/**
* Extract images URL from a blog HTML content.
*
* @access public
* @param String HTML content that can content fom <img /> XHTML element
* @return Array Return an array of the images URL founds.
*/
public static function getMediaURLfromHTML($text = NULL)
{
$media_ = array();
if (strlen(trim($text)) > 0) {
$tt = preg_match_all('/<(source|iframe|embed|param|img)[^>]+src=[\'"]([^\'"]+)[\'"].*>/i', str_replace('><', '>
<', FeedValidator::removeBreaklines($text, '
')), $matches);
if ($tt > 0 && count($matches[2]) > 0) {
foreach ($matches[2] as $i => $value) {
$sb1 = array();
$sb2 = array();
$sb3 = array();
$sb4 = array();
if (FeedValidator::isValidURL($value)) {
$simple_tag = str_replace("'", "\"", strtolower(stripcslashes($matches[0][$i])));
$sb1 = explode('title="', $simple_tag);
if (count($sb1) > 1) {
$sb3 = explode('"', $sb1[1]);
}
$sb2 = explode('alt="', $simple_tag);
if (count($sb2) > 1) {
$sb4 = explode('"', $sb2[1]);
}
$media_type = FeedValidator::getMediaType($value);
array_push($media_, array('uri' => $value, 'type' => $media_type, 'name' => isset($sb3[0]) ? $sb3[0] : (isset($sb4[0]) ? $sb4[0] : $media_type . " - " . $i)));
}
}
}
// Strip HTML content and analyzed individual world to find URL in the text. (CF: MediaWiki content).
$text_split = explode(' ', FeedValidator::getOnlyText($text, self::CHARSET));
if (count($text_split) > 0) {
foreach ($text_split as $value) {
foreach (array('image', 'sound', 'video') as $media_type) {
if (FeedValidator::isValidURL($value)) {
if (FeedValidator::isValidMediaURL($value, $media_type)) {
if (!in_array($value, $media_)) {
array_push($media_, array('uri' => $value, 'type' => $media_type, 'name' => $media_type));
}
}
}
}
}
}
}
return $media_;
}
