Fork me on GitHub

Hello, I have a little problem with a core.async reference in my code. I have core.async in my project.clj :dependencies, and all my core.async API calls are working fine. But in one file I have:

#?(:clj (:import (clojure.core.async.impl.channels ManyToManyChannel)))
and when I start up a debug REPL, sometimes it is fine, and sometimes I get a ClassNotFoundException! Not sure if this is something specific to Cursive but ???


Is there a way to reduce the number of spelling mistake markers to more relevant ones? I think Cursive should consider symbols and keywords, from the libraries a Clojure IntelliJ module depend on, spelled correctly, since we can't do anything about those (unless they are pointing to our code via :local/root) That way, we can't easily introduce typos into the new code we are writing anew, but we won't be bothered by the choices of the library developers whos' libs we depend on. Similarly, in deps.edn there is no reason to spell-check artifact names, like lambdaisland/kaocha-junit-xml or com.hyperfiddle/rcf. In fact, those should probably contribute to the spelling dictionary automatically too. Eg. I have the following alias in deps.edn:
  {:extra-deps {lambdaisland/kaocha-junit-xml {:mvn/version "0.0.76"}}
   :main-opts ["-m" "kaocha.runner" "--plugin" "junit-xml"]}
and these words are underlined: • lambdaislandkaocha, twice! 1. in the maven artifact name 2. in the 2nd arg of :main-opts while nothing should be underlined.


Possibly, I haven’t checked to see if there’s a way to programmatically specify exceptions - I assume so. Could you file an issue with your exception cases, please, and I’ll look at it?

👍 2