Sample with processing error about mathjax-a11y HOT 9 CLOSED

@zorkow should I rather file this at SRE?

from mathjax-a11y.

Yes, that is enrichment/SRE related. Must be a regression.

from mathjax-a11y.

There was a problem with inner structure in embellished fences. The patch to SRE solves this.

from mathjax-a11y.

Thanks. The processing error is fixed but now the collapse extension is broken (see the second collapse on the third line).

from mathjax-a11y.

The semantic tree and the embedding in this example are correct, however. It seems to be a problem when collapsing an punctuated/endpunct element.
Here's a minimised version of the example:
\left.a+b+c+d+e\right|^x_y

from mathjax-a11y.

It seems to be a problem when collapsing an punctuated/endpunct element.

Right. Should I open a separate issue?

Here's a minimised version of the example:

Thanks! Confirmed.

It looks like a problem of the HTML-CSS output.

from mathjax-a11y.

I've opened mathjax/MathJax#1332 upstream.

I'm leaving this open to discuss one small detail, I'm curious about: why does one of the | not get fence and stretchy added? (It's from the collapse extension, not enrichment I think.)

from mathjax-a11y.

I'm curious about: why does one of the | not get fence and stretchy added?

Because these are the defaults for the | character, so they are removed by the toMathML() call that is used to generate the MathML that is passed to the enrichment (since the internal format came from TeX rather than MathML originally, toMathML() strips any default attributes).

from mathjax-a11y.

Thanks, Davide.

from mathjax-a11y.