2196: Aborted drag and drop (drag but no drop) leaves page in disabled state

Greg.*****@gmai***** (Google Code) (Is this you? Claim this profile.)
Sept. 5, 2011
What version are you running?
1.6 RC2

What's the URL of the page containing the problem?
I noticed it on the review page and "view diff" page.

What steps will reproduce the problem?
1. Drag a file over the view port of the browser.
2. Don't drop it on the browser.
3. Must now refresh page to continue working.

What is the expected output? What do you see instead?
Once the mouse left the browser window I expected the page
to be returned to normal.

What operating system are you using? What browser?
Windows 7 Firefox 5.0
#1 chipx86
Hmm, it should be fading out after a brief delay. Maybe something regressed in FF5.
  • +Milestone-Release1.6
#2 Greg.*****@gmai***** (Google Code) (Is this you? Claim this profile.)
I confirmed the expected functionality with IE 8, Chrome, and Safari 5.1.

Here comes the crazy part...

I also confirmed that it works in FF 5 as expected, (most of the time[1], when the mouse leaves the browser window from the right (crossing the scroll bar), but almost always breaks when leaving from the left or bottom of the browser window.

I also find that crossing from one drop zone to another (Screenshot vs. File Attachment) will oftentimes keep things behaving correctly.

[1] Moving the mouse quickly out of the window from the right results in broken behavior more often than moving it slowly.
#3 chipx86
That sounds like a browser bug... There are some heuristics we can use, but I kind of hate to have to. Can you see if you're hitting this in FF6?
  • +NeedInfo
  • +chipx86
#4 Greg.*****@gmai***** (Google Code) (Is this you? Claim this profile.)
Yep, FF 6 is still behaving similarly.

Has anybody else been able t replicate the problem?  If not, I'll try again with an add-on free FF profile.
#5 chipx86
So this is a bug with Firefox, and not something we at this point have any control over, unfortunately...

See https://bugzilla.mozilla.org/show_bug.cgi?id=665704
#6 chipx86
Got some fixes and somewhat of a workaround.
  • -NeedInfo
#7 chipx86
Fixed on master (0fc184d)
  • -PendingReview