroups.com>.

> To view this discussion visit https://groups.google.com/d/msgid/pretext-dev/

> CALMYOnep2C%2BJZfBdvrNGFEw9tUASab7g8Pakn20rnuNk6U60Hw%40mail.gmail.com <https://
> groups.google.com/d/msgid/pretext-dev/
> CALMYOnep2C%2BJZfBdvrNGFEw9tUASab7g8Pakn20rnuNk6U60Hw%40mail.gmail.com?
> utm_medium=email&utm_source=footer>.

[Rob Beezer]

Dear Jeremy,

What would you think of:

figure
stack
figure
figure

The overall #figure would be Figure 9.3, and have its own overall #caption.
The two #figure in the #stack would be 9.3(a) and 9.3(b) each with its
own #caption.

In this scenario, the #stack would only hold a sequence of #figure. It would
not be the general-use #stack of a #sidebyside panel or an #interactive.

Debatable if the #stack is necessary: as markup or logistically for processing,
but I would lean that way at first blush.

Rob

[Jeremy Sylvestre]

On Sat, 5 Sept 2026 at 15:48, 'Rob Beezer' via PreTeXt development <prete...@googlegroups.com> wrote:

What would you think of:

figure
   stack
     figure
     figure

that would be great, thanks

[Rob Beezer]

OK, fully implemented as discussed - schema, Guide, all conversions. Only
"subfigures" inside a #figure, stacked vertically.

@Jeremy - please take this for a test drive. A pointer to an example use in
your project would be great (assuming it is public). Once I hear from you, I
can do a pretext-announce post. Thanks.

Note: we have implemented the bare minimum of just Jeremy's request. Start a
discussion if there is a reasonable case for some subnumbeed vertical flow with
more variety.

Website example once I get some technical problems sorted. See sample article
in distribution for source of an example.

https://github.com/PreTeXtBook/pretext/pull/3214

Rob


On 9/8/26 06:28, Jeremy Sylvestre wrote:
> On Sat, 5 Sept 2026 at 15:48, 'Rob Beezer' via PreTeXt development <pretext-

> d...@googlegroups.com <mailto:prete...@googlegroups.com>> wrote:
>
> What would you think of:
>
> figure
>    stack
>      figure
>      figure
>
>
> that would be great, thanks
>

> --
> You received this message because you are subscribed to the Google Groups
> "PreTeXt development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email
> to pretext-dev...@googlegroups.com <mailto:pretext-
> dev+uns...@googlegroups.com>.
> To view this discussion visit https://groups.google.com/d/msgid/pretext-dev/

> CALMYOnfua63R_yhEWDkHU1GVCJdqouSPcfW%3DUmv9pt%2BmFs-cww%40mail.gmail.com
> <https://groups.google.com/d/msgid/pretext-dev/
> CALMYOnfua63R_yhEWDkHU1GVCJdqouSPcfW%3DUmv9pt%2BmFs-cww%40mail.gmail.com?
> utm_medium=email&utm_source=footer>.

[Rob Beezer]

OK, flip-flopping some more on this one. ;-)

There are times you would want a row of a #sbsgroup to have just one panel, so
the consituent #sidebyside would need to allow just one panel. Discovered this
through my latest conversion of a PDF research article, which is heavy on
figures/graphics (and so stressing this situation, as intended).

Suppose you have 10 similar images, all the same size, and at about 1/3 the page
width each. Maybe you want to go 3, 3, 3, 1, and not be forced into something
like 3, 3, 2, 2.

So a lone #sidebyside could be required to have two or more panels, while a
#sbsgroup could be required to have at least two #sidebyside (that is the
situation now), but these consituent #sidebyside could be single panel.

You might want a row with an extra-wide plot (say) with three one-third-width
images on the row below (maybe close-ups of regions of the plot?).

I like the stack-of-figures solution just enacted, but it comes at a cost.
Subnumbering is tricky. Just found a latent (long-standing?) bug in LaTeX
output as a result of the most recent PDF conversion (needs more investigation,
see issue). So it would be good to contain subnumbering to
#sidebyside/#sbsgroup. And I think the proposal above for teh schema would
catch most tag abuse.

https://github.com/PreTeXtBook/pretext/issues/3218

So we would be back to Jeremy's

figure
sbsgroup
sidebyside
figure
sidebyside
figure

More verbose, but not solely un-semantic, at the benefit of rolling back the new
behavior of #stack inside #figure. We'd adjust the schema as indicated above,
so this would be going back to old (mature) code that has always accomodated
single panels.

As a friend said to me the other day, "I get to change my mind, that is why I
have one." But I think I would at least owe Jeremy a beer.

Comments? Other than any Bronx cheers?

Rob


On 9/8/26 6:28 AM, Jeremy Sylvestre wrote:
> On Sat, 5 Sept 2026 at 15:48, 'Rob Beezer' via PreTeXt development <pretext-

> d...@googlegroups.com <mailto:prete...@googlegroups.com>> wrote:
>
> What would you think of:
>
> figure
>    stack
>      figure
>      figure
>
>
> that would be great, thanks
>

> --
> You received this message because you are subscribed to the Google Groups
> "PreTeXt development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email
> to pretext-dev...@googlegroups.com <mailto:pretext-
> dev+uns...@googlegroups.com>.
> To view this discussion visit https://groups.google.com/d/msgid/pretext-dev/

[Sean Fitzpatrick]

There's a way to do this in the current schema, but it feels like cheating: for an sbsgroup with a 3 3 1 layout, make the last sidebyside p figure p

To unsubscribe from this group and stop receiving emails from it, send an email to pretext-dev...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pretext-dev/MTAwMDAzYlhWQXVjNkc.1789689544%40pnsh.

[Rob Beezer]

Yes, that would be cheating. And especially if you put #nbsp into the #p as its
entire content - just to get the LaTeX to behave. ;-)

Seriously, always think about what the reader is going to get in braille or via
a screenreader. In braille, a transcriber note will announce 9 panels (3 rows
of 3) before listing them sequentially. The empty #p might be nothing, or just
a blank line (whose indentation is not recognizable).

Rob

On 9/18/26 10:59 AM, Sean Fitzpatrick wrote:
> There's a way to do this in the current schema, but it feels like cheating: for
> an sbsgroup with a 3 3 1 layout, make the last sidebyside p figure p
>

> On Thu, Sep 17, 2026, 5:59 p.m. 'Rob Beezer' via PreTeXt development <pretext-
> d...@googlegroups.com <mailto:prete...@googlegroups.com>> wrote:
>
> OK, flip-flopping some more on this one.  ;-)
>
> There are times you would want a row of a #sbsgroup to have just one panel, so
> the consituent #sidebyside would need to allow just one panel.  Discovered this
> through my latest conversion of a PDF research article, which is heavy on
> figures/graphics (and so stressing this situation, as intended).
>
> Suppose you have 10 similar images, all the same size, and at about 1/3 the
> page
> width each.  Maybe you want to go 3, 3, 3, 1, and not be forced into something
> like 3, 3, 2, 2.
>
> So a lone #sidebyside could be required to have two or more panels, while a
> #sbsgroup could be required to have at least two #sidebyside (that is the
> situation now), but these consituent #sidebyside could be single panel.
>
> You might want a row with an extra-wide plot (say) with three one-third-width
> images on the row below (maybe close-ups of regions of the plot?).
>
> I like the stack-of-figures solution just enacted, but it comes at a cost.
> Subnumbering is tricky.  Just found a latent (long-standing?) bug in LaTeX
> output as a result of the most recent PDF conversion (needs more investigation,
> see issue).  So it would be good to contain subnumbering to
> #sidebyside/#sbsgroup.  And I think the proposal above for teh schema would
> catch most tag abuse.
>

> https://github.com/PreTeXtBook/pretext/issues/3218 <https://github.com/
> PreTeXtBook/pretext/issues/3218>

>
> So we would be back to Jeremy's
>
> figure
>    sbsgroup
>      sidebyside
>        figure
>      sidebyside
>        figure
>
> More verbose, but not solely un-semantic, at the benefit of rolling back the
> new
> behavior of #stack inside #figure.  We'd adjust the schema as indicated above,
> so this would be going back to old (mature) code that has always accomodated
> single panels.
>
> As a friend said to me the other day, "I get to change my mind, that is why I
> have one."  But I think I would at least owe Jeremy a beer.
>
> Comments?  Other than any Bronx cheers?
>
> Rob
>
>
> On 9/8/26 6:28 AM, Jeremy Sylvestre wrote:
> > On Sat, 5 Sept 2026 at 15:48, 'Rob Beezer' via PreTeXt development <pretext-

> > d...@googlegroups.com <mailto:d...@googlegroups.com> <mailto:pretext-

> d...@googlegroups.com <mailto:prete...@googlegroups.com>>> wrote:
> >
> >     What would you think of:
> >
> >     figure
> >         stack
> >           figure
> >           figure
> >
> >
> > that would be great, thanks
> >
> > --
> > You received this message because you are subscribed to the Google Groups
> > "PreTeXt development" group.
> > To unsubscribe from this group and stop receiving emails from it, send an
> email
> > to pretext-dev...@googlegroups.com <mailto:pretext-

> dev%2Bunsu...@googlegroups.com> <mailto:pretext- <mailto:pretext->
> > dev+uns...@googlegroups.com
> <mailto:dev%2Bunsu...@googlegroups.com>>.

> > To view this discussion visit https://groups.google.com/d/msgid/pretext-

> dev/ <https://groups.google.com/d/msgid/pretext-dev/>
> > CALMYOnfua63R_yhEWDkHU1GVCJdqouSPcfW%3DUmv9pt%2BmFs-cww%40mail.gmail.com
> <http://40mail.gmail.com>
> > <https://groups.google.com/d/msgid/pretext-dev/ <https://
> groups.google.com/d/msgid/pretext-dev/>
> > CALMYOnfua63R_yhEWDkHU1GVCJdqouSPcfW%3DUmv9pt%2BmFs-cww%40mail.gmail.com
> <http://40mail.gmail.com>?

> > utm_medium=email&utm_source=footer>.
>
> --
> You received this message because you are subscribed to the Google Groups
> "PreTeXt development" group.
> To unsubscribe from this group and stop receiving emails from it, send an

> email to pretext-dev...@googlegroups.com <mailto:pretext-
> dev%2Bunsu...@googlegroups.com>.

> To view this discussion visit https://groups.google.com/d/msgid/pretext-dev/

> MTAwMDAzYlhWQXVjNkc.1789689544%40pnsh <https://groups.google.com/d/msgid/
> pretext-dev/MTAwMDAzYlhWQXVjNkc.1789689544%40pnsh>.

>
> --
> You received this message because you are subscribed to the Google Groups
> "PreTeXt development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email

> to pretext-dev...@googlegroups.com <mailto:pretext-
> dev+uns...@googlegroups.com>.
> To view this discussion visit https://groups.google.com/d/msgid/pretext-dev/

> CAH%2BNcPaa3XKa%3DzvixUvv4AFXLi4%3Dd5pdNSHoTkaSiWBCiy27vw%40mail.gmail.com
> <https://groups.google.com/d/msgid/pretext-dev/
> CAH%2BNcPaa3XKa%3DzvixUvv4AFXLi4%3Dd5pdNSHoTkaSiWBCiy27vw%40mail.gmail.com?
> utm_medium=email&utm_source=footer>.