Re: lout, unicode & truetype/opentype

[Jeff Kingston]

Mark Summerfield wrote:

> I was wondering how much work it would be to make lout unicode friendly?
> I don't know the code (and my C is rusty) but there seem to me to be two
> key issues:
> 
> - using wchar_t instead of char.
> - being able to write the glyph corresponding to the wchar_t's code
>   point into the postscript.
>
> For the first of these I guess it would be a long tedious process of
> going through the code and changing all those char's that needed it, and
> anything that depends on their size (which hopefully isn't anything).

I moved "char" into a typedef long ago, but there are still quite a
few things to look at:  the fopens, the lexical analyser (you need
an optimized trie to classify the characters, instead of the old
256-element array), and I'm not sure what else.

> For the second of these the LASi project have come up with an ingenious
> solution...

That sounds interesting.  Lout is a mature system that does what it does
and takes up a few days of my time once or twice a year when I put out a
new release.  I'm really not interested in investing a lot of time
chasing new technologies.  I'm far enough behind with Nonpareil as it
is.  If someone else wants to add such a feature and wear the cost of
the development and maintenance, I'm ready to give advice and help.

Jeff