[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lojban] camxes request ...
Hello Frank,
I am currently working on the 4th baseline grammar; which is the
translation of the official grammar from YACC/BNF to PEG. I'm
building on work done by Robin and xorxes.
I'm in a quiet period for this work, being focused on fixing bugs
in the so-called Magic Words implementation in the PEG grammar.
There isn't a lot to report publically, but I could certainly upload
my very-much-a-work-in-progress nothing-useful-here implementation
if you ask. I doubt it is useful to you in it's current form,
however.
The public work I do have is here:
http://wiki.call-cc.org/eggref/4/jbogenturfahi
With the PEG grammar specifically being:
http://bugs.call-cc.org/browser/release/4/jbogenturfahi/trunk/rafske.peg
http://bugs.call-cc.org/browser/release/4/jbogenturfahi/trunk/rafske_gumgau.peg
http://bugs.call-cc.org/browser/release/4/jbogenturfahi/trunk/gerna.peg
These files are all derived from camxes, and have only minor
bugfixes until I merge in my work on Magic Words.
No one knows yet whether jbogenturfa'i will become the official
parser, or whether the PEG grammar it uses will be ported to another
parser which will become the official parser.
However, the PEG grammar in jbogenturfa'i is the most continuously
tested. One of my projects is to take all of the example Lojban in the
CLL and run in through the parser to assert it is producing a
correct parse tree. As well, I'm constructing other test cases to ensure
the PEG grammar is correct. I'm interested in asserting that every
production can be reached and matched; using the grammar to generate
example phrases that reparse to the same parse tree.
Unless you're interested in some very obscure boundary conditions with
Magic Words, the PEG grammar is the most accurate grammar we've
got. I am still finding the occasional bug, but Robin and I last week
found a bug in the YACC grammar!
What project are you working on? Is it something where we'd be able
to share work? Are you planning on making changes to the PEG that
should be merged upstream?
-Alan
On Thu, Jul 07, 2011 at 06:38:09PM -0400, Frank wrote:
> On 7 July 2011 16:48, Robin Lee Powell <rlpowell@digitalkingdom.org> wrote:
> > On Thu, Jul 07, 2011 at 04:44:53PM -0400, Frank wrote:
> >
> > http://www.digitalkingdom.org/~rlpowell/hobbies/lojban/grammar/
>
> ki'e
>
> > Honestly, though, now that there are *way* more decent PEG parsers
> > out there, I'd rather people stole the PEG and made their own, one
> > that sucks less.
>
> Ah, I see. I was under the impression, however, camxes was considered
> an adequate parser. No matter. You definitely have more experience in
> this area than I.
>
> BTW, the PEG in question is this one, yes? =>
> http://www.digitalkingdom.org/~rlpowell/hobbies/lojban/grammar/lojban.peg.txt
>
> > IIRC Alan Post is working on that already?
>
> Perhaps. I will ask Alan.
>
> ki'e
>
> --
> 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.
>
--
.i ma'a lo bradi ku penmi gi'e du
--
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.