CSS and slideshows

[Rob Beezer]

There are a lot of templates in the conversion to (RevealJS) slideshows that are
just getting in the way of our "regular" HTML. (See Alexandre's recent post on
-support about localizations for just one example.) The intent often seems to
be an extra enclosing #div to provide a handle for CSS styling. That CSS is in
the head of the output HTML and is hard-coded in the stylesheet for the conversion.

So I'd like to remove these templates. And then the CSS will bge useless, so it
can go. But the styling itself is likely worth keeping. And I think we already
have things like #article with a .proof-like that can be used for selectors

1. An "extra" pretext-reveal.css file to use to extend the regular CSS?

2. A slideshow is a single HTML file. Right inside #body we have a #div with
classes "reveal" and "pretext-content". Or, is this good enough to make new
rules inside the existing CSS? Should these classes be somewhere else, say on
#body?

3. Most important: answers to above are good, but only if there is somebody who
wants to work on the CSS side. I'm happy to do the XSL to create the HTML, but
there will need to be someone who wants to coordinate the CSS and perhaps take
long-term responsibility for it.

Thanks,
Rob

[Sean Fitzpatrick]

I was ready to say I can probably help with xsl but not CSS, and then I got to #3!

I'll see what I can do in any case

--
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 on the web visit https://groups.google.com/d/msgid/pretext-dev/MTAwMDAzOS5iZWV6ZXI.1720909676%40quikprotect.

[Andrew Scholer]

Rob -

I'd be happy to help fixup the CSS to match whatever changes you (or Sean) make. Allowing different targets to flexibly reuse parts of the CSS is one of the possible benefits of the CSS bundling I am working on. This is a good test case for that. Long-term, I would be happy to help with minor fixes. But, I don't see myself using the reveal target, so I can't promise to be an active long-term maintainer or to work on major new features for it.

Given that, here would be my answers for #1/#2

1. A pretext-reveal.css than bundles in the necessary CSS that is shared by the HTML target.

2. A top-level div that has class "reveal" should be enough to do targeting. Having it on the same container as "pretext" (i.e. "pretext reveal" instead of "pretext book") seems to make sense.

Andrew Scholer (he/him/his)

Computer Science Instructor/Program Chair

Chemeketa Community College

503.589.7649 

computerscience.chemeketa.edu/people/andrew-scholer/

[Rob Beezer]

#3 is mostly ripping out XSL (and consequent CSS). I'm happy to just advise.

I hope it wasn't the use of the word "responsibility" in #3 that gave you pause. ;-)

[Rob Beezer]

Dear Andrew,

Thanks for the interest and the helpful answers. I think you, Sean and I can make some good progress.

If I was flying solo I might adjust XSL to get new HTML and then try to adjust ad-hoc CSS to preserve appears and intent. Then better handling if the CSS could take place independently, all without too much disruption for authors.

We'll see what Sean and I can organize this week.

Rob

[Rob Beezer]

A little bit of luck last night ripping out templates and getting improvements.

To the original post, removing a "type-name" template hard-coding "Theorem"
opened up the correct localization. Maybe just that alone, but maybe in
conjumnction with more removals.

Removing all the theorem- and proof-related templates, got me a ".hiddenproof",
which is no good, and perhaps related to how we try to turn off knowlization
completely for slides.

[Alexandre Lymberopoulos]

Hi there!

I just comment several lines on "xsl/pretext-revealjs.xsl" and every
environment title appears in portuguese (but numbered, that's ok). As
side effects, The eviroment name (Definição, Teorema, Proposição, etc.)
is smaller and there is no box around it, which should be desirable.

To have things working here, I translated everything on the above file,
commented that parenthesis with source number. This is far from a decent
solution more like a selfish on, I know. I would have to dig a lot on
the CSS things (which I know a few things about it) and the XSL (which I
know nothing), understang how is the integration of this
pretext-revealjs.xsl file with the rest of the system.

How could I contribute to this (in the sense that I'm using it and I
want to give something back)?

Best, Alexandre

--
===============================================================================
Alexandre Lymberopoulos - lym...@gmail.com
===============================================================================

[Rob Beezer]

Dear Alexandre,

Thanks for the testing and the report. You've seen my other message today,
reporting changes I was able to make while at a PreTeXt workshop. I think
things will be much better for your project now.

It was a big help to have you report this, and as you develop slides for your
course, I hope you can continue to make suggestions for (necessary) improvements.

Rob

[Andrew Scholer]

Rob - 

Are you feeling the XSL is in good shape?  Please give a holler when you think you are done and I'll get in there to check out the CSS.

Andrew Scholer (he/him/his)

Computer Science Instructor/Program Chair

Chemeketa Community College

503.589.7649 

computerscience.chemeketa.edu/people/andrew-scholer/

To view this discussion on the web visit https://groups.google.com/d/msgid/pretext-dev/MTAwMDAzNy5iZWV6ZXI.1721179347%40quikprotect.

[Rob Beezer]

Yes, I'm done with the XSL for now. Some larger issues for proofs of theorems
(and theorem-like), so don't get too worked up there.

Feel free to suggest/implement a new home for the CSS (as already discussed).
And if you think the CSS (in the stylesheet that migrates to the head) has
redundant rules, they can be removed. In other words, if authors get minor
changes in appearance, that is OK. We don't guarantee backward-compatible
presentation.

Thanks,
Rob

On 7/16/24 21:26, Andrew Scholer wrote:
> Rob -
>
> Are you feeling the XSL is in good shape?  Please give a holler when you think
> you are done and I'll get in there to check out the CSS.
>
> Andrew Scholer (he/him/his)
> Computer Science Instructor/Program Chair
> Chemeketa Community College
> 503.589.7649
> computerscience.chemeketa.edu/people/andrew-scholer/

> <http://computerscience.chemeketa.edu/people/andrew-scholer/>

> >>> <andrew....@chemeketa.edu <mailto:andrew....@chemeketa.edu>> wrote:
> >>>> Rob -
> >>>>
> >>>> I'd be happy to help fixup the CSS to match whatever changes you (or Sean)
> >>>> make. Allowing different targets to flexibly reuse parts of the CSS is one
> >>>> of the possible benefits of the CSS bundling I am working on. This is a
> >>>> good test case for that. Long-term, I would be happy to help with minor
> >>>> fixes. But, I don't see myself using the reveal target, so I can't promise
> >>>> to be an active long-term maintainer or to work on major new features for
> >>>> it.
> >>>>
> >>>> Given that, here would be my answers for #1/#2
> >>>>
> >>>> 1. A pretext-reveal.css than bundles in the necessary CSS that is
> shared by
> >>>> the HTML target.
> >>>>
> >>>> 2. A top-level div that has class "reveal" should be enough to do
> >>>> targeting. Having it on the same container as "pretext" (i.e. "pretext
> >>>> reveal" instead of "pretext book") seems to make sense.
> >>>>
> >>>> Andrew Scholer (he/him/his)
> >>>> Computer Science Instructor/Program Chair
> >>>> Chemeketa Community College
> >>>> 503.589.7649
> >>>> computerscience.chemeketa.edu/people/andrew-scholer/

> <http://computerscience.chemeketa.edu/people/andrew-scholer/>

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

> >>>>> To view this discussion on the web visit
> >>>>>

> https://groups.google.com/d/msgid/pretext-dev/MTAwMDAzOS5iZWV6ZXI.1720909676%40quikprotect <https://groups.google.com/d/msgid/pretext-dev/MTAwMDAzOS5iZWV6ZXI.1720909676%40quikprotect>

> >>>>> .
> >>>>>
> >>>>
> >>>
> >>
> >
>
> --
> 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 on the web visit

> https://groups.google.com/d/msgid/pretext-dev/MTAwMDAzNy5iZWV6ZXI.1721179347%40quikprotect <https://groups.google.com/d/msgid/pretext-dev/MTAwMDAzNy5iZWV6ZXI.1721179347%40quikprotect>.

>
> --
> 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...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/pretext-dev/CACm44N_yjBNo0%3D7X0dX_eAN5toEZLttjaQpUQm27djMugOiT4w%40mail.gmail.com <https://groups.google.com/d/msgid/pretext-dev/CACm44N_yjBNo0%3D7X0dX_eAN5toEZLttjaQpUQm27djMugOiT4w%40mail.gmail.com?utm_medium=email&utm_source=footer>.

[kcri...@gmail.com]

Thanks for the testing and the report. You've seen my other message today,
reporting changes I was able to make while at a PreTeXt workshop. I think
things will be much better for your project now. 

Hi!  Can you (or someone else since you are at the workshop) point out where that message is?  (I'm assuming it was somewhere public, but I couldn't find it.)  I found the commits starting at 3d7c5f4 though.  Also (maybe Alexandre can answer this) what is the effect of the change "Slideshow: remove use of "source-number" attribute"?  It seems like this only was used in the slideshow conversions anyway :-(

 

It was a big help to have you report this, and as you develop slides for your
course, I hope you can continue to make suggestions for (necessary) improvements. 

I'm definitely interested in helping with some improvements the next two weeks.  (E.g. # 1888-90 but I will have to look at the most recent changes.)

[Rob Beezer]

I've removed some offending templates in the conversion to a RevealJS slideshow.

* Mostly I have preserved styling with adjustments, but still up in the head.

* Some "source-number" attribute experiment is gone, along with the parentheses
that were going to surround it.

* Numbers for blocks are back. Usual choices should be functional via the
publisher file.

* Localizations should be back!

I think this is ready for architectural improvements to the CSS.

Rob

On 7/15/24 11:30, Alexandre Lymberopoulos wrote:

[Rob Beezer]

On 7/17/24 12:02, kcri...@gmail.com wrote:
> Hi!  Can you (or someone else since you are at the workshop) point out where
> that message is?  (I'm assuming it was somewhere public, but I couldn't find
> it.)

Never made it through my outsourced Google frustrations. Reposted. I think.

[Rob Beezer]

On 7/17/24 12:02, kcri...@gmail.com wrote:

> answer this) what is the effect of the change "Slideshow: remove use of
> "source-number" attribute"?  It seems like this only was used in the slideshow
> conversions anyway :-(

Some unknown experiement by the early implementers of slideshows. Gone now.

[kcri...@gmail.com]

> answer this) what is the effect of the change "Slideshow: remove use of
> "source-number" attribute"?  It seems like this only was used in the slideshow
> conversions anyway :-(

Some unknown experiement by the early implementers of slideshows. Gone now.

Thanks on both counts!