[Webkit-unassigned] [Bug 158554] New: Using animation and transition to drive transform styles gives incorrect hitbox (hit testing)

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=158554

            Bug ID: 158554
           Summary: Using animation and transition to drive transform
                    styles gives incorrect hitbox (hit testing)
    Classification: Unclassified
           Product: WebKit
           Version: Safari 9
          Hardware: Macintosh
               URL: https://jsfiddle.net/MadLittleMods/pd7yk1cq/
                OS: OS X 10.11
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: Layout and Rendering
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: mad.little.mods at gmail.com
                CC: simon.fraser at apple.com

Tested and reproduced on Safari 9.1 (WebKit 601.5.17), Safari Technology Preview 9.1.2 (WebKit 602.1.35) on OSX 10.11.4 and with the mobile-Safari simulator on iOS 9.3 (WebKit 601.1.46).

When you have an animation and transition driving some `transform` styles the hitbox (hit-testing) is offset to where it was originally positioned before the animation.

Reproduction steps:

 1. Open the demo, https://jsfiddle.net/MadLittleMods/pd7yk1cq/
 2. Notice how the box turns green when you hover over it
 3. Click on the box to animate it into position.
 4. Notice how hovering over the box isn't turning it green
 5. Notice how hovering where the box used to be turns it green


---


But there seems a be a bit more going on. With that demo, toggling the `.is-active` class by listening to clicks on the body or box itself, reproduces the bug.

If you instead toggle the class with a separate button, the hitboxes are correct, https://jsfiddle.net/MadLittleMods/pd7yk1cq/11/

In my real world scenario, I am toggling a "next" button that slides in the next step but it still reproduces the bug. Hopefully these extra factors don't over-complicate things.


---


Here are some old reports but were the only reports I could find that are slightly relevant:

 - https://bugs.webkit.org/show_bug.cgi?id=119344
 - https://bugs.webkit.org/show_bug.cgi?id=153202
 - https://bugs.webkit.org/show_bug.cgi?id=124777

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20160609/3ae8f8ab/attachment.html>