ddrudik ... my hero, it works like a charm. You've saved med hours and hours of "trial an errror" ... 

I made a version of your code and my old to collect all the data in the three first TD's and put them all into one single Item, like this. 

<?php 
$text = file_get_contents('http://landris.hh.se/4DACTION/WebShowRoll/1-21?offset=4320&update=0&rows=0&page=0&branch=4&group=-21&start=yes&stop=yes&order=ascending&web_cols=1&web_numChars=-'); 
$regex = "# 
    <tr>\s* 
    <td\s+style=[^>]+>\s*<font\s*class='a_text'>\s*([^>]*)</font>\s*</td>\s* 
    <td\s+style=[^>]+>\s*<font\s*class='a_text'>\s*([^>]*)</font>\s*</td>\s* 
    <td\s+style=[^>]+>\s*<font\s*class='a_text'>\s*([^>]*)</font>\s*</td> 
#isx"; 
if(preg_match_all($regex, $text, $matches)) { 

foreach ($matches[0] as $title) {

Another question ... is it possible to add an custom message if no $matches where found. 

Nevermind ... solved that by adding:

 else { 

   echo "<item>\n\t<title>" . strip_tags($title) . "DET FINNS INGA LEKTIONER SCHEMALAGDA.</title>\n\t<description>" . strip_tags($title) . "DET FINNS INGA LEKTIONER SCHEMALAGDA.</description>\n\t<guid>http://www.campus.varberg.se/".md5(uniqid()).".html</guid>\n</item>\n"; 

} 

Thanks again for all your help 

Consider this code which includes your last 2 requirements:

<?php
header("Content-Type: text/xml; charset=ISO-8859-1");
echo "<?xml version=\"1.0\" encoding=\"ISO-8859-1\" ?>\n";
?>
<rss version="2.0"
  xmlns:dc="http://purl.org/dc/elements/1.1/"
  xmlns:content="http://purl.org/rss/1.0/modules/content/"
  xmlns:admin="http://webns.net/mvcb/"
  xmlns:atom="http://www.w3.org/2005/Atom"
  xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
<channel>
<title>Lektioner:</title>
<description>Campus Varberg</description>
<link>http://landris.hh.se</link>
<atom:link href="http://www.campus.varberg.se/dev/scrape/schema.php" rel="self" type="application/rss+xml" />
<language>se-SE</language>
<?php
$text = file_get_contents('http://landris.hh.se/4DACTION/WebShowRoll/1-21?offset=4320&update=0&rows=0&page=0&branch=4&group=-21&start=yes&stop=yes&order=ascending&web_cols=1&web_numChars=-');
$regex = "#
    <tr>\s*
    <td\s+style=[^>]+>\s*<font\s*class='a_text'>\s*([^>]*)</font>\s*</td>\s*
    <td\s+style=[^>]+>\s*<font\s*class='a_text'>\s*([^>]*)</font>\s*</td>\s*
    <td\s+style=[^>]+>\s*<font\s*class='a_text'>\s*([^>]*)</font>\s*</td>
#isx";
if(preg_match_all($regex, $text, $matches)) {
    for ($i = 0; $i < count($matches[0]); $i++) {
        $title=$matches[1][$i].' '.$matches[2][$i].' '.$matches[3][$i];
        echo "<item>\n\t<title>$title</title>\n\t<description>$title</description>\n\t<guid>http://www.campus.varberg.se/".md5(uniqid()).".html</guid>\n</item>\n";
    }
} else {  
   echo "<item>\n\t<title>DET FINNS INGA LEKTIONER SCHEMALAGDA.</title>\n\t<description>DET FINNS INGA LEKTIONER SCHEMALAGDA.</description>\n\t<guid>http://www.campus.varberg.se/".md5(uniqid()).".html</guid>\n</item>\n";   
}   
?>
</channel>
</rss> 

Thanks a lot