From - Thu Mar 14 11:53:55 1996 Received: from VMS.DC.LSOFT.COM (vms.dc.lsoft.com [205.186.43.2]) by locke.ccil.org (8.6.9/8.6.10) with ESMTP id PAA20744 for ; Wed, 13 Mar 1996 15:50:40 -0500 Message-Id: <199603132050.PAA20744@locke.ccil.org> Received: from PEACH.EASE.LSOFT.COM (205.186.43.4) by VMS.DC.LSOFT.COM (LSMTP for OpenVMS v1.0a) with SMTP id 84997609 ; Wed, 13 Mar 1996 15:52:16 -0500 Date: Wed, 13 Mar 1996 22:50:29 +0200 Reply-To: Veijo Vilva Sender: Lojban list X-UIDL: 826754219.001 From: Veijo Vilva Subject: Re: TECH: PROPOSED GRAMMAR CHANGE 2$i X-To: lojban@cuvmb.cc.columbia.edu To: John Cowan Status: U X-Mozilla-Status: 0011 Content-Length: 952 > Date: Wed, 13 Mar 1996 14:26:10 -0300 > From: "Jorge J. Llambias" > Subject: Re: TECH: PROPOSED GRAMMAR CHANGE 2$i >>> (Even better if the first NUhU could be avoided.) >> >> 1 shift/reduce conflict > > Is it possible to identify when this happens? When a {gi} is found, > the only possibility is that it is closing the latest open gek, > isn't it? In other words, under what circumstances would the first > nu'u not be elidable? After some juggling I found a solution at the YACC level: --- in the rules for term_set_83 replace GEK_807 terms_80 GIK_816 NUhU_gap_460 terms_80 NUhU_gap_460 with GEK_807 terms_80 term_81 GIK_816 terms_80 term_81 NUhU_gap_460 Now, a termset must have at least two terms. co'o mi'e veion --------------------------------- .i mi du la'o sy. Veijo Vilva sy. ---------------------------------