X-Authentication-Warning: delorie.com: mail set sender to djgpp-workers-bounces using -f
Message-ID: <403BE07F.178E187E@yahoo.com>
Date: Tue, 24 Feb 2004 18:38:39 -0500
From: CBFalconer <cbfalconer AT yahoo DOT com>
Organization: Ched Research
X-Mailer: Mozilla 4.75 [en] (Win98; U)
X-Accept-Language: en
MIME-Version: 1.0
To: djgpp-workers AT delorie DOT com
Subject: Re: strtoul bug (was Re: Fibonacci number)
References: <d7c3a0b2 DOT 0402220759 DOT 34d6435d AT posting DOT google DOT com>
	 <4038E8CA DOT 6491815E AT virginia DOT edu> <4039DD96 DOT 3F36F3B7 AT yahoo DOT com>
	 <200402231458 DOT i1NEwKwm020904 AT envy DOT delorie DOT com> <403A301C DOT E8F5FE65 AT yahoo DOT com>
	 <200402231751 DOT i1NHp5lv022894 AT envy DOT delorie DOT com> <403A5F55 DOT 7E608910 AT yahoo DOT com>
	 <200402232034 DOT i1NKYqrt024366 AT envy DOT delorie DOT com> <403ADB4A DOT ED9F77C2 AT yahoo DOT com> <403BD3E2 DOT 90602 AT cyberoptics DOT com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Reply-To: djgpp-workers AT delorie DOT com

Eric Rudd wrote:
> 
> 7.20.1.4, paragraph 3 in the C99 standard specifies the convertible
> part of the input string to consist of "a sequence of letters and
> digits...optionally preceded by a plus or minus sign", so signs
> appear to be acceptable.  Even if the objection is made that
> strtoul("-1", NULL, 10) doesn't lie within the representable range
> of an unsigned long, then the value of strtoul("-0", NULL, 10)
> surely does, in spite of its minus sign.
> 
> For me, the question whose answer is most unclear is about the
> proper thing to do when the string represents a number that is out
> of range.  From a reading of 7.20.1.4, it appears to me that
> ULONG_MAX should be returned, regardless of whether the number was
> out of range in the positive or negative direction, and regardless
> of how far, but I'm by no means sure of that.  Thus, the present
> strtoul() is correct for an input string of "-1", but it appears
> to treat other negative constants as if they were first converted
> to a signed long and then to an unsigned long, which I think is
> incorrect.

After having had various things in the standard pointed out to me,
I am of two minds about this.  The standard does say that out of
range values should return ULONG_MAX and set errno to ERANGE,
which the present implementation does not do for negative values.

At any rate, I have posted something to c.l.c and to comp.std.c to
attempt to resolve this.

-- 
Chuck F (cbfalconer AT yahoo DOT com) (cbfalconer AT worldnet DOT att DOT net)
   Available for consulting/temporary embedded and systems.
   <http://cbfalconer.home.att.net>  USE worldnet address!