Overflow Hidden Doesn't Work During a Transition
Reported by
danielhi...@gmail.com,
Apr 18 2016
|
|||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.11 Safari/537.36 Example URL: http://jsfiddle.net/mcNqc/2/ Steps to reproduce the problem: 1. http://jsfiddle.net/mcNqc/2/ 2. Toggle Hidden 3. See the overflow: hidden doesn't work in the middle of the transition. What is the expected behavior? Overflow: hidden should still be clipping the object. I worked a hack up that involves adding position: absolute; and z-index:0; that show how it reacts in other browsers and should react in Chrome. What went wrong? Overflow: hidden wasn't taking into account when transitioning. Does it occur on multiple sites: Yes Is it a problem with a plugin? No Did this work before? N/A Does this work in other browsers? Yes Chrome version: 50.0.2661.11 Channel: n/a OS Version: 10.0 Flash Version: Shockwave Flash 21.0 r0 Works correctly in Edge, IE, and Firefox. Seems to be an issue with Blink or Webkit. Seems to be related somewhat but not enough to merge with https://bugs.chromium.org/p/chromium/issues/detail?id=389185
,
Apr 19 2016
,
Apr 19 2016
,
Apr 19 2016
I can repro on Mac and @alancutter could repro on Linux. I tested it on Chrome 50 and 52. Sorry for leaving that out.
,
Apr 19 2016
This is a composited opacity animation, most likely interfering with paint clipping, most likely fixed by slimming paint v2.
,
Apr 19 2016
,
Apr 19 2016
==================================== Good Build: 30.0.1549.0 Base Position: 208818 Bad Build: 31.0.1600.0 Base Position: 217417 ===================================== Able to repro this issue on Windows 7, MAC (10.11.4) & Ubuntu Trusty (14.04) for the Google Chrome Stable Version - 50.0.2661.75 This is a regression issue broken in M30, below mentioned is the bisect info: Bisect Tool - CHANGELOG URL: https://chromium.googlesource.com/chromium/src/+log/1905badf9db988e48a008ec208229d8d11208f62..54043cdf14d102f9bc99cd517da37bd2b109121c Manual Bisect - CHANGELOG URL - https://chromium.googlesource.com/chromium/src/+log/30.0.1549.0..31.0.1600.0?pretty=fuller&n=10000 Manual Bisect - BLINK CHANGELOG URL - https://chromium.googlesource.com/chromium/blink/+log/e45308800edc818f9c2b72695c0f9a0552fdd5fb..f53a57175e032e4c7729b5e4a932c082bd31bbc9 Suspecting Commit: 501641268b78509523658877b120fa0fd0c0676a Review URL: https://codereview.chromium.org/21372005 @ajuma: Could you please look into the issue, and if it has nothing to do with your changes and if possible please do assign it to the concerned owner. Thank you.
,
Apr 19 2016
That CL only changes TestExpectations, so cannot cause a behavior change. This bug is another example of border-radius not being applied to composited descendants when the element with border-radius isn't a stacking context. Making that element a stacking context (e.g. by giving it position:absolute and z-index:0, or by adding will-change:transform) "fixes" the bug. |
|||||||
►
Sign in to add a comment |
|||||||
Comment 1 by dpranke@chromium.org
, Apr 18 2016