X-Recipient: archive-cygwin AT delorie DOT com
X-Spam-Check-By: sourceware.org
Date: Thu, 19 Mar 2009 21:30:46 +0100
From: Corinna Vinschen <corinna-cygwin AT cygwin DOT com>
To: cygwin AT cygwin DOT com
Subject: Re: Q: Is anybody here using the CYGWIN=codepage:oem setting?
Message-ID: <20090319203046.GF9322@calimero.vinschen.de>
Reply-To: cygwin AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
References: <20090319130909 DOT GZ9322 AT calimero DOT vinschen DOT de> <49C281F7 DOT 6080602 AT acm DOT org> <20090319181323 DOT GB1868 AT calimero DOT vinschen DOT de> <49C29366 DOT 8080708 AT acm DOT org> <20090319192031 DOT GB9322 AT calimero DOT vinschen DOT de> <20090319192229 DOT GC9322 AT calimero DOT vinschen DOT de> <loom DOT 20090319T193406-675 AT post DOT gmane DOT org> <20090319201144 DOT GE9322 AT calimero DOT vinschen DOT de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20090319201144.GE9322@calimero.vinschen.de>
User-Agent: Mutt/1.5.19 (2009-02-20)
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
Precedence: bulk
List-Id: <cygwin.cygwin.com>
List-Unsubscribe: <mailto:cygwin-unsubscribe-archive-cygwin=delorie DOT com AT cygwin DOT com>
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com

On Mar 19 21:11, Corinna Vinschen wrote:
> On Mar 19 19:41, Eric Blake wrote:
> > Corinna Vinschen <corinna-cygwin <at> cygwin.com> writes:
> > > ...unless Cygwin itself would call setlocale().
> > 
> > I'm not a fan of that.  POSIX is explicit that an application that 
> > intentionally avoids calling setlocale() shall behave as though it had called 
> > setlocale(LC_ALL,"C").
> > [...]
> But I admit that I'm not very happy with this idea either.  Still, we
> have to convert from MB to WC and vice-versa independently of the
> application, while other systems based on byte charsets simply don't
> have this problem.

Here's another idea:

If the codeset is not UTF-8, and if a filename contains wide chars not
representable in the current ANSI codeset, use the good old ASCII "SO/SI"
method.

Example:  Assuming the ANSI codepage is CP1252.  Assuming the filename
is in UTF-16

  /dir/to/foo\x1234bar
  
All chars except for \x1234 are convertible to the current ANSI code
page.  The convertible chars are converted as usual.  The
non-convertible characters are converted to an ASCII SO/SI sequence:

  /dir/to/foo\x0e\x12\x34\x0fbar

On the way back, Cygwin converts SO/SI sequences back to their
UTF-16 counterpart and converts everything else using the current\
codepage to UTF-16 conversion.

This would allow to manipulate all files on the disk regardless of
using characters invalid in the current CP.

Does that solution make sense?


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/