miniflux-legacy/vendor/fguillot/picofeed/lib/PicoFeed/Scraper/RuleParser.php

94 lines
2.0 KiB
PHP
Raw Normal View History

2015-04-28 18:08:42 +02:00
<?php
namespace PicoFeed\Scraper;
use DOMXPath;
use PicoFeed\Parser\XmlParser;
/**
* Rule Parser
*
* @author Frederic Guillot
* @package Scraper
*/
class RuleParser implements ParserInterface
{
private $dom;
private $xpath;
private $rules = array();
/**
* Constructor
*
* @access public
* @param string $html
* @param array $rules
*/
public function __construct($html, array $rules)
{
$this->rules = $rules;
$this->dom = XmlParser::getHtmlDocument('<?xml version="1.0" encoding="UTF-8">'.$html);
$this->xpath = new DOMXPath($this->dom);
}
/**
* Get the relevant content with predefined rules
*
* @access public
* @return string
*/
public function execute()
{
$this->stripTags();
return $this->findContent();
}
/**
* Remove HTML tags
*
* @access public
*/
public function stripTags()
{
if (isset($this->rules['strip']) && is_array($this->rules['strip'])) {
foreach ($this->rules['strip'] as $pattern) {
$nodes = $this->xpath->query($pattern);
if ($nodes !== false && $nodes->length > 0) {
foreach ($nodes as $node) {
$node->parentNode->removeChild($node);
}
}
}
}
}
/**
* Fetch content based on Xpath rules
*
* @access public
*/
public function findContent()
{
$content = '';
if (isset($this->rules['body']) && is_array($this->rules['body'])) {
foreach ($this->rules['body'] as $pattern) {
$nodes = $this->xpath->query($pattern);
if ($nodes !== false && $nodes->length > 0) {
foreach ($nodes as $node) {
$content .= $this->dom->saveXML($node);
}
}
}
}
return $content;
}
}