[lojban] Re: Python module for parsing lojban

[Marek Rogalski <mafikpl@gmail.com>]

I have added some macros, so it builds also under python2.
http://stud.ics.p.lodz.pl/~mafik/pylojban-1.0.1.tar.bz2

On Sep 26, 6:17 am, purpleposeidon <purpleposei...@gmail.com> wrote:
> Bugs!
> * It can only parse once

Yes. It stores a lot of its parsing state in global and static
variables. Also some garbage stays in the memory. After parsing one
string its state needs to be reversed.

> * Can't parse "dabrOda"
> * Doesn't support erasures
> * Most gregarious of all: It's written in C! It'll probably run faster
> than mine in python! This is a very bad bug! It makes zirsam look
> bad....
>

99,9% (or something like that) of c code is taken from parser-3.0.00.
As far as i know it was using different approach than zirsam (faster
but a lot less clean).

Parser-3.0.00 seemed to me a lot cleaner and faultless two days ago...
It seems to have been bad choice.

> Could you give the parse a better interface than a dict?
>

I am currently writing other utility (peg parser) which hopefully
might be used as backend for zirsam. I think that pylojban is too
buggy to fix it in reasonable amount of time. Only possible use for
it, I can imagine, are short command-line utilities for which one
parsing will be enough.

> mu'omi'e.djeims.

--
Marek Rogalski

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.