Received: (from majordom@localhost) by hyperreal.org (8.8.5/8.8.5) id EAA29869; Fri, 22 Aug 1997 04:13:14 -0700 (PDT) Received: from eastwood.aldigital.algroup.co.uk (eastwood.aldigital.algroup.co.uk [194.128.162.193]) by hyperreal.org (8.8.5/8.8.5) with SMTP id EAA29863 for ; Fri, 22 Aug 1997 04:13:09 -0700 (PDT) Received: from freeby.ben.algroup.co.uk (freeby.ben.algroup.co.uk [193.133.15.6]) by eastwood.aldigital.algroup.co.uk (8.6.12/8.6.12) with ESMTP id LAA26788 for ; Fri, 22 Aug 1997 11:12:54 GMT Received: from naughty (naughty.ben.algroup.co.uk [193.133.15.107]) by freeby.ben.algroup.co.uk (8.6.12/8.6.12) with ESMTP id MAA21030 for ; Fri, 22 Aug 1997 12:12:36 +0100 Message-ID: <33FD741A.66CB5A37@algroup.co.uk> Date: Fri, 22 Aug 1997 12:12:26 +0100 From: Ben Laurie Organization: A.L. Digital Ltd. X-Mailer: Mozilla 4.01 [en] (WinNT; I) MIME-Version: 1.0 To: new-httpd@apache.org Subject: Re: Spaces in Win32 executable names X-Priority: 3 (Normal) References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: new-httpd-owner@apache.org Precedence: bulk Reply-To: new-httpd@apache.org Paul Sutton wrote: > > On Thu, 21 Aug 1997, Ben Laurie wrote: > > Well, I finally got VC++ 5 to debug Apache (I had to set the working > > directory to ...\apachen\CoreD, or it complained that it couldn't find > > ApacheCore.dll [which is not what the manual sez]), to look into the > > spaces in executables thing. > > Heh, I can never get that to work in 4.2 or 5.0. Even if the full path to > the DLL was added to the "extra DLLs" section in the debug configuration. > > > BTW, the reason I was doing this at all was testing name > > canonicalisation on CGIs. The result is this (on a request of > > .../scripts/copy_o~1.bat): > > > > SCRIPT_FILENAME=c:/apachetest/scripts/copy_of_t1.bat > > REQUEST_URI=/scripts/copy_o~1.bat > > SCRIPT_NAME=/scripts/copy_o~1.bat > > I wonder if trying to expand filenames is really the right thing to do. > Couldn't we just convert all filenames to short form and *always* work > internally with the short form? So the first stage of parsing ...>, ScriptAlias ... etc is to munge the filename or path to a short > form. And when a request URI is transformed to a filename also convert it > to a short form. I think this is a bad idea, coz it requires people to write regexes for the short form. > I am getting worried about the amount of effort required to ensure that > things like security checks will work when there are multiple possible > pathnames for the same directory or resource. The effort required for this was fairly small, really (one extra function call, and a few tens of lines of code). It does need some cleanup, though. Cheers, Ben. -- Ben Laurie |Phone: +44 (181) 994 6435|Apache Group member Freelance Consultant |Fax: +44 (181) 994 6472|http://www.apache.org and Technical Director|Email: ben@algroup.co.uk |Apache-SSL author A.L. Digital Ltd, |http://www.algroup.co.uk/Apache-SSL London, England. |"Apache: TDG" http://www.ora.com/catalog/apache