Re: [rox-devel] Re: OroboROX on 64-bit

[Tony H. <h...@re...>]

In <pan...@ec...>, Thomas Leonard wrote:

> On Tue, 03 May 2005 14:07:21 +0200, Guido Schimmels wrote:
> 
> > This part is was a bit surprising at first:
> > -	CARD32 xy[2] = { x, y };
> > +	long xy[2] = { x, y };
> >  	XChangeProperty(dpy, w, a, XA_CARDINAL, 32, PropModeReplace, (unsigned char *) xy, 2);
> 
> I don't think that can be right. sizeof(long) depends on the architecture
> of the machine running the window manager. Clients would see different
> values depending on where the window manager was run from!

Doesn't the window manager always run on the same machine as the clients
though? I know it doesn't seem right, and I did try leaving it as
CARD32, but it didn't work properly (Pager couldn't see the windows or
get them the right size), while it does with long. I chose the above
solution after trying metacity and finding that worked correctly and
checking its source. 

-- 
TH * http://www.realh.co.uk