The docs for Promise.all(iterable) mention several times and also give examples that if iterable is empty (e.g. the array ) then it resolves synchronously, and if it's not empty but doesn't contain promises (e.g. the array [1,2,3]) then it resolves a-synchronously. As far as I can tell this info is completely inconsequential, and therefore confusing. I don't think the user can use this info for anything. For instance, let's say a is an array which is either empty or contains only numbers (without any promises), but the developer doesn't know which is it. Say they write this code: Promise.all(a).then(foo, bar); I assert that whether or not Promise.all(a) resolves sync or a-sync has no impact whatsoever. The only thing which does have an impact the user might care about is when foo/bar are invoked. However, this is well defined. Because Promise.all returns a Promise, then calling its 'then' method is well defined - foo/bar will be called with "a clean stack" once the current execution context exits. Whether or not a Promise is "resolved sync" or "resolved async" has no impact because there's no way to inspect its state. This is implementation details which don't affect the user. Therefore I think that all the places where it mentions that it resolves sync or async should be removed from the docs - it's not information the user can use and doesn't help understand how Promise.all() works. It's enough to say that it returns a Promise, and that's it.
Also, if you think there is a difference between "resolves sync" and "resolves a-sync" in this context, then I'd appreciate a code example which shows how it behaves differently depending on whether it resolves sync or async. Here's an example where it _doesn't_ make a difference. var a = ; if (Math.random() > 0.5) a.push(1); Promise.all(a).then(() => console.log("Clean stack")); console.log("Dirty stack"); Which will always print: Dirty stack Clean stack Regardless of whether it was "Resolved sync" or "Resolved async". So how can a user use this information to their advantage? Are there pitfalls they can use this information to avoid? Is there a way at all (other than logging a Promise object at the console) that it can differentiate between two different ways their code might behave?
You need to log in before you can comment on or make changes to this bug.