Will Lachance

> Basically I'd recommend the following to make things easier for us all to comprehend and pick up again if it takes a while: > ... Hi yeah, that more...

@jayaddison Hey, revised this PR to reference a specific issue as you suggested. It includes changes from other PRs for ease-of-testing, but that should go away once those are merged....

Update: @jayaddison and I had a back and forth about different approaches in https://github.com/wlach/sphinx/pull/1. After all that discussion, I feel pretty ok about the approach in this PR and feel...

Ok, I rebased this on top of all the work that landed so I think this is ready for review/landing on its own now. In addition to the unit test,...

> Personally I have a preference for fixing this in the Python code so that the JavaScript clients don't have to do repeat work here. But I'm OK with the...

> I've read your comments. AFAICT, everything is ready for that PR to land. Since we merged some related PRs, I'll let you guys tell me whether it's fine to...

Some improvements: on Firefox nightly, pandas now only takes about 6 seconds to load: https://definitely-staging.iodide.io/notebooks/44/ Still a little slow for my liking.

Have been looking at this a bit today during the pyodide sprint. A few ideas have come up: * We could use a space-saving optimization when compiling pandas specifically. @rth...

> Also, maybe delay loading all those 41 so's somehow? And make sure that everything indeed loads from cache (chrome won't cache anything larger than ~40Mb) Ah right! There's actually...

> I spent some time optimizing pandas's import time in in late 2019-early 2020, got most of the low-hanging fruit. There's a decent chance more has accumulated since then. For...