Use MathJax -full config.

[robrap]

TNL-5767

Description

Use the -full config for MathJax to see if it resolves the loading issue.

Sandbox

• https://robrap.sandbox.edx.org (rebuilding now...)

Testing

• Unit, integration, acceptance tests as appropriate
• Performance (no - skipping due to urgency)

Reviewers

If you've been tagged for review, please check your corresponding box once you've given the 👍.

• Code review: @andy-armstrong
• Code review: @cahrens

FYI: @nasthagiri

Post-review

• Rebase and squash commits

[robrap]

@cahrens @nasthagiri: I feel like I have limited time to determine which of these configs is needed and which are not. The options are:

1. Go with the 5 updates from Adam's PR: https://github.com/edx/edx-platform/pull/11585/files
2. Go with the 6 I found via search.
3. Go with just the lms pages we think were reported with issues.
   I voted for Fix js that prevented single step OE problems (ie only self assessment) ... #2, but want to know your thoughts since this would be going with the release if it works.

[robrap]

@cahrens provided the following information:

This is what "full" adds: The -full version also loads the SVG and PreviewHTML output jax main code, plus the SVG mtable extension, which is normally loaded on demand.
http://docs.mathjax.org/en/latest/config-files.html#the-tex-mml-am-svg-configuration-file

Upgrading to 2.7 would also be good. It seems they fixed some bugs with CommonHTML, which means we could possibly go back to CommonHTML instead of SVG.
http://docs.mathjax.org/en/latest/whats-new-2.7.html

There is likely to be a performance impact of this change, but we are in a bind. Someone could look into a follow-up ticket for 2.7 upgrade and dropping -full.

FYI: @ormsbee @adampalay

[robrap]

@cahrens @nasthagiri: I would need 2 thumbs to push this through and test on stage. I set up a sandbox. You can see it working (but it was never broken) here: https://robrap.sandbox.edx.org/xblock/block-v1:HarvardX+Robert-SPU30x+2016+type@html+block@bce6035b21b941a6ad7e7a164ad7c454

At least it didn't regress. @cahrens - not sure if you could help test some of the other locations?

@cahrens: @nasthagiri would not like to interrupt the release, so this will go as part of the release candidate. As long as it doesn't break anything, even if it doesn't fix the issue, it will go to Production as we determine what to do next. Hopefully, it does fix the issue though. :)

[andy-armstrong]

👍 thanks for pushing this through. I agree with updating all six occurrences.

[cahrens]

👍

[robrap]

jenkins run bokchoy

[robrap]

I had a typo. :) Just a test, but didn't see a failure.