Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 02 Nov 2001 17:36:04 +0900 (JST)
From:      NINOMIYA Hideyuki <nin@Jp.FreeBSD.org>
To:        ache@FreeBSD.org
Cc:        hackers@FreeBSD.org, current@FreeBSD.org
Subject:   About stscasestr() prototyped with string.h of current lib
Message-ID:  <20011102.173604.78790653.nin@smtp.shikoku.ne.jp>

next in thread | raw e-mail | index | archive | help
Hello

# please Cc: to me in reply time. because, i doesn't subscribe
# in to hackers and current.

There is a little problem in ports build on current, and I send an
email to you with its having been the function that you committed to
because the cause seemed to be lib function of current.

include/string.h cvslog:
---8<------8<------8<------8<------8<------8<------8<------8<------8<---
1.8 Wed Oct 10 2:19:09 JST 2001 UTC by ache
Implement strcasestr() which many others (f.e. Linux) already have
---8<------8<------8<------8<------8<------8<------8<------8<------8<---

Ports mail/mew2 has that problem, in current

ports build error message:
---8<------8<------8<------8<------8<------8<------8<------8<------8<---
mewls.c:128: conflicting types for `strcasestr'
/usr/include/string.h:86: previous declaration of `strcasestr'
---8<------8<------8<------8<------8<------8<------8<------8<------8<---

It is reported in ports-jp@jp.freebsd.org that there is the problem
that mewls.c which this ports includes don't build.
If Linux has the same problems in build of mew2, mew2 should solve it
as a problem of mew2. But he had the report that he had on to build
without a problem with Linux (gcc-2.95.4 or 3.0.2, glibc-2.2.4, PPC
etc...) in a mew-dist mailing list (This is a thing for Mew develop-
ment. http://www.mew.org/).

That report said, Linux's /usr/include/string.h
---8<------8<------8<------8<------8<------8<------8<------8<------8<---
#ifdef __USE_GNU
/* Similar to `strstr' but this function ignores the case of both strings.  */
extern char *strcasestr (__const char *__haystack, __const char *__needle)
     __THROW __attribute_pure__;
#endif
---8<------8<------8<------8<------8<------8<------8<------8<------8<---

This seems to be the prototyping that becomes effective when special
definitions of __USE_GNU is done. On this account a problem seems not
to have been caused by general program in that this prototyping was
not referred to, and mewls.c can build.
In implementation with current, even if you implemented it for the
reason that Linux included, there is the problem that behavior is
different from Linux in about prototyping reference.

I consider a method to solve this problem shuts up this function
in a cage of special definitions similar to __USE_GNU of Linux,
or remove it.
----
***************************************************************
* This message is a thing by output of automatic translation. *
* Therefore it will be for there to be a lot of funny parts.  *
* Please approve it.                                          *
***************************************************************
NINOMIYA(family name) Hideyuki(fast name) @ ehime japan
mailto:nin@shikoku.ne.jp mailto:nin@Jp.FreeBSD.org
PGP-Fingerprint:6C59 EC08 5B23 6490 44D0  7CD3 DA40 219F 7114 8553
PGP-Public-Key:  http://user.shikoku.ne.jp/~nin/pgp/public-key.txt

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20011102.173604.78790653.nin>