New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 686391 link

Starred by 3 users

Issue metadata

Status: Duplicate
Merged: issue 669039
Owner: ----
Closed: Feb 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Mac
Pri: 1
Type: Bug-Regression



Sign in to add a comment

Absolute position child of display:flex element doesn't update position when parent size changes

Reported by corrie...@gmail.com, Jan 28 2017

Issue description

UserAgent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.95 Safari/537.36

Steps to reproduce the problem:
1. Put an absolute position element with "bottom: 0" or similar in a relative position element with "height: auto" and put all of that under an element with "display: flex"
2. Do something that causes the height of the intermediary (position relative) node to change height

What is the expected behavior?
In the attached example, the scrollbar should be updating to stay at the bottom of the codemirror editor.

What went wrong?
Note that now the position absolute element renders in the wrong spot.  Doing something like changing the browser window size or position absolute style to "bottom: 0.1px" instead of "bottom: 0" causes it to jump to the correct position

Did this work before? N/A 

Does this work in other browsers? Yes

Chrome version: 55.0.2883.95  Channel: n/a
OS Version: OS X 10.12.2
Flash Version: Shockwave Flash 24.0 r0

We *think* it might have regressed with Chrome 52 but are unsure since that was a while ago.
 
testcase.html
2.0 KB View Download
capture.webm
2.6 MB View Download
Components: -Blink>CSS Blink>Layout>Flexbox

Comment 2 by ajha@chromium.org, Jan 31 2017

Labels: Needs-Milestone

Comment 3 by e...@chromium.org, Feb 2 2017

Cc: cbiesin...@chromium.org
Status: Available (was: Unconfirmed)
Labels: Needs-Bisect
Cc: sureshkumari@chromium.org
Labels: -Type-Bug -Pri-2 -Needs-Bisect hasbisect-per-revision M-58 Pri-1 Type-Bug-Regression
Owner: msten...@opera.com
Status: Assigned (was: Available)
Able to reproduce the issue on  Mac-10.12.2 using chrome stable version 56.0.2924.87 and Reported version 55.0.2883.95.
And issue not observed in Canary 58.0.3005.2 and Beta 57.0.2987.21

This is regression issue fixed in M57. Please find the Reverse bisect information as below

Using the per-revision bisect providing the bisect results,
Bad:: 57.0.2942.0 --   (build revision 436203)
Good :57.0.2943.0 --   (build revision 436483)

ChangeLog: 
================
https://chromium.googlesource.com/chromium/src/+log/915e461af0dcb90d0d98b1aab614f409d8770379..9d96126f67850daf3c07586dc09fa334d592529c


Review-Url: https://codereview.chromium.org/2553833002

mstensho@ could you please look into this issue if it is related to your change,else please help us in finding the appropriate owner for this issue.

Note:Issue is specific to Mac.

Thanks.

Comment 6 by msten...@opera.com, Feb 7 2017

Cc: msten...@opera.com
Owner: ----
Status: Available (was: Assigned)
I don't have access to any Mac, so I cannot help. If someone could come up with a testcase that fails on all platforms (or at least Linux), I can take a look.

Comment 7 by phistuck@gmail.com, Feb 13 2017

#6 - your changes fixes it - is it safe to merge it to Chrome 56?

Comment 8 by msten...@opera.com, Feb 13 2017

Thanks for testing! I don't think it's a good idea to merge this. The CL itself is a quite heavy refactoring, and it also depends on preceding refactoring CLs, that most likely don't exist in M56. Better wait for M57.

Comment 9 by phistuck@gmail.com, Feb 13 2017

#8 - I did not test it, I just stated the result of the reverse bisect.

Got it, thank you.

I guess you can mark this as fixed, then, since it is fixed in Chrome 57+ (according to the bisect).

I can try BrowserStack for you, if you want to verify it before you mark it as fixed (or maybe just merge it to  issue 669039 ?).

Comment 10 by msten...@opera.com, Feb 13 2017

Mergedinto: 669039
Status: Duplicate (was: Available)
Since this was apparently broken for so long (since Chrome 52?) before it got fixed, and since it's apparently fixed now, I think it's okay to just forget about Chrome 56. I trust the bisect info above, so I'll merge it into 669039.

Sign in to add a comment