902827 - CSS: origin of "position:absolute" ignores a non-static positioned box with display:table-cell

Status: RESOLVED DUPLICATE of bug 63895

(Core :: Layout, defect)

Description

[ducksteven]

User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/29.0.1547.32 Safari/537.36

Steps to reproduce:

Repositioning a box by "position:absolute" inside a box with "position:relative" and "display: table-cell".

Test case: http://jsbin.com/uxaset/6/edit


Actual results:

The origin of repositioned box should be on the nearest box with non-static position.

In the test case, the box "C" should be inside box "B".



Expected results:

The origin of repositioned box is not based on the nearest box with non-static position and display set to "table-cell".

In the test case, for example, if C is wrapped in another div (.b-inner), then the origin of C is set on the origin of .b-inner. According to such behavior, I concluded that when finding the nearest ascendant box for origin calculation, it ignores the box with display: table-cell, even if it is the nearest one that has non-static position.

I found a forum thread talking about this behavior. Someone said that the spec does not define a behavior when position:relative is set on a display:table-cell element. I'm not sure if it is related to this issue: http://www.sitepoint.com/forums/showthread.php?775540-Firefox-does-not-obey-position-relative-on-display-table-cell-elements

Tested with the following browsers:

Problem:

- Firefox 24 Aurora: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:24.0) Gecko/20130805 Firefox/24.0
- Firefox 23 Stable: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:23.0) Gecko/20100101 Firefox/23.0

No such problem:

- Chrome Canary m30: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/30.0.1589.2 Safari/537.36
- Chrome Stable m28: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/29.0.1547.32 Safari/537.36
- Safari 6.1 Beta (8537.48.2)
- IE10 on Windows 7
- IE9 on Windows 7
- IE8 on Windows 7

Comment 1

[Tim Guan-tin Chien [:timdream] (please needinfo)]

Thanks for filing! It would be better if the testcase is uploaded to Bugzilla.

Comment 2

[ducksteven]

Attachment: test case

Comment 3

[ducksteven]

I've attached the test case.

By the way, the description fields was filled incorrectly. The "Actual results" section is actually "Expected results", and "Expected results" is actually "Actual results". I filled them in wrong order. Sorry if anyone is confused when reading the description.