home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
NetNews Offline 2
/
NetNews Offline Volume 2.iso
/
news
/
comp
/
std
/
c
/
494
< prev
next >
Wrap
Internet Message Format
|
1996-08-06
|
3KB
Path: solon.com!not-for-mail
From: seebs@solutions.solon.com (Peter Seebach)
Newsgroups: gnu.misc.discuss,comp.std.c
Subject: Re: Coding Standards are ignorant
Date: 4 Mar 1996 10:17:22 -0600
Organization: Usenet Fact Police (Undercover)
Message-ID: <4hf52i$15m@solutions.solon.com>
References: <4gum82$14v4@info4.rus.uni-stuttgart.de> <1996Mar403.23.06.8316@koobera.math.uic.edu> <4he37i$a0u@solutions.solon.com> <1996Mar408.48.01.11162@koobera.math.uic.edu>
NNTP-Posting-Host: solutions.solon.com
In article <1996Mar408.48.01.11162@koobera.math.uic.edu>,
D. J. Bernstein <djb@koobera.math.uic.edu> wrote:
>Peter Seebach <seebs@solutions.solon.com> wrote:
>> D. J. Bernstein <djb@koobera.math.uic.edu> wrote:
>> > Seebach has been making a bunch of false statements on gnu.misc.discuss
>> > about portability, speed, and various other things.
>> Example false statements:
>``There are no interesting programs you can't write entirely within
>POSIX and ANSI.''
(I didn't realize POSIX sockets don't exist yet. :) )
>``[ispunct()] has no excuse to assume that char is 8 bits.'' (Of course
>it has an excuse. Whether it's a _good_ excuse is a side issue.)
(Mostly, my point is that ispunct is a library function, and may assume
whatever it pleases. A user-land equivalent, I believe, has no call to
assume char is 8 bits.)
>``It will hurt because it's incorrect by the only standard applicable.''
>(False hypothesis, false logic, false conclusion.)
I don't see any of these; code which is not strictly conforming, or even
just code which invokes undefined behavior, is worse off than code that
doesn't.
>``void main [has] NO BENEFIT WHATSOEVER.''
>``If declaring main as void really shuts gcc up, there is a bug in
>either gcc or your code.''
Both of these I stand by, and I consider them to be error-free.
>Referring to isalnum || strchr: ``... this version is likely to be
>within 5% of your version, because isalnum is normally blindingly fast,
>and most characters in addresses will be isalnum.''
Blatantly false, except that it might be very close on a machine with
a different implementation. Becomes true if you include the rest of the
code.
>Referring to RFC 822: ``... '.' must also be quoted.'' (No, not always.)
So I just skimmed the grammar, and saw a list of characters marked "must
be quoted". Followups on this one to comp.mail.headers.
-s
--
Peter Seebach - seebs@solon.com - Copyright 1996 Peter Seebach.
C/Unix wizard -- C/Unix questions? Send mail for help. No, really!
FUCK the communications decency act. Goddamned government. [literally.]
The *other* C FAQ - http://www.solon.com/~seebs/c/c-iaq.html