Re: mouse cursor problems

søn, 08.02.2004 kl. 17.42 skrev Phil:
> Hi,
> 
> On Sunday 08 February 2004 17:18, David Hammerton wrote:
> 
> > As for  the mouse issue, the snap-back thing is so we can fake relative
> > mouse input, we generally only do this when the app requested relative
> > app movements..
> 
> Is there a possibility to use winex3-3.1's mouse code in current cvs?

Not unless you could identify *which* mouse code. There's mouse code all
over the place, all with different purposes and levels of abstraction
(for example, one to read events from X, one to alter the cursor, one to
lock it inside the fullscreen window, one to interpret incoming events,
one to dispatch them to hooks, one to dispatch them to ordinary Win32
windows, one to adapt them to dinput requirements, one to queue them as
dinput device events, and so on), and scattered all over the source
tree.

And in any case it'd be best to figure out what is wrong with the mouse
code now. Perhaps you could put a log with, for example, -debugmsg
+dinput,+cursor,+input,+hook,+event,+message,+x11drv online somewhere
for analysis.