Various Updates

[ozobi]

js
highlightJS -> https://highlightjs.org/download/ => update with desired languages, default: https://cdnjs.cloudflare.com/ajax/libs/highlight.js/9.14.2/highlight.min.js
modernizr -> https://modernizr.com/download/?csstransforms3d-domprefixes-prefixes-setclasses-shiv-testallprops-testprop-teststyles
lunrJS -> https://cdnjs.cloudflare.com/ajax/libs/lunr.js/2.3.5/lunr.min.js
clipboard -> https://cdnjs.cloudflare.com/ajax/libs/clipboard.js/2.0.4/clipboard.min.js
jquery -> https://cdnjs.cloudflare.com/ajax/libs/jquery/3.3.1/jquery.min.js
featherlight -> https://cdnjs.cloudflare.com/ajax/libs/featherlight/1.7.13/featherlight.min.js
search.js -> updated for lunrJS 2.3.5
learn.js -> updated for clipboard

css
Added highlight.js theme
Update theme.css to match highlight.js theme
Update featherlight.css

[blackmichael]

@matcornic is there any plan to merge this soon? Or provide feedback for changes? I believe this jQuery upgrade will resolve an XSS concern I currently have with my hugo site.

[coliff]

This is a great PR. I hope to see this merged soon.
FYI; Lunr 2.3.6 is available now if you want to update that in this PR (https://github.com/olivernn/lunr.js/blob/master/CHANGELOG.md)

[matalo33]

This was held up because I wanted to quickly validate checksums of the javascript, however this has proven much more difficult than expected because a couple of the files come from dynamic generators. Here's what I found:

• modernizer - md5 differs and LARGE amount of changes ⛔️
• highlight - md5 differs and LAEGE amount of changes ⛔️
• lunr - md5 mismatch, but diff is trivial 🆗
• clipboard.min.js - md5 match 🆗
• jquery-3.3.1.min.js - md5 match 🆗
• featherlight.min.js - md5 match 🆗

We don't really have a method for validating these files so I'll continue and merge for now. Maybe in the future dynamically generated JS shouldn't get supplied in PRs?

On the other hand, if only maintainers could update dynamic JS files why should anyone trust that we haven't tampered with the JS? The end user should be able to md5 validate the files against upstream. Will give this some thought.

In the meantime, thanks again for your extensive contribution! Apologies it took so long to merge.