[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lojban] zoi bug in camxes?
On Tue, Jan 25, 2011 at 09:19:57AM -0800, Robin Lee Powell wrote:
> On Tue, Jan 25, 2011 at 09:01:34AM -0300, Jorge Llambías wrote:
> > On Tue, Jan 25, 2011 at 4:07 AM, Robin Lee Powell
> > <rlpowell@digitalkingdom.org> wrote:
> > >
> > > it shouldn't be processing stuff inside the zoi quote at all.
> > > "gyrate" or "gyrations" or whatever shouldn't be able to be
> > > matched as the delimiter because "gy" isn't *seperated out as a
> > > word* there.
> >
> > The text inside the quote "zoi gy gy" is the empty string.
> > Following this with the Lojban words "ra", "te", "gy" makes no
> > difference, there is no "processing stuff inside the quote" going
> > on here since there is no stuff inside the quote to process, other
> > than a space.
>
> I understand what you're saying, I simply disagree that zoi should
> work that way; the ending particle should have to be completely
> delimited.
>
> I don't actually *care* all that much, it's just my feeling on the
> matter.
>
From the conversation, I can summarize three separate proposals:
1) leave the PEG grammar alone and correct the CLL to describe the
way this grammar is behaving.
2) replace the rule for zoi-word to match non-lojban-word rather
than any-word, so 'gyrate' won't be divided into three words,
satisfying Robin's consistency argument. We'll still need to
update the CLL for the behavior, I think.
3) Replace the PEG grammar with something that reads the stuff
between the ZOI delimiter a character at a time. Either require
a pause before the final delimiter or not.
I like 2, 1, and then 3 in that order.
3 I don't like because the rest of the grammar doesn't really work
that way. The grammar is defined by token streams and the
morphology file handles composing the input stream into tokens.
3 Is not completely unprecedented, however, FAhO does something
similar.
2 I like because because of the argument Robin raised about
consistency. I think it is surprising that gyrate is invalid
but gyration is valid, and I don't think that surprise is a useful
feature. I can't think of a reason to dislike 2, please help me
with that.
1 Has the advantage of working that way right now. It also has the
advantage of prefering Lojbanic text to non-Lojbanic text. ZOI
isn't special from that perspective, the parser does what the parse
does ZOI or not.
Will you give me your preference ordering? I'd like to know whether
I should:
a) update the CLL errata.
b) update the PEG grammar.
c) raise an issue with the BPFK to make a decision.
I'll use the preference ordering to make a choice.
-Alan
--
.i ko djuno fi le do sevzi
--
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.