Simon Parten

@tgodzik I haven't managed to spend enough time with this, but here is an initial (horrible) take on this. Which currently, doesn't work. https://github.com/scalameta/metals/compare/main...Quafadas:metals:jupyter_notebooks https://github.com/scalameta/metals/compare/main...Quafadas:metals:jupyter_notebooks The idea is to add...

I've created a PR... I hope to the right place... both marked WIP

@tanishiking I'm not sure how relevant this is to the discussion, but I had a look at almond 0.13.1 (which is awesome) and following the instruction here https://jupyterlab-lsp.readthedocs.io/en/latest/Configuring.html with metals...

I don't understand why this hasn't gone through :-(... time to down tools for now.... trying too reproduce locally.

As far as I can tell, the failing task simply takes longer than 90 minutes - but then how can this work normally in CI?

@lolgab Thankyou for being willing to look 🙏 ... I'm very grateful! The map actually should always be there - the new property of the `ScalaJsModule` has type `Map[String, String]`...

I don't understand how that change, could have made it pass CI? I'm happy it's green though ...

@lolgab With reference to the implementation as a map, I agree with your commentary, although I would be willing to defend the current implementation for a second look. Arman and...

@lolgab I am agreed, but fearful of adding extra complexity. Jumping straight to regexing things would not be my personally preferred solution - I would bow to democratic will though....

@lolgab Would it be possible to re-run the failing test? I think it should pass... I'm not 100% clear why sometimes that suite times out .