User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b3) Gecko/20050709 Firefox/1.0+ Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b3) Gecko/20050712 Firefox/1.0+ Problem: Create a parent container which is relativley positioned with overflow hidden and padding.Create a child within the container which is relatively positioned to the right (right:0).The child should touch the right side of the container, but DOESNT if overflow is set to hidden All very clearly shown in the attached test case. This displays correctly in Firefox 1.0.x and Deer Park Alpha 1, it regressed before Deer Park Alpha 2 was released. Reproducible: Always Steps to Reproduce: 1. try the test case 2. OR go to the website http://phonophunk.phreakin.com/news/ 3.

You can see the problem in the header image on http://phonophunk.phreakin.com/news/ The header image is the container div. The shadows inside the container are relatively positioned. The RIGHT side shadow is not touching the right side of the aimge as it should.

Further investigation on regression dates: It appears Deer Park Alpha 1 does not display the testcase correctly, meaning the regression is older than that. This increases the possibility this bug is a dupe. A build from 5/7/05 displayed both the testcase and website incorrectly. A build from 3/1/05 displayed the website correctly but the testcase incorrectly (but different) -- the blue line was shifted to the left right beside the green line. Quiting for tonight, more testing later...

The testcase works in Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.7.9) Gecko/20050709 Firefox/1.0.5, but not in Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8b2) Gecko/20050606 Firefox/1.0+ Probably need to find a regression date.

Okay, regression time frame is narrowed down. Testcase works on build 20050207, broke on build 20050208. It appears the checkin for Bug 277420 caused this bug. However, further investigation makes me wonder if this bug isn't simply a duplicate of 294934, I am CC'ing Robert for confirmation as this stuff is WAY above my head :( One other note, the website issue that caused this bug to be reported is NOT the same as the testcase, the website issue works in later builds and that regression was at a later date. I am planning to file a seperate bug there once I narrow down a regression window and hopefully someone can provide a testcase.

I think this is a duplicate of bug 294934.

I just attached a more accurate testcase resembling http://phonophunk.phreakin.com/news/ more closely. As Darin Grimm stated, the website issue was not exactly the same as the first testcase.

Both testcases now work, as well as the original website. Fairly certian patch for Bug 294934 fixed this as well. Thanks Robert!