User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8a3) Gecko/20040817 Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8a3) Gecko/20040817 Go to http://bugzilla.mozilla.org/query.cgi and scroll down to the "Bug Changes" box. Click to the right of the select box that is right under "where one or more of the following changed:", but at the same height. On Mozilla versions prior to 1.8 alpha 3, the correct thing happens -- the select box does not get focus, since the mouse click missed it. On Mozilla 1.8a3 (and newer -- I've also tested nightly 20040819) the select box gets the focus; when you try to scroll, it is the select box that scrolls, not the page. Reproducible: Always Steps to Reproduce: 1. Go to http://bugzilla.mozilla.org/query.cgi 2. Click to the right of the middle select box in "Bug Changes" 3. Press Up/Down or use the mouse scroll wheel. Actual Results: The select box in "Bug Changes" scrolls because it incorrectly gets the focus. Expected Results: The entire page should have scrolled, because the click was delivered to the background, not the select box.

Worksforme with 1.8a3 on Win98.... Is this a GFX issue on Linux of some sort?

I can reproduce it on Windows XP with 1.8a3, but I was wrong about the bug appearing when using the mouse scrollwheel -- it only happens when you scroll with the arrow keys!

Hmm... that worksforme as well...

It's probably easier to use a picture to describe what I'm doing: http://folk.uio.no/hakonrk/mozbug256276.png (Btw, I just noticed that the click has to be outside the Bug Changes frame.)

> (Btw, I just noticed that the click has to be outside the Bug Changes frame.) Ah, ok. There we go. Mats, any idea what's up here?

This will be "fixed" by bug 254966 (which will change kbd scroll target from caret to focus). Perhaps the caret really ends up inside the list when the body is clicked in that area or nsEventStateManager::GetDocSelectionLocation() could have a bug... Anyway, the only way to see that something is wrong is by scrolling and that symptom will disappear when bug 254966 is checked in so I wouldn't worry too much over this one.

The problem as stated in comment 0 is not reproducible anymore (bug 254966 removed the method to reproduce it). The caret placement is still the same though and a bit strange in my opinion - testcase coming up...

I am interested to work on this bug and I would I like to know more details about this bug. I would also like to know where I should get started.