© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
35a32c5
(Get the code!)
The regression is due to the timestamp used to start the drag changing from the time of the last button press to the time of the mouse motion, after the button release, that started the drag.
The last button release time used in code added in bug 307184 to end the drag is older than the mouse motion timestamp, and so the pointer ungrab fails. GTK still thinks it has ended the drag and so things are in a bad state.
The code added in bug 307184 was intended to handle the situation where the button release event is in the event queue at the time the drag started. It still functions as intended in that situation.
The regression is that the code added in bug 307184 no longer happens to save us from bug 648477, which is when the drag begins after the button release has been processed.