Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!dali.cs.montana.edu!milton!sumax!polari!kentfo
From: kentfo@polari.UUCP (Kent Forschmiedt)
Newsgroups: comp.sources.d
Subject: Re: v01INF1: Status - Status of comp.sources.reviewed
Message-ID: <3840@polari.UUCP>
Date: 23 Apr 91 16:34:08 GMT
References: <B9UA6DD@xds13.ferranti.com> <MEISSNER.91Apr20151434@curley.osf.org> <6_WAABB@xds13.ferranti.com>
Organization: Seattle Online Public Unix (206) 328-4944
Lines: 27

In article <MEISSNER.91Apr20151434@curley.osf.org> meissner@osf.org (Michael Meissner) writes:
> This can produce the wrong error message if sprintf sets errno (for
> example if it calls malloc and malloc fails).

In article <6_WAABB@xds13.ferranti.com> peter@ficc.ferranti.com (Peter da Silva) writes:
>
>I don't know of any implementations where sprintf can set errno.
>Does *anyone* know of one?

Who cares?  If there is one, or somebody someday makes one, somebody will
eventually be bitten.  Many years ago, I began adding two lines of code to
my versions of these functions, and stopped losing sleep.

errorthing(char *whatever)
{
	char buf[200];
	int etmp = errno;

	/* whatever */

	errno = etmp;
	perror(buf);
}


-- 
Kent Forschmiedt ... kentfo@polari ... uw-beaver!sumax!polari!kentfo