Re: PangoXft and PangoFT2 patch from hell



On Wed, 21 Nov 2001, Tor Lillqvist wrote:

> Hans Breuer writes:
>  > I'm not sure if it really should _not_ create a monochrome bitmap, 
>  > given that the resulting bitmap already was huge.
> 
> True, there must have been some other related problem in that case,
> that caused the image to come out blank (on Win2k, at least).
> 
> Actually, it is better to use at least 16bpp bitmaps here, as
> pango-win32 should be able to have Windows produce anti-aliased
> characters. (See the docs for CreateFont().)  sets the lfQuality field
> in the LOGFONT to ANTIALIASED_QUALITY. At some point I do remember
> that this used to work in (the pre-Pango) GTK+ 1.3.0 at least, and
> IIRC in pango-win32, too. I don't know why it doesn't seem to work now
> in pango-win32.
> 
> Umm, now I notice that the code that uses ANTIALIASED_QUALITY setting
> is not active, it is inside #if _WE_WANT_GLOBAL_ALIASES_ (alexl
> 2001-08-09). Time for some hacking to see where else
> ANTIALIASED_QUALITY could be turned on...

The WE_WANT_GLOBAL_ALIASES code can be removed now btw. I implemented 
global aliases this weekend. It was part of the huge churn that involved 
mini-xft.

The default pango_fontmap_load_fontset() implementation (that pango-win32 
uses) loads a pango_get_sysconf_subdirectory()/pango.aliases or 
g_get_home_dir()/.pango.aliases which is a key-value pair file like this:

sans=arial,foo,bar
serif=helvetica,gazonk

/ Alex





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]