Fork me on GitHub
#cider
<
2016-01-28
>
malabarba00:01:37

@pandeiro @bozhidar actually, yes it changed. In clojure-mode. I'll fix it as soon as I can.

malabarba00:01:13

@isaac @nicola the simplest way to have a dual clojure/ClojureScript is to start an nrepl server, connect to it twice (each connection is a Clojure REPL), and transform one of those connections into a ClojureScript REPL.

malabarba00:01:43

Cider-jack-in-clojurescript does all of that for you

malabarba00:01:36

It's certainly possible to use two processes instead of one, but it's not necessary.

malabarba00:01:20

@isaac > Is there any way make repl can eval both clj and cljs code according the file type? Just to make sure we're on the same page. Are you talking about in-file evaluation (like C-x C-e), or are you talking about evaluating code at the REPL buffer?

jeremyraines00:01:23

what do I do to make sure my code is loaded when I use cider-jack-in-clojurescript?

jeremyraines00:01:12

with leiningen, specifically

isaac06:01:07

@malabarba: yeah! I am talk about in file evaluation(C-x C-e)

nicola09:01:46

@malabarba: start one repl, two connections and one turned into cljs - works for me - thx!

slipset11:01:08

I keep playing around with Planck, but I haven't gotten around to try to do so with Cider. Planck does support Socket REPL, https://twitter.com/mfikes/status/669247547169488896 so I was wondering if that is usable for Cider?

malabarba13:01:24

@pandeiro: @bozhidar I've reverted the indentation change.

malabarba13:01:14

Unrelated, I remember someone here had asked about showing stack traces from the debugger. There's now a key for that on the current snapshots.

malabarba13:01:26

Would be great to hear if it works well.

bozhidar13:01:27

we want to support the socket repl

bozhidar13:01:35

but no one has expressed the desire to work on this

bozhidar13:01:52

I’m extremely busy at work and have next to 0 time for cider these days

cichli14:01:21

I’ve thought about the socket REPL a bit and have a fairly good idea of what needs to be done to implement it - we’d most likely end up re-implementing a fair bit of nREPL (sessions; async eval) and wouldn’t really gain anything

cichli14:01:00

As I understand it one of the benefits of the socket REPL is that you can start it without including any new deps, just setting the system property - but you’d still need to include a dep on cider-socket or whatever to make CIDER work, so why not just include the nREPL dep too?

jonpither14:01:36

hi guys, #390 JVM process leak on windows... is there any workaround (aside from one mentioned in #390). Three of us windows problems are OK, one has this problem

bozhidar15:01:49

@cichli: you’re right to a large extent

bozhidar15:01:58

that’s why I’m not rushing to work on this

bozhidar15:01:15

but in theory the ClojureScript stuff will be easier to implement

bozhidar15:01:22

using the other repl

bozhidar15:01:30

as opposed to using piggieback

bozhidar15:01:38

which might be the big takeaway for us

malabarba16:01:54

@cichli perfectly mimics my thoughts on this. It's something I'd like to support, but it would be a lot of work just to get up to feature parity with nrepl

malabarba16:01:27

Although an easier ClojureScript support would be a big gain for us.