Fork me on GitHub
#cider
<
2018-09-04
>
zane03:09:01

What's the right way to control the Java version that's used when you -jack-in? cider-jdk-src-paths?

bartuka04:09:48

Hi ppl, I closed a cider repl buffer and now when I try to M-x cider-jack-in I get cider--gather-session-params: Selecting deleted buffer any help on that?

đź‘Ť 4
bozhidar07:09:14

@iagwanderson Please, file a ticket about this. Looks like some sesman bug.

bozhidar07:09:27

@stardiviner Likely some issue with content-type middleware.

bozhidar07:09:38

You might want to file a ticket about this as well.

bozhidar07:09:33

@vigilancetech Or file a ticket a well with more details about your problems so we can analyze them properly.

bozhidar07:09:51

I’m reasonably sure the support for cljs has never been better than now.

đź‘Ť 4
stardiviner07:09:44

@bozhidar Thanks for checking this out. I will submit an issue.

bozhidar09:09:10

@zane Nope. CIDER will just use whatever Java happens to be first on your exec-path (Emacs’s version of $PATH, which is often derived from it).

Timo Freiberg10:09:09

not sure if this is a CIDER issue or a spacemacs issue... but i get the following messages after running cider-jack-in-clj:

[nREPL] Starting server via /usr/bin/lein update-in :dependencies conj \[org.clojure/tools.nrepl\ \"0.2.13\"\ \:exclusions\ \[org.clojure/clojure\]\] -- update-in :plugins conj \[refactor-nrepl\ \"2.4.0\"\] -- update-in :plugins conj \[cider/cider-nrepl\ \"0.18.0\"\] -- repl :headless :host ::...
[nREPL] server started on 37195
[nREPL] Establishing direct connection to localhost:37195 ...
[nREPL] Direct connection to localhost:37195 established
error in process filter: cider-repl--history-read: End of file during parsing
error in process filter: End of file during parsing
after which i have an nrepl-server buffer containing some stdout of my defs and nREPL server started on port 37195 on host :: - nrepl://:::37195 as well as an empty cider-uninitialized-nreplbuffer

Timo Freiberg10:09:22

i see the nREPL process and cider-connect-clj suggests its port, but i can't connect.

bozhidar10:09:26

@timo.freiberg Seems to me your repl history file is corrupted.

bozhidar10:09:37

Likely deleting it is going to fix the issue.

bozhidar10:09:59

> cider-repl--history-read: End of file during parsing

bozhidar10:09:07

That’s the important part of the log.

bozhidar10:09:24

The unhandled error during startup messes up everything else.

Timo Freiberg10:09:45

ah, thanks! that was it - it works now

bozhidar10:09:19

Not sure what caused the corruption, though. And maybe we should add some error handling around this.

Timo Freiberg10:09:10

maybe i shouldn't have deleted the file... i'll store it somewhere if it happens again

borkdude11:09:20

Are there any limitations to the usage of #break? In a simple function it works, but in some other it doesn’t

borkdude11:09:26

now it works, not sure what I did wrong

borkdude11:09:24

and now it doesn’t. when I eval the whole namespace it doesn’t seem to work, but when I only eval the function with the #break and then call it, it works

borkdude11:09:46

(it’s a namespace with clojure.test, maybe it does some tricks)

bozhidar12:09:38

To be honest - I can’t remember. @malabarba was working on the debugger mostly by himself and I haven’t really had much time to work on improvements there, so I don’t recall how things are implemented exactly.

bozhidar12:09:14

(plus I usually instrument functions with C-u C-M-x)

bozhidar12:09:26

Unfortunately after Artur’s “retirement” the debugger hasn’t seen much love. I hope to get to working on it one of those days, but I doubt it’s going to happen in the foreseeable future.

dpsutton12:09:29

i think vspinu gave it a once-over and made the expanded macro forms much smaller

dpsutton12:09:11

that was an incredible amount of work from malabarba. got my hopes up with that issue he opened recently

bozhidar12:09:35

Yeah, that’s true. The debugger is not complex, but it seems most people think it works like magic and are afraid to touch it. 🙂

mike_ananev12:09:00

@gfredericks Hi! I recently wrote some tips, how to run nREPL process w/o lein with Rebel capabilities http://middlesphere-1.blogspot.com/2018/09/clojure-cli-repl-rebel.html

bozhidar13:09:06

@mike1452 In nREPL 0.4.4+ you can just use --middleware/-m to pass the vector of middlewares, so the handcrafted -e for this is no longer needed.

bozhidar13:09:40

Can’t speak about using rebel-readline with the built-in client, but if it’s as simple as you describe it to be - @bhauman will be pleased. 🙂

mike_ananev14:09:49

@bozhidar yeah! second way in my tips uses -m. Full cmd is: clj -R:repl -m nrepl.cmdline -p 7888 -i -m "[refactor-nrepl.middleware/wrap-refactor,cider.nrepl/cider-middleware]"

mike_ananev15:09:03

@bozhidar is it right way to separate middlewares by comma?

bozhidar15:09:12

Or spaces. It’s just a Clojure vector. 🙂

bozhidar15:09:59

Ah, I didn’t see the second way. Generally the first way is going to be removed at some point, as this predated nREPL 0.4.

qle-guen16:09:42

hello, I can't seem to find how to reload lein dependencies in cider. To reproduce: 1. change [org.clojure/clojure "1.8.0"] => [org.clojure/clojure "1.9.0"] 2. M-x cider-restart or M-x cider-quit then M-x cider-jack-in 3. clojure version still is the same

pauld17:09:57

if using leiningen, I would do lein deps :tree to see if clojure versions are conflicting

pauld17:09:30

If so, you can put in an exclude in your project.clj dependency to resolve the conflict.

qle-guen18:09:38

it also occurs with any new dependency added (even with clj-add-project-dependency from clj-refactor)

bozhidar20:09:52

I can’t imagine how something like this could happen - after all cider-jack-in simply invokes whatever build tool you’re using.

hlolli22:09:09

shadow+cider 20180827.1827 + nrepl 0.4.5 + cider/nrepl 0.18.0, starting cider cljs repl the second time, be it connect or jack-in

Debugger entered--Lisp error: (error "Selecting deleted buffer")
  cider--gather-connect-params(nil #<killed buffer>)
  cider--gather-session-params(("Music/erkitid_2018:localhost:37119" #<killed buffer>))
as I wrote this, I noticed version 0.19.0, going to try that...