LaTeX: first pass at texstyle infrastructure

[oscarlevin]

There is a LOT going on here, and there will be more to come. Right now:

• Inside journals a folder of texstyles contains two different texstyle files describing the latex layout for ams journals and springer nature journals. A third dependent style is in dependents, with only the documentclass changed.
• The xsl that manages all this is xsl/latex/pretext-latex-texstyle.xsl. The goal is that this will replace all the recent journal-specific xsl files. It reads in the correct texstyle file and does all the processing.
• A very small change to the python script is merely cosmetic (name of stringparam).

The xsl file really pushed me to understand how xsl works. There will absolutely be better ways to do some of these things I think. The merging of a dependent and base texstyle, using techniques from assembly, was particularly fun. Also the processing of authors: we bounce back and forth between the source pretext and texstyle file a couple times so we can "loop over" the authors and the author data in the correct order. Fun fun fun.

My plan:

• Fix a few things that are broken (keywords and their separators)
• Implement the other journal styles as texstyles
• Redo most of it when I get feedback and see the more obvious and better way to do the xsl.

So no rush, but it's here if you are currious.

[rbeezer]

Fun fun fun.

;-)

More at a more convenient hour.

[oscarlevin]

I did quite a bit of refactoring since you last looked at this. It is in a state that would be ready to merge, but I'm not being held back at this point if it doesn't for a while. I will keep adding new texstyle files, and with each, will likely need to refactor some anyway.

Three journals: ams-general, taylor-francise-general, and spring-natural-general. Two dependents of ams-general.

Should the logic to get the cls/sty files required for a build be inside core or the cli do you think? I haven't started on that.

[rbeezer]

Sounds good. I'll checkin when I'm ready to take another look. Can grabbing class/style files be a discrete function that the pretext/pretext script could execute on-demand? Could they be fetched once, placed with external files and copied into temporary working directory as needed? Guess it depends on your vision, which is likely very different from last paragraph.

…

On March 19, 2025 2:59:11 PM PDT, Oscar Levin ***@***.***> wrote: oscarlevin left a comment (PreTeXtBook/pretext#2433) I did quite a bit of refactoring since you last looked at this. It is in a state that would be ready to merge, but I'm not being held back at this point if it doesn't for a while. I will keep adding new texstyle files, and with each, will likely need to refactor some anyway. Three journals: ams-general, taylor-francise-general, and spring-natural-general. Two dependents of ams-general. Should the logic to get the cls/sty files required for a build be inside core or the cli do you think? I haven't started on that.