[Mplayer-cvslog] CVS: main/libvo x11_common.c,1.47,1.48 vo_gl.c,1.22,1.23 vo_gl2.c,1.15,1.16 vo_xv.c,1.54,1.55 vo_x11.c,1.62,1.63

[Alban Bedel]

Hi Felix Buenemann,

on Tue, 12 Feb 2002 00:22:55 +0100 you wrote:

> On Monday, 11. February 2002 12:44, Alban Bedel CVS wrote:
> > Index: vo_gl.c
> > ===================================================================
> > RCS file: /cvsroot/mplayer/main/libvo/vo_gl.c,v
> > retrieving revision 1.22
> > retrieving revision 1.23
> > diff -u -r1.22 -r1.23
> > --- vo_gl.c     9 Feb 2002 01:21:48 -0000       1.22
> > +++ vo_gl.c     11 Feb 2002 11:44:50 -0000      1.23
> > @@ -207,7 +207,11 @@
> >  //     bpp = myximage->bits_per_pixel;
> >  
> >         //XSelectInput(mDisplay, mywindow, StructureNotifyMask); // !!!!
> > -        XSelectInput(mDisplay, mywindow, StructureNotifyMask |
> > KeyPressMask ); +        XSelectInput(mDisplay, mywindow,
> > StructureNotifyMask | KeyPressMask +#ifdef HAVE_NEW_INPUT
> > +                    | ButtonPressMask | ButtonReleaseMask
> > +#endif
> > +        );
> >  
> >  //  printf("Window setup ok\n");
> 
> IMHO this is bad coding style and can lead to problems, if called function is 
> really a macro. You should replace this with two line one function call for 
> normal and one for new input (->see problematic printf like that, in gcc 
> 2.95.3 it's function, but in gcc3 it's macro, so such code will only cause 
> problem with gcc3 in that case).
	Your right, but in this case I doubt XSelectInput become a macro, at least
	before X5 ;)
		Albeu