Add tests for microtask checkpoints and custom element constructors #5208
Conversation
Firefox (nightly channel)Testing web-platform-tests at revision 1c31f4d All results1 test ran/custom-elements/microtasks-and-constructors.html
|
Chrome (unstable channel)Testing web-platform-tests at revision 1c31f4d All results1 test ran/custom-elements/microtasks-and-constructors.html
|
|
Notifying @alsemenov, @deepak-sa, @dominiccooney, @hayatoito, @kojiishi, @rniwa, @sgrekhov, and @TakayoshiKochi. (Learn how reviewing works.) |
|
These tests are now available on w3c-test.org |
Firefox (nightly channel)Testing web-platform-tests at revision 8db354b All results1 test ran/custom-elements/microtasks-and-constructors.html
|
Chrome (unstable channel)Testing web-platform-tests at revision 8db354b All results1 test ran/custom-elements/microtasks-and-constructors.html
|
Fixes part of whatwg/html#2381. The other part is covered by whatwg/html#2457. Tests: web-platform-tests/wpt#5208
This fixes part of #2381 (the other part is in DOM's "create an element"). Note this has two consequences, as discussed there: * Properly managing the entry and incumbent concepts. This is technically unobservable in the current spec landscape. * Ensuring microtasks are run, if the construct is initiated with an empty stack (such as during the parser). Tests: web-platform-tests/wpt#5208
| const xBad = document.querySelector("x-bad"); | ||
| assert_false(xBad.hasAttribute("attribute"), "The attribute must not be present"); | ||
| assert_true(xBad instanceof HTMLUnknownElement, "The element must be a HTMLUnknownElement"); | ||
| }, "Microtasks evaluate immediately when the stack is empty inside the parser, causing the " + |
There was a problem hiding this comment.
I don't understand this description, is it in reference to a check somewhere in https://html.spec.whatwg.org/multipage/scripting.html#concept-upgrade-an-element?
AFAICT, the test is either asserting that the doMicrotasks bit above hasn't run yet?
There was a problem hiding this comment.
It's referring to the last step in https://html.spec.whatwg.org/#clean-up-after-running-script.
There was a problem hiding this comment.
OK, so the microtask is in fact run, the attribute is added, then there's some "checks on no attributes" that I guess happens as part of the upgrade. I couldn't find an explicit attribute-related check anywhere.
I played a bit with the test and figured out that I don't understand custom elements. The this in in this.setAttribute isn't the same thing as xBad is later. I suppose that's only natural, one can't turn an instance of HTMLUnknownElement into an instance of something else.
Some links to help guide the less familiar reader (me) might be nice, but I'll merge this and that can be an optional follow-up.
domenic
force-pushed
the
custom-elements-construct
branch
from
d6b5696
to
8f933fe
Compare
LintPassed |
This fixes part of whatwg#2381 (the other part is in DOM's "create an element"). Note this has two consequences, as discussed there: * Properly managing the entry and incumbent concepts. This is technically unobservable in the current spec landscape. * Ensuring microtasks are run, if the construct is initiated with an empty stack (such as during the parser). Tests: web-platform-tests/wpt#5208
Follows whatwg/html#2457 and whatwg/dom#430.
/cc @rniwa
This change is