New issue
Advanced search Search tips

Issue 603835 link

Starred by 20 users

Issue metadata

Status: Duplicate
Merged: issue 603507
Owner:
Closed: Apr 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Windows
Pri: 2
Type: Bug-Regression



Sign in to add a comment

Chrome 50 Change in implicit table-cell height behaviour

Reported by bpoz...@gmail.com, Apr 15 2016

Issue description

UserAgent: Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.75 Safari/537.36

Steps to reproduce the problem:
1. See jsFiddle https://jsfiddle.net/zc21pg9q/1/
2. Open in Chrome 49, the div fills the table cell
3. Open in Chrome 50, the div does not fill the table cell

What is the expected behavior?
I would expect the div to fill the table cell.

What went wrong?
Since Chrome 50, table cells do not implicitly inherit their parents percentage heights.

Did this work before? Yes Chrome 49

Chrome version: 50.0.2661.75  Channel: stable
OS Version: 10.0
Flash Version: Shockwave Flash 21.0 r0

This has been an issue for us in Firefox and earlier versions of IE.

Chrome, Safari and IE11 have always shown the correct behaviour.

Chrome 50 this is not the case.
 

Comment 1 by yurk...@gmail.com, Apr 15 2016

Looks like this is predictable behavior as parent's height should have defined value.
Maybe related issue - https://bugs.chromium.org/p/chromium/issues/detail?id=603963.

I can confirm the behavior on  Chrome 50.0.2638.0 (64-bit) on OS X El Capitan.
Fix ASAP please, affects everything!! 

Comment 4 by angie1...@gmx.net, Apr 16 2016

I disagree that this is expected behavior. All other browser handle it as chrome did in 49 and earlier. We have a big company platform web application in  production that is affected by this at the moment. Would appreciate it, if this could be looked at.
Thanks,
Janett 

Comment 5 by bpoz...@gmail.com, Apr 16 2016

The justification in this bug report is that the change is "more spec compliant".

https://bugs.chromium.org/p/chromium/issues/detail?id=583670&q=label%3ACr-Blink-Layout-Table&colspec=ID%20Pri%20M%20Stars%20ReleaseBlock%20Cr%20Status%20Owner%20Summary%20OS%20Modified

However IE11 and Edge moved away from this behaviour to join the same behaviour as Chrome (previously), Safari and Opera. Leaving Firefox behind. Mozilla appear to acknowledge this behaviour as a bug in this open report since 1999:

https://bugzilla.mozilla.org/show_bug.cgi?id=10212

I'm surprised there hasn't been huge uproar surrounding this change as I'm sure there will be thousands of sites broken by this. Even Google Hangouts has been affected.

The offered workaround is fine in isolation, however for those of us dealing with large scale legacy applications this is a disaster.

I have provided a temporary workaround for others who may need an emergency fix for this issue. However I hope that this will not be permanently required.

http://brettpostin.com/2016/04/16/chrome-50-and-table-cell-percentage-heights/
Components: Blink>Layout>Table

Comment 7 by andrew...@gmail.com, Apr 18 2016

This "fix" to be more compliant is causing a major disaster for our production SPA app of over 1000 users. While we can add display: table-cell to affected divs, this is not enough; the fixed height of the parent table is ignored when height: 100% is added to a table-cell, and whereas previously the cell's inner content would scroll, keeping the page always at 100% height, now the entire page expands so that a particular cell can show 100% of its content. The various overflow properties don't seem to override this behavior. This is causing quite a problem for our users who are upgrading to the latest Chrome.

Comment 8 by andrew...@gmail.com, Apr 18 2016

I'm rather dumbfounded that automatic updates to Chrome 50 are happening for our users. A very popular browser rendering service we use, Browserstack, lists Chrome 50 as a development branch, and it's so new that the service doesn't even offer Chrome 50 for any versions of Windows yet, only OSX. Why is Google automatically moving everyone to this release when it's still in beta? This was a significant error to force everyone into such an upgrade.

Comment 9 by e...@chromium.org, Apr 18 2016

Cc: cbiesin...@chromium.org dgro...@chromium.org tabatkins@chromium.org
 Issue 604396  has been merged into this issue.

Comment 10 by e...@chromium.org, Apr 18 2016

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

Comment 11 by e...@chromium.org, Apr 18 2016

Cc: rnimmagadda@chromium.org
 Issue 604173  has been merged into this issue.
andrew006:  Chrome 50 was promoted to stable 5 days ago.  You can check what version of Chrome is stable here:  http://omahaproxy.appspot.com/.
Cc: -rnimmagadda@chromium.org -tabatkins@chromium.org
Mergedinto: 603507
Status: Duplicate (was: Assigned)
Another issue suggests that related behavior to what I and several others mentioned might indeed by reverted in the most recent version of Chrome: https://bugs.chromium.org/p/chromium/issues/detail?id=605257
indonesia
30 Apr 2016 20.58, "andrew006@gmail.com via Monorail" <monorail@chromium.org>
menulis:

Sign in to add a comment