[Groff] Re: What's missing for Unicode support of groff?

[Werner LEMBERG]

> In short - I proposed a simple hack to accept UTF8 input encoding,
> triggered by -mutf8 on the command line,

This is not possible, I think.  `-m' only handles macro stuff.

> a quick and dirty solution.

I don't see how this shall work.

> gpreconv is just a digest of this thread (and "man pages encoding"
> one), translated from English to C.  It suggest how to do it
> properly.

Yes.  Your idea of just using a preprocessor without actually
implementing real UTF8 support in troff is the very idea I like.

> Integrating gpreconv in the way you talk about is a step towards 2.0
> (review the messages of Bruno Haible in this thread),

Yes, it is.

> while code from uni2groff.c hidden behind -mutf8 is just a small step
> towards 1.19.3

Please tell me how to do that.

> The gpreconv approach is quite opposite to what I proposed.

No, it isn't.

> gpreconv is written along the lines of the discussion in this tread
> of the PROPER internationalization of groff.  Nobody has come up
> with a reason good enough for you to start coding and all us get
> ready for lots of testing.

Look, gpreconv.c has already been written!  It just need to be
integrated into the groff pipeline.  This isn't much work, and it
doesn't require deep knowledge of GNU troff internals.

> Let us make minimal changes.

This *is* the minimal change.  Perhaps I'm misunderstanding you, so
please give more details what you really would like to see, and how
integration should work.

> I propose a HACK - in case of -m UTF8 call internal conversion
> function, not instructions from some external UTF8.tmac file.

You mean by adding a new request?


    Werner