Message-ID: <435FEEEC.4080901@route-add.net>
Date: Wed, 26 Oct 2005 23:02:36 +0200
From: Alessandro Selli <dhatarattha@route-add.net>
User-Agent: Mozilla/5.0 (X11; U; Linux i586; en-US; rv:1.7.12) Gecko/20051014
MIME-Version: 1.0
To: gopher@complete.org
Subject: [gopher] Illegal or simply strange characters in gophermap
Content-type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 8bit
Sender: gopher-bounce@complete.org
Errors-to: gopher-bounce@complete.org
Precedence: bulk
Reply-to: gopher@complete.org

  I recently developed a script to automate the conversion and copying
of files from an IMAPD server's directory where I store Usenet messages
and a tree of directories where these files are fetched to be served by
the Gopher server.  The output filenames are the same as the Subject:
line in the Usenet messages, plus a numerical index if more like
Subject: lines are present.  The script also generates or updates the
gophermap file in each directory.
  Well, I found out I cannot access files whose name contains a
question mark, if i try I get this error message (bucktooth-0.1pr6):

0'/Utenti/Alessandro/Usenet/it.comp.os.linux.sys/Re: Server Alpha,
distro ' doesn't exist!		error.host	1
.

  The filename is: "Re: Server Alpha, distro ?" and the corresponding
gophermap line is:

0Re: Server Alpha, distro ?     Re: Server Alpha, distro ?

This is a nuisance considering the many subject: lines that are
encoded in RFC-1522 MIME format, such as:

=?ISO-8859-1?Q?Re=3A_=5BOT=5D_Linux_kernel_FALLATO=3A_?=

  Is this the right way a Gopher server is supposed to handle such
filenames?

  I'm considering editing the script to convert all question marks
into something else; are there any more characters that I should
catch before creating the files and updating the gophermap?

-- 
Alessandro Selli
Tel: 340.839.73.05
http://alessandro.route-add.net