NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/lang/c-part2/13149

Text File | 1996-08-05 | 1.9 KB | 53 lines

Path: sun001.spd.dsccc.com!jmccarty
From: [e-mail] (Mike McCarty)
Newsgroups: comp.programming,comp.sys.sgi.misc,comp.lang.c
Subject: Re: C pointer question.
Date: 4 Apr 1996 22:53:38 GMT
Organization: DSC Communications Corporation
Message-ID: <4k1jti$[e-mail]>
References: <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]>
NNTP-Posting-Host: sun1307.spd.dsccc.com

In article <[e-mail]>,
Kazimir Kylheku <[e-mail]> wrote:
)In article <[e-mail]>,
)Max Waterman  <[e-mail]> wrote:
)>Ie what are the arguments for :
)>
)>char    name1,
)>    name2,
)>    name3;
)>
)>over :
)>
)>char name1;
)>char name2;
)>char name3;
)>
)>?
)>
)>Aren't they identical?
)>
)>Any compiler boffs out there know about this?
)
)These are identical. But if you put too many declarators onto one declaration
)specifier string, an ANSI-compliant compiler may fail, because the standard
)says that a compiler is only required to process declarations that have at
)least 12 (AFAIR) declarators (if I interpreted the confusing paragraph
)correctly: I'm sure this article will be followed up with some clarifications
)and/or flames). This might be intended as a relief to parsers that build n-ary
)trees, where the number of children that a parse tree node can have is a fixed
)array contained in the node data structure:

They are not guaranteed to be identical. I haven't seen anything in the
standard that says that the same program must be produced by these
differing sources. I have, in fact, used compilers which -did- produce
different programs depending on such declarations.

Mike
--
----
char *p="char *p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);}

I don't speak for DSC.         <- They make me say that.