On Mon, Oct 25, 2010 at 05:22:29PM +0200, Christian Ehrlicher wrote:
> Thomas Adam schrieb:
> >On Sun, Oct 24, 2010 at 08:34:03PM +0200, Christian Ehrlicher wrote:
> >>Dominik Vogt schrieb:
> >>>On Sun, Oct 24, 2010 at 07:39:09PM +0200, Christian Ehrlicher wrote:
> >>>>I've a problem with fvwm2 window positioning when using Qt4 in fvwm2. It 
> >>>>works
> >>>>fine with KDE. You can reproduce it with the attached example. You can 
> >>>>compile
> >>>>it with the following steps:
> >>>>
> >>>>- create a subdir 'example'
> >>>>- place main.cpp in there
> >>>>- execute 'qmake -project'
> >>>>- then 'qmake'
> >>>>- and then 'make'
> >>>>
> >>>>When you start the resulting executable without an argument and then with 
> >>>>an
> >>>>argument (doesn't matter what you give as argument) you can see that the
> >>>>resulting window position differs.
> >>>>It looks like when the window is moved before it gets shown the windeco 
> >>>>isn't
> >>>>took into account which leads to a displacement in the y-axis.
> >>>
> >>>Without going into the details:  It's not a bug, it's a feature.
> >>>If you don't like that behaviour, use this line in your .fvwm2rc:
> >>>
> >>>   style * MoveByProgramMethod UseGravity
> >>>
> >>Wow - never thought about such a solution ;)
> >>But I've another bug but I currently can't provide a testcase which
> >>is simple enough for a mailing list. I'll try to explain - maybe
> >>you've also such an easy solution for it (but I doubt).
> >>It's a normal toplevel window which shows information about an
> >>entity. When no entity is selected I hide the window. Then it's
> >>later shown when a another entity is selected. Now I've a problem
> >>when I hide the window twice it doesn't show up afterwards again -
> >>no matter what I try.
> >>I did not test it with .31, only with .26 because the solution was
> >>simple - destroy isntead hide the window.
> >
> >Yes, I've seen this and have a patch for it already -- it's to do with the
> >deferred mapping logic in events.c -- can you confirm for me if the window
> >is always in the WithDrawn state when this happens?
> >
> Can you give me a short hint how to check for this attribute?

Fixed in CVS.

-- Thomas Adam

Reply via email to