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!linus!philabs!cmcl2!seismo!brl-tgr!gwyn
From: gwyn@brl-tgr.ARPA (Doug Gwyn <gwyn>)
Newsgroups: net.unix-wizards
Subject: Re: NULL problems
Message-ID: <10713@brl-tgr.ARPA>
Date: Wed, 15-May-85 16:48:44 EDT
Article-I.D.: brl-tgr.10713
Posted: Wed May 15 16:48:44 1985
Date-Received: Thu, 16-May-85 22:03:01 EDT
References: <10550@brl-tgr.ARPA> <5369@ucla-cs.ARPA> <240@ucbcad.UUCP>
Distribution: net
Organization: Ballistic Research Lab
Lines: 11

> Can anybody think of any problems with making "NULL" a part of the C
> language, specifically as a pointer to anything with a value of 0? As
> has been pointed out, it is a pain to have to remember to cast it all
> the time, and I tend to think of NULL as a pointer, not an integer.

Pointer to WHAT????  That is the whole point (n.p.i.) of casting NULL
when using it as an argument to a function.  (You don't need to cast
it for nearly any other use.)  The ANSI solution to the general
argument coercion issue will work for 0 just as well.

Shouldn't this discussion be in net.lang.c (a.k.a. INFO-C)?