1403055 - [Pointer Event] Revise the implementation to fire boundary events by handling pointercancel and pointerup

Status: RESOLVED FIXED

(Core :: DOM: Events, enhancement)

Description

[Ming-Chou Shih [:stone]]

Now we fire boundary events by handling pointerup and pointercancel in ESM::PostHandleEvent. We call GenerateMouseEnterExit twice when handling pointercancel.

Comment 1

[Ming-Chou Shih [:stone]]

Now we call GenerateMouseEnterExit twice when handling pointercancel. The second one is redundant since we do some checks in GenerateMouseEnterExit. And we should call GenerateMouseEnterExit before removing the entry in mPointersEnterLeaveHelper, which is used in GenerateMouseEnterExit.

Trying to fix the incorrect handling of pointercancel and pointerup and merge them together.

Comment 2

[Ming-Chou Shih [:stone]]

Attachment: Bug 1403055 Part1: Revise the implementation to fire boundary events by handling pointercancel and pointerup.

We don't need to call GenerateMouseEnterExit twice when handling pointercancel. The second one is redundant since we do some checks in GenerateMouseEnterExit. And we should call GenerateMouseEnterExit before removing the entry in mPointersEnterLeaveHelper, which is used in GenerateMouseEnterExit.

Review commit: https://reviewboard.mozilla.org/r/185424/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/185424/

Comment 3

[Masayuki Nakano [:masayuki] (he/him)(JST, +0900)]

Comment on attachment 8914030 [details]
Bug 1403055 Part1: Revise the implementation to fire boundary events by handling pointercancel and pointerup.

https://reviewboard.mozilla.org/r/185424/#review190410

Thank you for separating the bug for this change. Let's try this. I wonder, can or do we have automated tests for this? If you'll add automated tests for this, please add new patch and request review to me.

Comment 4

[Ming-Chou Shih [:stone]]

Attachment: Bug 1403055 Part2: Add a test to make sure we fire pointerout and pointerleave before pointercancel.

Review commit: https://reviewboard.mozilla.org/r/185930/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/185930/

Comment 5

[Ming-Chou Shih [:stone]]

Comment on attachment 8914030 [details]
Bug 1403055 Part1: Revise the implementation to fire boundary events by handling pointercancel and pointerup.

Review request updated; see interdiff: https://reviewboard.mozilla.org/r/185424/diff/1-2/

Comment 6

[Masayuki Nakano [:masayuki] (he/him)(JST, +0900)]

Comment on attachment 8914584 [details]
Bug 1403055 Part2: Add a test to make sure we fire pointerout and pointerleave before pointercancel.

https://reviewboard.mozilla.org/r/185930/#review190890

Looks good and please check if this runs as expected in <iframe> of the test harness before landing. (Perhaps, running per directory is enough to check it.)

Comment 7

[Ming-Chou Shih [:stone]]

https://treeherder.mozilla.org/#/jobs?repo=try&revision=f22c7449df4a77dc622f563b37a2cd990fe98ab3

Comment 8

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/a10c13ddb8b7
Part1: Revise the implementation to fire boundary events by handling pointercancel and pointerup. r=masayuki
https://hg.mozilla.org/integration/mozilla-inbound/rev/49820a586d6d
Part2: Add a test to make sure we fire pointerout and pointerleave before pointercancel. r=masayuki

Comment 9

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/a10c13ddb8b7
https://hg.mozilla.org/mozilla-central/rev/49820a586d6d