Fork me on GitHub

Q: is there a new exception behaviour for the test integration. this doesn’t report an exception any more….

(deftest foo
  (/ 1 0)
  (is (= 1 1)))


seems like all exceptions, including assertion failures, are no longer reported in the console


the red failure popup is correct. just no details of why




IntelliJ IDEA 2021.3.2 (Ultimate Edition) Build #IU-213.6777.52, built on January 28, 2022 Runtime version: 11.0.13+7-b1751.25 aarch64 VM: OpenJDK 64-Bit Server VM by JetBrains s.r.o.


Cursive 1.12.4-eap1-2021.3


I’ll check that, thanks. In general I tidied things up a lot, and was wondering if I’d done too much.


But when I checked the stacktrace still came out in the REPL, that might have been broken by a subsequent change.

🙏 1

Thanks. I've switched back to stable for now


Cool, I should have a fix for this out in a day or two.

🙏 3

That’s out now.


thx. I’ll give it a roll


@U0567Q30W this EAP release handles exceptions a little too well. they appear as red gutter markers but there’s no obvious way to see the stack trace. Am I missing a stack trace feature or is there another explanation?


The stack trace should come out in the REPL console.


Do you see it there? That was what happened in releases prior to the current EAPs - would you also expect to see it in the tooltip?


it did not appear in the console. console would be my preference as there’s plenty of room and I can reach it with keybindings


That’s definitely on 1.12.4-eap2, correct? That’s odd. And you’re seeing the exception marked as a failure in the editor?


it was the most recent EAP. I’ve gone back to stable because I’m testing a lot today


yes the exception showed a red marker in the left gutter


Ok, I’ll see if I can figure out any situations where that wouldn’t happen.


thx. maybe one of the other interested users could confirm? it could be specific to me


could also demo if you want to see/capture the behaviour. the advantage of similar timezones shouldn’t be wasted 🙂


That would be great, thanks.


I’d like to see that.


ok. taking a brekkie break, then open all morning


Sure, me too - let me know when you’re back.


back now but it’s a false alarm. while getting ready to demo, I’ve update to EAP and restarted a few times. now I see stack traces in console


must have been IDE in an odd state (or my brain without caffiene)


I just spent some time staring at the code and couldn’t figure that one out.


The only think I can think of that might have caused it would be if there were an exception which provoked an assertion failure in clojure.test, rather than provoking an error. I think the logic there might have changed slightly in this EAP.


sorry for the confusion. I’ll restrict bug reports till after morning cuppa in future


No worries. If you do see anything like that let me know, I’m planning to get another build out this afternoon.


Hi there, where is this information stored and can I store it in the project somehow?

Jordan Robinson17:05:19

hey, from my notes, this is at Settings → Languages & Frameworks → Clojure → Symbol Resolution

Jordan Robinson17:05:31

hopefully that hasn't changed in the versions since I wrote those notes

Jordan Robinson17:05:45

I also wanted to store them in the project but couldn't find a way, hopefully someone else knows


ah found it, there seems to be a way, i’ll check where this info is stored


(your notes where right, thank you!)

👍 1
Jordan Robinson17:05:10

yeah, I think from memory it stores it with .idea but I was kind of hoping to have it store somewhere in the root so it could be shared with others

Jordan Robinson17:05:45

but like I say this is all from memory so I could be wrong, hope it helped though 🙂


so that’s how you store it, nice

AJ Jaro18:05:09

I think there is a way to export the data as well


Yes, various people have come up with scripts to fiddle with that data. I don’t explicitly choose where to store it, it’s just using IntelliJ’s standard config mechanism.

R.A. Porter18:05:33

I use parinfer, but when I paste code, I sometimes need to set structural style to off, paste, then turn back on. I looked to see if there was an option under Copy/Paste Special... to paste as though structural were off, but there was not. Is there a way to do that? If not, I'd request it as a very low-importance feature.


Is it possible to declare somehow a custom resolve form? Particularly a combination of def+let.

(some-ns/def some-sym 
  [a 0
   b 1
  (+ a b)
This snippet wouldn't work if mark some-ns/def as def or let because it's both of them. defn isn't suitable too. I know it's uncommon and rare someone would need this stuff, however I really-really need this.


Just curious, have I asked some well-known non-issue or is this just how support works with selective ignoring?


Support is just busy, sorry. There’s no good way to do this currently, sorry. I’m gradually working towards it, but it’s not there yet.


@U44T5FWA3 is some-ns/def defined by your project, or is it a 3rd-party one? if it's yours, then you might want to consider being less tricky with your some-ns/def and just compose it via the built-in clojure.core/def and your custom macro, which then u can set to be resolved as clojure.core/let. eg.:

(def some-sym
    [a 0
     b 1
    (+ a b)


it's probably a small price to pay, in exchange for having symbol resolution available and hence code navigation and refactoring too.


All bindings are unresolved but actually are easily resolvable 😞