[XeTeX] The future of XeTeX

Shriramana Sharma samjnaa at gmail.com
Tue Jul 31 02:34:13 CEST 2012

On Tue, Jul 31, 2012 at 1:35 AM, Adam Twardoch (List)
<list.adam at twardoch.com> wrote:
> optional) than to write full-scale OpenType Layout shapers in Lua from
> scratch.

... which would still leave people who are heavily banking on Graphite
for their work into old Indic scripts (which are unlikely to be
supported [well] in any OT implementation) high and dry. The
probability of Graphite support being written in Lua is much lesser
than a successful support for OT.

(Do I understand correctly that the Lua code has to itself parse the
font tables? No external library linkups allowed? So simple bindings
to the SIL-distributed Graphite library will not be possible?)

> HarfBuzz is still not perfect but it definitely developing
> towards becoming the only sensible OpenType Layout engine within the
> open-source realm.

While HB has its own OT shapers, it also supports the Gr2 library so
for people like me HB support is the only answer. HB also supports the
Uniscribe backend (for those who want it) and from Behdad's recent
comment, the latest Mac backend too. So it's not only "YAOTI" (yet
another OpenType implementation) but a global solution -- kudos to
Behdad and Jonathan and others who are working on it!

> Developing some of the shapers, especially for Indic
> languages, is not trivial per se (it is taking the HarfBuzz developers
> quite a long time, and it is after all an intensive effort). Doing
> everything from scratch in Lua will be a similarly long-winded effort:
> Lua is not a very popular language, and even if an complete OTL solution
> is created in Lua some day, it will need maintenance.

Hm -- over on the ConTeXT list I wrote "Please don't tell me you are
writing yet another OT engine", and I got the semi-jocular (?) reply
"OK we don't tell you"! :-) Frankly as a Indian, and a Sanskrit/Vedic
scholar, my use cases (esp complicated svara-markup in Vedic) test the
limit of rendering engines, and I really don't want to keep submitting
bug reports for each and every open-source OT engine (doesn't that in
some way go against the concept of open-source by taking out the
collaboration aspect?) -- so switching to ConTeXT mkiv which supports
only LuaTeX is NOT an option for me.

> The great thing about the XeTeX concept is that it doesn't try to do
> everything itself. It uses the best available components to do the
> heavy-lifting.

And that's what is in line with the Unix philosophy write? Code-reuse?

I'd have been happy if ConTeXT with its provisions for fine
typographic control would have continued to support XeTeX in its
further development, but well the developers have decided to make it
monolithic and based on Lua, so that's their call.

Shriramana Sharma

More information about the XeTeX mailing list