I updated yesterday (4/15/2015) and am noticing that when I click on links to inline comments for a revision, a new tab opens, scrolls to the comment, then immediately scrolls to the bottom of the page. This seems to happen almost every time - maybe one-in-ten it will stay on the right inline comment.
For inline comment links that do not open a new tab (I think this happens only with a single diff or on most recent diff?), it properly jumps to the right location.
I can't reproduce this as described. Can you give us more details on your computer, os, browser, if you're using a mouse/trackpad, etc.
D12347 here has a comment on a previous diff, it loads for me in a new window (expected) and scrolls properly. I'm on Chrome/Trackpad.
This seems to work fine on secure.phabricator.com. I tried a few revisions and am seeing the same behavior. I just started noticing it yesterday evening/this morning. Is it possible a recent change caused this which isn't up on secure.phabricator.com yet?
I just confirmed another user sees this on our instance with "Chrome 41.0.2272.118 m" on Windows.
It's a few days behind HEAD, maybe @epriestley can roll it for further testing. I can't get to a mouse right now.
OK It looks like it's working here too.
I did some more testing on our instance to try and nail down what circumstance this happens on. The auto-scrolling seems to go to the proper location when the inline comment:
Any inline comment that is not on the newest diff and also not the last inline comment of its diff seems to jump to end of page (actually, it might be jumping to the last inline comment which happens to be bottom of page on my system). I'm not sure whether these circumstances are correct, but it's the case for the revisions I see this happen.
Just seeing if I can go 2/2 here: are you an administrator on your install, with a persistent "unresolved setup warnings" banner showing? Specifically, do all pages which exhibit this behavior for you have the banner on them?
I am an administrator on my install. There are no banners on any page I visit.
There are 2 warnings that I have had set to ignore for a while:
The first one was turned off since we have a small local install.
The second one was because a user had used an outdated arcanist to upload a diff at one point.
The issues themselves wouldn't cause a problem, but T7830 was a similar issue with UI/positional glitches related to the banner itself.
Second guess is that you have a large image on the page somewhere above the inline comment, and this is an interaction between image stabilization and maybe Differential content stabilization.
@epriestley - there is definitely a large image on the diff above the inline comment. Where is that super sleuth badge you mentioned?