Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!husc6!cmcl2!brl-adm!brl-smoke!gwyn
From: gwyn@brl-smoke.ARPA (Doug Gwyn )
Newsgroups: comp.lang.c
Subject: Re: ANSI standard and open()
Message-ID: <6049@brl-smoke.ARPA>
Date: Fri, 3-Jul-87 22:39:37 EDT
Article-I.D.: brl-smok.6049
Posted: Fri Jul  3 22:39:37 1987
Date-Received: Sun, 5-Jul-87 00:16:56 EDT
References: <18346@ucbvax.BERKELEY.EDU> <8042@utzoo.UUCP>
Reply-To: gwyn@brl.arpa (Doug Gwyn (VLD/VMB) <gwyn>)
Organization: Ballistic Research Lab (BRL), APG, MD.
Lines: 19

In article <2343@hoptoad.uucp> farren@hoptoad.UUCP (Mike Farren) writes:
>...  He says that in persuing H&S's discussion of the ANSI
>standard, the low-level I/O functions (such as open(), read(), etc.)
>do not seem to exist.

I'm posting the response rather than mailing it, because this is an issue
worth understanding.

The (proposed) ANS for C (X3.159) limits itself to the "raw" language and
to that portion of the hosted run-time environment (library routines etc.)
that are believed to be usefully implementable on as wide a range of host
architectures and operating systems as possible.  The UNIX functions
open(), read(), etc. can be adequately implemented on some systems but not
on others.  Consequently the C standard will define the STDIO interface in
a way believed to be universally implementable, rather than a UNIX-like
set of primitive I/O functions.  The latter, however, are being
standardized by IEEE 1003.1 (POSIX), which has its scope specifically
limited to UNIX-like systems.  ANS X3.159-198x should be implementable on
any POSIX-supporting system, but not the converse.