[cairo] Problem with Pango & Cairo on Win32

[Owen Taylor]

On Wed, 2006-04-12 at 07:22 +0200, Bogdan Nicula wrote:
> 
> > > I'm using Pango's rendering with the Cairo backend on Windows and I'm
> > > hitting an assertion inside cairo-win32-font.c, line 1096 in
> > > _cairo_win32_scaled_font_glyph_init ():
> > >
> > >     if ((info & CAIRO_SCALED_GLYPH_INFO_SURFACE) != 0) {
> > >     ASSERT_NOT_REACHED;
> > >     }
> > >
> > > Is this a known problem or work in progress?
> 
> >What surface are you rendering to? Some of the recent reworking
> >of the font code paths in Cairo really aren't right for the way that
> >the win32 font backend works, and certainly haven't been tested
> >there, but I don't see anything offhand that would cause problems
> >rendering to the win32 or image surfaces.
> 
> The drawing context is acquired from a GdkPixmap with gdk_cairo_create (GTK+ 
> 2.8.17).
> 
> >Can you provide a backtrace?
> 
> I will try to come up with a reduced testcase next week.

While a reduced test case is always a good thing to have and
appreciated, I don't have a cairo HEAD build around at the moment 
for Windows, so it wouldn't tell me all that much.

What I was asking for really was a debugger backtrace when the
assertion is triggered - so I could see what function was calling
what function - this may be hard if you are using mingw, since gdb
isn't that reliable on Windows, but sometimes you get lucky.

Regards,
					Owen