Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
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
Message-ID: <43094F28.70405@nc.rr.com>
Date: Mon, 22 Aug 2005 00:06:00 -0400
From: Joe Krahn <jkrahn AT nc DOT rr DOT com>
User-Agent: Mozilla Thunderbird 1.0.2 (Windows/20050317)
MIME-Version: 1.0
To: cygwin AT cygwin DOT com
Subject: Re: Path processing bug
References: <082120052231 DOT 12264 DOT 430900D300063BF300002FE822007601800A050E040D0C079D0A AT comcast DOT net>
In-Reply-To: <082120052231.12264.430900D300063BF300002FE822007601800A050E040D0C079D0A@comcast.net>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Eric Blake wrote:
>>Cygwin will accept the path "dir/../file" as being the same as "file", 
>>regardless of whether "dir" exists. Apprently, someone decided that a 
>>simple path-trimming rule would speed things up, but it is wrong. For 
>>example, it breaks building of xedit/lisp, where "lisp/../xedit.h" is 
>>not the same as "xedit.h".
>>
>>This occurs from bash and tcsh, so it must be in some low-level 
>>unix-to-win32 path name processing.
> 
> 
> I've raised the issue of this bug in the past, and the response
> was that fixing it would likely slow down the normal case.  I too
> would like to see it fixed, because it is contrary to POSIX.
> 
Well, it is obvious that supporting an invalid path is just plain wrong,
and not a matter of opinion. However, I just checked in a DOS prompt and
found that DOS/Windows allows removal of "<bogus_dir>/..". So, the real
bug apparently comes from MS (big surprise?)

I can see where path validation could be a significant slow-down, but
"dir/.." is not part of most path names, so doing it 'right' should have
a minimal performance impact.

Joe Krahn


--
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/