Open sosukesuzuki opened 4 months ago
Sounds great. I'd hope basically all implementation bugs end up with test262 regression tests :-)
Thank you, I will try to submit a PR.
Following up from #4038, I think it might help to have additional coverage for ASCII strings (to help debuggability in case an implementation gets single characters right but surrogate pairs wrong) and grapheme clusters (like 👨👦 being split into 👨, ZWJ, and 👦.)
Object.groupBy
andMap.groupBy
take an iterable, such as an array, as their first argument. This can be any iterable, so they can also accept a string.However, it was discovered that in some implementations, a TypeError was thrown when the first argument was not of object type, so this was not working^1.
What do you think about adding test cases to test262 to cover this?