This is the mail archive of the
cygwin-patches@sourceware.cygnus.com
mailing list for the Cygwin project.
Re: preliminary patch for incorporating internationalizing facilities
- To: cygpatch <cygwin-patches at sourceware dot cygnus dot com>
- Subject: Re: preliminary patch for incorporating internationalizing facilities
- From: Kazuhiro Fujieda <fujieda at jaist dot ac dot jp>
- Date: 29 Jun 2000 06:50:57 +0900
- References: <s1sr99ho8cf.fsf@jaist.ac.jp> <395A676F.F78E67A6@cygnus.com>
>>> On Wed, 28 Jun 2000 23:00:31 +0200
>>> Corinna Vinschen <vinschen@cygnus.com> said:
> > /* The two paths were the same length, so just determine normal
> > lexical sorted order. */
> > - res = strcasecmp (ap->posix_path, bp->posix_path);
> > + res = strcmp (ap->native_path, bp->native_path);
>
> I don't understand that part. It seems to be a clear mistake that
> the posix paths are compared but why are you using `strcmp' now?
> What's about the case?
I believe this change causes no serious problem. It affects only
the case that there are two or more representations of the same
native path in the mount table.
strcasecmp is a locale-dependent function in ISO C. I don't want
to use it if it can be safely replaced with another function. I
judged strcasecmp can be safely replaced with strcmp in this case.
____
| AIST Kazuhiro Fujieda <fujieda@jaist.ac.jp>
| HOKURIKU School of Information Science
o_/ 1990 Japan Advanced Institute of Science and Technology