[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Recent NSWindow and NSMenuView commits (was: Re: [ 102031]mproving
From: |
Michael Hanni |
Subject: |
Re: Recent NSWindow and NSMenuView commits (was: Re: [ 102031]mproving application startup visual appe) |
Date: |
Sun, 20 Apr 2003 16:33:40 -0700 (PDT) |
Hi,
--- Alexander Malmberg <alexander@malmberg.org> wrote:
>
> Fixing the problems would be better, but trickier. :)
Immensely better, I agree!
> back/x11/XGServerWindow.m, -placewindow::, in particular around line
> 1807, seems suspicious, but I'm not really sure what it's supposed to
> accomplish, and I can't reproduce the problems here. Someone who can
> reproduce them should probably experiment with that code (eg. commenting
> it out to see if that helps).
I've commented this out to no effect. I've also checked the events that
typically get processed in this loop and they tend be AppKitLeaveWindow events,
so I'm not sure our problem is that loop.
The problem, as far as I can trace it, is that the AppKitWindowMoved events are
not being processed fast enough for the origin point to be always considered
correct. So what we end up with is:
(ideally)
Drag events start
AppKitWindowMoved
LeftMouseDragged event
set new window frame origin
(repeat, and so forth.)
(actually)
Drag events start
AppKitWindowMoved
LeftMouseDragged event
set new window frame origin
LeftMouseDragged event
LeftMouseDragged event
LeftMouseDragged event
LeftMouseDragged event
set new window frame origin
AppKitWindowMoved
AppKitWindowMoved
AppKitWindowMoved
AppKitWindowMoved
(and progressively, as the movement increases in speed, the origin gets more
and more distorted.)
Can anyone confirm or refute this? I'm not sure where else to look.
Thanks,
Michael
=====