Test for unhandled promise rejection tracking

[keirlawson]

As per https://github.com/tc39/ecma262/pull/76

[ljharb]

fwiw, that tracking is not part of the language spec. It's something browsers can choose to opt in to - and those hooks you linked to are just so that the browsers don't have to write overriding spec text.

I don't think it's something we should be testing for, because it's not required by ES - but nor is it something we can test for in any implementation I've seen (unhandled rejection tracking is console-only)

[zloirock]

I disagree. Promise rejection tracking events can be added as a part of a new table for ES-related HTML features, like #636.

[ljharb]

Sure, if it's in a new table that makes sense. The tests shouldn't be added until there's at least one browser implementation, though.

[zloirock]

It's implemented in Chrome 49+.

[ljharb]

Good to know - once that ships stable then my conditions are met :-)

[zloirock]

Strange condition. Why do you think so? :) It's a spec.

[Daniel15]

Since Chrome supports this now, I'll see if I can add it to the table.

[zloirock]

To which table? :) We still haven't a table for it.

[Daniel15]

Haha. That's a good point. I guess it doesn't make sense on the ES6 page since it's specific to browsers. ¯_(ツ)_/¯

In any case, 'onunhandledrejection' in window seems to be a sufficient test for it. Looks like only Chrome supports it right now.

[kangax]

What about non-standard table?

[zloirock]

@kangax it's already a part of HTML standard. As I wrote before, makes sense add a new table for ES-related HTML features.

[chicoxyzzy]

IMO this should be on caniuse website instead. or we'll end up creating a copy of caniuse in a new tab of compat-table