search

bokeh / jupyter_bokeh

An extension for rendering Bokeh content in JupyterLab notebooks
BSD 3-Clause "New" or "Revised" License
251 stars 48 forks source link

Decide on distro channels #125

Closed bryevdv closed 3 years ago

bryevdv commented 3 years ago

Specifically: Do we support NPM / manual installation in addition to pre-built pip/conda packages? It's unclear that the current 3.0 NPM package works. I am personally OK with only having the pre-built packages as options as it is less for us to worry about.

If we do want to support NPM install for this extension then the build or other issues with it need to be sorted out.

cc @bokeh/dev

mattpap commented 3 years ago

If jlab doesn't require a npm package, then we should not maintain it, especially given it wouldn't be of any use on its own.

bryevdv commented 3 years ago

It's not required but it is possible to offer one for installation using jupyter labextension directly. [1] I just don't see why that would be valuable enough to bother doing though.

[1] Though the current 3.0 published on NPM may or may not work (might be broken somehow, or might be local issues) But it's alot more fidgety and maybe that's the point.

bryevdv commented 3 years ago

Unless @philippjfr has any objection I think should just proceed with pip/conda only going forward. I'll close this issue EOD today unless there is further comment.

philippjfr commented 3 years ago

I'm a strong +1 on offering pip/conda only. If we already published 3.0 to npm I think we should pull it since it doesn't seem to work anyway.

bryevdv commented 3 years ago

👍 I will close this now then

bryevdv commented 3 years ago

I deprecated all the 3.0.0 versions (including dev) with the message

Use pre-built pip or conda packages instead