From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 19777 invoked from network); 22 Feb 1999 09:38:15 -0000 Received: from sunsite.auc.dk (130.225.51.30) by ns1.primenet.com.au with SMTP; 22 Feb 1999 09:38:15 -0000 Received: (qmail 27026 invoked by alias); 22 Feb 1999 09:37:56 -0000 Mailing-List: contact zsh-workers-help@sunsite.auc.dk; run by ezmlm Precedence: bulk X-No-Archive: yes X-Seq: 5462 Received: (qmail 27019 invoked from network); 22 Feb 1999 09:37:55 -0000 Message-Id: <9902220921.AA11759@ibmth.df.unipi.it> To: probin@qnx.com, zsh-workers@sunsite.auc.dk (Zsh hackers list) Subject: Re: zsh on QNX4 In-Reply-To: ""Philippe Robin""'s message of "Fri, 19 Feb 1999 14:15:04 NFT." <000001be5c12$3f07e000$089d9e9f@philippe.qssl> Date: Mon, 22 Feb 1999 10:21:57 +0100 From: Peter Stephenson "Philippe Robin" wrote: > I got zsh running on QNX 4.25 operating system. I have included the few > modifications I did to get it to work at the end of this mail if you want to > incorporate this... This looks pretty easy to include, but it ought to be decided automatically. It's possible to put ad hoc code in configure if necessary, though if there were some predefined preprocessor symbol that would be better. > #define GETPGRP_VOID Does the configure test for this not work? You should get something like checking whether getpgrp takes no argument... yes and config.log should say why it failed, if it did. > #ifdef QNX > oxtabs = 0; > #else > oxtabs = ((SGTTYFLAG & SGTABTYPE) == SGTABTYPE); > #endif Is it enough to test #if defined(SGTTYFLAG) && defined(SGTABTYPE) ? > # ifdef OXTABS > ti.tio.c_oflag &= ~OXTABS; > # else > // ti.tio.c_oflag &= ~XTABS; > # endif > # endif > // ti.tio.c_oflag |= ONLCR; Likewise, can we just test directly for these flags being defined? > #ifndef QNX > char * > zgetcwd(void) > { > return getcwd(NULL, 0); > } > #else This looks like something we need a preprocessor definition for. If there really isn't one (_QNX_, _QNX, __QNX__, ...?) the system as printed out when configure starts should be enough to work on. > halloc() is already defined on QNX with the following prototype: That's an annoyance, but we can easily rename it to zhalloc(). -- Peter Stephenson Tel: +39 050 844536 WWW: http://www.ifh.de/~pws/ Dipartimento di Fisica, Via Buonarroti 2, 56127 Pisa, Italy