/usr/share/doc/pygopherd/pygopherd.html

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<HTML
><HEAD
><TITLE
>PyGopherd Manual</TITLE
><META
NAME="GENERATOR"
CONTENT="Modular DocBook HTML Stylesheet Version 1.7"><STYLE
TYPE="text/css"
>.synopsis, .classsynopsis {
    background: #eeeeee;
    border: solid 1px #aaaaaa;
    padding: 0.5em;
}
.programlisting {
    background: #eeeeff;
    border: solid 1px #aaaaff;
    padding: 0.5em;
}
.variablelist {
    padding: 4px;
    margin-left: 3em;
}
.navigation {
    background: #ffeeee;
    border: solid 1px #ffaaaa;
    margin-top: 0.5em;
    margin-bottom: 0.5em;
}
.navigation a {
    color: #770000;
}
.navigation a:visited {
    color: #550000;
}
.navigation .title {
    font-size: 200%;
}</STYLE
></HEAD
><BODY
CLASS="REFERENCE"
BGCOLOR="#FFFFFF"
TEXT="#000000"
LINK="#0000FF"
VLINK="#840084"
ALINK="#0000FF"
><DIV
CLASS="REFERENCE"
><A
NAME="PYGOPHERD"
></A
><DIV
CLASS="TITLEPAGE"
><H1
CLASS="TITLE"
>I. PyGopherd Manual</H1
><DIV
CLASS="TOC"
><DL
><DT
><B
>Table of Contents</B
></DT
><DT
><A
HREF="pygopherd.html#AEN3"
>pygopherd</A
>&nbsp;--&nbsp;Multiprotocol Information Server</DT
></DL
></DIV
></DIV
><H1
><A
NAME="AEN3"
></A
>pygopherd</H1
><DIV
CLASS="REFNAMEDIV"
><A
NAME="AEN15"
></A
><H2
>Name</H2
>PyGopherd&nbsp;--&nbsp;Multiprotocol Information Server</DIV
><DIV
CLASS="REFSYNOPSISDIV"
><A
NAME="AEN18"
></A
><H2
>Synopsis</H2
><P
><B
CLASS="COMMAND"
>pygopherd</B
>  [<VAR
CLASS="REPLACEABLE"
>configfile</VAR
>]</P
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="DESCRIPTION"
></A
><H2
>Description</H2
><P
>	Welcome to <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>.  In a nutshell, <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>
	is a modern dynamic
	multi-protocol hierarchical information server with a pluggable
	modularized extension system, 
	full flexible caching, virtual files and
	folders, and autodetection of file types -- all with support for
	standardized yet extensible per-document metadata. Whew! Read on for
	information on this what all these buzzwords mean.
      </P
><DIV
CLASS="REFSECT2"
><A
NAME="DESCRIPTION.FEATURES"
></A
><H3
>Features</H3
><P
>	  Here are some of <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>'s features:
	</P
><P
></P
><UL
><LI
><P
>Provides built-in support for multiple protocols:
	      HTTP (Web), Gopher+, Gopher (RFC1436), Enhanced Gopher0,
	      and WAP (mobile phones).  Protocols can be enabled or
	      disabled as desired.
	    </P
></LI
><LI
><P
>Provides protocol autodetection.  That is,
	      <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> can listen for all the above protocols
	      <SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>on a single port</I
></SPAN
> and will
	      automatically respond using the protocol it detects the
	      client is using.  Practical effects of this are that you
	      can, for instance, give out a single URL and have it
	      viewable normally on desktop Web browsers and in WAP
	      mode on mobile phones -- and appropriately in various
	      Gopher browsers.
	    </P
></LI
><LI
><P
>Metadata and site links can be entered in a
	      variety of formats, including full UMN dotfile metadata
	      formats as well as Bucktooth gophermap files.  Moreover,
	      gophermap files are not limited to Gopher protocols, and
	      can be used for all protocols.
	    </P
></LI
><LI
><P
>Support for inter-protocol linking (linking
	  from Gopher sites to web sites)</P
></LI
><LI
><P
>Virtual folder system lets you serve up
	  anything as if it were regular files and directories.
	  PyGopherd comes with the following virtual folder systems
	      built in:</P
><P
></P
><UL
><LI
><P
>Can present any Unix MBOX, MMDF box, MH
		  directory, Maildir directory, or Babyl mailbox as a
		  virtual folder, the contents of which are the
		  messages in the mailbox.
		</P
></LI
><LI
><P
>Can use a configurable separator to
		  split a file into multiple parts, the first line of each
		  becoming the name for the virtual folder.</P
></LI
><LI
><P
>Can peek inside a ZIP file and serve it
	      up as first-class site citizens -- metadata can even be
	      stored in the ZIP files.
		</P
></LI
><LI
><P
>Can serve up the contents of a dictd
	      server as a filesystem.
		</P
></LI
></UL
></LI
><LI
><P
>	      Modular, extensible design: you can use PyGopherd's own
	      PYG extension format, or UMN- or Bucktooth-style
	      executables.
	    </P
></LI
><LI
><P
>	      Runs on any platform supported by Python 2.2 or 2.3.
	      This includes virtually every past and current flavor of
	      Unix (Linux, *BSD, Solaris, SunOS), Windows, MacOS 9.x
	      and X, and more.  Some features may not be available on
	      non-Unix platforms.
	    </P
></LI
><LI
><P
>Runs on any platform supported by Java 1.1
	  via the Jython Python implementation.</P
></LI
><LI
><P
>Tunable server types via configuration
	  directive -- forking or threading.</P
></LI
><LI
><P
>Secure design with support for chrooted execution.</P
></LI
><LI
><P
>Feature-complete, full implementations of:
	  Gopher0 (RFC1435), Gopher+, HTTP, and WAP.</P
></LI
><LI
><P
>Support for automatically finding the titles
	  of HTML documents for presentation in a directory.</P
></LI
><LI
><P
>Versatile configuration file format is both
	  extensible and nicely complementary of the module system.</P
></LI
><LI
><P
>Protocol-independant, handler-dependant
	  caching.  This increases performance by letting handlers
	  cache dynamically-generated information -- currently used by
	  the directory handlers.  This can improve performance of
	  directories by several orders of magnitude.  Because this is
	  a handler cache only, all protococls share the single
	  cache.  Since the processing time for the protocols is
	  negligable, this works out very well.</P
></LI
><LI
><P
>Autosensing of MIME types and gopher0 item
	  types.  Both are completely configurable.  MIME type
	  detection is done using a standard mime.types file, and
	  gopher0 types are calculated by using a configurable
	  regexp-based MIME-to-gophertype map.</P
></LI
><LI
><P
>Heavy support of regular expressions in configuration.</P
></LI
><LI
><P
>ProtocolMultiplexer and HandlerMultiplexer
	  let you choose only those protocols and handlers that you
	  wish your server to support and the order in which they are
	  tried when a request comes in.
	    </P
></LI
><LI
><P
>Full logging via syslog.</P
></LI
></UL
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="DESCRIPTION.ABOUTGOPHER"
></A
><H3
>About Gopher</H3
><P
>	  <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> started life as a server for the Gopher Internet
	  protocol. With Gopher, you can mount a filesystem (viewing files and
	  folders as if they were local),
	  browse Gopherspace with a web browser,
	  download files, and be interactive with searching.
	</P
><P
>	  But this is only part of the story. The world of Gopher is more
	  expansive than this. There are two major gopher protocols: Gopher0
	  (also known as RFC1436) and Gopher+. Gopher0 is a small, simple,
	  lightweight protocol that is very functional yet also extremely easy
	  to implement. Gopher0 clients can be easily places in small embedded
	  devices or in massive environments like a modern web browser.
	</P
><P
>	  Gopher+ is based on Gopher0 but extends it by providing document
	  metadata such as file size and MIME type. Gopher+ allows all sorts of
	  neat features, such as configurable metadata (serving up a bunch of
	  photos? Add a Subject field to your metadata to let
	  a customized photo
	  browser display who is pictured) and multiple
	  views of a file (let the
	  user select to view your photos as PNG or JPEG).
	</P
></DIV
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="QUICKSTART"
></A
><H2
>Quick Start</H2
><P
>	If you have already installed <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> system-wide, or your
	administrator has done that for you, your task for setting up
	<SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> for the first time is quite simple.  You just need
	to set up your configuration file, make your folder directory,
	and run it!
      </P
><P
>	You can quickly set up your configuration file.  The
	distribution includes two files of interest:
	<TT
CLASS="FILENAME"
>conf/pygopherd.conf</TT
> and
	<TT
CLASS="FILENAME"
>conf/mime.types</TT
>.  Debian users will find
	the configuration file pre-installed in
	<TT
CLASS="FILENAME"
>/etc/pygopherd/pygopherd.conf</TT
> and the
	<TT
CLASS="FILENAME"
>mime.types</TT
> file provided by the system
	already.
      </P
><P
>	Open up <TT
CLASS="FILENAME"
>pygopherd.conf</TT
> in your editor and
	adjust to suit.  The file is heavily commented and you can
	refer to it for detailed information.  Some settings to take a
	look at include: <SPAN
CLASS="PROPERTY"
>detach</SPAN
>,
	<SPAN
CLASS="PROPERTY"
>pidfile</SPAN
>, <SPAN
CLASS="PROPERTY"
>port</SPAN
>, 
	<SPAN
CLASS="PROPERTY"
>usechroot</SPAN
>, <SPAN
CLASS="PROPERTY"
>setuid</SPAN
>,
	<SPAN
CLASS="PROPERTY"
>setgid</SPAN
>, and <SPAN
CLASS="PROPERTY"
>root</SPAN
>.
	These may or may not work at their defaults for you.  The
	remaining ones should be fine for a basic setup.
      </P
><P
>	Invoke <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> with <B
CLASS="COMMAND"
>pygopherd
	  path/to/configfile</B
> (or
	<B
CLASS="COMMAND"
>/etc/init.d/pygopherd start</B
> on Debian).
	Place some files in the location specified by the
	<SPAN
CLASS="PROPERTY"
>root</SPAN
> directive in the config file and
	you're ready to run!
      </P
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="INSTALLATION"
></A
><H2
>Installation</H2
><P
>	If  you  are  reading this document via the "man" command, it is likely
	that you have no installation tasks to perform; your system administra-
	tor  has already installed <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>.  If you need to install it yourself, you
	have three options: a system-wide installation with Debian, system-wide
	installation  with  other systems, and a single-user installation.  You
	can    download    the    latest    version    of    PyGopherd
	from
	<A
HREF="http://quux.org/devel/gopher/pygopherd/"
TARGET="_top"
>http://quux.org/devel/gopher/pygopherd/</A
>
      </P
><DIV
CLASS="REFSECT2"
><A
NAME="INSTALLATION.DEBIAN"
></A
><H3
>Debian System-Wide Installation</H3
><P
>	  If you are tracking Debian unstable, you may install
	  <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> by simply running this command as root:
	</P
><P
>	  <B
CLASS="COMMAND"
>apt-get install pygopherd</B
>
	</P
><P
>	  If you are not tracking Debian unstable, download the .deb
	  package from the <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> website and then run
	  <B
CLASS="COMMAND"
>dpkg -i</B
> to install the downloaded
	  package.  Then, skip to the configuration section below.
	  You will use <B
CLASS="COMMAND"
>/etc/init.d/pygopherd start</B
>
	  to start the program.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="INSTALLATION.OTHER"
></A
><H3
>Other System-Wide Installation</H3
><P
>	  Download the tar.gz version of the package from the website.  Make
	  sure you have Python 2.2 or above installed; if now, download and
	  install it from <A
HREF="http://www.python.org/"
TARGET="_top"
>http://www.python.org/</A
>.  Then run these
	  commands:
	</P
><PRE
CLASS="PROGRAMLISTING"
>	  <B
CLASS="COMMAND"
>tar -zxvf pygopherd-x.y.z.tar.gz</B
>
	  <B
CLASS="COMMAND"
>cd pygopherd-x.y.z</B
>
	  <B
CLASS="COMMAND"
>python2.2 setup.py</B
>
	</PRE
><P
>	  Some systems will use <B
CLASS="COMMAND"
>python</B
> or
	  <B
CLASS="COMMAND"
>python2.3</B
> in place of
	  <B
CLASS="COMMAND"
>python2.2</B
>.
	</P
><P
>	  Next, proceed to configuration.  Make sure that the
	  <TT
CLASS="FILENAME"
>/etc/pygopherd/pygopherd.conf</TT
> file
	  names valid users (<SPAN
CLASS="PROPERTY"
>setuid</SPAN
> and
	  <SPAN
CLASS="PROPERTY"
>setgid</SPAN
> options) and a valid document
	  root (<SPAN
CLASS="PROPERTY"
>root</SPAN
> option).
	</P
><P
>	  You will type <TT
CLASS="FILENAME"
>pygopherd</TT
> to invoke the
	  program.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="INSTALLATION.SINGLE"
></A
><H3
>Single-Account Installation</H3
><P
>	  Download the tar.gz version of the package from the website.  Make
	  sure you have Python 2.2 or above installed; if now, download and
	  install it from <A
HREF="http://www.python.org/"
TARGET="_top"
>http://www.python.org/</A
>.  Then run these
	  commands:
	</P
><PRE
CLASS="PROGRAMLISTING"
>	  <B
CLASS="COMMAND"
>tar -zxvf pygopherd-z.y.z.tar.gz</B
>
	  <B
CLASS="COMMAND"
>cd pygopherd-x.y.z</B
>
	</PRE
><P
>	  Modify <TT
CLASS="FILENAME"
>conf/pygopherd.conf</TT
> as follows:
	</P
><P
></P
><UL
><LI
><P
>Set <SPAN
CLASS="PROPERTY"
>usechroot = no</SPAN
></P
></LI
><LI
><P
>Comment out (add a # sign to the start of
	      the line) the <SPAN
CLASS="PROPERTY"
>pidfile</SPAN
>,
	      <SPAN
CLASS="PROPERTY"
>setuid</SPAN
>, and
	      <SPAN
CLASS="PROPERTY"
>setgid</SPAN
> lines.</P
></LI
><LI
><P
>Set <SPAN
CLASS="PROPERTY"
>root</SPAN
> to osomething appropriate.</P
></LI
><LI
><P
>Set <SPAN
CLASS="PROPERTY"
>port</SPAN
> to a number
	      greater than 1024.</P
></LI
></UL
><P
>	  When you want to run <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>, you will issue the
	  <B
CLASS="COMMAND"
>cd</B
> command as above and then type
	  <B
CLASS="COMMAND"
>PYTHONPATH=. bin/pygopherd</B
>.  There is no
	  installation step necessary.
	</P
></DIV
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="CONFIGURATION"
></A
><H2
>Configuration</H2
><P
>	<SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> is regulated by a configuratoin file normally
	stored in <TT
CLASS="FILENAME"
>/etc/pygopherd/pygopherd.conf</TT
>.
	You can specify an alternate configuration file on the command
	line.  The <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> distribution ships
	with a sample <TT
CLASS="FILENAME"
>pygopherd.conf</TT
> file that
	thoroughly documents the configuration file options and
	settings.
      </P
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="OPTIONS"
></A
><H2
>Options</H2
><P
>	All <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> configuratoin is done via the configuration
	file.  Therefore, the program has only one command-line
	option:
      </P
><P
></P
><TABLE
CLASS="variablelist"
BORDER="0"
CELLSPACING="0"
CELLPADDING="4"
><TBODY
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN184"><SPAN
STYLE="white-space: nowrap"
><VAR
CLASS="REPLACEABLE"
>configfile</VAR
></SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>This option argument specifies the location
	      of the configuration file that <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> is to use.</P
></TD
></TR
></TBODY
></TABLE
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="HANDLERS"
></A
><H2
>Handlers</H2
><P
>	<SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> defines several handlers which are responsible for
	finding data on your server and presenting it to the user.  The
	handlers are used to generate things like links to other documents and
	directory listings.  They are also responsible for serving up regular
	files and even virtual folders.
      </P
><P
>	Handlers are specified with the <SPAN
CLASS="PROPERTY"
>handlers</SPAN
>
	option in <TT
CLASS="FILENAME"
>pygopherd.conf</TT
>.  This option is
	a list of handlers to use.  For each request that arrives,
	<SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> will ask each handler in
	turn whether or not it can handle the request, and will handle the
	request according to the first handler that is capable of doing so.
	If no handlers can handle the request, a file not found error is
	generated.  See the example configuration file for an example.
      </P
><P
>	The remaining parts of this section describe the different
	handlers that ship with <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>.  Please note that some
	versions of this manual may show the handlers in all caps;
	however, their names are not all caps and are case-sensitive.
      </P
><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.DIRHANDLER"
></A
><H3
>dir.DirHandler</H3
><P
>	  This handler is a basic one that generates menus based
	  on the contents of a directory.  It is used for
	  directories that contain neither a
	  <TT
CLASS="FILENAME"
>gophermap</TT
> file nor UMN-style links
	  files, or situations where you have no need for either
	  of those.
	</P
><P
>	  This handler simply reads the contents of your on-disk
	  directory, determines the appropriate types of each file,
	  and sends the result to the client.  The descriptions of
	  each item are usually set to the filename, but the
	  <SPAN
CLASS="PROPERTY"
>html.HTMLFileTitleHandler</SPAN
> may override
	  that.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.BUCKGOPHERMAP"
></A
><H3
>gophermap.BuckGophermapHandler</H3
><P
>	  This handler is used to generate directory listings
	  based on <TT
CLASS="FILENAME"
>gophermap</TT
> files.  It will
	  not read the directory on-disk, instead serving content
	  from the <TT
CLASS="FILENAME"
>gophermap</TT
> file only.
	  Gophermaps are useful if you want to present a directory
	  in which the files do not frequently change and there is
	  general information to present.  Overall, if you only
	  wish to present information particular to certain files,
	  you should consider using the abstract feature of
	  UMN.UMNDirHandler.
	</P
><P
>	  The <TT
CLASS="FILENAME"
>gophermap</TT
> files contain two
	  types of lines, which are described here using the same
	  convention normally used for command line arguments.  In
	  this section, the symbol \t will be used to indicate a
	  tab character, Control-I.
	</P
><P
> <VAR
CLASS="REPLACEABLE"
>full line of informational
	      text</VAR
> </P
><P
> <VAR
CLASS="REPLACEABLE"
>gophertypeDESCRIPTION</VAR
>  [\t<VAR
CLASS="REPLACEABLE"
>selector</VAR
> [\t<VAR
CLASS="REPLACEABLE"
>host</VAR
> [\t<VAR
CLASS="REPLACEABLE"
>port</VAR
>]]]</P
><P
>	  Note: spaces shown above are for clarity only and should
	  not actually be present in your file.
	</P
><P
>	  The informational text must not contain any tab
	  characters, but may contain spaces.  Informational text
	  will be rendered with gopher type
	  <SPAN
CLASS="PROPERTY"
>i</SPAN
>, which will cause it to be
	  displayed on a client's screen at its particular
	  position in the file.
	</P
><P
>	  The second type of line represents a link to a file or
	  directory.  It begins with a single-character Gopher
	  type (see Gopher Item Types below) followed immediately
	  by a description and a tab character.  There is no space
	  or other separator between the gopher type and the
	  description.  The description may contain spaces but not
	  tabs.
	</P
><P
>	  The remaining arguments are optional, but only to the
	  extent that arguments may be omitted only if all
	  arguments after them are also omitted.  These arguments
	  are:
	</P
><P
></P
><TABLE
CLASS="variablelist"
BORDER="0"
CELLSPACING="0"
CELLPADDING="4"
><TBODY
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN231"><SPAN
STYLE="white-space: nowrap"
><VAR
CLASS="REPLACEABLE"
>selector</VAR
></SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The <SPAN
CLASS="PROPERTY"
>selector</SPAN
> is
		the name of the file on the server.  If it begins
		with a slash, it is an absolute path; otherwise,
		it is interpreted relative to the directory that
		the gophermap file is in.  If no selector is
		specified, the description is also used as the
		selector.
	      </P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN237"><SPAN
STYLE="white-space: nowrap"
><VAR
CLASS="REPLACEABLE"
>host</VAR
></SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The <SPAN
CLASS="PROPERTY"
>host</SPAN
>
		specifies the host on which this resource is
		located.  If not specified, defaults to the
		current server.
	      </P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN243"><SPAN
STYLE="white-space: nowrap"
><VAR
CLASS="REPLACEABLE"
>port</VAR
></SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The <SPAN
CLASS="PROPERTY"
>port</SPAN
>
		specifies the port on which the resource is
		located.  If not specified, defaults to the port
		the current server is listening on.
	      </P
></TD
></TR
></TBODY
></TABLE
><P
>	  An example of a gophermap to help illustrate the concept
	  is included with the <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> distribution in the
	  file <TT
CLASS="FILENAME"
>examples/gophermap</TT
>.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.COMPRESSEDFILEHANDLER"
></A
><H3
>file.CompressedFileHandler</H3
><P
>	  In order to save space, you might want to store
	  documents on-disk in a compressed format.  But then
	  clients would ordinarily have to decompress the files
	  themselves.  It would be nice to have the server
	  automatically decompress the files on the fly, sending
	  that result to the client.  That's where
	  <SPAN
CLASS="PROPERTY"
>file.CompressedFileHandler</SPAN
> comes
	  in.
	</P
><P
>	  This handler will take compressed files, pipe them
	  through your chosen decompression program, and send the
	  result directly to clients -- completely transparently.
	</P
><P
>	  To use this handler, set the
	  <SPAN
CLASS="PROPERTY"
>decompressors</SPAN
> option in the
	  configuration file.  This option defines a mapping from
	  MIME encodings (as defined with the
	  <SPAN
CLASS="PROPERTY"
>encoding</SPAN
> option) to decompression
	  programs.  Files that are not encoded, or which have an
	  encoding that does not occur in the
	  <SPAN
CLASS="PROPERTY"
>decompressors</SPAN
> map, will not be
	  decompressed by this handler.
	</P
><P
>	  Please see the sample configuration file for more
	  examples and details about the configuration of this
	  handler.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.FILEHANDLER"
></A
><H3
>file.FileHandler</H3
><P
>	  The <SPAN
CLASS="PROPERTY"
>file.FileHandler</SPAN
> is just that
	  -- its duty is to serve up regular files to clients.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.HTMLFILETITLEHANDLER"
></A
><H3
>html.HTMLFileTitleHandler</H3
><P
>	  This handler is used when generating directories and
	  will set the description of HTML files to the HTML title
	  defined in them rather than let it be the default
	  filename.  Other than that, it has no effect.  UMN
	  gopherd implements a similar policy.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.MAILBOXES"
></A
><H3
>mbox handlers</H3
><P
>	  There are four mailbox handlers:
	</P
><P
></P
><UL
><LI
><P
>mbox.MaildirFolderHandler</P
></LI
><LI
><P
>mbox.MaildirMessageHandler</P
></LI
><LI
><P
>mbox.MBoxMessageHandler</P
></LI
><LI
><P
>mbox.MBoxFolderHandler</P
></LI
></UL
><P
>	  These four handlers provide a unique "virtual folder"
	  service.  They allow you to present mailboxes as if they
	  were folders, the items of the folders being the
	  messages in the mailbox, organized by subject.  This is
	  useful for presenting mail archives or just making
	  e-mail accessible in a nice and easy fashion.
	</P
><P
>	  To use these handlers, all you have to do is enable them
	  in your <SPAN
CLASS="PROPERTY"
>handlers</SPAN
> section.  They
	  will automatically detect requests for mailboxes and
	  handle them appropriately.
	</P
><P
>	  The different handlers are for traditional Unix mbox
	  mailboxes (all messages in a single file) and new
	  qmail-stype Maildir mailboxes.  You can enable only the
	  two handlers for the specific mailbox type that you use,
	  if desired.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.PYGHANDLER"
></A
><H3
>pyg.PYGHandler</H3
><P
>	  PYG (short for PYGopherd) is a mechanism that provides a
	  tremendous amount of flexibility.  Rather than just
	  letting you execute a script like other Gopher or HTTP
	  servers, PYGs are actually loaded up into PyGopherd and
	  become fully-capable first-class virtual handlers.  Yet
	  they need not be known ahead of time, and are loaded
	  dynamically.
	</P
><P
>	  With a PYG handler, you can generate gopher directories,
	  handle searches, generate files, and more on the fly.
	  You can create entire virtual directory trees (for
	  instance, to interface with NNTP servers or with DICT
	  servers), and access them all using the standard Gopher
	  protocol.  All of this without having to modify even one
	  line of <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> code.
	</P
><P
>	  If enabled, the <SPAN
CLASS="PROPERTY"
>pyg.PYGHandler</SPAN
> will
	  look for files with the extension .pyg that are marked
	  executable.  If found, they will be loaded and run as
	  PYGs.
	</P
><P
>	  Please note: this module provides the capability to
	  execute arbitrary code.  Please consider the security
	  ramifications of that before enabling it.
	</P
><P
>	  See the <SPAN
CLASS="PROPERTY"
>virtual.Virtual</SPAN
> handler for
	  more information about passing data to your scripts at
	  runtime.
	</P
><P
>	  At present, documentation on writing PYGs is not
	  provides, but you may find examples in the
	  <SPAN
CLASS="PROPERTY"
>pygfarm</SPAN
> directory included with the
	  <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> distribution.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.EXECHANDLER"
></A
><H3
>scriptexec.ExecHandler</H3
><P
>	  This handler implements "old-style" script execution;
	  that is, executing arbitrary programs and piping the
	  result to the client.  It is, for the most part,
	  compatible with both scripts written for UMN gopherd and
	  the Bucktooth gopher server.  If enabled, it will
	  execute any file that is marked executable in the
	  filesystem.  It will normally list scripts as returning
	  plain text, but you may create a custom link to the
	  script that defines it as returning whatever kind of
	  file you desire.  Unlike PYGs, this type must be known
	  in advance.
	</P
><P
>	  The <SPAN
CLASS="PROPERTY"
>scriptexec.ExecHandler</SPAN
> will set
	  environment variables for your scripts to use.  They are
	  as follows:
	</P
><P
></P
><TABLE
CLASS="variablelist"
BORDER="0"
CELLSPACING="0"
CELLPADDING="4"
><TBODY
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN304"><SPAN
STYLE="white-space: nowrap"
>SERVER_NAME</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The name of this server as defined in
		the configuration file or detected from the
		operating system.</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN308"><SPAN
STYLE="white-space: nowrap"
>SERVER_PORT</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The port this server is listening on.</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN312"><SPAN
STYLE="white-space: nowrap"
>REMOTE_ADDR</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The IP address of the client.</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN316"><SPAN
STYLE="white-space: nowrap"
>REMOTE_PORT</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The port number of the client.</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN320"><SPAN
STYLE="white-space: nowrap"
>REMOTE_HOST</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The same value as <SPAN
CLASS="PROPERTY"
>REMOTE_ADDR</SPAN
></P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN325"><SPAN
STYLE="white-space: nowrap"
>SELECTOR</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The file that was requested; that is,
		the relative path to this script.  If the selector
		included additional parameters after a |, they
		will be included in this string as well.
	      </P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN329"><SPAN
STYLE="white-space: nowrap"
>REQUEST</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The "base" part of the selector; that
		is, the part leading up to the |.</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN333"><SPAN
STYLE="white-space: nowrap"
>SEARCHREQUEST</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Included only if the client specified
		search data, this is used if the client is
		searching for something.
	      </P
></TD
></TR
></TBODY
></TABLE
><P
>	  See the <SPAN
CLASS="PROPERTY"
>virtual.Virtual</SPAN
> handler for
	  more information about passing data to your scripts at
	  runtime.
	</P
><P
>	  Please note: this module provides the capability to
	  execute arbitrary code.  Please consider the security
	  ramifications of that before enabling it.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.UMN"
></A
><H3
>UMN.UMNDirHandler</H3
><P
>	  This is one of the most powerful workhorse handlers in
	  <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>.  It is designed to emulate most of the ways
	  in which the UMN gopherd distribution generates
	  directories, even going so far as to be bug-compatible
	  in some cases.  Generating directories with this handler
	  is often the best general-purpose way to make nice
	  directories in gopherspace.
	</P
><P
>	  The remainder of the description of the
	  <SPAN
CLASS="PROPERTY"
>UMN.UMNDirHandler</SPAN
>, except for the
	  Abstracts and Info section, is lifted directly from the
	  original UMN gopherd documentation, with light editing,
	  because this handler implements it so exactly that there
	  was no point in rewriting all that documentation :-)
	</P
><DIV
CLASS="REFSECT3"
><A
NAME="HANDLERS.UMN.LINKS"
></A
><H4
>Links</H4
><P
>	    You can override the default view of a directory as
	    generated by <SPAN
CLASS="PROPERTY"
>dir.DirHandler</SPAN
> by
	    creating what are known as <SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>Links</I
></SPAN
> in
	    the data tree.
	  </P
><P
>	    The ability to make links to other hosts is how gopher
	    distributes itself among multiple hosts.  There are two
	    different ways to make a link.  The first and simplest is
	    to create a link file that contains the data needed by the
	    server.  By default all files in the gopher data directory
	    starting with a period are taken to be link files.  A link
	    file can contain multiple links.  To define a link you
	    need to put five lines in a link file that define the
	    needed characteristics for the document.  Here is an
	    example of a link.
	  </P
><PRE
CLASS="PROGRAMLISTING"
>Name=Cheese Ball Recipes
Numb=1
Type=1
Port=150
Path=1/Moo/Cheesy
Host=zippy.micro.umn.edu
	  </PRE
><P
>	    The Name= line is what the user will see when cruising
	    through the database.  In this case the name is "Cheese
	    Ball Recipes".  The "Type=" defines what kind of document
	    this object is.  For a list of all defined types, see
	    Gopher Item Types below.  For Gopher+ and HTTP, a MIME
	    type is also used, which is determined automatically based
	    on the type you specify.
	  </P
><P
>	    The "Path=" line contains the selector string that the
	    client will use to retrieve the actual document.  The
	    Numb= specifies that this entry should be presented first
	    in the directory list (instead of being alphabetized).
	    The "Numb=" line is optional.  If it is present it cannot
	    be the last line of the link.  The "Port=" and "Host="
	    lines specify a fully qualified domain name (FQDN) and a
	    port respectively.  You may substitute a plus '+' for
	    these two parameters if you wish.  The server will insert
	    the current hostname and the current port when it sees a
	    plus in either of these two fields.
	  </P
><P
>An easy way to retrieve links is to use the Curses
	    Gopher Client.  By pressing '=' You can get information
	    suitable for inclusion in a link file.
	  </P
></DIV
><DIV
CLASS="REFSECT3"
><A
NAME="HANDLERS.UMN.OVERRIDING"
></A
><H4
>Overriding Defaults</H4
><P
>	    The server looks for a directory called
	    <TT
CLASS="FILENAME"
>.cap</TT
> when parsing a directory.  The
	    server then checks to see if the <TT
CLASS="FILENAME"
>.cap</TT
>
	    directory contains a file with the same name as the file
	    it's parsing.  If this file exists then the server will
	    open it for reading.  The server parses this file just
	    like a link file.  However, instead of making a new
	    object, the parameters inside the
	    <TT
CLASS="FILENAME"
>.cap/</TT
> file are used to override any
	    of the server supplied default values.
	  </P
><P
>	    For instance, say you wanted to change the Title of a text
	    file for gopher, but don't want to change the filename.
	    You also don't want it alphabetized, instead you want it
	    second in the directory listing.  You could make a
	    set-aside file in the <TT
CLASS="FILENAME"
>.cap</TT
> directory with the same
	    filename that contained the following lines:
	  </P
><PRE
CLASS="PROGRAMLISTING"
>Name=New Long Cool Name
Numb=2
	  </PRE
><P
>	    An alternative to <TT
CLASS="FILENAME"
>.cap</TT
> files are
	    extended link files.  They work just the same as the files
	    described in Links above, but have a somewhat abbreviated
	    format.  As an example, if the name of the file was
	    <TT
CLASS="FILENAME"
>file-to-change</TT
>, then you could create
	    a file called <TT
CLASS="FILENAME"
>.names</TT
> with the
	    following contents:
	  </P
><PRE
CLASS="PROGRAMLISTING"
>Path=./file-to-change
Name=New Long Cool Name
Numb=2
	  </PRE
></DIV
><DIV
CLASS="REFSECT3"
><A
NAME="HANDLERS.UMN.COOL"
></A
><H4
>Adding Cool Links</H4
><P
>	    One cool thing you can do with .Links is to add neato
	    services to your gopher server.  Adding a link like this:
	  </P
><PRE
CLASS="PROGRAMLISTING"
>Name=Cool ftp directory
Type=h
Path=/URL:ftp://hostname/path/
Host=+
Port=+
 
Name=Cool web site
Type=h
Path=/URL:http://hostname/
Host=+
Port=+
	  </PRE
><P
>	    Will allow you to link in any FTP or Web site to your
	    gopher.  (See url.URLHandler for more details.)
	  </P
><P
>	    You can easily add a finger site to your gopher server thusly:
	  </P
><PRE
CLASS="PROGRAMLISTING"
>Name=Finger information
Type=0
Path=lindner
Host=mudhoney.micro.umn.edu
Port=79
	  </PRE
></DIV
><DIV
CLASS="REFSECT3"
><A
NAME="HANDLERS.UMN.HIDING"
></A
><H4
>Hiding an Entry</H4
><P
>	    This kind of trick may be necessary in some cases,
	    and thus for
	    object "fred", the overriding .names file entry would be:
	  </P
><PRE
CLASS="PROGRAMLISTING"
> Type=X
 Path=./fred
	  </PRE
><P
>	    by overriding default type to be "X".  This kind of
	    hideouts may be usefull, when for some reason there are
	    symlinks (or whatever) in the directory at which
	    <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> looks at, and those entries are not desired to
	    be shown at all.
	  </P
></DIV
><DIV
CLASS="REFSECT3"
><A
NAME="HANDLERS.UMN.ABSTRACTS"
></A
><H4
>Abstracts and Info</H4
><P
>	    Many modern gopher server maintainers like to intersperse
	    gopher directory listings with other information -- often,
	    additional information about the contents of files in the
	    directory.  The gophermap system provides one way to do
	    that, and abstracts used with UMN gopher directories
	    provides another.
	  </P
><P
>	    Subject to the <SPAN
CLASS="PROPERTY"
>abstract_headers</SPAN
> and
	    <SPAN
CLASS="PROPERTY"
>abstract_entries</SPAN
> configuration file
	    options, this feature allows you to define that extra
	    information.  You can do that by simply creating a file
	    named <TT
CLASS="FILENAME"
>filename.abstract</TT
> right
	    alongside the regular file in your directory.  The file
	    will be interpreted as the abstract.  For a directory,
	    create a file named <TT
CLASS="FILENAME"
>.abstract</TT
> in the
	    directory.  Simple as that!
	  </P
></DIV
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.HTMLURLHANDLER"
></A
><H3
>url.HTMLURLHandler</H3
><P
>	  <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> provides ways for you to link to pages outside
	  Gopherspace -- that is, web pages, FTP sites, and the like.
	  This is accomplished according to the <A
HREF="http://lists.complete.org/gopher@complete.org/2002/02/msg00033.html.gz"
TARGET="_top"
>Links
	  to URL</A
> specification (see Conforming To below for
	  details).  In order to link to a URL (EXCEPT gopher URLs)
	  from a menu, you create a link of type h (regardless of the
	  actual type of the resource that you are linking to) in your
	  <TT
CLASS="FILENAME"
>gophermap</TT
> or
	  <TT
CLASS="FILENAME"
>.Links</TT
>
	  file that looks like this:
	</P
><PRE
CLASS="PROGRAMLISTING"
>/URL:http://www.complete.org/
	</PRE
><P
>	  Modern Gopher clients that follow the Links to URL
	  specification will automatically follow that link when you
	  select it.  The rest need some help, and that's where this
	  handler comes in.
	</P
><P
>	  For Gopher clients that do not follow the Links to URL
	  specification, the <SPAN
CLASS="PROPERTY"
>url.HTMLURLHandler</SPAN
>
	  will automatically generate an HTML document for them on the
	  fly.  This document includes a refresh code that will send
	  them to the proper page.  You should not disable this
	  handler.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.URLTYPEREWRITER"
></A
><H3
>url.URLTypeRewriter</H3
><P
>	  Some people wish to serve HTML documents from their Gopher
	  server.  One problem with that is that links in Gopherspace
	  include an extra type character at the beginning, whereas
	  links in HTTP do not.  This handler will remove the extra
	  type character from HTTP requests that come in, allowing a
	  single relative-to-root link to work for both.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.VIRTUAL"
></A
><H3
>virtual.Virtual</H3
><P
>	  This handler is not intended to ever be used directly, but
	  is used by many other handlers such as the mbox support, PYG
	  handlers, and others.  It is used to generate virtual
	  entries in the directory hierarchy -- that is, entries that
	  look normal to a client, but do not actually correspond to a
	  file on disk.
	</P
><P
>	  One special feature of the
	  <SPAN
CLASS="PROPERTY"
>virtual.Virtual</SPAN
> handler is that you can
	  send information to it at runtime in a manner similar to a
	  CGI script on the web.  You do this by adding a question
	  mark after the regular selector, followed by any arbitrary
	  data that you wish to have sent to the virtual request
	  handler.
	</P
></DIV
><HR><DIV
CLASS="REFSECT2"
><A
NAME="HANDLERS.ZIP"
></A
><H3
>ZIP.ZIPHandler</H3
><P
>	  Using zip.ZIPHandler, you can save space on your server by
	  converting part or all of your site into a ZIP file.
	  <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> can use the contents of that ZIP file as the
	  contents of your site -- completely transparently.
	</P
><P
>	  The ZIP file handler must be enabled in the configuration
	  file for this to work.
	</P
></DIV
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="TYPES"
></A
><H2
>Gopher Item Types</H2
><P
>	When you construct links to files via
	<TT
CLASS="FILENAME"
>.Links</TT
> or <TT
CLASS="FILENAME"
>gophermap</TT
>
	files, or modify the <SPAN
CLASS="PROPERTY"
>mapping</SPAN
> in the
	configuration file, you will need to know these.  Items
	bearing the "not implemented" text are not served up by
	<SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> as it ships, generally due to requirements of
	customized per-site software, but may be served up via PYG
	extension modules or other gopher servers.
      </P
><P
>	This list was prepared based on RFC1436, the UMN gopherd(1) manpage,
	and best current practices.
      </P
><P
></P
><TABLE
CLASS="variablelist"
BORDER="0"
CELLSPACING="0"
CELLPADDING="4"
><TBODY
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN424"><SPAN
STYLE="white-space: nowrap"
>0</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Plain text file</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN428"><SPAN
STYLE="white-space: nowrap"
>1</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Directory</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN432"><SPAN
STYLE="white-space: nowrap"
>2</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>CSO phone book server (not implemented by <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>)</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN437"><SPAN
STYLE="white-space: nowrap"
>3</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Error condition; text that follows is plain text</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN441"><SPAN
STYLE="white-space: nowrap"
>4</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Macintosh file, BinHex format</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN445"><SPAN
STYLE="white-space: nowrap"
>5</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>DOS binary archive (not implemented by
	      <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>; use type 9 instead)</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN450"><SPAN
STYLE="white-space: nowrap"
>6</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>uuencoded file; not directly generated by
	      <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> automatically, but can be linked to
	      manually.  Most gopher clients will handle this better
	      as type 1.
	    </P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN455"><SPAN
STYLE="white-space: nowrap"
>7</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Search</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN459"><SPAN
STYLE="white-space: nowrap"
>8</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Telnet link</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN463"><SPAN
STYLE="white-space: nowrap"
>9</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Binary file</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN467"><SPAN
STYLE="white-space: nowrap"
>+</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Redundant server (not implemented by <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>)</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN472"><SPAN
STYLE="white-space: nowrap"
>c</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Calendar (not implemented by <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>)</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN477"><SPAN
STYLE="white-space: nowrap"
>e</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Event (not implemented by <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>)</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN482"><SPAN
STYLE="white-space: nowrap"
>g</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>GIF-format graphic</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN486"><SPAN
STYLE="white-space: nowrap"
>h</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>HTML file</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN490"><SPAN
STYLE="white-space: nowrap"
>I</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Any kind of graphic file other than GIF</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN494"><SPAN
STYLE="white-space: nowrap"
>i</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Informational
	text included in a directory that is displayed but does not
	link to any actual file.</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN498"><SPAN
STYLE="white-space: nowrap"
>M</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>MIME multipart/mixed file</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN502"><SPAN
STYLE="white-space: nowrap"
>s</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Any kind of sound file</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN506"><SPAN
STYLE="white-space: nowrap"
>T</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>tn3270 link</P
></TD
></TR
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
><A
NAME="AEN510"><SPAN
STYLE="white-space: nowrap"
>X</SPAN
>, <SPAN
STYLE="white-space: nowrap"
>-</SPAN
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>	UMN-specific -- signifies that this entry should not be
	displayed in a directory entry, but may be accessed via a
	direct link.  This value is never transmitted in any Gopher
	protocol.
	    </P
></TD
></TR
></TBODY
></TABLE
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="CONFORMING"
></A
><H2
>Conforming To</H2
><P
></P
><UL
><LI
><P
>	    The Internet Gopher Protocol as specified in RFC1436
	  </P
></LI
><LI
><P
>	    The Gopher+ upward-compatible enhancements to the Internet Gopher
	    Protocol from the University of Minnesota as laid out at
	    <A
HREF="gopher://gopher.quux.org/0/Archives/mirrors/boombox.micro.umn.edu/pub/gopher/gopher_protocol/Gopher+/Gopher+.txt"
TARGET="_top"
>gopher://gopher.quux.org/0/Archives/mirrors/boombox.micro.umn.edu/pub/gopher/gopher_protocol/Gopher+/Gopher+.txt</A
>.
	  </P
></LI
><LI
><P
>	    The gophermap file format as originally implemented in the
	    Bucktooth gopher server and described at
	    <A
HREF="gopher://gopher.floodgap.com/0/buck/dbrowse%3Ffaquse%201"
TARGET="_top"
>gopher://gopher.floodgap.com/0/buck/dbrowse%3Ffaquse%201</A
>.
	  </P
></LI
><LI
><P
>	    The Links to URL specification as laid out by John Goerzen
	    at
	    <A
HREF="gopher://gopher.quux.org/0/Archives/Mailing%20Lists/gopher/gopher.2002-02%3f/MBOX-MESSAGE/34"
TARGET="_top"
>gopher://gopher.quux.org/0/Archives/Mailing%20Lists/gopher/gopher.2002-02%3f/MBOX-MESSAGE/34</A
>.
	  </P
></LI
><LI
><P
>	    The UMN format for specifying object attributes and links
	    with .cap, .Links, .abstract, and similar files as specified elsewhere
	    in this document and implemented by UMN gopherd.
	  </P
></LI
><LI
><P
>	    The PYG format for extensible Python gopher objects as created for
	    <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>.</P
></LI
><LI
><P
>	    Hypertext Transfer Protocol HTTP/1.0 as specified in
	    RFC1945
	  </P
></LI
><LI
><P
>	    Hypertext Markup Language (HTML) 3.2 and 4.0
	    Transitional as specified in RFC1866 and RFC2854.
	  </P
></LI
><LI
><P
>	    Maildir as specified in
	    <A
HREF="http://www.qmail.org/qmail-manual-html/man5/maildir.html"
TARGET="_top"
>http://www.qmail.org/qmail-manual-html/man5/maildir.html</A
> and
	    <A
HREF="http://cr.yp.to/proto/maildir.html"
TARGET="_top"
>http://cr.yp.to/proto/maildir.html</A
>.
	  </P
></LI
><LI
><P
>	    The mbox mail storage format as specified in
	    <A
HREF="http://www.qmail.org/qmail-manual-html/man5/mbox.html"
TARGET="_top"
>http://www.qmail.org/qmail-manual-html/man5/mbox.html</A
>.
	  </P
></LI
><LI
><P
>	    Registered MIME media types as specified in RFC2048.
	  </P
></LI
><LI
><P
>	    Script execution conforming to both UMN standards as laid out in UMN
	    gopherd(1) and Bucktooth standards as specified at
	    <A
HREF="gopher://gopher.floodgap.com:70/0/buck/dbrowse%3ffaquse%202"
TARGET="_top"
>gopher://gopher.floodgap.com:70/0/buck/dbrowse%3ffaquse%202</A
>, 
	    so far as each can be implemented consistent with secure
	    design principles.
	  </P
></LI
><LI
><P
>	    Standard Python 2.2.1 or above as implemented on
	    POSIX-compliant systems.
	  </P
></LI
><LI
><P
>	    WAP/WML as defined by the WAP Forum.</P
></LI
></UL
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="BUGS"
></A
><H2
>Bugs</H2
><P
>	Reports of bugs should be sent via e-mail to the <SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>
	bug-tracking system (BTS) at
	<CODE
CLASS="EMAIL"
>&#60;<A
HREF="mailto:pygopherd@bugs.complete.org"
>pygopherd@bugs.complete.org</A
>&#62;</CODE
> or submitted online
	using the Web interface at <A
HREF="http://bugs.complete.org/"
TARGET="_top"
>http://bugs.complete.org/</A
>.
      </P
><P
>	The Web site also lists all current bugs, where you can check their
	status or contribute to fixing them.
      </P
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="COPYRIGHT"
></A
><H2
>Copyright</H2
><P
>	<SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> is Copyright (C) 2002, 2003 John Goerzen.
      </P
><P
>	This program is free software; you can redistribute it and/or
	modify it under the terms of the GNU General Public License as
	published by the Free Software Foundation; version 2 of the
	License.
      </P
><P
>	This program is distributed in the hope that it will be useful,
	but WITHOUT ANY WARRANTY; without even the implied warranty of
	MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
	GNU General Public License for more details.
      </P
><P
>	You should have received a copy of the GNU General Public License
	along with this program; if not, write to:
      </P
><PRE
CLASS="PROGRAMLISTING"
>Free Software Foundation, Inc.
59 Temple Place
Suite 330
Boston, MA  02111-1307
USA
      </PRE
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="AUTHOR"
></A
><H2
>Author</H2
><P
>	<SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
>, its libraries, documentation, and all included
	files (except where noted) was written by John Goerzen
	<CODE
CLASS="EMAIL"
>&#60;<A
HREF="mailto:jgoerzen@complete.org"
>jgoerzen@complete.org</A
>&#62;</CODE
>
	and copyright is held as stated in the
	Copyright section.
      </P
><P
>	Portions of this manual (specifically relating to certian UMN gopherd
	features and characteristics that PyGopherd emulates) are modified
	versions of the original
	gopherd(1) manpage accompanying the UMN gopher distribution.  That
	document is distributed under the same terms as this, and
	bears the following copyright notices:
      </P
><PRE
CLASS="PROGRAMLISTING"
>Copyright (C) 1991-2000  University of Minnesota
Copyright (C) 2000-2002  John Goerzen and other developers
      </PRE
><P
>	<SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> may be downloaded, and information found, from its
	homepage via either Gopher or HTTP:
      </P
><P
>	<A
HREF="gopher://quux.org/1/devel/gopher/pygopherd"
TARGET="_top"
>gopher://quux.org/1/devel/gopher/pygopherd</A
></P
><P
>	<A
HREF="http://quux.org/devel/gopher/pygopherd"
TARGET="_top"
>http://quux.org/devel/gopher/pygopherd</A
></P
><P
>	<SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> may also be downloaded using Subversion.  Additionally,
	the distributed tar.gz may be updated with a simple "svn update"
	command; it is ready to go.  For information on getting
	<SPAN
CLASS="APPLICATION"
>PyGopherd</SPAN
> with Subversion, please visit <A
HREF="http://svn.complete.org/"
TARGET="_top"
>http://svn.complete.org/</A
>.
      </P
></DIV
><DIV
CLASS="REFSECT1"
><A
NAME="AEN586"
></A
><H2
>See Also</H2
><P
>	python (1).
      </P
></DIV
></DIV
></BODY
></HTML
>
pygopherd 2.0.18.3+nmu3 / usr / share / doc / pygopherd / pygopherd.html
