Re: Some misprints and problems

[Ethan M. <merritt@u.washington.edu>]

On Monday 08 September 2008 11:06:50 Ethan Merritt wrote:
> On Sunday 07 September 2008 23:48:20 Shigeharu TAKENO wrote:

> > 
> > I found some points that seem to be misprints and problems in CVS
> > version. 
> > 
>       
> > diff -uN term/cairo.trm.ORG term/cairo.trm
> > --- term/cairo.trm.ORG	Mon Sep  8 10:40:53 2008
> > +++ term/cairo.trm	Mon Sep  8 13:31:18 2008
> > @@ -505,7 +505,11 @@
> >    int stride = cairo_image_surface_get_stride(surface);
> >    int  i, j, x1 = 0, y1 = 0, x2 = width, y2 = height;
> >  
> > +#ifndef __sun
> >    typedef u_int32_t uint32;
> > +#else
> > +  typedef uint32_t uint32;
> > +#endif
> >    uint32 BG = ~0x0;
> >    uint32 *row;
> 
> Yeah.  That's what I was worried about when I commented on the
> pngcairo cropping patch.
>   
> I don't think it is sufficient to test specifically for __sun.
> It will break for other compilers and other platforms as well.
> That typedef is simply wrong.
> It needs to take the type from somewhere in the cairo headers.
> Or, at worst, the configure script needs to check for legal types.
> 
> Note that initializing BG = ~0x0 is also wrong, since it assumes
> that the background is always solid white.

I have now fixed these in CVS by adding tests in configure.in
Cropping will fail silently for platforms that both
- do not use autoconf
- have (sizeof(int) != 4)
This can be fixed by adding an explicit definition of GP_UINT32_T
in the appropriate platform-specific configuration file.

-- 
Ethan A Merritt