home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
NetNews Offline 2
/
NetNews Offline Volume 2.iso
/
news
/
comp
/
std
/
c
/
607
< prev
next >
Wrap
Internet Message Format
|
1996-08-06
|
2KB
Path: life.ai.mit.edu!mib
From: mib@gnu.ai.mit.edu (Michael I. Bushnell, p/BSG)
Newsgroups: gnu.misc.discuss,comp.std.c
Subject: Re: Coding Standards are ignorant
Date: 19 Mar 1996 15:54:55 GMT
Organization: Free Software Foundation, Cambridge, MA
Message-ID: <MIB.96Mar19105455@gnu.ai.mit.edu>
References: <4gum82$14v4@info4.rus.uni-stuttgart.de>
<MIB.96Mar16174948@gnu.ai.mit.edu> <4ifq40$i87@sundog.tiac.net>
<MIB.96Mar18105957@gnu.ai.mit.edu> <4ikh3o$2kv@sundog.tiac.net>
NNTP-Posting-Host: duality.ai.mit.edu
In-reply-to: stanr@tiac.net's message of 18 Mar 1996 20:29:44 GMT
In article <4ikh3o$2kv@sundog.tiac.net> stanr@tiac.net (Stan Ryckman) writes:
At the risk of re-starting the ages-old "long long" debate, I
can't find anything in the ANSI standard that permits this.
I don't have a Posix standard.
To the contrary, you have to find a statement that size_t is required
to be one of the specified list of types, or it can be anything that
integer operations work on correctly.
(All that's needed for printing whatever_t types would be confirmation
that they are no larger than unsigned long. I think that allowing
them to be larger would break ANSI even if "long long" were an
allowed extension.)
This implies that ANSI is inherently unbroken. The standard might
actually have a flaw, you know! In fact, that's exactly the *point*
here. Prohibiting systems from using more than three widths of
integers would be a flaw; allowing it but not solving the printf
problem is a flaw.
Michael