Fork me on GitHub

I think it was cider-words-of-inspiration or something like this. It's basically used to display an inspirational message on connect.


yea I did remember something with inspiration 😛 Guess it's out of scope. But cool instead of lorem ipsum


After upgrading cider today to the latest version I've got a weird error when trying to start a clojure repl in one of my leiningen projects. At first, thought it's a leiningen issue because I also upgrade lein, but it doesn't seem so now (`lein repl` works fine). Here's the error: Here's the stacktrace: Does anyone know what might be causing this? I'm using macOS 12.1.

$ lein -v
Leiningen 2.9.8 on Java 1.8.0_302 OpenJDK 64-Bit Server VM
It seems that after upgrading cider today I face some weird errors -


what's twarc.core? Do you have anything funny in your user.clj? I think #1 thing you should determine is what is invoking tools.analyzer. Generally that's not something Lein or cider-nrepl do. refactor-nrepl does, however it's not present in the stacktrace If you're using refactor-nrepl make sure to use version > 3, older versions have less graceful err handling


checking out again the stacktrace, my bet would be that this has nothing to do with tooling, but with dep resolution (which might have changed with the Lein update?)


basically core.async depends on tools.analyzer, so if you had two deps depending on different tools.analyzer versions, maybe the wrong one is winning


you can just bump core.async and t.ana both to $latest and see what happens


twarc is a quartz wrapper: It's funny that lein repl just works but cider-jack-in fails with the exception


From the deps, this is relevant to core.async and tools analyzer

[org.clojure/core.async "1.3.618"]
   [org.clojure/tools.analyzer.jvm "1.1.0"]
     [org.clojure/tools.analyzer "1.0.0"]
     [org.ow2.asm/asm "5.2"]


I'd try to use latest, for example core.a is now in 1.5.x I don't think cider requires core.a, but the more stuff you give to a dep resolution algo, the funnier it can get :)


Just bumping up tools.analyzer.jvm to 1.2.2 helped. @U45T93RA6 thanks a ton!


I found another thing which I believe worked fine before - cider-pprint-eval-last-sexp and cider-pprint-eval-defunt-at-point now fail with this:

apply: Cannot open load file: No such file or directory, cider-compat
Before, a new buffer would be open with pretty-printed result.


I use spacemacs so it might be related to do - wanted to ask here first to be sure.


known one, I believe the latest cider shapshot fixed cider-compat whoopsies?


cider-compat was remove (its now redundant) at the end of 2021. A package update in Spacemacs should get the latest snapsots of cider and related packages (i.e. helm-cider if using helm)


Updating all my emacs packages and spacemacs to latest develop helped. Thanks @U05254DQM !


Hi all, wondering about the cider nrepl port auto detection functionality. I work with both a both a clj repl, and a cljs repl via shadow, but if a clj repl is already started, it will fail to autodetect the shadow nrepl server port. My best guess from a quick glance at the cider code is that nrepl-extract-port in nrepl-client.el will always choose the clj port over the shadow port, even if both may be valid. Should nrepl-extract-port return multiple ports if they exist? Wondering from someone with more cider expertise if this is issue worthy, or if there is some other configuration approach I'm missing. Thanks!


I'm having some trouble debugging a defrecord in cider (1.2.0snapshot) . When instrumenting the top level form with C-u C-M-x I get the following error: Unable to resolve symbol: STATE__ in this context -- has anyone run into something similar? My error trace is . The form compiles fine, I only run into this failure when trying to debug the top level form. Happy to make a bug report if that is appropriate.

Ben Sless07:01:12

I don't think you can debug defrecords. I tried but got the same exception. What people usually do is pull out the implementation if the method they want to debug to a defn

Ben Sless07:01:38

Then directly dispatch to it in the record


I see. I can do that then, thanks.