Xml SAX implementation example

01_example.php

<?php

$reader = new ExampleDataSource("posts.xml");

foreach ($reader as $post) {
  // $post e' un array con chiavi 'title' e 'body'
}

02_posts.xml

<?xml>
<results>
  <post>
    <titleStrangeTag>Title</titleStrangeTag>
    <content>Body of the post</content>
  </post>
  <post>
    <titleStrangeTag>Title 2</titleStrangeTag>
    <content>Body of the post 2</content>
  </post>
</results>

04_AbstractXmlDataSource.php

<?php

/**
 * Description of AbstractXmlDataSource
 *
 * @author flevour
 */
abstract class AbstractXmlDataSource implements \Iterator
{

    protected $reader;
    protected $tagName;

    public function __construct($dataFile)
    {
        $this->tagName = $this->getTagName();
        $this->reader = new \XMLReader;
        $this->reader->open($dataFile);
        // move to the first <TAGNAME /> node
        while ($this->reader->read() && $this->reader->name !== $this->tagName)
            ;
    }

    /**
     * Defines the tag to loop for in the XML file.
     * 
     * @return string the tag name (case sensitive) containing a single information in the XML.
     */
    abstract public function getTagName();

    /**
     * Defines a mapper from a SimpleXMLElement to an array containing the extracted information.
     * 
     * @return array a keyed array containing the data extracted from the current node.
     */
    abstract protected function mapData(\SimpleXMLElement $node);

    /**
     * Implementazione Iterator
     */
    public function current()
    {
        $doc = new \DOMDocument;
        $node = simplexml_import_dom($doc->importNode($this->reader->expand(), true));

        $data = $this->mapData($node);

        return $data;
    }

    public function next()
    {
        // go to next <TAG_NAME />
        $this->reader->next($this->tagName);
    }

    /**
     * Quando l'xml e' stato passato tutto, reader->name sara' nullo pertanto anche
     * questo iteratore deve terminare.
     * @return boolean
     */
    public function valid()
    {
        return $this->reader->name === $this->tagName;
    }

    /**
     * Non implementato, non ha senso parlare di una chiave in questo contesto.
     * Il curorse nell'xml puo' solo avanzare.
     */
    public function key()
    {
        
    }

    /**
     * Il curorse nell'xml puo' solo avanzare.
     */
    public function rewind()
    {
        
    }

}

04_ExampleDataSource.php

<?php
/**
 * Description of ExampleDataSource
 *
 * @author flevour
 */
class ExampleDataSource extends AbstractXmlDataSource
{

    public function getTagName()
    {
        return 'post';
    }

    protected function mapData(\SimpleXMLElement $node)
    {
        // now you can use $node without going insane about parsing
        return array(
                'title' => trim((string) $node->titleStrangeTag),
                'body' => (string) $node->content,
        );
    }

}