From mboxrd@z Thu Jan  1 00:00:00 1970
From: Geoff Keating <geoffk@ozemail.com.au>
To: hjl@lucon.org
Cc: libc-hacker@sourceware.cygnus.com
Subject: Re: Setenv/putenv changes break existing code.
Date: Mon, 02 Aug 1999 18:48:00 -0000
Message-id: <199908030123.LAA00651@gluttony.geoffk.wattle.id.au>
References: <19990803010000.E96D457BA@ocean.lucon.org>
X-SW-Source: 1999-08/msg00007.html

> Date: Mon, 2 Aug 1999 18:00:00 -0700 (PDT)
> From: hjl@lucon.org (H.J. Lu)

> The recent setenv/putenv change break existing code like:
> 
> 	p = malloc (100);
> 	.....
> 	putenv (p);
> 	free (p);
> 
> sice p is used directly in putenv. However, from the Unix standard:
> 
> The putenv() function uses the string argument to set environment
> variable values. The string argument should point to a string of the
> form "name=value". The putenv() function makes the value of the
> environment variable name equal to value by altering an existing
> variable or creating a new one. In either case, the string pointed to
> by string becomes part of the environment, so altering the string will
> change the environment. The space used by string is no longer used
> once a new string-defining name is passed to putenv(). 
> 
> It seems to me that it is ok to free 'p' since it is no longer used
> when returning from putenv ().

Surely freeing 'p' counts as altering *p?  And altering it "will
change the environment."

The last sentence means that if you do

p = malloc(100);
strcpy (p, "NAME=value");
putenv (p);
putenv ("NAME=some_other_value");

_now_ you can free `p', because you have redefined `NAME'.

-- 
Geoffrey Keating <geoffk@cygnus.com>