Date: Wed, 3 Jan 2001 23:12:58 +0000 From: Tony Finch <dot@dotat.at> To: Bruce Evans <bde@zeta.org.au> Cc: Matthew Jacob <mjacob@feral.com>, =?iso-8859-1?Q?G=E9rard_Roudier?= <groudier@club-internet.fr>, Peter Wemm <peter@netplex.com.au>, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/dev/isp isp.c Message-ID: <20010103231258.F6550@hand.dotat.at> In-Reply-To: <Pine.BSF.4.21.0101031929540.13047-100000@besplex.bde.org> References: <20010102133120.M47732@hand.dotat.at> <Pine.BSF.4.21.0101031929540.13047-100000@besplex.bde.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Bruce Evans <bde@zeta.org.au> wrote:
>
>Actually, literal strings are arrays of plain chars.
Gah!
>Their non-modifyability is specified explicitly since it doesn't
>follow from const'ness.
Why doesn't it follow? The C standard uses pretty much the same
language to say that string literals and const objects shouldn't be
modified ("behaviour is undefined"). [I'm looking at the C99 draft.]
>Enforcement of const'ness is also optional. The non-const'ness of the
>chars in literal strings gives implementors even more freedom. E.g.:
>
>---
>void foo(char *);
>char *bar;
>...
> foo(bar); /* Diagnostic required. */
> foo("string literal"); /* Diagnostic not required. */
>---
Was there supposed to be a const in there somewhere?
>Use `gcc -Wwrite-strings' if you don't want warnings for possibly
>modifying string literals.
Surely "... if you do want ..."?
Tony.
--
f.a.n.finch fanf@covalent.net dot@dotat.at
"Plan 9 deals with the resurrection of the dead."
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010103231258.F6550>