RESOLVED FIXED 47975
Max width of a floated container with floated children calculated incorrectly
https://bugs.webkit.org/show_bug.cgi?id=47975
Summary Max width of a floated container with floated children calculated incorrectly
Bruno Couto
Reported 2010-10-20 02:16:22 PDT
Please see the demo url - http://jsfiddle.net/Trn8M/ - for details. It doesn't happen on Firefox and in any IE version, only on webkit browsers like Chrome and Safari. The issue is clear, when removing (or setting another value to letter-spacing on '.label'. Please check behavior on a webkit browser, like chrome, and others.
Attachments
Patch (4.72 KB, patch)
2012-12-22 10:57 PST, Robert Hogan
no flags
Patch (4.61 KB, patch)
2013-01-04 11:33 PST, Robert Hogan
no flags
Robert Hogan
Comment 1 2012-10-08 11:40:20 PDT
This looks fine in Chromium20. Closing - but please comment here if you can still recreate it.
Bruno Couto
Comment 2 2012-10-08 11:47:50 PDT
Still happens on Chrome Version 22.0.1229.79
Robert Hogan
Comment 3 2012-10-08 12:48:54 PDT
Can you be more specific about the problem so? I don't know what the broken rendering looks like.
Bruno Couto
Comment 4 2012-10-08 12:53:37 PDT
Just go to the jsfiddle, and change letter-spacing:normal; in .label class to another value or even remove it. You will notice that the width of it's container change to weird values without reason. Only happens on webkit browsers.
Robert Hogan
Comment 5 2012-10-08 13:09:32 PDT
I see it now - thanks for the clarification.
Bruno Couto
Comment 6 2012-10-08 13:19:09 PDT
Happy to help! Hope it's an easy fix.
Robert Hogan
Comment 7 2012-12-22 10:57:27 PST
Levi Weintraub
Comment 8 2013-01-03 13:38:01 PST
Comment on attachment 180605 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=180605&action=review Patch doesn't apply to ToT :( r=me after the nits are addressed. > Source/WebCore/ChangeLog:3 > + letter-spacing set's weird width when clearing floats Update the title to describe the actual bug, since letter-spacing isn't the issue here. > LayoutTests/fast/block/float/max-width-clear-float-with-overflow-hidden.html:1 > +<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd"> Why the long doctype instead of just !DOCTYPE html?
Robert Hogan
Comment 9 2013-01-04 11:33:25 PST
WebKit Review Bot
Comment 10 2013-01-05 12:27:38 PST
Comment on attachment 181345 [details] Patch Clearing flags on attachment: 181345 Committed r138899: <http://trac.webkit.org/changeset/138899>
WebKit Review Bot
Comment 11 2013-01-05 12:27:41 PST
All reviewed patches have been landed. Closing bug.
Dimitri Glazkov (Google)
Comment 12 2013-01-07 10:25:01 PST
Dimitri Glazkov (Google)
Comment 13 2013-01-09 13:50:29 PST
(In reply to comment #12) > For some reason, the added test has been failing on MoLion ever since it was added: http://test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=fast%2Fblock%2Ffloat%2Fmax-width-clear-float-with-overflow-hidden.html Anyone? Bueller?...
Robert Hogan
Comment 14 2013-01-09 14:05:12 PST
(In reply to comment #13) > (In reply to comment #12) > > For some reason, the added test has been failing on MoLion ever since it was added: http://test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=fast%2Fblock%2Ffloat%2Fmax-width-clear-float-with-overflow-hidden.html > > Anyone? Bueller?... I looked and figured it must be something to do with Ahem on that platform. Is it OK to skip it for now and assign a bug to me to turn it into a text test?
Dimitri Glazkov (Google)
Comment 15 2013-01-10 15:01:56 PST
(In reply to comment #14) > (In reply to comment #13) > > (In reply to comment #12) > > > For some reason, the added test has been failing on MoLion ever since it was added: http://test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=fast%2Fblock%2Ffloat%2Fmax-width-clear-float-with-overflow-hidden.html > > > > Anyone? Bueller?... > > I looked and figured it must be something to do with Ahem on that platform. Is it OK to skip it for now and assign a bug to me to turn it into a text test? Okay.
Note You need to log in before you can comment on or make changes to this bug.