Use jquery to make the javascript code nicer

[jasongrout]

This ticket splits off part of #4267.

CC: @boothby @timothyclemans @robertwb

Component: notebook

Issue created by migration from https://trac.sagemath.org/ticket/4704

[jasongrout]

comment:1

Attachment: jquery-javascript-cleanup.patch.gz

This depends on the patch at #4700.

[jasongrout]

comment:4

Ignore the dependency on #4700. Instead, this depends on #3767.

[williamstein]

comment:6

This definitely needs to be commented on by Tom Boothby, since this patch simply deletes the entire optimized-for-us javascript AJAX "framework" he wrote, and replaces it by jQuery's. Is jQuery's actually better?

[jasongrout]

comment:7

Yep, I agree. CCing boothby.

[boothby]

comment:8

I'm not sure I see the point of this. I'm inclined to say, "If it ain't broke, don't fix it." One might accuse me of being biased... but I really don't think I am here -- I'd be happy to see my code go if it made the rest of the notebook code significantly better.

There's one benefit that I see:

'newcell=0' + '&id=' + id + '&input='+escape0('%__sage_interact__\n' + input))

becomes

{newcell: 0, id: id, input: '%__sage_interact__\n' + input}

and this is more readable (I'm hoping that the escape0 functionality is preserved). I'd like to see what was, bradshaw, and tclemans say about this. I'll test after finals are over.

[jasongrout]

comment:9

I like the cleaner syntax. The other thing that may be nice about jquery is that it offloads maintaining to a third party that presumably has more time to focus on it. That may be a non-issue for our code right now, though.

[robertwb]

comment:10

If there isn't any behavior or performance degradation, I think cleaner syntax is worth having, especially if it makes it easier for more people to work on the notebook code. Also, assuming that jquery can do all we need it to, the more we can offload javascript ajax and browser compatibility code to a larger community the more time we can focus on writing interesting stuff.