@harmmatan: You’re lookinng for resources on vscode extension development? It’s a bit lacking, I would say. The API docs are both good and bad. I can recommend joining the vscode dev slack and the extension channel there: https://vscode-dev-community.slack.com/messages/C74CB59NE

I was thinking about helping out Calva and taking some issues. Do you still think https://github.com/BetterThanTomorrow/calva/issues/89 is an easy one or do you have some other recommendations, Peter?

I don't think it is particularly easy, actually. But it would be awesome to have it done.

I could take a look at this 👀 Do you have insights how I could approach this?

@janne.sauvala Oh, sorry, with your anon avatars I thought your question came from @harmmatan... and was about the issue he had mentioned. Confusion. So those compiler errors... might be a good first issue, actually.

Take a look at the test runner to see how we squiggle things there. And then it will be a matter of reading that ex-data to pick up file path and line numbers and such. Might be that the biggest problem is figuring out when to clear the problem marks...

Slack apparently changed our avatars to be even more confusing 😄 I’ll take a look at that then, thanks!

i don't have println in any of my tests, so was just an observation @pez

@gerred: But I don’t follow what it means “I don’t have println in my tests”…

I was using his example (deftest foo (is (= (println 1) (println 2) (println 3) (println 4) (println 5))))

jepp

and then used Calva: Run current test

gotcha. trying it now

since I just default to that chord while I'm developing

(instead of doing what he did, which was (foo))

i don't think I'd want to see the printlns though, so I think this is behaving as I'd expect it to.

For me it prints:

Running test: foo…
1
2
3
4
5

1 tests finished, all passing 👍, ns: 1, vars: 1

really? odd

clearing my console to make sure I'm not red herring myself.

Can you try with latest release build? Or dev.

oh, it sure does.

PEBCAK

I had to Google that 😃

😄

@pez i've been doing some thinking about the whole REBL in Calva thing, since not everyone has REBL necessarily.

it'd be nice to be able to configure (dunno if there's a ~/.clojure/deps.edn way to do this?) additional nrepl middleware in Calva

thinking about this issue: https://github.com/BetterThanTomorrow/calva/issues/240 and this project: https://github.com/RickMoynihan/nrebl.middleware

@gerred need to run. Will get back to you.

👍

I think it should be a setting in Calva, a path to the REBL.jar and some way to choose if it should be included in the jack-in, then Calva injects everything that is needed. (Even if it looks like a huge dependency list for deps.edn…)

I need to either find some time to experiment with it or get a demo of how it works when it is set up manually. Then I might understand enough to know about how to do it and wether Calva should have some more support, beyond channeling evaluations like that.

The whole thing would benefit from you spearheading it, anyway, @gerred. Just sayin’. 😄

I actually just forked to play with it 😉

Awesome.

hey side note, if you sign up for Github Actions, I'd happily port over my workflow that generates builds on every commit for the dev branch.

I’m available for a zoom session if you like.

and then we could add them as draft/preview releases to kick out the vsix

I need to google Github Actions now.

We already have builds for every commit though.

oh do you?

oh nm

Circle CI.

oh via circle

👍

We create github releases and even publish to the marketpkace with circle as well.

oh perfect.

yeah I think there's a separate "evaluate and send to REBL" command as well. I have REBL starting up just fine, but that's the next step, thankfully @seancorfield has a great command for that over in his Atom setup with Chlorine, so I was able to at least spike it out and test it while jacked in.

Yeah, It is so sweet. Very new, and I am still just in love with it.

though the nrebl middleware is just magical.

I’d love to see it in action.