Mailing-List: contact new-httpd-help@apache.org; run by ezmlm
Precedence: bulk
Reply-To: new-httpd@apache.org
Errors-To: <wrowe@rowe-clan.net>
Message-ID: <015501c089be$ee99c3e0$93c0b0d0@wroweviao>
From: "William A. Rowe, Jr." <wrowe@rowe-clan.net>
To: <new-httpd@apache.org>
References: <20010129062142.63059.qmail@apache.org>
 <014901c089be$33324d20$93c0b0d0@wroweviao>
Subject: Re: cvs commit: apr/file_io/win32 filestat.c dir.c
Date: Mon, 29 Jan 2001 00:44:29 -0600
MIME-Version: 1.0
Content-Type: text/plain;
	charset="Windows-1252"
Content-Transfer-Encoding: 7bit

From: "William A. Rowe, Jr." <wrowe@rowe-clan.net>
Sent: Monday, January 29, 2001 12:39 AM


> I'm seriously asking myself if we need all three access info fields, however.
> Each one costs buku cpu, so if we can reduce from three to simply world, we would
> be in much better shape.  I'll be researching tommorow.  One option would be to
> set up some option in the MPM to disable groups, users, or specific protections
> from the server's processing, and that's my first likely target.  [Many Win32
> installations will never create the group or set up group permissions, although
> they can.  For most, these two are worthless.]

Tommorow is a big grep, by the way, for all the places we tested finfo->protection
as our guiding star that the file exists.  This is a -very- lousy test on any system 
that doesn't implement the one true unix security schema.  It doesn't even work on 
several unix-like implementations that can have another group or user permission 
granting access while the 'official' three perms are zero.

Bill