Re: Could I get me a parser?

[cowan@snark.thyrsus.com (John Cowan)]

> First, I've thrown out the source code to the parser,

Don't do that any more.  You think sending e-mail across continents is
free?  .ianai zo'ocai

> and I'll need to run
> it on another machine(s) now, so I wonder if you could shell me down the
> source again.

On its way.  This is "2;0;30moi ke lojbo gersurfa'i", so I'll ship the
latest techfix document as well.  Note that some proposals, notably 28,
are subject to change or revocation.

> Second, I wonder if I could be allowed to toy with the source code.

Of course.  Naturally, you have to send us back your changes so we can
fold them in again.

> I had
> in mind using yacc as a preprocessor to the prolog, inserting all the
> terminators, and trying to minimise some work --- identifying grammemes,
> for example, so I'd like yacc to spit out slightly different stuff to
> what it currently does.

Go right ahead.  However, you should also check out the -t and -r options;
I don't know if the parser you have now supports -r, but -t has been in since
almost the beginning.  -t in particular was designed to make it easy for post-
processors to reconstruct the fully annotated parse tree in a single pass,
without requiring crufty conventions about passing binary stuff between
programs.

> I know there'll be huge duplication of effort in reparsing into PROLOG,
> but I guess these things can't be helped *sigh*

Such is life.  BTW, what will be your implementation language for the project?

On a different note, I think you should consider supporting two additional
things:  universal quantification a la simple Prolog variables, and
imperatives.  It would be way cool if a "ko" triggered a lookup so that
"ko ciska le broda" came out "print(le_broda)." or the like.  That way
actual Prolog programming in Lojban would be possible!

-- 
John Cowan	cowan@snark.thyrsus.com		...!uunet!lock60!snark!cowan
			e'osai ko sarji la lojban.