Summary: | wrong stacking context with mixed fixed and absolute positioning | ||
---|---|---|---|
Product: | [Applications] konqueror | Reporter: | mah |
Component: | khtml renderer | Assignee: | Konqueror Developers <konq-bugs> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | djoham, kde-bugs, temp.addr |
Priority: | NOR | ||
Version: | unspecified | ||
Target Milestone: | --- | ||
Platform: | Compiled Sources | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: | ||
Sentry Crash Report: | |||
Attachments: |
testcase
real world example of xhtml-strict page demonstrating the problem stripped down testcase |
Description
mah
2002-06-05 09:06:41 UTC
Here's a real world test case that is displayed correctly in Opera and Mozilla: http://www.mathA.rwth-aachen.de/index_bg.html In section 9.9 of the CSS2 specs (http://www.w3.org/TR/REC-CSS2/visuren.html#q30) is described that the stack level depends on the stacking context of the parent's box and not on the stacking context of the containing block (which is what Konqueror seems to assume). This means that 'absolute' and 'fixed' positioned elements have to be treated the same way with respect to the stack level. *** Bug 47871 has been marked as a duplicate of this bug. *** Hello, I'm running into a similar problem with css position: relative on some web pages I'm building for our LaserJets (I work for Hewlett-Packard). Konqueror does not respect the z-index in situations where divs are positioned with css position: relative. I'll attach a test case as well at the real-world XHTML that I'm generating to demonstrate the issue. Created attachment 90 [details]
testcase
this test case is a very small recreation of the problem. Mozilla IE and Opera
render correctly.
Created attachment 91 [details]
real world example of xhtml-strict page demonstrating the problem
Here's a snippit of code that I'm developing for our next generation laserjets.
It demonstrates a real-world example of the problem.
I've got the same problem. I'm eager to see it solved. Created attachment 2060 [details]
stripped down testcase
Here is a much stripped down version of the real world test case. It's as small
as I can make it (it's pretty small) and still have it relevant to the XHTML I
use. Also note that while Konq's behavior has changed in CVS HEAD, the first
test case is still mis-rendering
Lastly, just so that there is no confusion, the correct rendering would be no black line under the word "information" in the stripped down testcase. still buggy in kde 3.2a2. This seems to be fixed with 3.2 I can confirm that my test case (cf. comment #1) and all the others are now rendered correctly. |