These are chat archives for ensime/ensime-atom

28th
Feb 2016
Matthew de Detrich
@mdedetrich
Feb 28 2016 02:48
@hedefalk Is there an update on ensime/ensime-atom#177
Would love to start using enimse properly :smile:
Chip Senkbeil
@chipsenkbeil
Feb 28 2016 03:06
@magnusart sorry I didn't see your question about the debugger rewrite earlier. The mention didn't notify me because of a typo. Anyway, I've been quite busy the past two weeks, but can hopefully resume work this coming week. Progress of the ensime-server update boils down to me first getting 1.1.0 of the debugger out the door. That introduces the last abstractions I think I need to update all of ensime-server with the new debugger API. The actual conversion of the ensime-server should, I think, be pretty quick. The issue for ensime-server can be found here: ensime/ensime-server#1223
Viktor Hedefalk
@hedefalk
Feb 28 2016 09:49

@Ciantic Awesome! I actually spent yesterday pulling out stuff into ensime-node thingy. I can't really push anything out yet either so I guess we're gonna have to merge our explorations later, but that's ok I think. What's important is learning at this point.

I'm all for typescript, np. I think ensime-node could probably be written in typescript for better documentation of the API and there'll be no problem calling that from ensime-atom using coffescript or ensime-vscode using typescript.

Come to think of it, I really really hate those function() boilerplate and return-statements :) But I'd definitely would liked types… Maybe we can argee on Purescript in the end? :)
Viktor Hedefalk
@hedefalk
Feb 28 2016 10:06

@mdedetrich No, sorry. Comments in ticket. Please try to isolate with really clean state and more focused reports. I haven't been able to reproduce so it's really hard to do much. Issues are definitely server or presentation compiler (SOE and scala.reflect.internal.FatalError in pres. compile), but if ensime-atom provokes it and not ensime-sublime or ensime-emacs, I would very much like to help in nailing what's the provocation.

But please try to create a reproducable instruction from zero state. That's really the big work of nailing these things…

Jari Pennanen
@Ciantic
Feb 28 2016 10:18
@hedefalk here is all the code I've written yet: https://gist.github.com/Ciantic/e61f95a01cc918827d09
async await is very neat, that's the reason why your coffeescript looks funky as you've mentioned in the comments
now I'm trying to figure out how to create a .ensime file, docs ain't helping
Jari Pennanen
@Ciantic
Feb 28 2016 10:23
there is a broken link here: http://ensime.github.io/editors/emacs/install/ and another broken link here: ensime/ensime-server#885 basically, I can't find a working link at the moment
Ghost
@ghost~540393fe163965c9bc2018ce
Feb 28 2016 10:31
"a broken link" which one?
Ghost
@ghost~540393fe163965c9bc2018ce
Feb 28 2016 10:36
somebody was asking about creating custom .ensime earlier this week. The examples are in the ensime-server source code under EnsimeConfigSpec and schema defined in config.scala
but you should only need to do this if you have an in-house build tool or something, why do you need to do that?
Jari Pennanen
@Ciantic
Feb 28 2016 10:39
@fommil on the section "Manual" : http://ensime.github.io/editors/emacs/install/#manual
I'm doing this to create tests against, but I can figure this out, there is more than enough tools to generate one
Ghost
@ghost~540393fe163965c9bc2018ce
Feb 28 2016 10:41
@Ciantic I just updated the docs
Jari Pennanen
@Ciantic
Feb 28 2016 10:41
I want to have testdata directory where I run tests
with .ensime and perhaps some scala files
Ghost
@ghost~540393fe163965c9bc2018ce
Feb 28 2016 10:43
that whole page needs a rewrite
Magnus Andersson
@magnusart
Feb 28 2016 11:16
@chipsenkbeil thanks very much for info!
Viktor Hedefalk
@hedefalk
Feb 28 2016 12:08
I'm trying to break out ensime-node now, but I'm having git issues. I'm trying this approach: http://stackoverflow.com/a/17864475/312873 but it simply isn't working. Since the soon to be ensime-node code is split in two folders /lib and /spec of ensime-atom, I move them first to a common parent folder.
mkdir ensime-node
mkdir ensime-node/spec
git mv lib/ensime-client ensime-node
git mv spec/ensime-client ensime-node/spec
git commit -m "move ensime-client to single folder before breakout"
git log --follow ensime-node/ensime-client/client.coffee
gives me the full history of client.coffee, one of the files that has been hanging on from the beginning…
But then I do:
git subtree split -P ensime-node -b ensime-node-breakout
git checkout ensime-node-breakout
git log --follow ensime-client/client.coffee
This only gives me this single last commit, no history preserved.
Viktor Hedefalk
@hedefalk
Feb 28 2016 12:19
Any tips? Gotta be some git wizs here? @fommil ?
Ghost
@ghost~540393fe163965c9bc2018ce
Feb 28 2016 12:20
lemme find a commit that might help
there was an atlassian blog post I used to get it set up initially
I'd personally go for a repo clone and preserve full history
I assume you don't have any binary blobs
there is an app to clean those out
Viktor Hedefalk
@hedefalk
Feb 28 2016 12:22
Yeah, hm. Your right. No big stuff here, so should probably just clone.
Good idea! I'll just do that!
Thanks!
Matthew de Detrich
@mdedetrich
Feb 28 2016 13:29
@hedefalk I thought @rorygraves knew what the reason was behind the issue?
Matthew de Detrich
@mdedetrich
Feb 28 2016 15:12
It seems weird that you can’t reproduce it though, I have done clean runs and everything
The latest repo I gave with the screenshot is a reproduction from zero state
I will update the ticket with a complete reproduction of my steps to see if that helps, will also disable global plugins