From jcowan@reutershealth.com Thu Apr 08 22:10:52 2004 Return-Path: X-Sender: jcowan@reutershealth.com X-Apparently-To: lojban@yahoogroups.com Received: (qmail 74400 invoked from network); 9 Apr 2004 05:10:52 -0000 Received: from unknown (66.218.66.217) by m23.grp.scd.yahoo.com with QMQP; 9 Apr 2004 05:10:52 -0000 Received: from unknown (HELO mail.reutershealth.com) (65.246.141.36) by mta2.grp.scd.yahoo.com with SMTP; 9 Apr 2004 05:10:51 -0000 Received: from skunk.reutershealth.com (mail [65.246.141.36]) by mail.reutershealth.com (Pro-8.9.3/Pro-8.9.3) with SMTP id BAA00612; Fri, 9 Apr 2004 01:04:12 -0400 (EDT) Received: by skunk.reutershealth.com (sSMTP sendmail emulation); Fri, 9 Apr 2004 01:10:24 -0400 Date: Fri, 9 Apr 2004 01:10:24 -0400 To: Bob LeChevalier , rlpowell@digitalkingdom.org Cc: lojban@yahoogroups.com Message-ID: <20040409051024.GA2886@skunk.reutershealth.com> References: <20040407124110.46977.qmail@web41903.mail.yahoo.com> <5.2.0.9.0.20040407062416.03375890@pop.east.cox.net> <20040407124110.46977.qmail@web41903.mail.yahoo.com> <5.2.0.9.0.20040408203833.037595d0@pop.east.cox.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <5.2.0.9.0.20040408203833.037595d0@pop.east.cox.net> User-Agent: Mutt/1.4.1i X-eGroups-Remote-IP: 65.246.141.36 From: jcowan@reutershealth.com Subject: Re: [lojban] Re: Official parser and "lo ni'a zu crino" X-Yahoo-Group-Post: member; u=8122456 X-Yahoo-Message-Num: 21985 Bob LeChevalier scripsit: > I realized that, but wanted to point out what investigation showed if it > helps in any attempts Cowan may be making to identify and fix bugs in the > official parser (which unfortunately even with bugs has to remain a > standard unless you can prove that your alternate parser has the same > grammar and that it is unambiguous to the same or higher degree than the > YACC grammar) I am not going to do any further actual work on the official parser; using Yacc is just too bug-prone and unreliable. Further, I am going to support the substitution of the PEG grammar as official; not yet, but when it's further debugged and I've convinced myself that it's equivalent. I will not support, however, any structural changes that *could* be made as a result of going to infinite-lookahead grammar: no A/JA merger, e.g. For one thing, human beings don't support infinite lookahead. But I am okay with accepting things like "le broda joi le brodi", since that is not truly an ambiguity but just the result of smarter resolution of elidable terminators than Yacc allows. I'll also do whatever I'm able to make an actual PEG parser available, with the same kinds of features (at least) that the official parser supports. I don't know what this will involve at present. > Given what we know, the bug must be the tense lexer which somehow is not > accepting ni'a zu as a valid string. I never looked at the lexer internals > so I can't debug it. Checking the source of lexruleo.c shows that the space-before-time option was never put in. Here's a patch for Robin's benefit: In the function tense_C_979, add the line "TRY(tense_C_979_4);" just after "TRY(tense_C_979_1);". This makes the order time-space, time, space-time, space. Then add a new function tense_C_979_4, which is a copy of tense_C_979_3 except that the two CALL lines are in reverse order. -- Andrew Watt on Microsoft: John Cowan "Never in the field of human computing jcowan@reutershealth.com has so much been paid by so many http://www.ccil.org/~cowan to so few!" (pace Winston Churchill) http://www.reutershealth.com