Cleanup: Fix core-test Deps Order, Remove Unused, etc.
#798
Conversation
| #?@(:cljs [] | ||
| :clj [179769313486231570000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000N r/max-double] | ||
| :clj [179769313486231570000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000N r/max-double]) | ||
| :cljr [179769313486231570814527423731704356798070567525844996598917476803157260780028538760589558632766878171540458953514382464234321326889464182768467546703537516986049910576551282076245490090389328944075868508455133942304583236903222948165808559332123348274797826204144723168738177180919299881250404026184124858368N r/max-double] |
There was a problem hiding this comment.
Added cljr specific largest double
| (ns clojure.core-test.bound-fn | ||
| (:require [clojure.test :as t] | ||
| [clojure.core-test.portability #?(:cljs :refer-macros :default :refer) [when-var-exists]])) | ||
| (:require [clojure.test :as t :refer [deftest is testing]] |
There was a problem hiding this comment.
converted to refer for consistency with the rest of the project
| (t/is (= @fut :unset) "bound-fn stays bound even in other thread")))) | ||
| (testing "Threaded/future cases" | ||
| (let [f (bound-fn [] *x*) | ||
| fut (future (f))] |
There was a problem hiding this comment.
removed redundant let
| (t/is (= @fut :unset) "bound-fn stays bound even in other thread")))) | ||
| (testing "Threaded/future cases" | ||
| (let [f (bound-fn* test-fn) | ||
| fut (future (f))] |
There was a problem hiding this comment.
removed redundant let
E-A-Griffin
changed the title
core-test Deps Order, Remove Unused, etc.
| (is (ratio? (- 0 1/3))) | ||
| (is (ratio? (- 0N 1/3))) | ||
| (is (ratio? (- 1 1/3))) | ||
| (is (ratio? (- 1N 1/3))) |
There was a problem hiding this comment.
No reason not to use this over the more verbose instance? ...
| false true | ||
| true false | ||
| #?(:clj (Object.) | ||
| :cljr (Object.) |
There was a problem hiding this comment.
This seemed like an oversight so added it
| :clj Thread/sleep | ||
| :cljr Thread/sleep) | ||
| n)) | ||
| :cljs #(js/setTimeout identity %) |
There was a problem hiding this comment.
:cljr was repeated twice for some reason and there was no cljs option so added something
|
Thanks, Emma. :) I worked through all of the emails for PRs and discussions today, which ultimately led to a conflict for this PR. I will review the PR now, but we'll need to fix that conflict before we can merge. |
| @@ -1,8 +1,9 @@ | |||
| #_{:clj-kondo/ignore [:unused-referred-var]} | |||
There was a problem hiding this comment.
This isn't something we should have to put in every file which isn't using a referred var. If we want kondo to be clean, we should disable these diagnostics at a project level. Same goes for things like inline defs and unused values.
There was a problem hiding this comment.
How do we catch genuine issues then (since there are cases where we'd want this to warn)? Or would it be better just to live with those?
There was a problem hiding this comment.
Given the odd nature of this code base, where we're specifically writing incorrect code, I'd rather we just live with them.
2 participants
This is a large PR full of a ton of small refactorings, mostly for consistency, some just general cleanup.
Given how large this already is, I'll handle
string-testin a separate PRMost changes fit into one of the following, and if they don't I've left a comment to make this easier to review:
:referlisting to be alphabetical:defaultor be treated the same between platforms (most often this is collapsing:cljand:cljrinto:default)clj-kondoignore comments so the files are easier to read for folks usingclj-kondoand so that actual warnings appear differently and don't get lostclojure.core/xinstead ofx)xto_x