redundant loading of certain MJ config settings, and false testing

[Alex Jordan]

If I am reading pretext-html.xsl correctly, it uses MJ4, and always loads one of two CDN configurations for MJ4.

Then there is

pretext/js/mathjax_startup.js

with some modifications to the MJ config. Some things about this file are redundant if it's given we are loading one of those CDN configurations. They already load the amscd and color packages. So is there a reason to keep these redundantly added here? Maybe there is (or will be) a time to load MJ without one of those two CDN configurations. But at first glance they look like cruft that carried over from older implementations of MJ in PTX.

More significant, I think. The sample-article "tests" loading an additional package using the "cancel" package. But the cancel package is again already loaded by these two CDN configurations. So I don't think we are truly testing things by using "cancel" in the sample article. We would need to use an MJ extension that is not already loaded. If you agree, which one to use? For example the "physics" extension could be used. It defines a new thing: \abs{}, and we could use that somewhere in the SA for testing. But "physics" changes some default macro definitions, and maybe we don't want that.

Here is the list of extensions:

https://docs.mathjax.org/en/latest/input/tex/extensions/index.html

Many, like cancel, color, and amscd, are loaded by the autoload extension, which is loaded by those two CDN configurations. So many on this list would not help us actually test things.

[Rob Beezer]

Caveat: MJ4 seemed to be working great, we ironed out a few things, and went
for it. I can't say that anybody did a real careful evaluation of clean-up
potential. And I can't say I understand when we need to typset stuff -
page-load, knowl reveal, Runestone assignment page, student free-response text
box - you name it. So there is probably lots of room for improvement if it
doesn't break anything.

I like the "cancel" package! It *was* a great test. The example should stay,
but you are right, we need to find a better way to demo additional packages.
The "extpfeil" package did a whole lot of re-definitions on the LaTeX side.
That was a real mess, that I think we have finally wiggled out of. Let's be
*very* careful about what we decide to support.

Maybe not your question here, but do you thiunk we should be doing things like
loading "cacel" on teh LaTeX side since authors will be using them
"automatically" on the MathJax side? Can we refuse some bad-actor packages like
"physics"?

Rob

On 6/10/26 10:15, Alex Jordan wrote:
> If I am reading pretext-html.xsl correctly, it uses MJ4, and always loads one of
> two CDN configurations for MJ4.
>
> Then there is
> pretext/js/mathjax_startup.js
> with some modifications to the MJ config. Some things about this file are
> redundant if it's given we are loading one of those CDN configurations. They
> already load the amscd and color packages. So is there a reason to keep these
> redundantly added here? Maybe there is (or will be) a time to load MJ without
> one of those two CDN configurations. But at first glance they look like cruft
> that carried over from older implementations of MJ in PTX.
>
> More significant, I think. The sample-article "tests" loading an additional
> package using the "cancel" package. But the cancel package is again already
> loaded by these two CDN configurations. So I don't think we are truly testing
> things by using "cancel" in the sample article. We would need to use an MJ
> extension that is not already loaded. If you agree, which one to use? For
> example the "physics" extension could be used. It defines a new thing: \abs{},
> and we could use that somewhere in the SA for testing. But "physics" changes
> some default macro definitions, and maybe we don't want that.
>
> Here is the list of extensions:

> https://docs.mathjax.org/en/latest/input/tex/extensions/index.html <https://
> docs.mathjax.org/en/latest/input/tex/extensions/index.html>

> Many, like cancel, color, and amscd, are loaded by the autoload extension, which
> is loaded by those two CDN configurations. So many on this list would not help
> us actually test things.
>

> --
> 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/
> CA%2BR-jrdpz7OREF0kfBrE85tF2MJTiyWDLesbVR-%3DRm7g08EvEQ%40mail.gmail.com
> <https://groups.google.com/d/msgid/pretext-dev/CA%2BR-
> jrdpz7OREF0kfBrE85tF2MJTiyWDLesbVR-%3DRm7g08EvEQ%40mail.gmail.com?
> utm_medium=email&utm_source=footer>.