use Friendica\Database\DBA;
use Friendica\DI;
use Friendica\Model\Contact;
+use Friendica\Model\GServer;
use Friendica\Model\Profile;
use Friendica\Protocol\ActivityNamespace;
use Friendica\Protocol\ActivityPub;
"community", "keywords", "location", "about", "hide",
"batch", "notify", "poll", "request", "confirm", "poco",
"following", "followers", "inbox", "outbox", "sharedinbox",
- "priority", "network", "pubkey", "baseurl"];
+ "priority", "network", "pubkey", "baseurl", "gsid"];
$newdata = [];
foreach ($fields as $field) {
if (isset($data[$field])) {
$newdata[$field] = $data[$field];
- } else {
+ } elseif ($field != "gsid") {
$newdata[$field] = "";
+ } else {
+ $newdata[$field] = null;
}
}
$ssl_connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
if ($curlResult->isSuccess()) {
$xml = $curlResult->getBody();
- $xrd = XML::parseString($xml, false);
- $host_url = 'https://'.$host;
+ $xrd = XML::parseString($xml, true);
+ if (!empty($url)) {
+ $host_url = 'https://' . $host;
+ } else {
+ $host_url = $host;
+ }
} elseif ($curlResult->isTimeout()) {
Logger::info('Probing timeout', ['url' => $ssl_url], Logger::DEBUG);
self::$istimeout = true;
}
$xml = $curlResult->getBody();
- $xrd = XML::parseString($xml, false);
+ $xrd = XML::parseString($xml, true);
$host_url = 'http://'.$host;
}
if (!is_object($xrd)) {
$data[] = ["@attributes" => $link];
}
- if (is_array($webfinger["aliases"])) {
+ if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) {
foreach ($webfinger["aliases"] as $alias) {
$data[] = ["@attributes" =>
["rel" => "alias",
$data['url'] = $uri;
}
- if (!empty($data['photo']) && !empty($data['baseurl'])) {
- $data['baseurl'] = Network::getUrlMatch(Strings::normaliseLink($data['baseurl']), Strings::normaliseLink($data['photo']));
- } elseif (empty($data['photo'])) {
+ if (empty($data['photo'])) {
$data['photo'] = DI::baseUrl() . '/images/person-300.jpg';
}
$data['baseurl'] = self::$baseurl;
}
+ if (!empty($data['baseurl']) && empty($data['gsid'])) {
+ $data['gsid'] = GServer::getID($data['baseurl']);
+ }
+
if (empty($data['network'])) {
$data['network'] = Protocol::PHANTOM;
}
}
// If it is not JSON, maybe it is XML
- $xrd = XML::parseString($data, false);
+ $xrd = XML::parseString($data, true);
if (!is_object($xrd)) {
Logger::log("No webfinger data retrievable for ".$url, Logger::DEBUG);
return false;
}
/**
- * Check page for feed link
+ * Checks HTML page for RSS feed link
*
- * @param string $url Page link
- *
- * @return string feed link
+ * @param string $url Page link
+ * @param string $body Page body string
+ * @return string|false Feed link or false if body was invalid HTML document
*/
- private static function getFeedLink($url)
+ public static function getFeedLink(string $url, string $body)
{
- $curlResult = Network::curl($url);
- if (!$curlResult->isSuccess()) {
- return false;
- }
-
$doc = new DOMDocument();
- if (!@$doc->loadHTML($curlResult->getBody())) {
+ if (!@$doc->loadHTML($body)) {
return false;
}
- $xpath = new DomXPath($doc);
+ $xpath = new DOMXPath($doc);
- //$feeds = $xpath->query("/html/head/link[@type='application/rss+xml']");
- $feeds = $xpath->query("/html/head/link[@type='application/rss+xml' and @rel='alternate']");
- if (!is_object($feeds)) {
- return false;
- }
+ $feedUrl = $xpath->evaluate('string(/html/head/link[@type="application/rss+xml" and @rel="alternate"]/@href)');
- if ($feeds->length == 0) {
- return false;
+ $feedUrl = $feedUrl ? self::ensureAbsoluteLinkFromHTMLDoc($feedUrl, $url, $xpath) : '';
+
+ return $feedUrl;
+ }
+
+ /**
+ * Return an absolute URL in the context of a HTML document retrieved from the provided URL.
+ *
+ * Loosely based on RFC 1808
+ *
+ * @see https://tools.ietf.org/html/rfc1808
+ *
+ * @param string $href The potential relative href found in the HTML document
+ * @param string $base The HTML document URL
+ * @param DOMXPath $xpath The HTML document XPath
+ * @return string
+ */
+ private static function ensureAbsoluteLinkFromHTMLDoc(string $href, string $base, DOMXPath $xpath)
+ {
+ if (filter_var($href, FILTER_VALIDATE_URL)) {
+ return $href;
}
- $feed_url = "";
+ $base = $xpath->evaluate('string(/html/head/base/@href)') ?: $base;
- foreach ($feeds as $feed) {
- $attr = [];
- foreach ($feed->attributes as $attribute) {
- $attr[$attribute->name] = trim($attribute->value);
- }
+ $baseParts = parse_url($base);
+
+ // Naked domain case (scheme://basehost)
+ $path = $baseParts['path'] ?? '/';
+
+ // Remove the filename part of the path if it exists (/base/path/file)
+ $path = implode('/', array_slice(explode('/', $path), 0, -1));
+
+ $hrefParts = parse_url($href);
- if (empty($feed_url) && !empty($attr['href'])) {
- $feed_url = $attr["href"];
+ // Root path case (/path) including relative scheme case (//host/path)
+ if ($hrefParts['path'] && $hrefParts['path'][0] == '/') {
+ $path = $hrefParts['path'];
+ } else {
+ $path = $path . '/' . $hrefParts['path'];
+
+ // Resolve arbitrary relative path
+ // Lifted from https://www.php.net/manual/en/function.realpath.php#84012
+ $parts = array_filter(explode('/', $path), 'strlen');
+ $absolutes = array();
+ foreach ($parts as $part) {
+ if ('.' == $part) continue;
+ if ('..' == $part) {
+ array_pop($absolutes);
+ } else {
+ $absolutes[] = $part;
+ }
}
+
+ $path = '/' . implode('/', $absolutes);
}
- return $feed_url;
+ // Relative scheme case (//host/path)
+ $baseParts['host'] = $hrefParts['host'] ?? $baseParts['host'];
+ $baseParts['path'] = $path;
+ unset($baseParts['query']);
+ unset($baseParts['fragment']);
+
+ return Network::unparseURL($baseParts);
}
/**
return false;
}
- $feed_url = self::getFeedLink($url);
+ $feed_url = self::getFeedLink($url, $feed);
if (!$feed_url) {
return false;