Return to BSD News archive
Newsgroups: comp.unix.bsd Path: sserve!manuel.anu.edu.au!munnari.oz.au!news.hawaii.edu!ames!think.com!paperboy.osf.org!meissner From: meissner@osf.org (Michael Meissner) Subject: Re: 386BSD's non-standard C library In-Reply-To: dean@ksr.com's message of 6 Oct 92 18:12:03 EDT Message-ID: <MEISSNER.92Oct6231635@curley.osf.org> Lines: 60 Sender: news@osf.org (USENET News System) Organization: Open Software Foundation References: <1992Sep24.031603.21009@minyos.xx.rmit.oz.au> <id.58NT.RE4@ferranti.com> <rcskb.718045970@minyos.xx.rmit.OZ.AU> <16591@ksr.com> Date: Wed, 7 Oct 1992 03:16:39 GMT Lines: 60 In article <16591@ksr.com> dean@ksr.com (Dean Anderson) writes: | In article <rcskb.718045970@minyos.xx.rmit.OZ.AU> rcskb@minyos.xx.rmit.oz.au (Kendall Bennett) writes: | | >Oh? That is my problem? I assume that since this is a new standard (first | >proposed back in what - 1983?), and it is not supported on all systems | | The standard was not accepted until 1989 or so. My unapproved draft copy | is dated December 1988. ANSI C was approved in December of 1989. ISO C was approved in 1990 (December again I think). ANSI and ISO C are technically the same. POSIX came out in 1988 (trail use I think), and again in 1990. | >I should not use it, but write my code in pre-historic K&R C with lotsa | >#defines since non K&R system is ever comapatable with another? | | If you want to be portable(ie, run on lots of systems), this is what | you have to do. Maybe you should just stick with DOS. There are a | lot of those systems. Now that it is three or four years from the approval, I am less and less enthused about the crutches of supporting systems whose vendors are asleep at the wheel. ... | Go ahead and write ansi and posix only code. It will be a while before | it will run on vast numbers of systems. I personally don't like ansi | C because it violated its charter to document the existing language as | a standard. I don't think that all the decisions were correct, and I | don't think the community was permitted to test out the ideas well | enough. Innovation by committee usually produces very bad results. Documenting the existing language was only one of the goals. Another was to fix things. You can't satisfy conflicting goals completely. | The point here is that not everyone has an ansi compiler. Very little | code will work with an exclusively ansi compiler. Some people cannot | have both -traditional and -ansi built into their compiler. Then complain to the vendor, write your own compiler, and/or use GCC. As a compiler writer (who by the way has written a C front end from scratch), I seriously doubt providing both traditional vs. ansi support in any non-toy compiler is all that much code. Yeah, there are some dark corners, and such, and it may make you keep a tiny bit more state than if you have a traditional compiler or an ansi compiler alone, but if the compiler vendor can't hack it, they are obviously in the wrong profession. | >>Not shortcuts, exactly. The library *predates* the standard. For example, | >>the ANSI standard changed the behaviour of tolower for most compilers. Actually in this case, it depends on whether you came from the BSD universe or System V (the standard picked System V for the most part). -- Michael Meissner email: meissner@osf.org phone: 617-621-8861 Open Software Foundation, 11 Cambridge Center, Cambridge, MA, 02142 You are in a twisty little passage of standards, all conflicting.