From -6095286585281613473 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: f78e5,e94695845965b249 X-Google-Attributes: gidf78e5,public X-Google-ArrivalTime: 1992-12-19 15:59:34 PST Newsgroups: comp.std.c++ Path: sparky!uunet!uunet.ca!frumious!pat From: pat@frumious.uucp (Patrick Smith) Subject: Re: Zero-length structures and pointer comparisons Message-ID: Date: Fri, 18 Dec 1992 07:16:44 GMT Reply-To: uunet.ca!frumious!pat References: <1992Dec11.230534.10499@microsoft.com> <1992Dec15.162202.11231@ucc.su.OZ.AU> <1992Dec17.150914.2783@lpi.liant.com> Organization: None Lines: 44 pkt@lpi.liant.com (Scott Turner) writes: |In article <1992Dec15.162202.11231@ucc.su.OZ.AU> John (Max) Skaller writes: |It makes me uneasy, but I think Jim Adcock is correct: an implementation |can define an extension, such as memory mapping, which overrides the usual |execution time rules of C/C++, including the rule that pointers to the same |object compare equal. The rule that 1+1==2 would likewise be subject to |"extension". For C, this is my impression also. Although I find the terminology unfortunate; it's unsettling to think that a conforming implementation could compile and run a conforming program - which could then compute 1+1 and get 3. Has X3J16 adopted similar notions of conformity? |Nevertheless, the C standard implies that the onus is on the invoker of memory |mapping to be aware of libraries and other code which may rely on the usual |execution time rules of C/C++, such as the rule that pointers to the same |object compare equal. If we don't want libraries to make that assumption, |then John (Max) Skaller is also correct to say | |> So the question is really whether the requirement that pointers to |> the same object compare equal should be left implementation defined. Here I would disagree. I think the question is whether, if the C++ standard adopts the same notions of conformity as the C standard, pointers to the same object should be required to compare equal in a conforming program. And the answer is _no_. I hope this requirement would exist for _strictly_ conforming programs. And it might be _reasonable_ to expect it to hold in conforming programs. But in a conforming program which depends on behaviour undefined or unspecified by the standard, none of the rules apply, so the standard doesn't actually _require_ anything. (It's not clear to me whether all rules disappear if a program fails to be strictly programming because it depends on implementation-defined behaviour, but it doesn't depend on any undefined or unspecified behaviour.) -- Patrick Smith uunet.ca!frumious!pat pat%frumious.uucp@uunet.ca