Bug 147690 - horizontal margins of absolutely positioned, non-replaced elements (Uranio site)
Summary: horizontal margins of absolutely positioned, non-replaced elements (Uranio site)
Status: RESOLVED WORKSFORME
Alias: None
Product: konqueror
Classification: Applications
Component: khtml (show other bugs)
Version: 4.8.4
Platform: unspecified Linux
: NOR normal
Target Milestone: ---
Assignee: Konqueror Developers
URL: http://www.gtalbot.org/BrowserBugsSec...
Keywords: testcase
Depends on:
Blocks:
 
Reported: 2007-07-09 08:29 UTC by Chris Seufert
Modified: 2012-06-20 19:50 UTC (History)
1 user (show)

See Also:
Latest Commit:
Version Fixed In:
Sentry Crash Report:


Attachments
Test case (756 bytes, text/html)
2007-08-06 21:57 UTC, Allan Sandfeld
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Chris Seufert 2007-07-09 08:29:03 UTC
Version:           3.5.7 (using KDE 3.5.7, Debian Package 4:3.5.7.dfsg.1-1 (lenny/sid))
Compiler:          Target: x86_64-linux-gnu
OS:                Linux (x86_64) release 2.6.21-1-amd64

The following site doesnt line up the absolute positioned elements (the outer parts of the 'O' on the home page.

http://www.uranio.com.au/home.html

seems to have similar problems in Safari (not the same, the bottom part of the O is fine, but the top and right still exhibit the same behaviour, however it renders perfectly in Firefox and IE6 and IE7.
Comment 1 Allan Sandfeld 2007-08-06 21:56:37 UTC
Confirmed.

Seems to be something  with the ALIGN="right" tag. Possibly if the sum of margin-left and margin-right is smaller than the width of the container then the margin that wins depends on the alignment (margin-left becomes auto if right-align and vice-versa). 
Comment 2 Allan Sandfeld 2007-08-06 21:57:03 UTC
Created attachment 21337 [details]
Test case
Comment 3 Allan Sandfeld 2007-08-06 22:15:25 UTC
No, on second look it seems Iceweasel is just doing really spooky stuff with the negative margins.
Comment 4 Kevin Funk 2008-04-20 13:37:32 UTC
Still in KDE 4.0.3. Also the site renders really slowish here.
Comment 5 Gérard Talbot (no longer involved) 2011-06-25 20:36:22 UTC
> Seems to be something  with the ALIGN="right" tag. 

align="right" is a non-CSS presentational hint which browsers *may* honor; "if so, these attributes are translated to the corresponding CSS rules with specificity equal to 0"
http://www.w3.org/TR/CSS21/cascade.html#preshint

> Possibly if the sum of margin-left and margin-right is smaller than the width > of the container then the margin that wins depends on the alignment (margin-
> left becomes auto if right-align and vice-versa). 

It would be possible to assess accurately this bug report if a solid, accurate and, most importantly, a reduced & minimized testcase was provided or could be built from a working URL.

CSS 2.1 provides rules for determining width and horizontal margins in section 10.3.7 Absolutely positioned, non-replaced elements
http://www.w3.org/TR/CSS21/visudet.html#abs-non-replaced-width
and CSS 2.1 test suite provides several testcases.

Gérard
Comment 6 Gérard Talbot (no longer involved) 2011-07-17 05:22:15 UTC
Allan and Chris,

I worked several hours to find the webpage, then create a semi-reduced testcase and then a furthermore reduced testcase:

Semi-reduced testcase:
http://www.gtalbot.org/BrowserBugsSection/Konqueror4Bugs/Bug147690-Uranio.html

Reduced testcase:
http://www.gtalbot.org/BrowserBugsSection/Konqueror4Bugs/Bug147690-Uranio-2.html

Firefox 5.0, Opera 11.50 and Konqueror 4.6.5 (KHTML engine) render those testcases the same. Chrome 12.0.742.124 renders those testcases differently.

I sincerly am not sure which browsers is correct in rendering those testcases. What Chrome 12.0.742.124 makes sense in my mind; I need to carefully examine the spec on this.

KDE Platform Version: 4.6.5
Konqueror version: 4.6.5 (KHTML rendering engine)
Qt Version: 4.7.2
Operating System: Linux 2.6.38-10-generic-pae i686
Distribution: Kubuntu 11.04
here.

-----------


Correspondent bug report at webkit.org:
Bug 19124 - position: absolute handled incorrectly when text-align: right applied to parent element 
https://bugs.webkit.org/show_bug.cgi?id=19124

regards, Gérard
Comment 7 Myriam Schweingruber 2012-06-18 18:13:09 UTC
Gérard: your comment is a bit puzzling for me as a triager: does that mean it is still valid with Konqueror 4.6.5? Or is it not clear whether it is a bug or not?

In any case: does this still apply to Konqueor 4.8.4 or later?
Comment 8 Gérard Talbot (no longer involved) 2012-06-20 19:50:42 UTC
(In reply to comment #7)
> Gérard: your comment is a bit puzzling for me as a triager: does that mean
> it is still valid with Konqueror 4.6.5? Or is it not clear whether it is a
> bug or not?
> 
> In any case: does this still apply to Konqueor 4.8.4 or later?


I sincerly was not sure how the reduced testcase
http://www.gtalbot.org/BrowserBugsSection/Konqueror4Bugs/Bug147690-Uranio-2.html
was supposed to be rendered. But, now, after this comment 
https://bugs.webkit.org/show_bug.cgi?id=19124#c5
from Robert Hogan and after the fix in bug _47148_ in webkit rendering engine, all major browsers render the reduced testcase the same. So, this means that Konqueror did not have a bug originally. (I still have to re-examine that test and the spec later to see/reevaluate all this... but that's outside this bug report.)

So, based on all the preceeding research and tests, I am RESOLVING this bug as WORKSFORME

I have also updated the Summary field, Component field, URL field, Version field, Keywords field of this bug report.

Gérard