User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1b1) Gecko/20081007 Firefox/3.1b1 Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1b1) Gecko/20081007 Firefox/3.1b1 If I click the "autres sites" link at the bottom right of the aforementioned page, the link just jumps one line higher (some text goes in the left column) and I have to click on it again to follow the link. If I click somewhere else in the page, the text moves again to its normal state. Reproducible: Always Steps to Reproduce: 1. Go to http://jonathan.xulforum.org 2. Click on "autres sites", purple link at the bottom right corner 3. Admire the result ;-)
I also see this on Vista HP using today's trunk build: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.2a1pre) Gecko/20081202 Minefield/3.2a1pre Firefox/3.0.4 ID:20081202034510 No errors show in the Console2, and turning off JIT Content makes no difference.
Status: UNCONFIRMED → NEW
Ever confirmed: true
OS: Linux → All
Hardware: PC → All
Version: unspecified → Trunk
That sounds vaguely similar to bug 433375 (which I also reported quite some time ago). I ran some more tests. There is NO bug on Firefox 2 Windows. There IS a bug on Firefox 3 Windows. That's a regression (which makes it less likely to be related to bug 433375 which was present on firefox 2 too). Comment here if you need more information :)
Maybe you have a testcase jonathan? Regression window: http://bonsai.mozilla.org/cvsquery.cgi?module=PhoenixTinderbox&date=explicit&mindate=2008-04-17+20%3A00&maxdate=2008-04-19+02%3A00 So I think the cause could be bug 404215
Component: General → Layout
Product: Firefox → Core
QA Contact: general → layout
Problem is that I can't look into bug 404215 to see if there is a duplicate or to make it dependent.
I cannot look at it too. I haven't built a reduced testcase, I only have http://jonathan.xulforum.org/ which could be sufficient to make a test case ? I don't know what you expect exactly. I will try a 20080417 nightly and tell you whether the bug is still there or not.
Could this be related to bug 372772?
The regression window as cited in comment #3 is right : 20080416 nightly has no bug while 20080419 has. I do not think it is related to bug 372772 : does the test cases there give a bug on latest beta ? For me I see no difference between the two renderings.
404215 does seem likely. What's happening is that the outline rectangle on the link is triggering a rebalance of the columns. (You can get the same effect by tabbing focus through the links.) Column balancing after 404215 uses the nsLayoutUtils::CalculateContentBottom function. That uses data from GetRect().y for most boxes, and line->mBounds.YMost() for line boxes. If line->mBounds.YMost() is affected by 'outline' then that would explain the behavior here.
This is working correctly for me using the latest hourly Minefield build: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.2a1pre) Gecko/20090715 Minefield/3.6a1pre Firefox/3.0.11 ID:20090715134535 changeset: http://hg.mozilla.org/mozilla-central/rev/d190d9b6ccd1
Fixed on latest nightly (but NOT fixed on 3.5). Thanks for pointing it out. Marking as fixed.
Status: NEW → RESOLVED
Last Resolved: 9 years ago
Resolution: --- → FIXED
We don't know what fixdd it, so worksforme.
Resolution: FIXED → WORKSFORME
You need to log in before you can comment on or make changes to this bug.