Path: utzoo!attcan!uunet!tut.cis.ohio-state.edu!usenet.ins.cwru.edu!cwns1!chet From: chet@cwns1.CWRU.EDU (Chet Ramey) Newsgroups: comp.unix.questions Subject: Re: Compiling GNU Bourne Again SHell (bash) on SUN OS Message-ID: <1990Aug15.181554.19306@usenet.ins.cwru.edu> Date: 15 Aug 90 18:15:54 GMT References: <24173@adm.BRL.MIL> Sender: news@usenet.ins.cwru.edu Reply-To: chet@po.CWRU.Edu Followup-To: gnu.bash.bug Organization: Case Western Reserve Univ. Cleveland, Ohio, (USA) Lines: 49 In article <24173@adm.BRL.MIL> hsw@sparta.com (Howard Weiss) writes: $ I've got a copy of the Bourne Again SHell (bash) - version 1.05 that $ On the SUN, the make file says that its compiling for a SUN 3 running $ OS 4, but has problems with the bash-Makefile (fixed by commenting out $ the offending lines which are setting up stuff to build backup files) and $ then has problems with compiling y.tab.c (which is the result of running $ yacc on parse.y). I have already gotten and included all the latest $ patches for this version and have sent off a message to bug-bash but $ have not heard anything. $ $ Has anyone had this problem and/or does anyone know of a fix. I'm not $ versed in yacc (and don't really want to be). The yacc problem is solved by changing the lines in parse.y that read if (interactive) YYACCEPT; else YYABORT; to read if (interactive) { YYACCEPT; } else { YYABORT; } YYACCEPT and YYABORT became compound statements in SunOS 4.1 yacc. You will also need to use the SunOS malloc, because of the `localtime' bug (take USE_GNU_MALLOC out of the appropriate section in machines.h). There are a number of problems with bash 1.05. All have been fixed for the next release (whenever that is). Chet -- Chet Ramey ``Levi Stubbs' tears run down Network Services Group his face...'' Case Western Reserve University chet@ins.CWRU.Edu