Expresión regular que he utilizado para procesar el índice de telediarios de RTVE.

regex_telediarios.php

<?php

const REGEX = [
		'<span class="col_tit" id=".+?" name="progname">',
			'<a href="(?P<url>.+?)">',
			'(<em>Nuevo<\/em>&nbsp;)?',
			'(?P<title>.+?)',
			'<\/a>',
		'<\/span>',
		'<span class="col_tip">',
			'<span>(?P<type>.+?)<\/span>',
		'<\/span>',
		'<span class="col_dur">(?P<duration>.+?)<\/span>',
		'<span class="col_pop">.+?<\/span>',
		'<span class="col_fec">(?P<date>.+?)<\/span>',
		'<div id=".+?" class="tultip hddn">',
			'.*?',
			'<span class="detalle">',
			'(?P<description>.+?)',
			'<\/span>',
			'.*?',
		'<\/div>',
	'<\/li>',
];

$regex = sprintf('/%s/', implode('\s*', REGEX));