Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site brl-tgr.ARPA
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!think!harvard!seismo!brl-tgr!gwyn
From: gwyn@brl-tgr.ARPA (Doug Gwyn <gwyn>)
Newsgroups: net.lang.c
Subject: Re: ANSI
Message-ID: <1197@brl-tgr.ARPA>
Date: Fri, 3-Jan-86 08:52:57 EST
Article-I.D.: brl-tgr.1197
Posted: Fri Jan  3 08:52:57 1986
Date-Received: Sun, 5-Jan-86 01:13:11 EST
References: <1164@mtx5a.UUCP>
Distribution: net
Organization: Ballistic Research Lab
Lines: 59

> 1. ANSI supposed to come up with C standard at the end of 85.
>    I understand that it did not happen. What is the latest
>    info on it ? Where can I obtain a copy of their draft ?

The ANSI X3J11 committee is still very active, but they haven't
finished yet.  The latest draft standard that I have a copy of
is Doc. No. X3J11/85-138 (Nov. 11, 1985); it should be available
through the X3 Secretariat: CBEMA, 311 First St NW, Washington
DC 20001 (202)737-8888.

> 2. IEEE was working on their own C standard at one point.
>    What is the latest from them and how to obtain any
>    material on this subject ?

I don't think there is a separate IEEE C effort at present.
There is a Portable Operating System Environment standards
working group (P1003) that has just ballotted on a proposed
trial use (UNIX system interface) standard.

> 3. Are there any recent publications on AT&T internal
>    C standard ?

How about the C Reference Manual in the UNIX System V docs?

> a. How was the type equivalence resolved? According to
>    Rosler's article in BSTJ - 84 it is name equivalence.
>    According to Gahani ("Advanced C", 1985) it is
>    structural equivalence.

"Two types are the same if they have the same ordered set of
type specifiers and abstract declarators, either directly or
via typedefs.  Two structures or unions are different if they
have no names or different names (taking tags and typedef
synonyms into account), even if their members are identical."

> b. Is a pointer to void pointer to anything, like
>    it is done in C++ ?

(void *) is a type capable of being assigned a pointer to
anything and back without loss of information.

> c. Did aggregate's assignment statement make it into the
>    ANSI standard (draft) ?

Simple assignment requires both operands to be arithmetic
types or to have the same structure, union, or pointer type.

> d. Is specification of function's formal parameters
>    in declaration required by ANSI draft ?

Only if you want parameter type checking and coercion.

> Finally, what are the good books/articles on the
> new standard(s) ?

Since the standard has not yet been ratified, a book would
be premature.

See also the mod.std.c newsgroup.