General Remarks

You're website is located on the coli-webserver: httpd. Before you can start changing your pages you have to login to the server via ssh:

 $ ssh httpd.coli.uni-saarland.de

Please note that projects, courses and private web sites should be placed in appropriate directories as follows:

• Projects: /web/www.coli.uni-saarland.de/html/projects/

• Courses: /web/www.coli.uni-saarland.de/html/courses/

• Personal web pages: $HOME/public_html/ (esp. public_html/index.html or public_html/index.php)

Directories for courses and project web pages can be requested for at Systems Group. Please do not put any courses or project pages in your private home page.

Pagetree System

If you are setting up a complex web site, you might be interested in using the pagetree system that we use for the global Coli website. This system will allow you to specify the structure of your website in a configuration file, and then compute a navbar for you automatically.

The pagetree system consists of three parts:

• The file "pagetree.xml", which defines the structure of your site.
• HTML files in a subdirectory "contents", one for each page of your site, which contain the actual contents of your site.
• The file "page.php", which reads pagetree, computes the navbar, and displays the content for the selected page.

Setting up your web site

• Copy or link the file /web/www.coli.uni-saarland.de/html/page.php on httpd into the directory of your website
• Create a file "pagetree.xml" in that directory. The file must have a pagetree element that contains a section with an entry. It doesn't need a frontpage element.
• Create a subdirectory "contents" to place different html pages (i.e. contact.html)
• Make sure that all files and directories are readable by the webserver

• Now you can access the page with a URL like "http://..../page.php?id=contact"

pagetree.xml

The pagetree.xml should have the following construct:

 <?xml version="1.0"?>                                                                            
 <pagetree docroot="/HOMEPAGE_DIRECTORY/" title="HOMEPAGE TITLE">
 <section title="SECTION TITLE">
     <entry id="FILENAME" title="ENTRY TITLE" />
                 .
                 .
                 .
  </section>
  <section title="">
         .
         .
         .
  
 </pagetree>

FILENAME is the name of the html file in contents without .html, i.e. "contact".

The entries and sections of a pagetree.xml file can have additional attributes that are read as flags (i.e. you write <entry id=etc. etc. flag="true" />). Some useful flags are:

• local: This page is only visible from Coli, DFKI, or LSV
• hidden: This page exists, and can be accessed with page.php?id=..., but won't show up in the navbar.
• url: You can use this attributes to link an entry to any other page which is not placed in the "contents' directory.

The top-level "pagetree" element can also have attributes:

• title: This is the page title for the entire site under pagetree control
• docroot: Each link in the navbar will be prefixed by this string. That is, docroot can be the absolute or relative URL (not pathname on the web server!) for your pagetree site.

Example

 <?xml version="1.0"?>
 <pagetree docroot="/" title="Computational Linguistics and Phonetics">
  <section title="General">
    <entry id="whatis" title="What is...?"/>
    <entry id="coordinates" title="Coordinates"/>
  </section>
  <section title="Research">
    <entry id="projects" title="Research Projects"/>
    <entry id="publications" url="/publications.php" title="Publications"/>
    <entry id="dissertations" url="<nowiki>http://www.dfki.de/lt/diss.php</nowiki>" title="Dissertation Series"/>
    <entry id="foo" hidden="true"/>
  </section>
 </pagetree>