[lojban-beginners] Re: Lojban parser and graph generator

[John Leuner <jewel@subvert-the-dominant-paradigm.net>]

I have made some small updates to my lojban parser:

http://subvert-the-dominant-paradigm.net:9072/generator

You can now choose between a full parse tree and an edited one. It will
also output an png or svg file. With Mozilla 1.5 you can also view
inline svg.

I'm not sure what should be in the edited parse tree, can anyone help me
figure out which elements are the most important?

I would also like to ask if there is a plan for standardising the lojban
PEG grammar? It would be nice if we could create an official version and
track changes made to it.

At the moment I'm using the grammar as one file containing both the
lojban grammar and morphology. I know some of the other lojban PEG
utilities use the grammar in a different format, but maybe we can move
towards a common source format?

John Leuner


On Mon, 2006-01-23 at 23:28 +0700, John Leuner wrote:
> I have finished the first stage of a web front-end to my lojban parser.
> 
> You can visit it here:
> 
> http://subvert-the-dominant-paradigm.net:9072/generator
> 
> and enter some lojban text.
> 
> It uses my cl-peg package along with a lojban peg grammar based on the
> work by Robin and xorxes (and I suppose some other people too).
> 
> I process the output with the cl-dot package which feeds graphviz the
> result of the parse process. The result is a graphical representation of
> the parse nodes.
> 
> As yet there are no options to trim down the tree, so you see the whole
> morphological breakdown and all the intermediate labels.
> 
> Future work is to allow the user to control how much detail is displayed,
> as well as to lookup english glosses in wordlists and display them too.
> 
> John Leuner
> 
> Note:
> 
> There are some bugs in the grammar, I noticed that "zo" doesn't work. I
> haven't tested it thoroughly so I don't know what else doesn't work.
> 
> 
> 
> 
>