private static function mf2parse($content, $url)
{
$data = array();
$host = extract_domain_name($url);
switch ($host) {
case 'twitter.com':
$parsed = Mf2\Shim\parseTwitter($content, $url);
break;
default:
$parsed = Mf2\parse($content, $url);
}
if (mf2_cleaner::isMicroformatCollection($parsed)) {
$entries = mf2_cleaner::findMicroformatsByType($parsed, 'h-entry');
if ($entries) {
$entry = $entries[0];
if (mf2_cleaner::isMicroformat($entry)) {
foreach ($entry['properties'] as $key => $value) {
$data[$key] = mf2_cleaner::getPlaintext($entry, $key);
}
$data['published'] = mf2_cleaner::getPublished($entry);
$data['updated'] = mf2_cleaner::getUpdated($entry);
$data['name'] = mf2_cleaner::getPlaintext($entry, 'name');
$data['content'] = mf2_cleaner::getHtml($entry, 'content');
$data['summary'] = mf2_cleaner::getHtml($entry, 'summary');
$data['name'] = trim(preg_replace('/https?:\\/\\/([^ ]+|$)/', '', $data['name']));
$author = mf2_cleaner::getAuthor($entry);
if ($author) {
$data['author'] = array();
foreach ($author['properties'] as $key => $value) {
$data['author'][$key] = mf2_cleaner::getPlaintext($author, $key);
}
$data['author'] = array_filter($data['author']);
}
}
}
}
return array_filter($data);
}
private static function parseTwitter($html, $url)
{
$parsed = \Mf2\Shim\parseTwitter($html, $url);
if (!empty($parsed['items'][0]['properties']['content'][0]['html'])) {
$content = $parsed['items'][0]['properties']['content'][0]['html'];
$config = \HTMLPurifier_Config::createDefault();
$config->set('URI.Base', $url);
$config->set('URI.MakeAbsolute', true);
$config->set('HTML.Allowed', 'a[href],img[src],p,br');
$purifier = new \HTMLPurifier($config);
$content = $purifier->purify($content);
$parsed['items'][0]['properties']['content'][0]['html'] = $content;
}
return $parsed;
}
