Ah, ok. Thanks Martin. I knew there was something "not quite fixed," but I didn't have details. So I think this is the only real regression over 1.4. --Jeremy On Dec 3, 2009, at 12:27, Martin Otte wrote:
Jeremy,
I'm going to describe the last remaining problem the I see with border drawing in the recent 1.7.3 server, using the simple border.c example code that I sent a few months ago. I'll attach the simple code again so that you can easily have the code.
<border.c>
When I compile and run the code, it just creates a window with an internal window that has a border. Previously, the internal border was not shown, but you have fixed this in the recent servers. I have attached initial.png that shows the initial correct appearance when I first run the program:
<initial.png>
now when I drag the handle to make the window smaller to obscure the border (minimized.png):
<minimized.png>
and then drag the handle to make the window larger again, exposing the internal border again, the internal border becomes corrupted (reexposed.png):
<reexposed.png>
The only instance I can find when the internal border is always shown correctly is when the window is all the way in the upper-left corner of my display. The further the window is from the upper-left corner, the more corruption there is of the internal border.
With a larger program that I have which displays many buttons of various sizes, even the initial appearance of the borders can be slightly corrupted if the window is away from the upper-left corner.
In case this is system/graphic card dependent, I'm running this on a MacBook Pro. Can you verify the same behavior with my simple test case?
Thanks, Martin
_______________________________________________ Xquartz-dev mailing list Xquartz-dev@lists.macosforge.org http://lists.macosforge.org/mailman/listinfo.cgi/xquartz-dev