Re: [lojban] Re: Robin Confused (was Re: Re: "pu" versus "pu ku" and LR(1))

[Bob LeChevalier <lojbab@lojban.org>]

At 11:55 AM 4/6/04 -0700, Robin Lee Powell wrote:
> On Tue, Mar 30, 2004 at 10:24:30AM -0500, Bob LeChevalier wrote:
> > At 03:15 PM 3/29/04 -0800, Robin Lee Powell wrote:
> > >I'm sorry, I must be missing something.  In the last two cases,
> > >unless I'm seriously confused, that's just a tense binding to a
> > >selbri. Neither of them require a 'ku' to be inserted to parse,
> > >whereas the example I gave does:
> > >
> > >     mi pu ge klama le zarci gi tervecnu lo cidja
> > >
> > >versus
> > >
> > >     mi pu ku ge klama le zarci gi tervecnu lo cidja
> > >
> > >Only the latter is accepted by the current parser, but all of your
> > >examples are accepted, and using completely different functionality.
> > >I don't see how my examples relate to your examples.
> >
> > If the question is what the syntactic rule invalidating the first is,
> > then look at .300 gek_sentence_54, second line.  To tense a gek'd
> > bridi-tail, you need "pu ke" and not just bare "pu".
>
> Right, and my point was that the only *possible* valid interpretation of
> 'pu' there, without 'ke', is as 'pu ku'.

Appears to be a parser bug.  Using -d* option on the parser we see that 
when it sees pu ge klama, it *is* inserting the ku.

With a little fiddling with similar sentences however, it appears that 
while inserting the ku, it is losing the ge for some reason, and hence has 
no idea what to do with the gik portion when it has parsed it.

lojbab


--
lojbab                                             lojbab@lojban.org
Bob LeChevalier, Founder, The Logical Language Group
(Opinions are my own; I do not speak for the organization.)
Artificial language Loglan/Lojban:                 http://www.lojban.org