Re: Font Problems

[Owen Taylor <otaylor redhat com>]

Eric Mader <mader jtcsv com> writes:

> At 02:17 PM 8/20/2002, I wrote:
> > * Arabic text doesn't seem to be contextually forming. I'm trying to
> > debug this, but don't have any information yet, other than it does
> > seem to call arabic-xft.c to process the Arabic text. (Also, while
> > debuggin this, it looks like arabic-xft.c is getting called for
> > *each word* in the Arabic text. I guess that it's finding the space
> > character in another font. This is probably a performance hit...)
> > I'm seeing some error messages about invalid GPOS and GSUB tables,
> > so that may be some of the problem... did we perhaps loose some of
> > the bullet-proofing when things got updated to ft2?
> 
> I tracked down what's happening w/ Arabic text. Looks like
> arabic-xft.c is being invoked w/ the ClearlyU bitmap font! Since it's
> not an OT font, it doesn't have any GSUB/GPOS/GDEF tables, so not much
> processing is getting done.
> 
> Don't know why it should be calling arabic-xft with a bitmap font, and
> I don't know why it doesn't find the OT font that I have installed on
> my system and mentioned in /etc/fonts/fonts.conf

You probably have a non-gzipped pcf version of clearlyu somewhere
in /usr/share/fonts or /etc/fonts/fonts.conf.

Unfortunately, it's really hard to work around the problem of
fonts that have Arabic glyphs but no GSUB table, because fontconfig
works only by straight coverage, and it will have already said
"clearlyu has these glyphs, I don't need to add another font"
before Pango even gets involved at all.

The easiest way to deal with this is simply to list the fonts you
want used for such languages in fonts.conf explicitely; they
will then be found before other fonts with covering these
codepoints.

As for why fonts.conf isn't working for you, I suspect you are getting
confused (like most people) in that the first set of <alias>
declarations in fonts.conf aren't declaring the aliases, but rather
defining what fonts fallback to sans/serif/monospace.

We should probably reverse these fallback definitions and the real
alias definitions and put a big comment in.

Regards,
                                        Owen