[Webkit-unassigned] [Bug 23989] The cursor behavior in GTK is wrong

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=23989





------- Comment #6 from xan.lopez at gmail.com  2009-02-18 00:40 PDT -------
(In reply to comment #4)
> (In reply to comment #3)
> > I think this just fixes the bug by accident. By using the same cached cursor
> > for all widgets, and initializing it to 0 on each instance, you effectively
> > disable it once per widget created. Also, how does making the cache per-class
> > makes sense? Could you elaborate?
> 
> Cache per-class means cache per-application. One application hold one cursor
> type at one time. Make the cursor a static variable, we can trace the cursor
> type trough all the widgets of the application. I think this make sense. 
> Disable it once per widget created indeed make no sense. I forget to delete
> that line in the patch. I have upload a new patch, thanks. 
> 

Well, sure, I was hoping for the rationale of how this fixes the problem. I
think I get it now though: the issue is with widgets that get set a cursor A,
then the user moves elsewhere with the cursor changing, and when he tries to
come back and set cursor A again the cache says "no, I have it already", and
nothing happens. So making the cache global indeed fixes the issue (and saves
memory...).

The last patch looks good to me.


-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.