<?php
/**
- * @file src/Network/Probe.php
+ * @copyright Copyright (C) 2020, Friendica
+ *
+ * @license GNU AGPL version 3 or any later version
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as
+ * published by the Free Software Foundation, either version 3 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program. If not, see <https://www.gnu.org/licenses/>.
+ *
*/
-namespace Friendica\Network;
-/**
- * @file src/Network/Probe.php
- * @brief Functions for probing URL
- */
+namespace Friendica\Network;
use DOMDocument;
use DomXPath;
use Friendica\Core\Cache\Duration;
-use Friendica\Core\Config;
use Friendica\Core\Logger;
use Friendica\Core\Protocol;
+use Friendica\Core\System;
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;
use Friendica\Util\XML;
/**
- * @brief This class contain functions for probing URL
- *
+ * This class contain functions for probing URL
*/
class Probe
{
private static $istimeout;
/**
- * @brief Rearrange the array so that it always has the same order
+ * Remove stuff from an URI that doesn't belong there
+ *
+ * @param string $URI
+ * @return string Cleaned URI
+ */
+ public static function cleanURI(string $URI)
+ {
+ // At first remove leading and trailing junk
+ $URI = trim($URI, "@#?:/ \t\n\r\0\x0B");
+
+ $parts = parse_url($URI);
+
+ if (empty($parts['scheme'])) {
+ return $URI;
+ }
+
+ // Remove the URL fragment, since these shouldn't be part of any profile URL
+ unset($parts['fragment']);
+
+ $URI = Network::unparseURL($parts);
+
+ return $URI;
+ }
+
+ /**
+ * Rearrange the array so that it always has the same order
*
* @param array $data Unordered data
*
private static function rearrangeData($data)
{
$fields = ["name", "nick", "guid", "url", "addr", "alias", "photo", "account-type",
- "community", "keywords", "location", "about", "gender", "hide",
- "batch", "notify", "poll", "request", "confirm", "poco",
+ "community", "keywords", "location", "about", "hide",
+ "batch", "notify", "poll", "request", "confirm", "subscribe", "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;
}
}
}
/**
- * @brief Check if the hostname belongs to the own server
+ * Check if the hostname belongs to the own server
*
* @param string $host The hostname that is to be checked
*
}
/**
- * @brief Probes for webfinger path via "host-meta"
+ * Probes for webfinger path via "host-meta"
*
* We have to check if the servers in the future still will offer this.
* It seems as if it was dropped from the standard.
// Reset the static variable
self::$baseurl = '';
- $ssl_url = "https://".$host."/.well-known/host-meta";
- $url = "http://".$host."/.well-known/host-meta";
+ // Handles the case when the hostname contains the scheme
+ if (!parse_url($host, PHP_URL_SCHEME)) {
+ $ssl_url = "https://" . $host . "/.well-known/host-meta";
+ $url = "http://" . $host . "/.well-known/host-meta";
+ } else {
+ $ssl_url = $host . "/.well-known/host-meta";
+ $url = '';
+ }
- $xrd_timeout = Config::get('system', 'xrd_timeout', 20);
+ $xrd_timeout = DI::config()->get('system', 'xrd_timeout', 20);
- Logger::log("Probing for ".$host, Logger::DEBUG);
+ Logger::info('Probing', ['host' => $host, 'ssl_url' => $ssl_url, 'url' => $url, 'callstack' => System::callstack(20)]);
$xrd = null;
$curlResult = Network::curl($ssl_url, false, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']);
$ssl_connection_error = ($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;
- return false;
+ return [];
}
- if (!is_object($xrd)) {
+ if (!is_object($xrd) && !empty($url)) {
$curlResult = Network::curl($url, false, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']);
$connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
if ($curlResult->isTimeout()) {
Logger::info('Probing timeout', ['url' => $url], Logger::DEBUG);
self::$istimeout = true;
- return false;
+ return [];
} elseif ($connection_error && $ssl_connection_error) {
self::$istimeout = true;
- return false;
+ return [];
}
$xml = $curlResult->getBody();
- $xrd = XML::parseString($xml, false);
+ $xrd = XML::parseString($xml, true);
$host_url = 'http://'.$host;
}
if (!is_object($xrd)) {
}
/**
- * @brief Perform Webfinger lookup and return DFRN data
+ * Perform Webfinger lookup and return DFRN data
*
* Given an email style address, perform webfinger lookup and
* return the resulting DFRN profile URL, or if no DFRN profile URL
}
/**
- * @brief Check an URI for LRDD data
+ * Check an URI for LRDD data
*
- * this is a replacement for the "lrdd" function.
- * It isn't used in this class and has some redundancies in the code.
- * When time comes we can check the existing calls for "lrdd" if we can rework them.
- *
- * @param string $uri Address that should be probed
+ * @param string $uri Address that should be probed
*
* @return array uri data
* @throws HTTPException\InternalServerErrorException
*/
- public static function lrdd($uri)
+ public static function lrdd(string $uri)
{
$lrdd = self::hostMeta($uri);
$webfinger = null;
return [];
}
- $host = $parts["host"];
+ $host = $parts['scheme'] . '://' . $parts["host"];
if (!empty($parts["port"])) {
$host .= ':'.$parts["port"];
}
}
}
- if (!is_array($webfinger["links"])) {
+ if (empty($webfinger["links"])) {
Logger::log("No webfinger links found for ".$uri, Logger::DEBUG);
- return false;
+ return [];
}
$data = [];
$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",
}
/**
- * @brief Fetch information (protocol endpoints and user information) about a given uri
+ * Fetch information (protocol endpoints and user information) about a given uri
*
* @param string $uri Address that should be probed
* @param string $network Test for this specific network
$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;
}
}
/**
- * @brief Checks if a profile url should be OStatus but only provides partial information
+ * Checks if a profile url should be OStatus but only provides partial information
*
* @param array $webfinger Webfinger data
* @param string $lrdd Path template for webfinger request
}
/**
- * @brief Fetch information (protocol endpoints and user information) about a given uri
+ * Fetch the "subscribe" and add it to the result
+ *
+ * @param array $result
+ * @param array $webfinger
+ * @return array result
+ */
+ private static function getSubscribeLink(array $result, array $webfinger)
+ {
+ if (empty($webfinger['links'])) {
+ return $result;
+ }
+
+ foreach ($webfinger['links'] as $link) {
+ if ($link['rel'] === ActivityNamespace::OSTATUSSUB) {
+ $result['subscribe'] = $link['template'];
+ }
+ }
+
+ return $result;
+ }
+
+ /**
+ * Fetch information (protocol endpoints and user information) about a given uri
*
* This function is only called by the "uri" function that adds caching and rearranging of data.
*
$addr = $uri;
} else {
Logger::log("Uri ".$uri." was not detectable", Logger::DEBUG);
- return false;
+ return [];
}
$webfinger = false;
return self::feed($uri);
}
- $result = false;
+ $result = [];
- Logger::log("Probing ".$uri, Logger::DEBUG);
+ Logger::info("Probing", ['uri' => $uri]);
if (in_array($network, ["", Protocol::DFRN])) {
$result = self::dfrn($webfinger);
}
}
+ $result = self::getSubscribeLink($result, $webfinger);
+
if (empty($result["network"])) {
$result["network"] = Protocol::PHANTOM;
}
$result["baseurl"] = substr($result["url"], 0, $pos).$host;
}
}
+
return $result;
}
if (!empty($profile['description'])) {
$data['about'] = $profile['description'];
}
- if (!empty($profile['gender'])) {
- $data['gender'] = $profile['gender'];
- }
if (!empty($profile['keywords'])) {
$keywords = implode(', ', $profile['keywords']);
if (!empty($keywords)) {
if (!empty($profile['country'])) {
$loc['country-name'] = $profile['country'];
}
- if (!empty($profile['hometown'])) {
- $loc['locality'] = $profile['hometown'];
- }
$location = Profile::formatLocation($loc);
if (!empty($location)) {
$data['location'] = $location;
}
/**
- * @brief Perform a webfinger request.
+ * Perform a webfinger request.
*
* For details see RFC 7033: <https://tools.ietf.org/html/rfc7033>
*
* @return array webfinger data
* @throws HTTPException\InternalServerErrorException
*/
- private static function webfinger($url, $type)
+ public static function webfinger($url, $type)
{
- $xrd_timeout = Config::get('system', 'xrd_timeout', 20);
+ $xrd_timeout = DI::config()->get('system', 'xrd_timeout', 20);
$curlResult = Network::curl($url, false, ['timeout' => $xrd_timeout, 'accept_content' => $type]);
if ($curlResult->isTimeout()) {
self::$istimeout = true;
- return false;
+ return [];
}
$data = $curlResult->getBody();
$webfinger = json_decode($data, true);
- if (is_array($webfinger)) {
+ if (!empty($webfinger)) {
if (!isset($webfinger["links"])) {
Logger::log("No json webfinger links for ".$url, Logger::DEBUG);
- return false;
+ return [];
}
return $webfinger;
}
// 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;
+ return [];
}
$xrd_arr = XML::elementToArray($xrd);
if (!isset($xrd_arr["xrd"]["link"])) {
Logger::log("No XML webfinger links for ".$url, Logger::DEBUG);
- return false;
+ return [];
}
$webfinger = [];
}
/**
- * @brief Poll the Friendica specific noscrape page.
+ * Poll the Friendica specific noscrape page.
*
* "noscrape" is a faster alternative to fetch the data from the hcard.
* This functionality was originally created for the directory.
$curlResult = Network::curl($noscrape_url);
if ($curlResult->isTimeout()) {
self::$istimeout = true;
- return false;
+ return [];
}
$content = $curlResult->getBody();
if (!$content) {
Logger::log("Empty body for ".$noscrape_url, Logger::DEBUG);
- return false;
+ return [];
}
$json = json_decode($content, true);
if (!is_array($json)) {
Logger::log("No json data for ".$noscrape_url, Logger::DEBUG);
- return false;
+ return [];
}
if (!empty($json["fn"])) {
$data["about"] = $json["about"];
}
- if (!empty($json["gender"])) {
- $data["gender"] = $json["gender"];
- }
-
if (!empty($json["key"])) {
$data["pubkey"] = $json["key"];
}
}
/**
- * @brief Check for valid DFRN data
+ * Check for valid DFRN data
*
* @param array $data DFRN data
*
}
/**
- * @brief Fetch data from a DFRN profile page and via "noscrape"
+ * Fetch data from a DFRN profile page and via "noscrape"
*
* @param string $profile_link Link to the profile page
*
}
/**
- * @brief Check for DFRN contact
+ * Check for DFRN contact
*
* @param array $webfinger Webfinger data
*
}
if (!isset($data["network"]) || ($hcard_url == "")) {
- return false;
+ return [];
}
// Fetch data via noscrape - this is faster
}
/**
- * @brief Poll the hcard page (Diaspora and Friendica specific)
+ * Poll the hcard page (Diaspora and Friendica specific)
*
* @param string $hcard_url Link to the hcard page
* @param array $data The already fetched data
$curlResult = Network::curl($hcard_url);
if ($curlResult->isTimeout()) {
self::$istimeout = true;
- return false;
+ return [];
}
$content = $curlResult->getBody();
if (!$content) {
- return false;
+ return [];
}
$doc = new DOMDocument();
if (!@$doc->loadHTML($content)) {
- return false;
+ return [];
}
$xpath = new DomXPath($doc);
$vcards = $xpath->query("//div[contains(concat(' ', @class, ' '), ' vcard ')]");
if (!is_object($vcards)) {
- return false;
+ return [];
}
if (!isset($data["baseurl"])) {
}
/**
- * @brief Check for Diaspora contact
+ * Check for Diaspora contact
*
* @param array $webfinger Webfinger data
*
}
if (empty($data["url"]) || empty($hcard_url)) {
- return false;
+ return [];
}
if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) {
$data = self::pollHcard($hcard_url, $data);
if (!$data) {
- return false;
+ return [];
}
if (!empty($data["url"])
$data["notify"] = $data["baseurl"] . "/receive/users/" . $data["guid"];
$data["batch"] = $data["baseurl"] . "/receive/public";
} else {
- return false;
+ return [];
}
return $data;
}
/**
- * @brief Check for OStatus contact
+ * Check for OStatus contact
*
* @param array $webfinger Webfinger data
* @param bool $short Short detection mode
$data["addr"] = str_replace('acct:', '', $webfinger["subject"]);
}
- if (is_array($webfinger["links"])) {
+ if (!empty($webfinger["links"])) {
// The array is reversed to take into account the order of preference for same-rel links
// See: https://tools.ietf.org/html/rfc7033#section-4.4.4
foreach (array_reverse($webfinger["links"]) as $link) {
$curlResult = Network::curl($pubkey);
if ($curlResult->isTimeout()) {
self::$istimeout = true;
- return false;
+ return $short ? false : [];
}
$pubkey = $curlResult->getBody();
}
) {
$data["network"] = Protocol::OSTATUS;
} else {
- return false;
+ return $short ? false : [];
}
if ($short) {
$curlResult = Network::curl($data["poll"]);
if ($curlResult->isTimeout()) {
self::$istimeout = true;
- return false;
+ return [];
}
$feed = $curlResult->getBody();
$feed_data = Feed::import($feed);
if (!$feed_data) {
- return false;
+ return [];
}
if (!empty($feed_data["header"]["author-name"])) {
}
/**
- * @brief Fetch data from a pump.io profile page
+ * Fetch data from a pump.io profile page
*
* @param string $profile_link Link to the profile page
*
{
$curlResult = Network::curl($profile_link);
if (!$curlResult->isSuccess()) {
- return false;
+ return [];
}
$doc = new DOMDocument();
if (!@$doc->loadHTML($curlResult->getBody())) {
- return false;
+ return [];
}
$xpath = new DomXPath($doc);
}
/**
- * @brief Check for pump.io contact
+ * Check for pump.io contact
*
* @param array $webfinger Webfinger data
* @param string $addr
$data["network"] = Protocol::PUMPIO;
} else {
- return false;
+ return [];
}
$profile_data = self::pumpioProfileData($data["url"]);
if (!$profile_data) {
- return false;
+ return [];
}
$data = array_merge($data, $profile_data);
}
/**
- * @brief Check for twitter contact
+ * Check for twitter contact
*
* @param string $uri
*
}
/**
- * @brief Check page for feed link
- *
- * @param string $url Page link
+ * Checks HTML page for RSS feed 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);
}
/**
- * @brief Check for feed contact
+ * Check for feed contact
*
* @param string $url Profile link
* @param boolean $probe Do a probe if the page contains a feed link
$curlResult = Network::curl($url);
if ($curlResult->isTimeout()) {
self::$istimeout = true;
- return false;
+ return [];
}
$feed = $curlResult->getBody();
$feed_data = Feed::import($feed);
if (!$feed_data) {
if (!$probe) {
- return false;
+ return [];
}
- $feed_url = self::getFeedLink($url);
+ $feed_url = self::getFeedLink($url, $feed);
if (!$feed_url) {
- return false;
+ return [];
}
return self::feed($feed_url, false);
}
/**
- * @brief Check for mail contact
+ * Check for mail contact
*
* @param string $uri Profile link
* @param integer $uid User ID
private static function mail($uri, $uid)
{
if (!Network::isEmailDomainValid($uri)) {
- return false;
+ return [];
}
if ($uid == 0) {
- return false;
+ return [];
}
$user = DBA::selectFirst('user', ['prvkey'], ['uid' => $uid]);
$mailacct = DBA::selectFirst('mailacct', $fields, $condition);
if (!DBA::isResult($user) || !DBA::isResult($mailacct)) {
- return false;
+ return [];
}
$mailbox = Email::constructMailboxName($mailacct);
openssl_private_decrypt(hex2bin($mailacct['pass']), $password, $user['prvkey']);
$mbox = Email::connect($mailbox, $mailacct['user'], $password);
if (!$mbox) {
- return false;
+ return [];
}
$msgs = Email::poll($mbox, $uri);
Logger::log('searching '.$uri.', '.count($msgs).' messages found.', Logger::DEBUG);
if (!count($msgs)) {
- return false;
+ return [];
}
$phost = substr($uri, strpos($uri, '@') + 1);
}
/**
- * @brief Mix two paths together to possibly fix missing parts
+ * Mix two paths together to possibly fix missing parts
*
* @param string $avatar Path to the avatar
* @param string $base Another path that is hopefully complete
projects / friendica.git / blobdiff