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
Date: Sun, 2 Jan 2005 20:21:29 -0500 (EST)
From: Igor Pechtchanski <pechtcha AT cs DOT nyu DOT edu>
Reply-To: cygwin AT cygwin DOT com
To: Yitzchak Scott-Thoennes <sthoenna AT efn DOT org>
cc: cygwin AT cygwin DOT com
Subject: Re: check_case:adjust still an option?
In-Reply-To: <20050103011210.GA2112@efn.org>
Message-ID: <Pine.GSO.4.61.0501022019400.2820@slinky.cs.nyu.edu>
References: <20050102203221 DOT GB344 AT efn DOT org> <20050102213301 DOT GA552397 AT hpn5170> <20050102221844 DOT GA608 AT efn DOT org> <20050102231344 DOT GA35436409 AT hpn5170> <20050103011210 DOT GA2112 AT efn DOT org>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Sun, 2 Jan 2005, Yitzchak Scott-Thoennes wrote:

> On Sun, Jan 02, 2005 at 06:13:44PM -0500, "Pierre A. Humblet" wrote:
> > On Sun, Jan 02, 2005 at 02:18:44PM -0800, Yitzchak Scott-Thoennes wrote:
> > > On Sun, Jan 02, 2005 at 04:33:01PM -0500, "Pierre A. Humblet" wrote:
> > > > On Sun, Jan 02, 2005 at 12:32:21PM -0800, Yitzchak Scott-Thoennes wrote:
> > > > > /usr/share/doc/cygwin-doc-1.3/html/cygwin-ug-net/using-cygwinenv.html says:
> > > > >
> > > > >    check_case:level - Controls the behaviour of Cygwin when a user
> > > > > tries to open or create a file using a case different from the case of
> > > > > the path as asved on the disk. level is one of relaxed, adjust and
> > > > > strict.
> > > > >    ...
> > > > >    adjust behaves mostly invisible. The POSIX input path is internally
> > > > > adjusted in case, so that the resulting DOS path uses the correct case
> > > > > throughout. You can see the result when changing the directory using a
> > > > > wrong case and calling /bin/pwd afterwards.
> > > > >
> > > > > but I'm not seeing any difference between adjust and relaxed; is adjust
> > > > > still a valid option?
> > > >
> > > > Sure, but not as visibly as documented above. For example under sh
> > > > $: cd c:/HoME/PiErRe
> > > > $: /bin/pwd
> > > > /c/HoME/Pierre
> > > >
> > > > I has a look at the code and it is the Windows path that is adjusted,
> > > > not the Posix path. I think it was always like that.
> > > > In the case of chdir, the change in the Posix path was a result of former
> > > > transformations of the Windows path back to Posix form. Those can cause
> > > > trouble when several Posix paths point to the same Windows path (the mapping
> > > > Posix -- Windows is not bijective). In particular the final text/binary mode
> > > > need not be as expected, and this caused complaints.
> > >
> > > So you can expect it to "work" only outside of any mount point?
> >
> > It always "works" but it's hard to observe. I don't really care
> > about it and never tried to really make sense of it.
>
> I meant "work" as in have any user-visible effect at all.
>
> > > It doesn't seem to do this for me (using 20050101 snapshot):
> > >
> > > $ CYGWIN="$CYGWIN check_case:adjust " bash
> > > $ echo $CYGWIN
> > > tty ntsec title server check_case:adjust
> > > $ cd c:/SeTuP/CyGwIn
> > > $ /bin/pwd
> > > /cygdrive/c/SeTuP/CyGwIn
> > > $ ls -d c:/?e?u?/?y?w?n
> > > c:/setup/cygwin
> >
> > That's because you use bash. There are lots of Cygwin patches
> > in bash to transform paths (again, I never tried to understand
> > why) , so it's likely that the Cygwin chdir does not even see
> > what you give to bash, e.g.
> >
> > /c/home/PiERRe: strace bash -c 'cd c:/hOmE/PiErRe' | fgrep chdir
> >   351  524054 [main] bash 542197 chdir: dir '/c/hOmE/PiErRe'      <== NO MORE c:
> >   270  528790 [main] bash 542197 chdir: 0 = chdir() cygheap->cwd.posix '/c/hOmE/PiErRe' native 'c:\hOmE\Pierre'
> >
> > But as you see toward the end, the native path was indeed fixed!
> > Because chdir gets a Posix path from bash, it's happy and it does not
> > use the Windows path to get a Posix path (although it used to, which broke
> > various more important things).
>
> Thanks for the explanation.  A test program that does chdir and
> system("pwd") does show it having an effect (when chdir is given a
> windows path only).
>
> Seems kind of pointless, really.

It is pointless.  As I understood the documentation previously, it's the
POSIX path that should have been adjusted.  As is, though, it's useless.
FWIW, I have nothing against removing the check_case:adjust logic, as long
as the check_case:strict one is left in-place.  Unfortunately, most of the
check_case logic is associated with check_case:strict.
	Igor
-- 
				http://cs.nyu.edu/~pechtcha/
      |\      _,,,---,,_		pechtcha AT cs DOT nyu DOT edu
ZZZzz /,`.-'`'    -.  ;-;;,_		igor AT watson DOT ibm DOT com
     |,4-  ) )-,_. ,\ (  `'-'		Igor Pechtchanski, Ph.D.
    '---''(_/--'  `-'\_) fL	a.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

"The Sun will pass between the Earth and the Moon tonight for a total
Lunar eclipse..." -- WCBS Radio Newsbrief, Oct 27 2004, 12:01 pm EDT

--
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/