<?php
/**
- * @file src/Util/Network.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\Util;
+use DOMDocument;
+use DomXPath;
use Friendica\Core\Hook;
use Friendica\Core\Logger;
use Friendica\Core\System;
-use Friendica\Core\Config;
+use Friendica\DI;
use Friendica\Network\CurlResult;
-use DOMDocument;
-use DomXPath;
class Network
{
* Set the cookiejar argument to a string (e.g. "/tmp/friendica-cookies.txt")
* to preserve cookies from one request to the next.
*
- * @brief Curl wrapper
* @param string $url URL to fetch
- * @param boolean $binary default false
+ * @param bool $binary default false
* TRUE if asked to return binary results (file download)
- * @param integer $redirects The recursion counter for internal use - default 0
- * @param integer $timeout Timeout in seconds, default system config value or 60 seconds
+ * @param int $timeout Timeout in seconds, default system config value or 60 seconds
* @param string $accept_content supply Accept: header with 'accept_content' as the value
* @param string $cookiejar Path to cookie jar file
+ * @param int $redirects The recursion counter for internal use - default 0
*
* @return string The fetched content
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function fetchUrl($url, $binary = false, &$redirects = 0, $timeout = 0, $accept_content = null, $cookiejar = '')
+ public static function fetchUrl(string $url, bool $binary = false, int $timeout = 0, string $accept_content = '', string $cookiejar = '', int &$redirects = 0)
{
- $ret = self::fetchUrlFull($url, $binary, $redirects, $timeout, $accept_content, $cookiejar);
+ $ret = self::fetchUrlFull($url, $binary, $timeout, $accept_content, $cookiejar, $redirects);
return $ret->getBody();
}
* Inner workings and parameters are the same as @ref fetchUrl but returns an array with
* all the information collected during the fetch.
*
- * @brief Curl wrapper with array of return values.
* @param string $url URL to fetch
- * @param boolean $binary default false
+ * @param bool $binary default false
* TRUE if asked to return binary results (file download)
- * @param integer $redirects The recursion counter for internal use - default 0
- * @param integer $timeout Timeout in seconds, default system config value or 60 seconds
+ * @param int $timeout Timeout in seconds, default system config value or 60 seconds
* @param string $accept_content supply Accept: header with 'accept_content' as the value
* @param string $cookiejar Path to cookie jar file
+ * @param int $redirects The recursion counter for internal use - default 0
*
* @return CurlResult With all relevant information, 'body' contains the actual fetched content.
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function fetchUrlFull($url, $binary = false, &$redirects = 0, $timeout = 0, $accept_content = null, $cookiejar = '')
+ public static function fetchUrlFull(string $url, bool $binary = false, int $timeout = 0, string $accept_content = '', string $cookiejar = '', int &$redirects = 0)
{
return self::curl(
$url,
$binary,
- $redirects,
- ['timeout'=>$timeout,
- 'accept_content'=>$accept_content,
- 'cookiejar'=>$cookiejar
- ]
+ [
+ 'timeout' => $timeout,
+ 'accept_content' => $accept_content,
+ 'cookiejar' => $cookiejar
+ ],
+ $redirects
);
}
/**
- * @brief fetches an URL.
+ * fetches an URL.
*
* @param string $url URL to fetch
- * @param boolean $binary default false
+ * @param bool $binary default false
* TRUE if asked to return binary results (file download)
- * @param int $redirects The recursion counter for internal use - default 0
* @param array $opts (optional parameters) assoziative array with:
* 'accept_content' => supply Accept: header with 'accept_content' as the value
* 'timeout' => int Timeout in seconds, default system config value or 60 seconds
* 'nobody' => only return the header
* 'cookiejar' => path to cookie jar file
* 'header' => header array
+ * @param int $redirects The recursion counter for internal use - default 0
*
* @return CurlResult
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function curl($url, $binary = false, &$redirects = 0, $opts = [])
+ public static function curl(string $url, bool $binary = false, array $opts = [], int &$redirects = 0)
{
- $ret = ['return_code' => 0, 'success' => false, 'header' => '', 'info' => '', 'body' => ''];
-
$stamp1 = microtime(true);
- $a = \get_app();
+ $a = DI::app();
if (strlen($url) > 1000) {
Logger::log('URL is longer than 1000 characters. Callstack: ' . System::callstack(20), Logger::DEBUG);
return CurlResult::createErrorCurl(substr($url, 0, 200));
}
+ $parts2 = [];
$parts = parse_url($url);
- $path_parts = explode('/', defaults($parts, 'path', ''));
+ $path_parts = explode('/', $parts['path'] ?? '');
foreach ($path_parts as $part) {
if (strlen($part) <> mb_strlen($part)) {
$parts2[] = rawurlencode($part);
@curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
@curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
- $range = intval(Config::get('system', 'curl_range_bytes', 0));
+ $range = intval(DI::config()->get('system', 'curl_range_bytes', 0));
if ($range > 0) {
@curl_setopt($ch, CURLOPT_RANGE, '0-' . $range);
if (!empty($opts['timeout'])) {
@curl_setopt($ch, CURLOPT_TIMEOUT, $opts['timeout']);
} else {
- $curl_time = Config::get('system', 'curl_timeout', 60);
+ $curl_time = DI::config()->get('system', 'curl_timeout', 60);
@curl_setopt($ch, CURLOPT_TIMEOUT, intval($curl_time));
}
// by default we will allow self-signed certs
// but you can override this
- $check_cert = Config::get('system', 'verifyssl');
+ $check_cert = DI::config()->get('system', 'verifyssl');
@curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
if ($check_cert) {
@curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, 2);
}
- $proxy = Config::get('system', 'proxy');
+ $proxy = DI::config()->get('system', 'proxy');
if (strlen($proxy)) {
@curl_setopt($ch, CURLOPT_HTTPPROXYTUNNEL, 1);
@curl_setopt($ch, CURLOPT_PROXY, $proxy);
- $proxyuser = @Config::get('system', 'proxyuser');
+ $proxyuser = @DI::config()->get('system', 'proxyuser');
if (strlen($proxyuser)) {
@curl_setopt($ch, CURLOPT_PROXYUSERPWD, $proxyuser);
}
}
- if (Config::get('system', 'ipv4_resolve', false)) {
+ if (DI::config()->get('system', 'ipv4_resolve', false)) {
curl_setopt($ch, CURLOPT_IPRESOLVE, CURL_IPRESOLVE_V4);
}
$redirects++;
Logger::log('curl: redirect ' . $url . ' to ' . $curlResponse->getRedirectUrl());
@curl_close($ch);
- return self::curl($curlResponse->getRedirectUrl(), $binary, $redirects, $opts);
+ return self::curl($curlResponse->getRedirectUrl(), $binary, $opts, $redirects);
}
@curl_close($ch);
- $a->saveTimestamp($stamp1, 'network');
+ DI::profiler()->saveTimestamp($stamp1, 'network', System::callstack());
return $curlResponse;
}
/**
- * @brief Send POST request to $url
+ * Send POST request to $url
*
* @param string $url URL to post
* @param mixed $params array of POST variables
- * @param string $headers HTTP headers
- * @param integer $redirects Recursion counter for internal use - default = 0
- * @param integer $timeout The timeout in seconds, default system config value or 60 seconds
+ * @param array $headers HTTP headers
+ * @param int $redirects Recursion counter for internal use - default = 0
+ * @param int $timeout The timeout in seconds, default system config value or 60 seconds
*
* @return CurlResult The content
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function post($url, $params, $headers = null, &$redirects = 0, $timeout = 0)
+ public static function post(string $url, $params, array $headers = [], int $timeout = 0, int &$redirects = 0)
{
$stamp1 = microtime(true);
return CurlResult::createErrorCurl($url);
}
- $a = \get_app();
+ $a = DI::app();
$ch = curl_init($url);
if (($redirects > 8) || (!$ch)) {
curl_setopt($ch, CURLOPT_POSTFIELDS, $params);
curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent());
- if (Config::get('system', 'ipv4_resolve', false)) {
+ if (DI::config()->get('system', 'ipv4_resolve', false)) {
curl_setopt($ch, CURLOPT_IPRESOLVE, CURL_IPRESOLVE_V4);
}
if (intval($timeout)) {
curl_setopt($ch, CURLOPT_TIMEOUT, $timeout);
} else {
- $curl_time = Config::get('system', 'curl_timeout', 60);
+ $curl_time = DI::config()->get('system', 'curl_timeout', 60);
curl_setopt($ch, CURLOPT_TIMEOUT, intval($curl_time));
}
- if (defined('LIGHTTPD')) {
- if (!is_array($headers)) {
- $headers = ['Expect:'];
- } else {
- if (!in_array('Expect:', $headers)) {
- array_push($headers, 'Expect:');
- }
- }
- }
-
- if ($headers) {
+ if (!empty($headers)) {
curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
}
- $check_cert = Config::get('system', 'verifyssl');
+ $check_cert = DI::config()->get('system', 'verifyssl');
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
if ($check_cert) {
@curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, 2);
}
- $proxy = Config::get('system', 'proxy');
+ $proxy = DI::config()->get('system', 'proxy');
if (strlen($proxy)) {
curl_setopt($ch, CURLOPT_HTTPPROXYTUNNEL, 1);
curl_setopt($ch, CURLOPT_PROXY, $proxy);
- $proxyuser = Config::get('system', 'proxyuser');
+ $proxyuser = DI::config()->get('system', 'proxyuser');
if (strlen($proxyuser)) {
curl_setopt($ch, CURLOPT_PROXYUSERPWD, $proxyuser);
}
$s = @curl_exec($ch);
- $base = $s;
$curl_info = curl_getinfo($ch);
$curlResponse = new CurlResult($url, $s, $curl_info, curl_errno($ch), curl_error($ch));
curl_close($ch);
- $a->saveTimestamp($stamp1, 'network');
+ DI::profiler()->saveTimestamp($stamp1, 'network', System::callstack());
+
+ // Very old versions of Lighttpd don't like the "Expect" header, so we remove it when needed
+ if ($curlResponse->getReturnCode() == 417) {
+ $redirects++;
+
+ if (empty($headers)) {
+ $headers = ['Expect:'];
+ } else {
+ if (!in_array('Expect:', $headers)) {
+ array_push($headers, 'Expect:');
+ }
+ }
+ Logger::info('Server responds with 417, applying workaround', ['url' => $url]);
+ return self::post($url, $params, $headers, $redirects, $timeout);
+ }
Logger::log('post_url: end ' . $url, Logger::DATA);
}
/**
- * @brief Check URL to see if it's real
+ * Return raw post data from a post request
+ *
+ * @return string post data
+ */
+ public static function postdata()
+ {
+ return file_get_contents('php://input');
+ }
+
+ /**
+ * Check URL to see if it's real
*
* Take a URL from the wild, prepend http:// if necessary
* and check DNS to see if it's real (or check if is a valid IP address)
*
* @param string $url The URL to be validated
* @return string|boolean The actual working URL, false else
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function isUrlValid($url)
+ public static function isUrlValid(string $url)
{
- if (Config::get('system', 'disable_url_validation')) {
+ if (DI::config()->get('system', 'disable_url_validation')) {
return $url;
}
/// @TODO Really suppress function outcomes? Why not find them + debug them?
$h = @parse_url($url);
- if (!empty($h['host']) && (@dns_get_record($h['host'], DNS_A + DNS_CNAME) || filter_var($h['host'], FILTER_VALIDATE_IP) )) {
+ if (!empty($h['host']) && (@dns_get_record($h['host'], DNS_A + DNS_CNAME) || filter_var($h['host'], FILTER_VALIDATE_IP))) {
return $url;
}
}
/**
- * @brief Checks that email is an actual resolvable internet address
+ * Checks that email is an actual resolvable internet address
*
* @param string $addr The email address
* @return boolean True if it's a valid email address, false if it's not
*/
- public static function isEmailDomainValid($addr)
+ public static function isEmailDomainValid(string $addr)
{
- if (Config::get('system', 'disable_email_validation')) {
+ if (DI::config()->get('system', 'disable_email_validation')) {
return true;
}
$h = substr($addr, strpos($addr, '@') + 1);
// Concerning the @ see here: https://stackoverflow.com/questions/36280957/dns-get-record-a-temporary-server-error-occurred
- if ($h && (@dns_get_record($h, DNS_A + DNS_MX) || filter_var($h, FILTER_VALIDATE_IP) )) {
+ if ($h && (@dns_get_record($h, DNS_A + DNS_MX) || filter_var($h, FILTER_VALIDATE_IP))) {
return true;
}
if ($h && @dns_get_record($h, DNS_CNAME + DNS_MX)) {
}
/**
- * @brief Check if URL is allowed
+ * Check if URL is allowed
*
* Check $url against our list of allowed sites,
* wildcards allowed. If allowed_sites is unset return true;
* @param string $url URL which get tested
* @return boolean True if url is allowed otherwise return false
*/
- public static function isUrlAllowed($url)
+ public static function isUrlAllowed(string $url)
{
$h = @parse_url($url);
return false;
}
- $str_allowed = Config::get('system', 'allowed_sites');
+ $str_allowed = DI::config()->get('system', 'allowed_sites');
if (! $str_allowed) {
return true;
}
*
* @return boolean
*/
- public static function isUrlBlocked($url)
+ public static function isUrlBlocked(string $url)
{
$host = @parse_url($url, PHP_URL_HOST);
if (!$host) {
return false;
}
- $domain_blocklist = Config::get('system', 'blocklist', []);
+ $domain_blocklist = DI::config()->get('system', 'blocklist', []);
if (!$domain_blocklist) {
return false;
}
foreach ($domain_blocklist as $domain_block) {
- if (strcasecmp($domain_block['domain'], $host) === 0) {
+ if (fnmatch(strtolower($domain_block['domain']), strtolower($host))) {
return true;
}
}
}
/**
- * @brief Check if email address is allowed to register here.
+ * Check if email address is allowed to register here.
*
* Compare against our list (wildcards allowed).
*
* @param string $email email address
* @return boolean False if not allowed, true if allowed
- * or if allowed list is not configured
+ * or if allowed list is not configured
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function isEmailDomainAllowed($email)
+ public static function isEmailDomainAllowed(string $email)
{
$domain = strtolower(substr($email, strpos($email, '@') + 1));
if (!$domain) {
return false;
}
- $str_allowed = Config::get('system', 'allowed_email', '');
+ $str_allowed = DI::config()->get('system', 'allowed_email', '');
if (empty($str_allowed)) {
return true;
}
/**
* Checks for the existence of a domain in a domain list
*
- * @brief Checks for the existence of a domain in a domain list
* @param string $domain
* @param array $domain_list
* @return boolean
*/
- public static function isDomainAllowed($domain, array $domain_list)
+ public static function isDomainAllowed(string $domain, array $domain_list)
{
$found = false;
return $found;
}
- public static function lookupAvatarByEmail($email)
+ public static function lookupAvatarByEmail(string $email)
{
$avatar['size'] = 300;
$avatar['email'] = $email;
Hook::callAll('avatar_lookup', $avatar);
if (! $avatar['success']) {
- $avatar['url'] = System::baseUrl() . '/images/person-300.jpg';
+ $avatar['url'] = DI::baseUrl() . '/images/person-300.jpg';
}
Logger::log('Avatar: ' . $avatar['email'] . ' ' . $avatar['url'], Logger::DEBUG);
}
/**
- * @brief Remove Google Analytics and other tracking platforms params from URL
+ * Remove Google Analytics and other tracking platforms params from URL
*
* @param string $url Any user-submitted URL that may contain tracking params
* @return string The same URL stripped of tracking parameters
*/
- public static function stripTrackingQueryParams($url)
+ public static function stripTrackingQueryParams(string $url)
{
$urldata = parse_url($url);
if (!empty($urldata["query"])) {
}
/**
- * @brief Returns the original URL of the provided URL
+ * Returns the original URL of the provided URL
*
* This function strips tracking query params and follows redirections, either
* through HTTP code or meta refresh tags. Stops after 10 redirections.
*
- * @todo Remove the $fetchbody parameter that generates an extraneous HEAD request
+ * @todo Remove the $fetchbody parameter that generates an extraneous HEAD request
*
- * @see ParseUrl::getSiteinfo
+ * @see ParseUrl::getSiteinfo
*
* @param string $url A user-submitted URL
* @param int $depth The current redirection recursion level (internal)
* @param bool $fetchbody Wether to fetch the body or not after the HEAD requests
* @return string A canonical URL
+ * @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
- public static function finalUrl($url, $depth = 1, $fetchbody = false)
+ public static function finalUrl(string $url, int $depth = 1, bool $fetchbody = false)
{
- $a = \get_app();
+ $a = DI::app();
$url = self::stripTrackingQueryParams($url);
$http_code = $curl_info['http_code'];
curl_close($ch);
- $a->saveTimestamp($stamp1, "network");
+ DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
if ($http_code == 0) {
return $url;
$body = curl_exec($ch);
curl_close($ch);
- $a->saveTimestamp($stamp1, "network");
+ DI::profiler()->saveTimestamp($stamp1, "network", System::callstack());
if (trim($body) == "") {
return $url;
}
/**
- * @brief Find the matching part between two url
+ * Find the matching part between two url
*
* @param string $url1
* @param string $url2
* @return string The matching part
*/
- public static function getUrlMatch($url1, $url2)
+ public static function getUrlMatch(string $url1, string $url2)
{
if (($url1 == "") || ($url2 == "")) {
return "";
$i = 0;
$path = "";
do {
- $path1 = defaults($pathparts1, $i, '');
- $path2 = defaults($pathparts2, $i, '');
+ $path1 = $pathparts1[$i] ?? '';
+ $path2 = $pathparts2[$i] ?? '';
if ($path1 == $path2) {
$path .= $path1."/";
}
/**
- * @brief Glue url parts together
+ * Glue url parts together
*
* @param array $parsed URL parts
*
* @return string The glued URL
*/
- public static function unparseURL($parsed)
+ public static function unparseURL(array $parsed)
{
$get = function ($key) use ($parsed) {
return isset($parsed[$key]) ? $parsed[$key] : null;
(strlen($query) ? "?".$query : '') .
(strlen($fragment) ? "#".$fragment : '');
}
+
+
+ /**
+ * Switch the scheme of an url between http and https
+ *
+ * @param string $url URL
+ *
+ * @return string switched URL
+ */
+ public static function switchScheme(string $url)
+ {
+ $scheme = parse_url($url, PHP_URL_SCHEME);
+ if (empty($scheme)) {
+ return $url;
+ }
+
+ if ($scheme === 'http') {
+ $url = str_replace('http://', 'https://', $url);
+ } elseif ($scheme === 'https') {
+ $url = str_replace('https://', 'http://', $url);
+ }
+
+ return $url;
+ }
+
+ /**
+ * Adds query string parameters to the provided URI. Replace the value of existing keys.
+ *
+ * @param string $path
+ * @param array $additionalParams Associative array of parameters
+ * @return string
+ */
+ public static function appendQueryParam(string $path, array $additionalParams)
+ {
+ $parsed = parse_url($path);
+
+ $params = [];
+ if (!empty($parsed['query'])) {
+ parse_str($parsed['query'], $params);
+ }
+
+ $params = array_merge($params, $additionalParams);
+
+ $parsed['query'] = http_build_query($params);
+
+ return self::unparseURL($parsed);
+ }
+
+ /**
+ * Generates ETag and Last-Modified response headers and checks them against
+ * If-None-Match and If-Modified-Since request headers if present.
+ *
+ * Blocking function, sends 304 headers and exits if check passes.
+ *
+ * @param string $etag The page etag
+ * @param string $last_modified The page last modification UTC date
+ * @throws \Exception
+ */
+ public static function checkEtagModified(string $etag, string $last_modified)
+ {
+ $last_modified = DateTimeFormat::utc($last_modified, 'D, d M Y H:i:s') . ' GMT';
+
+ /**
+ * @see http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.26
+ */
+ $if_none_match = filter_input(INPUT_SERVER, 'HTTP_IF_NONE_MATCH');
+ $if_modified_since = filter_input(INPUT_SERVER, 'HTTP_IF_MODIFIED_SINCE');
+ $flag_not_modified = null;
+ if ($if_none_match) {
+ $result = [];
+ preg_match('/^(?:W\/")?([^"]+)"?$/i', $etag, $result);
+ $etagTrimmed = $result[1];
+ // Lazy exact ETag match, could check weak/strong ETags
+ $flag_not_modified = $if_none_match == '*' || strpos($if_none_match, $etagTrimmed) !== false;
+ }
+
+ if ($if_modified_since && (!$if_none_match || $flag_not_modified)) {
+ // Lazy exact Last-Modified match, could check If-Modified-Since validity
+ $flag_not_modified = $if_modified_since == $last_modified;
+ }
+
+ header('Etag: ' . $etag);
+ header('Last-Modified: ' . $last_modified);
+
+ if ($flag_not_modified) {
+ header("HTTP/1.1 304 Not Modified");
+ exit;
+ }
+ }
}