vemv

Hey there! I gave some thought to this one. Consider the following: ```clj user> (= (clojure.spec.alpha/nilable int?) (clojure.spec.alpha/nilable int?)) false user> (= (clojure.spec.alpha/nilable #{1}) (clojure.spec.alpha/nilable #{1})) false ``` So, in...

If the CI integration was analog to https://github.com/DeLaGuardo/clojure-lint-action , namely applying a diff over a PR's changes only, maybe https://github.com/nedap/formatting-stack/issues/132 wouldn't be that necessary. As of lately, I'm thinking that...

Given that by now, both clj-kondo and eastwood offer their own ignore 'syntaxes', f-s users can simply specify those and f-s will forward them to the underlying linters accordingly. That...

I was thinking, `def` could dispatch on type, expanding to either `clojure.core/def` or `clojure.spec.alpha/def` Would kill `def-with-doc`, which has a verbose name.

Could be. Depends on what all formatters/linters actually do. I think some just `println`, w/o offering a data-oriented API As you probably know `with-serialized-output` is a thing and it already...

Semi-related, `(def foo (java.text.Collator/getInstance))` is correct but potentially slow in a threaded context because it's a `synchronized` class. It's better but in a `defn`. http://hg.openjdk.java.net/jdk7/jdk7/jdk/file/00cd9dc3c2b5/src/share/classes/java/text/Collator.java

Also: insert :import statements if missing (people may refer to classes without importing them) Little caveat: speced.def uses e.g. `^int?` metadata, which may translate to `{:tag int?}` and confuse this...

> It's a good idea to make a definitive list of the supported / unsupported macros (especially interested in the latter). Good call, it's good to agree on a goal...

Also: https://github.com/stacksideflow/cljs-yagni https://github.com/borkdude/carve

(This branch has worked well for me in my personal setup. Will undust this PR soon enough)