math-jax widget: make it possible to suppress configuration #255
Conversation
|
After just now (very belatedly -- sorry!!) merging your PR for Unfortunately GitHub seems "stuck" wrt the Actions CI checks. After trawling through its UI for awhile, I can't figure out why. Both repos seem to have identical (in this regard) Maybe if you were to make some dummy amended commit, and force-push that (as happened for |
|
This happened to me before, and the fix is to rebase it on top of the current HEAD. |
I don't see that possibility here in the web UI. Maybe you mean if @tfeb does it? (Of course I can clone the branch locally and do anything, myself. And eventually if I need to I will. But atm I'm being lazy; just using some spare cycles to try to catch up on this from the web UI.) |
|
Yep, it looks like @tfeb needs to do it. |
|
Actually no, there's nothing to rebase... It's already up-to-date... Scratch everything that I said. Sorry for the noise :/ |
|
Today I have a bit more time so I figured I'd do this locally. I made a fresh local branch ( Alas the tests pass except on "current" Racket: https://github.com/greghendershott/frog/actions/runs/7223882584. This seems to be the recent change to the HTML emitted by Scribble for section headings:
```
--------------------
FAILURE
name: check-equal?
location: read-scribble.rkt:113:4
actual:
'((h1
()
"The Post"
rsquo
"s Title "
(span
((class "button-group"))
(a
((href "#(part._.The_.Post_s_.Title)")
(name "(part._.The_.Post_s_.Title)"))
(span ((class "heading-anchor") (title "Link here")) "🔗"))
" "
(a
((class "heading-source")
(title "Internal Scribble link and Scribble source"))
"ℹ")))
(h1
()
"1"
(tt () nbsp)
"Section 1 "
(span
((class "button-group"))
(a
((href "#(part._.Section_1)") (name "(part._.Section_1)"))
(span ((class "heading-anchor") (title "Link here")) "🔗"))
" "
(a
((class "heading-source")
(title "Internal Scribble link and Scribble source"))
"ℹ")))
(p () "Here is some text.")
(!HTML-COMMENT () "more")
(p () "Below the fold."))
expected:
'((h1 () (a ((name "(part._.The_.Post_s_.Title)"))) "The Post" rsquo "s Title")
(h1 () "1" (tt () nbsp) (a ((name "(part._.Section_1)"))) "Section 1")
(p () "Here is some text.")
(!HTML-COMMENT () "more")
(p () "Below the fold."))
```
This is a genuine problem to fix with IIRC I fixed a similar issue in Racket Mode (which parses doc HTML for I don't use Frog anymore and had been hoping to avoid more hacking on it. But I guess I need to find time to fix that (or merge a PR for that), then merge this... |
|
Actually I guess the test is just overly specific. I'll merge a commit to adjust that, then rebase this PR on that, then merge... |
Giving #:config #f will cause no configuration part to be added to the MathJax URL. This is needed for MathJax 3, which does configuration a different way, but is innocuous for MathJax 2 or earlier.
greghendershott
force-pushed
the
mathjax-3
branch
from
9043b40
to
d8787cf
Compare
|
Sorry, I'm away with only a phone so haven't been following this properly. I will look more closely when I get home.
|
|
@tfeb The TLDR is I (very belatedly) merged your PRs for markdown and frog repos. (The rest of it is noise re GitHub and CI -- which had nothing to do with your PRs per se. So please ignore until you're back from being mostly offline, and please continue hopefully enjoying that state. 😄) |
This makes a small, compatible, change to the
math-jaxwidget which helps support for MathJax 3. The change is to allow the#:configoption to be#f(previously it had to be a string). Doing this will cause no configuration part to be added to the MathJax URL, which is needed for MathJax 3, which does configuration a different way.This can't make any difference to any existing uses of the widget, since previously this option had to be a string, so
#fwould have been illegal.This is the first of two changes needed for #252, the other one being a change to markdown, which I need to check a bit more. As this is so simple and compatible I figured I should get this out of the way first so I can stop needing to rebase my branch.
I'm willing to handle any support problems which arise as a result of this change.