X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FUtil%2FNetwork.php;h=e4e9c5af5bb5a75673017425b6cc298057fa74ac;hb=eef85584c20121ed01b64401ac51ed279d19db8b;hp=aa118e3dd4fedb3158aaef28c97c66a30e52ec8d;hpb=66a103e36a42ab251fd9ca6c4b56b54380e4d535;p=friendica.git diff --git a/src/Util/Network.php b/src/Util/Network.php index aa118e3dd4..e4e9c5af5b 100644 --- a/src/Util/Network.php +++ b/src/Util/Network.php @@ -4,16 +4,13 @@ */ namespace Friendica\Util; -use Friendica\App; -use Friendica\Core\Addon; -use Friendica\Core\L10n; -use Friendica\Core\System; -use Friendica\Core\Config; -use Friendica\Network\Probe; -use Friendica\Object\Image; -use Friendica\Util\XML; use DOMDocument; use DomXPath; +use Friendica\Core\Config; +use Friendica\Core\Hook; +use Friendica\Core\Logger; +use Friendica\Core\System; +use Friendica\Network\CurlResult; class Network { @@ -34,12 +31,13 @@ class Network * @param string $cookiejar Path to cookie jar file * * @return string The fetched content + * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ - public static function fetchUrl($url, $binary = false, &$redirects = 0, $timeout = 0, $accept_content = null, $cookiejar = 0) + public static function fetchUrl($url, $binary = false, &$redirects = 0, $timeout = 0, $accept_content = null, $cookiejar = '') { $ret = self::fetchUrlFull($url, $binary, $redirects, $timeout, $accept_content, $cookiejar); - return $ret['body']; + return $ret->getBody(); } /** @@ -57,9 +55,10 @@ class Network * @param string $accept_content supply Accept: header with 'accept_content' as the value * @param string $cookiejar Path to cookie jar file * - * @return array With all relevant information, 'body' contains the actual fetched content. + * @return CurlResult With all relevant information, 'body' contains the actual fetched content. + * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ - public static function fetchUrlFull($url, $binary = false, &$redirects = 0, $timeout = 0, $accept_content = null, $cookiejar = 0) + public static function fetchUrlFull($url, $binary = false, &$redirects = 0, $timeout = 0, $accept_content = null, $cookiejar = '') { return self::curl( $url, @@ -86,48 +85,49 @@ class Network * 'novalidate' => do not validate SSL certs, default is to validate using our CA list * 'nobody' => only return the header * 'cookiejar' => path to cookie jar file + * 'header' => header array * - * @return array an assoziative array with: - * int 'return_code' => HTTP return code or 0 if timeout or failure - * boolean 'success' => boolean true (if HTTP 2xx result) or false - * string 'redirect_url' => in case of redirect, content was finally retrieved from this URL - * string 'header' => HTTP headers - * string 'body' => fetched content + * @return CurlResult + * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ public static function curl($url, $binary = false, &$redirects = 0, $opts = []) { - $ret = ['return_code' => 0, 'success' => false, 'header' => '', 'info' => '', 'body' => '']; - $stamp1 = microtime(true); - $a = get_app(); + $a = \get_app(); + if (strlen($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', '')); foreach ($path_parts as $part) { - if (strlen($part) <> mb_strlen($part)) { + if (strlen($part) <> mb_strlen($part)) { $parts2[] = rawurlencode($part); - } else { - $parts2[] = $part; - } + } else { + $parts2[] = $part; + } } - $parts['path'] = implode('/', $parts2); + $parts['path'] = implode('/', $parts2); $url = self::unparseURL($parts); if (self::isUrlBlocked($url)) { - logger('domain of ' . $url . ' is blocked', LOGGER_DATA); - return $ret; + Logger::log('domain of ' . $url . ' is blocked', Logger::DATA); + return CurlResult::createErrorCurl($url); } $ch = @curl_init($url); if (($redirects > 8) || (!$ch)) { - return $ret; + return CurlResult::createErrorCurl($url); } @curl_setopt($ch, CURLOPT_HEADER, true); - if (x($opts, "cookiejar")) { + if (!empty($opts['cookiejar'])) { curl_setopt($ch, CURLOPT_COOKIEJAR, $opts["cookiejar"]); curl_setopt($ch, CURLOPT_COOKIEFILE, $opts["cookiejar"]); } @@ -136,7 +136,7 @@ class Network // @curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true); // @curl_setopt($ch, CURLOPT_MAXREDIRS, 5); - if (x($opts, 'accept_content')) { + if (!empty($opts['accept_content'])) { curl_setopt( $ch, CURLOPT_HTTPHEADER, @@ -144,8 +144,12 @@ class Network ); } + if (!empty($opts['header'])) { + curl_setopt($ch, CURLOPT_HTTPHEADER, $opts['header']); + } + @curl_setopt($ch, CURLOPT_RETURNTRANSFER, true); - @curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent()); + @curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent()); $range = intval(Config::get('system', 'curl_range_bytes', 0)); @@ -158,15 +162,15 @@ class Network /// @todo We could possibly set this value to "gzip" or something similar curl_setopt($ch, CURLOPT_ENCODING, ''); - if (x($opts, 'headers')) { + if (!empty($opts['headers'])) { @curl_setopt($ch, CURLOPT_HTTPHEADER, $opts['headers']); } - if (x($opts, 'nobody')) { + if (!empty($opts['nobody'])) { @curl_setopt($ch, CURLOPT_NOBODY, $opts['nobody']); } - if (x($opts, 'timeout')) { + if (!empty($opts['timeout'])) { @curl_setopt($ch, CURLOPT_TIMEOUT, $opts['timeout']); } else { $curl_time = Config::get('system', 'curl_timeout', 60); @@ -203,8 +207,6 @@ class Network @curl_setopt($ch, CURLOPT_BINARYTRANSFER, 1); } - $a->set_curl_code(0); - // don't let curl abort the entire application // if it throws any errors. @@ -219,91 +221,20 @@ class Network $curl_info = @curl_getinfo($ch); } - if (curl_errno($ch) !== CURLE_OK) { - logger('error fetching ' . $url . ': ' . curl_error($ch), LOGGER_NORMAL); - } - - $ret['errno'] = curl_errno($ch); - - $base = $s; - $ret['info'] = $curl_info; - - $http_code = $curl_info['http_code']; - - logger($url . ': ' . $http_code . " " . $s, LOGGER_DATA); - $header = ''; - - // Pull out multiple headers, e.g. proxy and continuation headers - // allow for HTTP/2.x without fixing code - - while (preg_match('/^HTTP\/[1-2].+? [1-5][0-9][0-9]/', $base)) { - $chunk = substr($base, 0, strpos($base, "\r\n\r\n") + 4); - $header .= $chunk; - $base = substr($base, strlen($chunk)); - } - - $a->set_curl_code($http_code); - $a->set_curl_content_type($curl_info['content_type']); - $a->set_curl_headers($header); - - if ($http_code == 301 || $http_code == 302 || $http_code == 303 || $http_code == 307) { - $new_location_info = @parse_url($curl_info['redirect_url']); - $old_location_info = @parse_url($curl_info['url']); - - $newurl = $curl_info['redirect_url']; - - if (empty($new_location_info['path']) && !empty($new_location_info['host'])) { - $newurl = $new_location_info['scheme'] . '://' . $new_location_info['host'] . $old_location_info['path']; - } - - $matches = []; - - if (preg_match('/(Location:|URI:)(.*?)\n/i', $header, $matches)) { - $newurl = trim(array_pop($matches)); - } - if (strpos($newurl, '/') === 0) { - $newurl = $old_location_info["scheme"]."://".$old_location_info["host"].$newurl; - } - $old_location_query = @parse_url($url, PHP_URL_QUERY); - - if ($old_location_query != '') { - $newurl .= '?' . $old_location_query; - } - - if (filter_var($newurl, FILTER_VALIDATE_URL)) { - $redirects++; - @curl_close($ch); - return self::curl($newurl, $binary, $redirects, $opts); - } - } - - $a->set_curl_code($http_code); - $a->set_curl_content_type($curl_info['content_type']); - - $rc = intval($http_code); - $ret['return_code'] = $rc; - $ret['success'] = (($rc >= 200 && $rc <= 299) ? true : false); - $ret['redirect_url'] = $url; - - if (!$ret['success']) { - $ret['error'] = curl_error($ch); - $ret['debug'] = $curl_info; - logger('error: '.$url.': '.$ret['return_code'].' - '.$ret['error'], LOGGER_DEBUG); - logger('debug: '.print_r($curl_info, true), LOGGER_DATA); - } - - $ret['body'] = substr($s, strlen($header)); - $ret['header'] = $header; + $curlResponse = new CurlResult($url, $s, $curl_info, curl_errno($ch), curl_error($ch)); - if (x($opts, 'debug')) { - $ret['debug'] = $curl_info; + if ($curlResponse->isRedirectUrl()) { + $redirects++; + Logger::log('curl: redirect ' . $url . ' to ' . $curlResponse->getRedirectUrl()); + @curl_close($ch); + return self::curl($curlResponse->getRedirectUrl(), $binary, $redirects, $opts); } @curl_close($ch); - $a->save_timestamp($stamp1, 'network'); + $a->getProfiler()->saveTimestamp($stamp1, 'network', System::callstack()); - return($ret); + return $curlResponse; } /** @@ -315,31 +246,32 @@ class Network * @param integer $redirects Recursion counter for internal use - default = 0 * @param integer $timeout The timeout in seconds, default system config value or 60 seconds * - * @return string The content + * @return CurlResult The content + * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ public static function post($url, $params, $headers = null, &$redirects = 0, $timeout = 0) { $stamp1 = microtime(true); if (self::isUrlBlocked($url)) { - logger('post_url: domain of ' . $url . ' is blocked', LOGGER_DATA); - return false; + Logger::log('post_url: domain of ' . $url . ' is blocked', Logger::DATA); + return CurlResult::createErrorCurl($url); } - $a = get_app(); + $a = \get_app(); $ch = curl_init($url); if (($redirects > 8) || (!$ch)) { - return false; + return CurlResult::createErrorCurl($url); } - logger('post_url: start ' . $url, LOGGER_DATA); + Logger::log('post_url: start ' . $url, Logger::DATA); curl_setopt($ch, CURLOPT_HEADER, true); curl_setopt($ch, CURLOPT_RETURNTRANSFER, true); curl_setopt($ch, CURLOPT_POST, 1); curl_setopt($ch, CURLOPT_POSTFIELDS, $params); - curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent()); + curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent()); if (Config::get('system', 'ipv4_resolve', false)) { curl_setopt($ch, CURLOPT_IPRESOLVE, CURL_IPRESOLVE_V4); @@ -384,62 +316,29 @@ class Network } } - $a->set_curl_code(0); - // don't let curl abort the entire application // if it throws any errors. $s = @curl_exec($ch); - $base = $s; $curl_info = curl_getinfo($ch); - $http_code = $curl_info['http_code']; - - logger('post_url: result ' . $http_code . ' - ' . $url, LOGGER_DATA); - - $header = ''; - - // Pull out multiple headers, e.g. proxy and continuation headers - // allow for HTTP/2.x without fixing code - - while (preg_match('/^HTTP\/[1-2].+? [1-5][0-9][0-9]/', $base)) { - $chunk = substr($base, 0, strpos($base, "\r\n\r\n") + 4); - $header .= $chunk; - $base = substr($base, strlen($chunk)); - } - if ($http_code == 301 || $http_code == 302 || $http_code == 303 || $http_code == 307) { - $matches = []; - $new_location_info = @parse_url($curl_info['redirect_url']); - $old_location_info = @parse_url($curl_info['url']); + $curlResponse = new CurlResult($url, $s, $curl_info, curl_errno($ch), curl_error($ch)); - preg_match('/(Location:|URI:)(.*?)\n/', $header, $matches); - $newurl = trim(array_pop($matches)); - - if (strpos($newurl, '/') === 0) { - $newurl = $old_location_info["scheme"] . "://" . $old_location_info["host"] . $newurl; - } - - if (filter_var($newurl, FILTER_VALIDATE_URL)) { - $redirects++; - logger('post_url: redirect ' . $url . ' to ' . $newurl); - return self::post($newurl, $params, $headers, $redirects, $timeout); - } + if ($curlResponse->isRedirectUrl()) { + $redirects++; + Logger::log('post_url: redirect ' . $url . ' to ' . $curlResponse->getRedirectUrl()); + curl_close($ch); + return self::post($curlResponse->getRedirectUrl(), $params, $headers, $redirects, $timeout); } - $a->set_curl_code($http_code); - - $body = substr($s, strlen($header)); - - $a->set_curl_headers($header); - curl_close($ch); - $a->save_timestamp($stamp1, 'network'); + $a->getProfiler()->saveTimestamp($stamp1, 'network', System::callstack()); - logger('post_url: end ' . $url, LOGGER_DATA); + Logger::log('post_url: end ' . $url, Logger::DATA); - return $body; + return $curlResponse; } /** @@ -450,6 +349,7 @@ class Network * * @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) { @@ -469,7 +369,7 @@ class Network /// @TODO Really suppress function outcomes? Why not find them + debug them? $h = @parse_url($url); - if ((is_array($h)) && (@dns_get_record($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; } @@ -481,6 +381,7 @@ class Network * * @param string $addr The email address * @return boolean True if it's a valid email address, false if it's not + * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ public static function isEmailDomainValid($addr) { @@ -494,10 +395,11 @@ class Network $h = substr($addr, strpos($addr, '@') + 1); - if (($h) && (dns_get_record($h, DNS_A + DNS_MX) || filter_var($h, FILTER_VALIDATE_IP) )) { + // 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($h, FILTER_VALIDATE_IP) )) { return true; } - if (($h) && dns_get_record($h, DNS_CNAME + DNS_MX)) { + if ($h && @dns_get_record($h, DNS_CNAME + DNS_MX)) { return true; } return false; @@ -511,6 +413,7 @@ class Network * * @param string $url URL which get tested * @return boolean True if url is allowed otherwise return false + * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ public static function isUrlAllowed($url) { @@ -560,12 +463,12 @@ class Network public static function isUrlBlocked($url) { $host = @parse_url($url, PHP_URL_HOST); - if (! $host) { - return true; + if (!$host) { + return false; } $domain_blocklist = Config::get('system', 'blocklist', []); - if (! $domain_blocklist) { + if (!$domain_blocklist) { return false; } @@ -585,7 +488,8 @@ class Network * * @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) { @@ -595,7 +499,7 @@ class Network } $str_allowed = Config::get('system', 'allowed_email', ''); - if (!x($str_allowed)) { + if (empty($str_allowed)) { return true; } @@ -629,18 +533,18 @@ class Network public static function lookupAvatarByEmail($email) { - $avatar['size'] = 175; + $avatar['size'] = 300; $avatar['email'] = $email; $avatar['url'] = ''; $avatar['success'] = false; - Addon::callHooks('avatar_lookup', $avatar); + Hook::callAll('avatar_lookup', $avatar); if (! $avatar['success']) { - $avatar['url'] = System::baseUrl() . '/images/person-175.jpg'; + $avatar['url'] = System::baseUrl() . '/images/person-300.jpg'; } - logger('Avatar: ' . $avatar['email'] . ' ' . $avatar['url'], LOGGER_DEBUG); + Logger::log('Avatar: ' . $avatar['email'] . ' ' . $avatar['url'], Logger::DEBUG); return $avatar['url']; } @@ -699,23 +603,24 @@ class Network * 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) { - $a = get_app(); + $a = \get_app(); $url = self::stripTrackingQueryParams($url); if ($depth > 10) { - return($url); + return $url; } $url = trim($url, "'"); @@ -728,26 +633,24 @@ class Network curl_setopt($ch, CURLOPT_NOBODY, 1); curl_setopt($ch, CURLOPT_TIMEOUT, 10); curl_setopt($ch, CURLOPT_RETURNTRANSFER, true); - curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent()); + curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent()); curl_exec($ch); $curl_info = @curl_getinfo($ch); $http_code = $curl_info['http_code']; curl_close($ch); - $a->save_timestamp($stamp1, "network"); + $a->getProfiler()->saveTimestamp($stamp1, "network", System::callstack()); if ($http_code == 0) { - return($url); + return $url; } - if ((($curl_info['http_code'] == "301") || ($curl_info['http_code'] == "302")) - && (($curl_info['redirect_url'] != "") || ($curl_info['location'] != "")) - ) { - if ($curl_info['redirect_url'] != "") { - return(self::finalUrl($curl_info['redirect_url'], ++$depth, $fetchbody)); - } else { - return(self::finalUrl($curl_info['location'], ++$depth, $fetchbody)); + if (in_array($http_code, ['301', '302'])) { + if (!empty($curl_info['redirect_url'])) { + return self::finalUrl($curl_info['redirect_url'], ++$depth, $fetchbody); + } elseif (!empty($curl_info['location'])) { + return self::finalUrl($curl_info['location'], ++$depth, $fetchbody); } } @@ -758,12 +661,12 @@ class Network // if the file is too large then exit if ($curl_info["download_content_length"] > 1000000) { - return($url); + return $url; } // if it isn't a HTML file then exit - if (($curl_info["content_type"] != "") && !strstr(strtolower($curl_info["content_type"]), "html")) { - return($url); + if (!empty($curl_info["content_type"]) && !strstr(strtolower($curl_info["content_type"]), "html")) { + return $url; } $stamp1 = microtime(true); @@ -774,15 +677,15 @@ class Network curl_setopt($ch, CURLOPT_NOBODY, 0); curl_setopt($ch, CURLOPT_TIMEOUT, 10); curl_setopt($ch, CURLOPT_RETURNTRANSFER, true); - curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent()); + curl_setopt($ch, CURLOPT_USERAGENT, $a->getUserAgent()); $body = curl_exec($ch); curl_close($ch); - $a->save_timestamp($stamp1, "network"); + $a->getProfiler()->saveTimestamp($stamp1, "network", System::callstack()); if (trim($body) == "") { - return($url); + return $url; } // Check for redirect in meta elements @@ -805,7 +708,7 @@ class Network $pathinfo = explode(";", $path); foreach ($pathinfo as $value) { if (substr(strtolower($value), 0, 4) == "url=") { - return(self::finalUrl(substr($value, 4), ++$depth)); + return self::finalUrl(substr($value, 4), ++$depth); } } } @@ -827,8 +730,8 @@ class Network return ""; } - $url1 = normalise_link($url1); - $url2 = normalise_link($url2); + $url1 = Strings::normaliseLink($url1); + $url2 = Strings::normaliseLink($url2); $parts1 = parse_url($url1); $parts2 = parse_url($url2); @@ -899,7 +802,7 @@ class Network $match .= $path; - return normalise_link($match); + return Strings::normaliseLink($match); } /** @@ -932,4 +835,28 @@ class Network (strlen($query) ? "?".$query : '') . (strlen($fragment) ? "#".$fragment : ''); } + + + /** + * Switch the scheme of an url between http and https + * + * @param string $url URL + * + * @return string switched URL + */ + public static function switchScheme($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; + } }