php content class

Just some stuff for a scraper.

Basic content class

input : table [content, url, keywords]
input : ,, + [wordcount, new content]

$art = new Article(html_entity_decode($row['content']), $row['url']);
if($art->words>200) {
$art->FoulFilter();
$art->tags_nofollow();
$art->cut_text(0.80);
$art->tribute($row['keywords']);
}

That outputs a nice clean article cut off at N%, with all links on nofollow and one follow link back to the authors page, all foul language replaced with ****. Nice and simple.

Class Article {
var $original;
var $url;
var $title;
var $content;
var $keywords;
var $length;
var $words;
var $anchors;
var $tags=array();
var $trackbacks=array();
var $perc=.85;
var $publishdate='';
var $publishperdate=4;
var $foul='foulword.txt';
var $newcontent;
public function __construct($source, $url) {
$this->url = $url;
$this->original = $source;
$this->content=$source;
$this->get_stats();
$this->get_title();
}
public function get_title() {
//extract the title from any h1/2/3 tags, or from the url
if(strpos($this->url,'?p=')>0) {
//check content for title
$this->title=preg_match('/<h1>(.*?)</h1>/', $this->content);
if(!$this->title) $this->$title=preg_match('/<h2>(.*?)< \/h2>/', $this->content);
if(!$this->title) $this->$title=preg_match('/<h3>(.*?)< \/h3>/', $this->content);
if(!$this->title) $this->$title = substr( $this->content, 0, 20);
} else {
$x=substr($this->url, 0, strlen($this->url)-1);
$p=strrpos($x, '/');
$filenm=substr($this->url, $p);
$filenm=preg_replace('/\//', '', $filenm);
$ttl=preg_replace('/-/', ' ', $filenm);
$this->title = $ttl.' - '.parse_url($this->url, PHP_URL_HOST);
}
}
public function get_stats() {\
//rough stats : number of words, content length, number of anchors
$this->words=count(explode(' ', $this->content));
$this->length=strlen($this->content);
preg_match_all('/http/', $this->content, $matches);
$this->anchors=count($matches);
}
public function FoulFilter(){
//remove foul language based on 'foulword.txt'
$stopWords = mb_split( '[ \n]+', mb_strtolower(file_get_contents($this->foul), 'utf-8' ) );
for($i=0;$i<count ($stopWords);$i++) {
if(trim($stopWords[$i])<>''){
$this->content=preg_replace('/ '.trim($stopWords[$i]).' /', ' ***** ',$this->content);
}
}
}
public function tags_nofollow() {
//put all anchors in content on nofollow
$this->content=preg_replace('/ href=/', ' target="_blank" rel="nofollow" href=', $this->content);
}
public function get_div() {
//extract the first div section of the content snippet
$this->content=substr($this->original, 0, strpos($this->content, 'div>'));
}
public function tribute($k) {
//add a tribute link to the original author
$this->content .= '><br /><br />[<a href="'.$this->url.'" title="'.$k.', original wordpress article">more...</a>]<br />';
}
public function cut_text($perc=0.85) {
//cut off article at N%
if(!$perc) $perc=$this->perc;
$this->content = substr($this->content, 0 , round(strlen($this->content) * $perc));
}
}
</count></h3></h2>