*/
function header_extract_charset($header)
{
- preg_match('/charset="?([^; ]+)/i', $header, $match);
+ preg_match('/charset=["\']?([^; "\']+)/i', $header, $match);
if (! empty($match[1])) {
return strtolower(trim($match[1]));
}
{
$propertiesKey = ['property', 'name', 'itemprop'];
$properties = implode('|', $propertiesKey);
+ // We need a OR here to accept either 'property=og:noquote' or 'property="og:unrelated og:my-tag"'
+ $orCondition = '["\']?(?:og:)?'. $tag .'["\']?|["\'][^\'"]*?(?:og:)?' . $tag . '[^\'"]*?[\'"]';
// Try to retrieve OpenGraph image.
- $ogRegex = '#<meta[^>]+(?:'. $properties .')=["\']?(?:og:)?'. $tag .'["\'\s][^>]*content=["\']?(.*?)["\'/>]#';
+ $ogRegex = '#<meta[^>]+(?:'. $properties .')=(?:'. $orCondition .')[^>]*content=["\'](.*?)["\'].*?>#';
// If the attributes are not in the order property => content (e.g. Github)
// New regex to keep this readable... more or less.
- $ogRegexReverse = '#<meta[^>]+content=["\']([^"\']+)[^>]+(?:'. $properties .')=["\']?(?:og)?:'. $tag .'["\'\s/>]#';
+ $ogRegexReverse = '#<meta[^>]+content=["\'](.*?)["\'][^>]+(?:'. $properties .')=(?:'. $orCondition .').*?>#';
if (preg_match($ogRegex, $html, $matches) > 0
|| preg_match($ogRegexReverse, $html, $matches) > 0