sitemap.xml

<?xml version="1.0"?>
<!DOCTYPE book PUBLIC 
    "-//OASIS//DTD DocBook XML V4.1.2//EN" 
    "docbook/docbookxx.dtd" [
<!ENTITY % ISOnum PUBLIC
    "ISO 8879:1986//ENTITIES Numeric and Special Graphic//EN"
    "iso-num.ent">
%ISOnum;
<!ENTITY % ISOpub PUBLIC
    "ISO 8879:1986//ENTITIES Publishing//EN"
    "iso-pub.ent">
%ISOpub;
]>

<refentry id='sitemap.man'>

<refmeta>
<refentrytitle>sitemap</refentrytitle>
<manvolnum>1</manvolnum>
</refmeta>
<refnamediv id='name'>
<refname> sitemap</refname>
<refpurpose>make a site map from meta tags in an HTML tree</refpurpose>
</refnamediv>

<refsynopsisdiv id='synopsis'>
<cmdsynopsis>
  <command>sitemap</command>  
  <group choice='opt'>
     <arg choice='plain'><replaceable>start-dir</replaceable></arg>
     <arg choice='plain'><replaceable>config-file</replaceable></arg>
  </group>
</cmdsynopsis>
</refsynopsisdiv>

<refsect1><title>Description</title> 
<para><command>sitemap</command> indexes all pages under the start
directory and writes an HTML map page to standard output.  The code
looks for description information for each page in a META DESCRIPTION
header; if it doesn't find one, the page is omitted from the index.
That is, HTML pages to be indexed should have a <markup>meta</markup>
tag with its <markup>name</markup> attribute set to
<markup>description</markup> and its <markup>content</markup>
attribute set to a brief description ofthe contents.  For example,
</para>

<literallayout remap='Vb'>
 &lt;head&gt;
   &lt;title&gt;Sitemap documentation&lt;/title&gt;
   &lt;meta name="description" 
     content="Documentation for <command>sitemap</command> program to index HTML pages."&gt;
 &lt;/head&gt;
</literallayout>

<para>The output of <command>sitemap</command> is an HTML page that
contains a list of descriptions and links to the indexed pages.  This
output can be configured via an rc file (see below).</para>
</refsect1>

<refsect1><title>Arguments</title>
<para>If no options are supplied, the start directory is the directory
indicated by the DOCUMENT_ROOT or HOME environment variables, in that
order.  If neither variable is specified on a UNIX system, the
effective user's home directory (as indicated in the passwd file) will
be used.  If a <filename>start-dir</filename> directory is supplied as
an argument, then that directory will be used as the start directory.
In both of these cases, the configuration will be read from a file
named .sitemaprc in the start directory.  If the configuration file
does not exist, <command>sitemap</command> will run with a set of
default parameters, which is usually not what you want.</para>

<para>If a <filename>config-file</filename> configuration file is
specified, then the configuration for <command>sitemap</command> will
be read from that file.  In this case, the start directory will be the
directory containing the configuration file.</para>
</refsect1>

<refsect1><title>Configuration File</title>
<para><command>sitemap</command> is a Python script.  To configure the
strings used in the index page header and footer, you can create a
configuration file in your home directory called .sitemaprc (or as
indicated by the command-line parameter).  A skeleton of a
configuration file is provided with the program.  The file should
start with the text <markup>[sitemap]</markup> on a line
by itself.  Subsequent lines should be name=value pairs.  Lines
beginning with the # character are treated as comments and are
ignored. The possible field names in the configuration file are listed
below:</para>

<variablelist>

<varlistentry><term>Hometitle=<replaceable>title</replaceable></term>
<listitem><para>The title of your homepage.  The generated site map
will contain a link with this text.</para></listitem>
</varlistentry>

<varlistentry><term>Homepage=<replaceable>url</replaceable></term>
<listitem><para>The URL of your homepage.  The generated site map will
contain a link back to this page.</para></listitem>
</varlistentry>

<varlistentry><term>Indextitle=<replaceable>title</replaceable></term>
<listitem><para>The title for the generated site map
page.</para></listitem>
</varlistentry>

<varlistentry><term>Headinfo=<replaceable>any Html Text</replaceable></term> 
<listitem><para>Any additional HTML you want to include in the
&lt;head&gt; section of the site map.  Use with care - only certain
tags are legal in the &lt;head&gt; of a page.</para></listitem>
</varlistentry>

<varlistentry><term>Body=<replaceable>attributes</replaceable></term>
<listitem><para>Any additional attributes to be included in the
&lt;body&gt; tag.</para></listitem>
</varlistentry>

<varlistentry><term>Prefix=<replaceable>url</replaceable></term>
<listitem><para>An optional URL prefix to put before each
pathname. Normally, <command>sitemap</command> outputs each filename
as a site-relative path beginning with a '/', in the assumption that
the start-directory can be accessed with the URL '/'.  (That is, the
start directory would be the directory indciated by the web server's
DOCUMENT_ROOT.)  If this is incorrect (e.g. you are indexing a user's
home page whose URL begins with '/~username') you can supply the
alternative URL prefix here.</para></listitem>
</varlistentry>

<varlistentry><term>Dirtitle=<replaceable>title</replaceable></term>
<listitem><para>The title string to use for directories.  Directories
are listed and linked in the generated site map page with this
text.</para></listitem>
</varlistentry>

<varlistentry><term>Fullname=<replaceable>name</replaceable></term>
<listitem><para>Your full name.  This name will be included in one
corner of the generated site map page.  You may want to list a company
name or a copyright statement instead, for example. </para></listitem>
</varlistentry>

<varlistentry><term>Mailaddr=<replaceable>address</replaceable></term>
<listitem><para>E-mail address of a contact person.  Since the e-mail
address will be linked on the generated site map page, you may want to
set this parameter to the e-mail address of a contact person or a
webmaster.</para></listitem>
</varlistentry>

<varlistentry><term>Language=<replaceable>language</replaceable></term>
<listitem><para>The language for the boilerplate text included in the
output (Czech, English, French, German, Italian, Norwegian, Spanish,
or Swedish). </para></listitem>
</varlistentry>

<varlistentry><term>Icondirs=<replaceable>icon Path</replaceable></term>
<listitem><para>The path (relative to the start directory or a URL) of
the icon for directories.  The icon must be 33 pixels wide (or
scaleable to that size).  If omitted, no icon will be displayed next
to site map entries for directories.</para></listitem>
</varlistentry>

<varlistentry><term>Icontext=<replaceable>icon Path</replaceable></term>
<listitem><para>The path (relative to the start directory or a URL) of
the icon for HTML files.  The icon must be 33 pixels wide (or
scaleable to that size).  If omitted, no icon will be displayed next
to site map entries for HTML pages.</para></listitem>
</varlistentry>

<varlistentry><term>Indexfiles=<replaceable>file1 File2 File3</replaceable></term>
<listitem><para>A space-separated list of files to treat as index or
main pages for a directory.  Any file with a filename exactly equal to
one of the indicated filenames will be treated as an index page.
Index pages sort to the top of the list of files in a directory.  For
example, index.html or default.htm might be good candidates for this
parameter.</para></listitem>
</varlistentry>

<varlistentry><term>Exclude=<replaceable>word1 Word2</replaceable></term>
<listitem><para>A space-separated list of words to ignore when
scanning files and directories.  <command>sitemap</command> will skip
any file or entire subdirectories the contain any of the words in
their path.  For example, Test or CVS may be good candidates for this
parameter.</para></listitem>
</varlistentry>

<varlistentry><term>processalso=<replaceable>file1 file2 file3</replaceable></term>
<listitem><para>
A space-separated list of filenames, which will also be parsed by
<command>sitemap</command>. Here you can put your PHP, or Perl Pages, which
contains also a HTML header.
</para></listitem>
</varlistentry>

<varlistentry><term>webdevelop=<replaceable>y</replaceable></term>
<listitem><para>
If set to y, the output will also display the "Keywords" and also the lenght of
"Description" and "title". This is ideal while making some search engine
optimisation.  </para></listitem>
</varlistentry>

<varlistentry><term>ignoredirs=<replaceable>y</replaceable></term>
<listitem><para>
If set to y, there won't be any Directory Entry in the output. This is perfect,
if you actually have in each directory an index-file, or if you don't want to
have users browsing all directories in yout site tree.
</para></listitem>
</varlistentry>

<varlistentry><term>Debug=<replaceable>y</replaceable></term>
<listitem><para>Set this parameter to view the computed configuration
file name, start directory, document root, and prefix in the generated
site map page.  You'll need to view the source of the generated HTML
file because these values will be listed within and HTML comment.
Search for the word <emphasis>Debugging</emphasis> in the generated
HTML page.</para></listitem>
</varlistentry>
</variablelist>

</refsect1>

<refsect1><title>Use Under CGI</title>

<para>You can use <command>sitemap</command> to generate site maps on
the fly. Any command-line argument can be passed as the query string
(i.e. a string immediately following the URL of the CGI script and a
'?' character).</para>

<para><command>sitemap</command> will deduce that it is running under
the CGI by virtue of the fact that the REMOTE_ADDR environment
variable is defined. If so, it outputs a content-type header
(text/html) ahead of the HTML page.</para>

<para>When running as a CGI script, <command>sitemap</command> does
not assume that the document root is necessarily identical with the
start directory.  It inspects the DOCUMENT_ROOT environment variable
and constructs a prefix in an attempt to get from the server document
root to the start directory. This will fail if the start directory is
not a subdirectory under the document root, in which case the prefix
directive in the configuration file should be used.</para>
</refsect1>

<refsect1><title>Authors</title>

<para>Eric S. Raymond <email>esr@thyrsus.com</email>.</para>

<para>Immo Huneke <email>HunekeI@Logica.Com</email>.</para>

<para>Tom Bryan <email>tbryan@python.net</email>.</para>

<para>Claudio Clemens <email>asturio@gmx.net</email>.</para>
</refsect1>
</refentry>