garyb@abekrd.UUCP (Gary Bartlett) (12/20/90)
Does anyone have patches for zsh to enable it to compile with a K&R-C compiler, or know of a better Ansi-C -> K&R-C translator than ansi2knr (which can't hack it)? Thanks, Gary -- --------------------------------------------------------------------------- Gary C. Bartlett NET: garyb@abekrd.co.uk Abekas Video Systems Ltd. UUCP: ...!uunet!mcsun!ukc!pyrltd!abekrd!garyb 12 Portman Rd, Reading, PHONE: +44 734 585421 Berkshire. RG3 1EA. FAX: +44 734 567904 United Kingdom. TELEX: 847579
woods@eci386.uucp (Greg A. Woods) (12/22/90)
In article <1578@abekrd.UUCP> garyb@abekrd.UUCP (Gary Bartlett) writes: > Does anyone have patches for zsh to enable it to compile with a K&R-C compiler, > or know of a better Ansi-C -> K&R-C translator than ansi2knr (which can't > hack it)? I didn't think it could be done with a filter, which is why I spent a few hours doing it by hand. Of course while I was "fixing" the code, I was also trying to port it to SysVr3.0. The code conversion is done, and it compiles fine. There are some other lurking nasties beyond the parameter declarations, particularly with type/variable identifier clashes. As for running on a system without the bsd-4.2 signals (SysVr3 has something closer to bsd-4.1 or bsd-4.3), and without the bsd setpgrp(), the entire execpline() stuff needs re-design. It should be possible to provide background job monitoring for SysV without job control, since ksh does so quite nicely, and come csh ports also manage to work well enough. It may even be possible to patch in some shl support, though I won't be doing that, since I use a DMD both at work and at home (and I have ksh at home anyway :-). There are also some weird bugs that may, or may not, have something to do with my changes: aliases, command substitution (`cmd`), some csh history stuff, and some readline stuff; all don't work properly. There are also some annoying bugs in the shell itself, such as no Bourne shell compatible trap command, incompatibility with the ksh set command (i.e. set -o), it's impossible to turn off job control in an interactive shell, and so on. All in all, I think the guts of this shell are reasonably good. The manual is excellent. On a quick comparison, I'd rate it well above bash, though I've never looked at the guts of bash, only the bug reports! :-) Anyway, I'm not about to release my version (patches are impossibly large) without first discussing plans with the author, and certainly not before I either get a chance to try it on a bsd box, or fix it properly for SysV. -- Greg A. Woods woods@{eci386,gate,robohack,ontmoh,tmsoft}.UUCP ECI and UniForum Canada +1-416-443-1734 [h] +1-416-595-5425 [w] VE3TCP Toronto, Ontario CANADA Political speech and writing are largely the defense of the indefensible-ORWELL
jc@mips.COM (J.C. Webber III) (12/25/90)
In article <1578@abekrd.UUCP> garyb@abekrd.UUCP (Gary Bartlett) writes: >Does anyone have patches for zsh to enable it to compile with a K&R-C compiler, >or know of a better Ansi-C -> K&R-C translator than ansi2knr (which can't >hack it)? ^^^^^^^^ > |||||||| Does anybody know where to get the sources for this |||||||| ? thx..jc@mips
andreess@mrlaxs.mrl.uiuc.edu (Marc Andreessen) (12/25/90)
In article <44321@mips.mips.COM> jc@mips.COM (J.C. Webber III) writes: >In article <1578@abekrd.UUCP> garyb@abekrd.UUCP (Gary Bartlett) writes: >>or know of a better Ansi-C -> K&R-C translator than ansi2knr (which can't >>hack it)? ^^^^^^^^ >> |||||||| >Does anybody know where to get the sources for this |||||||| ? It's part of the GhostScript 2.0 distribution, available at an FTP site near you (prep.ai.mit.edu for sure). Marc -- Marc Andreessen___University of Illinois Materials Research Laboratory __________________andreessen@uimrl7.mrl.uiuc.edu______________________