Re: Pango change, first^Wsecond baby step !

[Cyrille Chepelov <cyrille chepelov org>]

Le Sat, Jun 22, 2002, à 09:04:04PM +0200, Hans Breuer a écrit:

> At 19:28 22.06.02 +0200, Hans Breuer wrote:
> >At 18:25 22.06.02 +0200, Cyrille Chepelov wrote:
> >>Le Sat, Jun 22, 2002, à 04:40:53PM +0200, Hans Breuer a écrit:
> >Fine with me. It'll probably take me an hour to compile and crash the
> >current version. Will go online than again.
> >
> It crashed at only one place : don't try to g_object_unref unitialized
> memory :-)

:-) 

good catch !

 
> There is something wrong with bounding box calculation [1] and
> we need to find a way to get on the right pango_context. It
> doesn't matter that much on win32 yet (there currently is only 
> one), but uncontroled mixing of pangox_context, pango_xft_context,
> or pango_ft2_context appears as asking for trouble to me.

Indeed. Oh, there is a pango_context_list_families() function ? Wonderful...

I guess (I'll try when you had updated) that this will help a lot my
problems with selecting the right font...
 
> Required patch for win32 attached. Ok to commit ?

Please do !!

I think we can now at this point resume the 'normal' flow of operation -- if
some details change, like tweaking the parameter lists of the dia_font_*()
functions, that should not lead to unsolvable CVS conflicts.

	-- Cyrille

-- 
Grumpf.