Re: Fonts in GNU Classpath

[Sascha Brawer]

Salut Pierrick,

>I am very interested in your "Fonts in GNU Classpath" project.
>Unfortunately, your blog indicates that you won't maintain this project
>anymore :-(
>
>I've also noticed that your submission to Savannah is still pending :
>http://mail.gnu.org/archive/html/savannah-hackers/2003-11/msg00966.html
>What about an integration in GNU Classpath ?

There are good arguments against my small library and using FreeType (and
Pango) instead: Their codebase is well maintained, and these libraries
are widely used. Any bugs are likely to be found and fixed soon.
Therefore, most people want to have FreeType (and Pango) as font backends
for GNU Classpath.

However, there does exist a minority of Classpath-derived projects that
that could't use Pango/FreeType. To this rather small group of people, my
library would indeed be useful. I've heard that my library got considered
for JNode.org, but I don't know what decision they made.

We had some discussions about pluggable backends for font engines: it
would be up to the Toolkit to create implementations of a FontPeer
interface. If I remember correctly, this was actually checked into
Classpath at one time, but it doesn't get used much. For details, you
should probably have a look at the code of RedHat's version of Classpath.
They are doing lots of AWT-related changes internally, and plan to send
things back to GNU Classpath once their implementation is finished.

>FYI : I'd like to implement the old Unicode Draft for egyptian
>hieroglyphs (http://www.dkuug.dk/JTC1/SC2/WG2/docs/n1944.pdf). This
>draft will certainly need some revisions but I think it is doable.
>
>I have fonts (http://www.yare.org/egypt/gardiner%20fonts.htm), but they
>are old (i.e. not designed for Unicode) and I can't physically change
>them (http://pharaoh.heavengames.com/downloads/descpages/misc/18.shtml).
>
>So, I'd like to use the concept of "memory fonts", i.e. a dynamic
>generation of a font object from files (fonts, mapping tables,
>positioning tables...).

I would strongly recommend against patching copyrighted font files inside
a renderer. Maybe you could talk to Michael Everson, the author of the
Unicode extension proposal that you mentioned. As far as I know, he has
designed some very large free fonts, and he might be able to assist you
with getting to a suitable font file.

>Since the main problem problem with hieroglyphs is the glyph
>positionning, I'd like to do it in the Graphite
>(http://scripts.sil.org/cms/scripts/page.php?
>site_id=nrsi&cat_id=RenderingGraphite)
>way. Technically, it consists in adding 3 tables to a true type font
>(see the *open sourced* Graphite Binary File Format) and in implementing
>the powerful font description language.

Of course, I'd find it great it someone took charge of my library. But to
be honest, I don't think that this would be the best way for getting
broad support for egyptian hieroglyphs. If I were you, I'd rather work on
supporting Graphite in Pango or OpenOffice.

>What about "some people around Sivan Toledo of Tel Aviv University think
>about enhancing the code by complex glyph substitution (GSUB) and
>positioning (GPOS) tables" ?

You might want to have a look at Sivan Toledo's TAUX project. If they are
indeed using my library for accessing font files, you could extend it by
Graphite and get egyptian glyph positioning in TeX.

Best regards, and good luck with your project,

-- Sascha | address@hidden | www.brawer.ch

>--
>Pierrick Brihaye, informaticien
>Service régional de l'Inventaire
>DRAC Bretagne
>mailto:pierrick.brihaye at culture.gouv.fr