Bug 204543 - macCatalyst: REGRESSION (r251320): WebKit-native media controls do not respond to hover or click
Summary: macCatalyst: REGRESSION (r251320): WebKit-native media controls do not respon...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Tim Horton
URL:
Keywords: InRadar
Depends on:
Blocks:
 
Reported: 2019-11-22 18:10 PST by Tim Horton
Modified: 2019-12-07 03:22 PST (History)
17 users (show)

See Also:


Attachments
Patch (29.67 KB, patch)
2019-11-22 18:11 PST, Tim Horton
no flags Details | Formatted Diff | Diff
Patch (29.67 KB, patch)
2019-11-22 18:13 PST, Tim Horton
no flags Details | Formatted Diff | Diff
Patch (30.86 KB, patch)
2019-12-04 14:55 PST, Tim Horton
no flags Details | Formatted Diff | Diff
Patch (30.85 KB, patch)
2019-12-04 18:38 PST, Tim Horton
graouts: review+
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Tim Horton 2019-11-22 18:10:28 PST
macCatalyst: REGRESSION (r251320): WebKit-native media controls do not respond to hover or click
Comment 1 Tim Horton 2019-11-22 18:11:15 PST
Created attachment 384218 [details]
Patch
Comment 2 Tim Horton 2019-11-22 18:11:17 PST
<rdar://problem/57208621>
Comment 3 Tim Horton 2019-11-22 18:13:10 PST
Created attachment 384219 [details]
Patch
Comment 4 Tim Horton 2019-12-04 14:55:43 PST
Created attachment 384849 [details]
Patch
Comment 5 Tim Horton 2019-12-04 18:38:45 PST
Created attachment 384871 [details]
Patch
Comment 6 Maciej Stachowiak 2019-12-04 23:06:18 PST
Comment on attachment 384871 [details]
Patch

Is it feasible to make test cases for this?
Comment 7 Antoine Quint 2019-12-05 02:32:17 PST
Comment on attachment 384871 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=384871&action=review

Very nice to see those #if !PLATFORM(IOS_FAMILY) go away!

> Source/WebCore/Modules/modern-media-controls/controls/auto-hide-controller.js:63
> +            this._mediaControls.element.addEventListener("pointerout", this);

I suppose we didn't have `touchcancel` before, but I wonder if we need to handle `pointercancel` here too.

> Source/WebCore/Modules/modern-media-controls/controls/auto-hide-controller.js:68
> +            this._mediaControls.element.removeEventListener("pointerout", this);

We add a `pointerleave` event listener that we don't seem to remove.

> Source/WebCore/Modules/modern-media-controls/controls/auto-hide-controller.js:172
> +        let disableAutoHiding = this._pointerIdentifiersPreventingAutoHide.size || this._pointerIdentifiersPreventingAutoHideForHover.size;

`const` would do here.
Comment 8 Antoine Quint 2019-12-05 02:33:38 PST
(In reply to Maciej Stachowiak from comment #6)
> Comment on attachment 384871 [details]
> Patch
> 
> Is it feasible to make test cases for this?

We already have tests for auto-hide of media controls, so we should be able to extend those to work for the macCatalyst case. However, I don't know what constraints may exist due to this being macCatalyst.
Comment 9 Tim Horton 2019-12-05 03:01:49 PST
(In reply to Antoine Quint from comment #8)
> (In reply to Maciej Stachowiak from comment #6)
> > Comment on attachment 384871 [details]
> > Patch
> > 
> > Is it feasible to make test cases for this?
> 
> We already have tests for auto-hide of media controls, so we should be able
> to extend those to work for the macCatalyst case. However, I don't know what
> constraints may exist due to this being macCatalyst.

The existing tests found a mistake I made, and we don't currently have layout testing for macCatalyst. Possible we could come up with something crazy though.
Comment 10 Tim Horton 2019-12-07 03:22:25 PST
https://trac.webkit.org/changeset/253248/webkit