Clojurians Log v2

@jonurnieta I found one false positive here:

I'll make an issue for it

https://github.com/clj-kondo/clj-kondo/issues/2704

I'm now looking into it. Seems an issue with defmulti

Ah, found the issue. The linter complains because in the cljs part of the .cljc file the deref var isn't used.

(ns scratch
  {:no-doc true}
  (:refer-clojure :exclude [#?(:clj deref) -deref])
  (:require
   [sci.impl.types :as types]))

(defmulti #?(:clj deref :cljs -deref) types/type-impl)

Maybe we should limit the warning if it's true for both branches only?

Yes, that could be a solution. Or maybe suggest excluding only in clj part. I think is cleaner limiting the warning if it’s true for both branches

That's not consistent with the :refer-clojure-exclude-unresolved-var linter

> I think is cleaner limiting the warning if it’s true for both branches ah yes

okey, I will try to look at both branches then

Thanks for the reproduction examples

I'm already looking at this, no PR needed

👌🏿 thank you

pushed to master

Nice! 🙌🏿

Also updated the findings on master now... it gets rid off a whole number of new findings :)

I now realize that the names of :unused-excluded-var and :refer-clojure-exclude-unresolved-var are not consistent. Maybe we should rename the latter to :unresolved-excluded-var?

I used :unused-… based on others like unused-value, unused-import etc

makes sense but would be nice if names could be consistent

What about unused-refer-clojure-exclude ?

I was suggesting to change the name of the other linter

I like unresolved-excluded-var then

me too

If you want to do a PR, that would be great. Else I'll do it later

Okey, I will do it now

Done!

thanks!