New issue
Advanced search Search tips

Issue 614481 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Sep 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux
Pri: 3
Type: Bug



Sign in to add a comment

Test for releasing pointers when pending_capture != actual_capture

Project Member Reported by mustaq@chromium.org, May 24 2016

Issue description

With lazy capture updates through pendingPointerCapture, releasePointerCapture can have two interpretations when pending capture is different from actual capture. We need a test to cover such cases.

But most importantly, is the spec precise enough? I think we should shift "pending + actual capture" targets upon releasing, to be precise. Is it right?

Here are the cases I have in my mind:
capture@A, capture@B, release@A (no PEs in between)
capture@A, capture@B, release@B (no PEs in between)

 
Owner: nzolghadr@chromium.org
Status: Assigned (was: Unconfirmed)

Comment 2 by mustaq@chromium.org, Aug 30 2016

Summary: Test for releasing pointers when pending_capture != actual_capture (was: Releasing pointers when pending_capture != actual_capture)
Spec has no ambiguity now I believe, after Navid's PR:
https://github.com/w3c/pointerevents/pull/137

We still want to add a test. Let me clarify the cases I meant above, looks like the "no PEs" above are confusing:
capture@A, pointermove, capture@B, (no PEs in between) release@A 
capture@A, pointermove, capture@B, (no PEs in between) release@B



I'd like to work on a wpt test directly for this one. Is it okay to add something like this:
capture@A, pointermove, capture@B, (no PEs in between) release@A (nothing should happen) release@B (capture should be released?)

Comment 4 by mustaq@chromium.org, Aug 30 2016

Yes, this covers both cases.
I updated one of the existing tests to cover this case here:
https://github.com/w3c/web-platform-tests/pull/3665

ptal whenever you get a chance.
Labels: PointerEvent
Status: Fixed (was: Assigned)

Sign in to add a comment