Rename `HTMLOrSVGOrXULElementMixin` to `HTMLOrForeignElement`
Categories
(Core :: DOM: Core & HTML, task)
Tracking
()
Tracking | Status | |
---|---|---|
firefox71 | --- | fixed |
People
(Reporter: e7358d9c, Assigned: fredw)
References
(Blocks 1 open bug, )
Details
Attachments
(1 file)
This change isn’t detectable by JavaScript code running on the web and both Chromium and WebKit have already done so.
Assignee | ||
Comment 1•5 years ago
|
||
Note that it's currently named HTMLOrSVGOrXULElementMixin, and it's actually not a mixin and it actually does not contain the whole IDL attributes (most are duplicated in each interface):
https://searchfox.org/mozilla-central/source/dom/webidl/Element.webidl#165
So there is a bit more work to do...
Comment 2•5 years ago
|
||
and it's actually not a mixin
That can just get fixed. We support the mixin syntax now, but we just haven't converted things over to it yet.
But yes, it's not just a rename to align with the spec here. That said, the changes should be non-web-observable until we start including that mixin in MathMLElement.
Assignee | ||
Updated•5 years ago
|
Assignee | ||
Comment 3•5 years ago
|
||
Untested WIP patch
Assignee | ||
Comment 4•5 years ago
|
||
@Boris: Can you please take a quick look at https://phabricator.services.mozilla.com/D44879 before I try to do more on this?
I think extended attributes are a bit inconsistent between SVG/HTML/XUL and the XUL class lacks the dataset attribute. So these are likely behavior changes. Not sure how important it is and whether that should be tested...
Comment 5•5 years ago
|
||
Adding dataset
to XULElement should be fine. Yes, it's a behavior change. I wouldn't worry about it.
The [Constant]
vs [SameObject]
on dataset
on HTMLElement
and SVGElement
doesn't matter: those are aliases for practical purposes if the type is a type SameObject
can apply to. [Constant]
predates the existence of [SameObject]
in specs.
The lack of [CEReactions]
on tabIndex
on SVG/XUL is likely a bug we have right now, and yes, we may want to add some tests for that. The lack of [Pure]
on the XUL version is not web-observable.
Assignee | ||
Comment 6•5 years ago
|
||
(In reply to Boris Zbarsky [:bzbarsky, bz on IRC] from comment #5)
The lack of
[CEReactions]
ontabIndex
on SVG/XUL is likely a bug we have right now, and yes, we may want to add some tests for that.
Thanks. Regarding this one, is it actually testable? AFAIK one cannot create custom SVG/XUL elements? I took a quick look at testing/web-platform/tests/custom-elements/reactions/ and IIUC all the tests end up creating custom HTML elements in testReflectAttributeWithContentValues().
Comment 7•5 years ago
|
||
You can definitely create custom XUL elements; that's what we're using to replace XBL.
I think you're correct that right now you can't create custom SVG elements. So maybe this is not worth worrying about.
Assignee | ||
Comment 8•5 years ago
|
||
(In reply to Boris Zbarsky [:bzbarsky, bz on IRC] from comment #7)
You can definitely create custom XUL elements; that's what we're using to replace XBL.
OK I've learned something today :-)
I think you're correct that right now you can't create custom SVG elements. So maybe this is not worth worrying about.
Thanks. Then I think I only need XUL tests for tabIndex reaction & dataset.
Comment 9•5 years ago
|
||
I wouldn't worry about those too much, since XUL is not web-exposed and we are working on getting rid of it anyway.
Updated•5 years ago
|
Assignee | ||
Updated•5 years ago
|
Comment 10•5 years ago
|
||
Pushed by nerli@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/85efdc31cf55
Rename HTMLOrSVGOrXULElementMixin
to HTMLOrForeignElement
. r=bzbarsky
Comment 11•5 years ago
|
||
bugherder |
Description
•