Following on your 3rd point about reducing code patterns that lead to
TIMEOUTs, I'm wondering if the test harness has a function that allows you
to fail the WPT entirely, without running the remaining tests?
Here's an example of what I mean: suppose this function is named fail(),
and we have the following test:
test(function() {
if (failingCondition)
fail();
else
// continue test as normal
}, "test 1");
test(function() { ... }, "test 2");
Test 2 would never be run if the failingCondition is true, and all
subsequent tests after test 2 would also not be run, should there be any.
- Keith
On Wed, Sep 26, 2018, 8:22 AM Josh Bowman-Matthews <j...@joshmatthews.net>
wrote:
> I'll be meeting with various WPT stakeholders at the Web Engines
> Hackfest next week. Do we have any thoughts or concerns about the future
> of WPT as a platform to drive Servo's web compatibility and testing web
> features across all browsers? Are there any web platform features that
> are still hard to test that we should be talking about? Are there any
> concerns about the quality of tests, or feature sets that Servo does not
> support that we should discuss rewriting in existing tests?
>
> On my list right now:
> * PRs can merge upstream that fail the lint without being detected; this
> breaks our nightly syncing
> (https://github.com/servo/servo/pull/21814#issuecomment-424737163 and
> https://github.com/web-platform-tests/wpt/issues/13222).
> * Unnecessary use of the named window getter in tests, which Servo does
> not support and is more complex than it sounds
> * Reduce code patterns in tests that lead to TIMEOUT rather than ERROR
> when a feature is unsupported
>
> Please let me know if there is anything I'm missing!
>
> Cheers,
> Josh
> _______________________________________________
> dev-servo mailing list
> dev-servo@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-servo
>
_______________________________________________
dev-servo mailing list
dev-servo@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-servo
