From rlpowell@digitalkingdom.org Fri May 07 13:47:31 2004 Received: with ECARTIS (v1.0.0; list lojban-list); Fri, 07 May 2004 13:47:31 -0700 (PDT) Received: from rlpowell by chain.digitalkingdom.org with local (Exim 4.31) id 1BMCFh-0008UD-Bx for lojban-list@lojban.org; Fri, 07 May 2004 13:47:25 -0700 Date: Fri, 7 May 2004 13:47:25 -0700 To: lojban-list@lojban.org Subject: [lojban] Re: My parser, SI, SA, and ZOI Message-ID: <20040507204725.GB27947@digitalkingdom.org> Mail-Followup-To: lojban-list@lojban.org References: <20040507185509.GN7020@digitalkingdom.org> <20040507203819.68997.qmail@web41901.mail.yahoo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20040507203819.68997.qmail@web41901.mail.yahoo.com> User-Agent: Mutt/1.5.5.1+cvs20040105i From: Robin Lee Powell X-archive-position: 7685 X-ecartis-version: Ecartis v1.0.0 Sender: lojban-list-bounce@lojban.org Errors-to: lojban-list-bounce@lojban.org X-original-sender: rlpowell@digitalkingdom.org Precedence: bulk Reply-to: lojban-list@lojban.org X-list: lojban-list On Fri, May 07, 2004 at 01:38:19PM -0700, Jorge Llamb?as wrote: > So both words turn what follows into special tokens, but remain > themselves as separate tokens. Oh, I'm sorry, you are correct. I thought both 'zo' and the following token were eaten, but it's just the following token. > That's not good. It causes a lot of problems. Yeah, no kidding. I don't do token replacement of any kind, of course, as I have no preprocessor. > All of the following should give error if grammar .300 is followed to > the letter: > > {zo da si de} > > si will erase the previous token, 'any_word_698', and then > ZO followed by KOhA should give an error. Correct. We can't test this because the official parser doesn't have si handling. > {zo da zei de} > > zei will join 'any_word_698' and KOhA and turn everything > into BRIVLA, but then ZO followed by BRIVLA should cause > an error. No. This fails, but not for that reason. ZO processing happens long before ZEI, so ZEI tries to bind any_word_698 and KOhA, any_word_698 is not allowed as an argument to ZEI in grammar.300 It does, in fact, fail in the official parser. > {zo da bu} > > bu will turn 'any_word_698' into BY, but then ZO followed > by BY should cause an error. Again, ZO happens first, but BU can't handle any_word_698, so it fails. It does, in fact, fail in the official parser. > Similar things will happen with zoi: > > {zoi gy sth gy si} > > will give an error because si will swallow the 'any_word_698' > token No; anything_699 (which is what you meant) does *NOT* swallow the delimiter, so si will swallow gy, and you end up with "zoi gy anything_699", which won't parse. Again, not testable. > I think that the Right Thing is to treat {zo } and {zoi > } as single tokens of selmaho KOhA. Why KOhA? > In any case, I don't see any justification for treating {zo} in one > way and {zoi} in another. Neither do I, as zo doesn't actually work the way I thought it did. -Robin -- http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin. "Many philosophical problems are caused by such things as the simple inability to shut up." -- David Stove, liberally paraphrased. http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi