[Lazarus] BUG: Gtk MouseExit, MouseUp

[Vojtěch Čihák]

Yes, already reported by me, 3 years ago :-)
 
http://bugs.freepascal.org/view.php?id=21982
 
V.
 
______________________________________________________________
> Od: Anthony Walter <sysrpl at gmail.com>
> Komu: Lazarus mailing list <lazarus at lists.lazarus.freepascal.org>
> Datum: 07.04.2015 23:16
> Předmět: [Lazarus] BUG: Gtk MouseExit, MouseUp
>
I can't believe this bug hasn't been reported,so I think I'll ask before submitting a mantis report.On Linux Gtk2 with TGraphicControl derived classes there seems to be a problem with reliable MouseExit and MouseUp firing.I had to write some work arounds which are still unreliable, but here's the basic gist of the problem.With TGraphicControl derived classes on Gtk2 sometimes MouseExit or MouseUp do not fire. If you mouse the mouse over a graphic control, press the left mouse button, mouse out of the control and release the left button, the OnMouseUp event is not invoked nor is the virtual MouseUp method.Example:procedure TForm1.Shape1MouseUp(Sender: TObject; Button: TMouseButton;  Shift: TShiftState; X, Y: Integer);begin  Tag := Tag + 1;  Caption := IntToStr(Tag);end;  With a TShape the MouseUp will only fire if the mouse is released while over the control. On Windows and Macintosh this is not the standard behavior. MouseCapture has no effect in fixing this problem. Cur
 iously TSpeedButton seems to have some special logic that gets around the problem on Gtk2 which I've copied to my TGraphicControls but the problem still exists with other graphic controls.And then there is the MouseExit problem. It's not reliably being fired on Gtk2 graphic controls. If you have 2 graphic controls close together or you mouse the mouse too fast between controls one of them occasionally doesn't receive the MouseExit. This causes issues with graphic controls which might display different states based on whether the mouse is in them or out of them.Have these issues been reported before? They seem to be quite obvious and I'm struggling to find that these problems haven't been reported or discussed before.

----------

--
_______________________________________________
Lazarus mailing list
Lazarus at lists.lazarus.freepascal.org
http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus <http://lists.lazarus.freepascal.org/mailman/listinfo/lazarus>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lazarus-ide.org/pipermail/lazarus/attachments/20150408/39fca8c9/attachment-0003.html>