no-array-reduce - enable reporting of optional chaining and calling of .reduce()
#2721
Conversation
ipanasenko
changed the title
no-array-reduce - add test cases for optional reduceno-array-reduce - enable reporting of optional chaining and calling of .reduce()
| @@ -12,6 +12,8 @@ test({ | |||
| 'a[b.reduce]()', | |||
| 'a(b.reduce)', | |||
| 'a.reduce()', | |||
| 'a?.reduce()', | |||
| 'a.reduce?.()', | |||
There was a problem hiding this comment.
It doesn't make sense to allow optional call.
There was a problem hiding this comment.
Added it because I saw it in no-array-sort and no-array-reverse 🤷♂️
https://github.com/sindresorhus/eslint-plugin-unicorn/blob/main/test/no-array-sort.js#L14
https://github.com/sindresorhus/eslint-plugin-unicorn/blob/main/test/no-array-reverse.js#L14
There was a problem hiding this comment.
They are valid cases in these two rules, and they both use optionalCall: false,
There was a problem hiding this comment.
I mean a?.reduce() should be invalid, but a.reduce?.() should be valid.
There was a problem hiding this comment.
Sorry, I comment this in a misleading place, I mean L112
There was a problem hiding this comment.
I don't understand, sorry :(
Why one should be valid and another invalid? I thought this rule finds usages of reduce, and reports it (except trivial callback cases, if corresponding option is provided). Why reporting of a?.reduce() should be different to a.reduce?.() and a.reduce()?
Or array.reduce((str, item) => str += item, "") different to array?.reduce((str, item) => str += item, "") and array.reduce?.((str, item) => str += item, "") if we talk about L112?
There was a problem hiding this comment.
Because it's doesn't make sense to check Array#reduce, if user using foo.reduce?.() then foo can be a non-array object.
Explained in this commit 8240f4f (#2721)
|
@sindresorhus I plan to go through all rules to check if |
Fixes #2720
This PR enables checking for optionally chainged
.reduce()calls forno-array-reducerule.