Fork me on GitHub
#cider
<
2018-03-07
>
iwannaseethelight00:03:30

i’m going to try and rebuild to see if that fixes things

iwannaseethelight00:03:30

no luck, man I’m sure this used to work in this big project

emoarmy00:03:39

@iwannaseethelight if you're using figwheel, have you opened up your app in a browser?

iwannaseethelight00:03:47

cider-find-ns works in one project but not another, I don’t see anything unusual in *Messages*

iwannaseethelight00:03:02

it will show all the namespaces, but navigating to any of them just says “unable to find …”

emoarmy00:03:37

ahh sorry, I'm not much help then

iwannaseethelight00:03:54

is there some kind of “find uses of” functionality in cider?

bozhidar00:03:48

But there’s find-usages in clj-refactor.el (which is an extension of CIDER).

bozhidar00:03:38

> but the prompt is still there on cider-jack-in-clojurescript.

bozhidar00:03:42

That’s very odd.

bozhidar01:03:48

Seems I found the problem.

bozhidar01:03:55

(and fixed it)

xiongtx04:03:33

@gonewest818 What the cljfmt error here? The output doesn't make that clear. https://travis-ci.org/clojure-emacs/orchard/jobs/350143199

gonewest81804:03:01

— just getting home from a long day. Will look in a little while.

gonewest81804:03:42

The output shows a diff of what’s currently in the file versus what cljfmt produces. The former is prefixed with “-“ and the latter with “+”. Specifically I think the keyword :doc is indented one character to the left of :name on the line above, and subsequently the rest of that map also needs to shift right.

xiongtx05:03:20

@bozhidar Seems that orchard: - Allows merge before CI has finished running - Creates a merge commit by default, which differs from cider (which does a rebase and merge by default) You can fix these in the GitHub settings.

gonewest81806:03:57

I see settings to allow a fast-forward merge (rather than a merge commit) for pull requests, but I don’t see any way to declare a default.

bozhidar08:03:24

I didn’t even know there was a setting for that. I just rebase PRs when I merge them, as that’s my preference.

xiongtx05:03:39

Seems that the Deps Versions badge for orchard is broken; https://versions.deps.co/clojure-emacs/orchard/status.svg gives Deps Internal Server Error Perhaps @danielcompton can help?

danielcompton21:03:43

@ the issue is coming up because the Orchard project.clj uses a read-eval which is disabled in Deps Versions for security reasons

xiongtx21:03:44

☝️ @gonewest818 Do we need the #=(eval ...)?

gonewest81821:03:25

It’s not critical for right now. It’s just used to direct codox output to named directories for instance when there are tagged releases and “master” versions having different APIs. I would stop using the read-eval if Leiningen offered more generalized access to :env/variable_name syntax, but at the moment I’m pretty sure that’s hard coded to specific map keys in project.clj.

xiongtx18:03:10

> if Leiningen offered more generalized access to :env/variable_name syntax @ ☝️ is this possible? I’m not familiar w/ this aspect of leiningen at all.

hypirion19:03:38

I don't think this is possible in general as mentioned, but I don't remember why we ended up doing it for only some keys.

gonewest81819:03:34

I touched that code when I implemented unsupervised gpg signatures (e.g. for lein deploy), and all I saw was support for specific keys. All I did was add another special case for :env/gpg_passphrase. It wasn’t obvious to me why that wasn’t general already.

danielcompton05:03:30

Hey I saw that come through the exception tracker recently. I think it’s an issue with the dependency parser

manuel06:03:36

@bozhidar latest CIDER gets the value of cider-default-cljs-repl from .dir-locals.el correctly. Thank you!

stardiviner06:03:11

I found command cider-doc on (. clojure.lang.RT (next x))'s clojure.lang.RT open URL http://www.clojure.com/lang/RT.html is invalid. Where to find doc for this clojure.lang.RT ?

arrdem06:03:14

@stardiviner there are no docs for RT

arrdem06:03:26

that is unfortunately the official answer

arrdem06:03:56

that is however probably a bug in cider-doc that it even tries to do that.

stardiviner06:03:47

It's not a bug, seems cider-doc provide "Java doc" link for all similar functions. I guess CIDER can improve on this. If it is not available, shouldn't show it.

arrdem06:03:55

Java does not retain javadocs as source metadata in any remotely usable form, so there’s not really a good way for CIDER to answer that in general.

arrdem06:03:10

hummm got an example that works for?

stardiviner07:03:56

Not found workable example yet.

arrdem07:03:59

Yeah there’s probably shouldn’t be. cider-javadoc should work, cider-doc is designed only for Cloure vars AFAIK

bozhidar08:03:14

> Yeah there’s probably shouldn’t be. cider-javadoc should work, cider-doc is designed only for Cloure vars AFAIK

bozhidar08:03:33

Originally yes, at some point we augmented it to work with Java classes as well.

bozhidar08:03:47

I guess that’s a bug that we should look into. I was under the impression that we were rendering this link only if if was part of the metadata.

gganley20:03:39

@bozhidar I’ve been reading through cider-interaction.el and went into subr-x.el and found the following

gganley20:03:57

> ;; NB If you want to use this library, it’s almost always correct to use: > ;; (eval-when-compile (require ’subr-x))

gganley20:03:11

is there a reason that CIDER doesn’t use this convention?