Closed michaelficarra closed 5 months ago
What I meant was actually the case in zipToObjects
, where we do
i. For each element key of keys, do
1. Let value be Completion(Get(paddingOption, key)).
2. IfAbruptCloseIterators(value, iters).
3. Append value to padding.
without first having checked that paddingOption
is an object. (That's actually an editorial error because Get
only works on objects.)
With regards to this change, we didn't get to consensus on the question of whether iterable-taking things should reject strings, so we don't formally have guidance here. I could go either way.
As written this change also has the effect of allowing bare iterators which lack a Symbol.iterator
. I don't think we should do that, the same way we don't for the iterables
argument.
@bakkot Thanks. Opened an alternative: https://github.com/tc39/proposal-joint-iteration/pull/24
Fixes #22. /cc @bakkot