bla.php

<?php

    /**
     * The following two methods implements plain text parsing.
     * @see https://wiki.zegnat.net/media/textparsing.html
     **/
	public function textContent(DOMElement $element)
	{
        return preg_replace(
            '/( +(?=\n)|(?<=\n) +|^[\t\n\f\r ]+|[\t\n\f\r ]+$| +(?= ))/',
            '',
            $this->elementToString($element)
        );
	}
	private function elementToString(DOMElement $input)
	{
	    $output = '';
	    foreach ($input->childNodes as $child) {
	        if ($child->nodeType === XML_TEXT_NODE) {
	            $output .= str_replace(array("\t", "\n", "\r") , ' ', $child->textContent);
	        } else if ($child->nodeType === XML_ELEMENT_NODE) {
	            $tagName = strtoupper($child->tagName);
	            if (in_array($tagName, array('SCRIPT', 'STYLE'))) {
	                continue;
	            } else if ($tagName === 'IMG') {
	                if ($child->hasAttribute('alt')) {
	                    $output .= ' ' . trim($child->getAttribute('alt'), "\t\n\f\r ") . ' ';
	                } else if ($child->hasAttribute('src')) {
	                    $output .= ' ' . $this->resolveUrl(trim($child->getAttribute('src'), "\t\n\f\r ")) . ' ';
	                }
	            } else if ($tagName === 'BR') {
	                $output .= "\n";
	            } else if ($tagName === 'P') {
	                $output .= "\n" . $this->elementToString($child);
	            } else {
	                $output .= $this->elementToString($child);
	            }
	        }
	    }
	    return $output;
	}