X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FNetwork%2FHTTPRequest.php;h=e4ff041039ed8ba225022000faa6af8f2c159af7;hb=722aada460aa023212b7d97932fe338fe54861a8;hp=f9279fa6028473711a543e4c12ead24cab0e484b;hpb=7029012f27e9f5199617ae1ec2c3aae10f0d1fc8;p=friendica.git diff --git a/src/Network/HTTPRequest.php b/src/Network/HTTPRequest.php index f9279fa602..e4ff041039 100644 --- a/src/Network/HTTPRequest.php +++ b/src/Network/HTTPRequest.php @@ -33,7 +33,7 @@ use Psr\Log\LoggerInterface; /** * Performs HTTP requests to a given URL */ -class HTTPRequest +class HTTPRequest implements IHTTPRequest { /** @var LoggerInterface */ private $logger; @@ -53,22 +53,10 @@ class HTTPRequest } /** - * fetches an URL. + * {@inheritDoc} * - * @param string $url URL to fetch - * @param bool $binary default false - * TRUE if asked to return binary results (file download) - * @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 - * 'http_auth' => username:password - * '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 - * @param int $redirects The recursion counter for internal use - default 0 + * @param int $redirects The recursion counter for internal use - default 0 * - * @return CurlResult * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ public function get(string $url, bool $binary = false, array $opts = [], int &$redirects = 0) @@ -149,6 +137,8 @@ class HTTPRequest @curl_setopt($ch, CURLOPT_NOBODY, $opts['nobody']); } + @curl_setopt($ch, CURLOPT_CONNECTTIMEOUT, 10); + if (!empty($opts['timeout'])) { @curl_setopt($ch, CURLOPT_TIMEOUT, $opts['timeout']); } else { @@ -202,7 +192,7 @@ class HTTPRequest $curlResponse = new CurlResult($url, $s, $curl_info, curl_errno($ch), curl_error($ch)); - if ($curlResponse->isRedirectUrl()) { + if (!Network::isRedirectBlocked($url) && $curlResponse->isRedirectUrl()) { $redirects++; $this->logger->notice('Curl redirect.', ['url' => $url, 'to' => $curlResponse->getRedirectUrl()]); @curl_close($ch); @@ -211,21 +201,16 @@ class HTTPRequest @curl_close($ch); - $this->profiler->saveTimestamp($stamp1, 'network', System::callstack()); + $this->profiler->saveTimestamp($stamp1, 'network'); return $curlResponse; } /** - * Send POST request to $url + * {@inheritDoc} * - * @param string $url URL to post - * @param mixed $params array of POST variables - * @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 + * @param int $redirects The recursion counter for internal use - default 0 * - * @return CurlResult The content * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ public function post(string $url, $params, array $headers = [], int $timeout = 0, int &$redirects = 0) @@ -255,6 +240,8 @@ class HTTPRequest curl_setopt($ch, CURLOPT_IPRESOLVE, CURL_IPRESOLVE_V4); } + @curl_setopt($ch, CURLOPT_CONNECTTIMEOUT, 10); + if (intval($timeout)) { curl_setopt($ch, CURLOPT_TIMEOUT, $timeout); } else { @@ -293,7 +280,7 @@ class HTTPRequest $curlResponse = new CurlResult($url, $s, $curl_info, curl_errno($ch), curl_error($ch)); - if ($curlResponse->isRedirectUrl()) { + if (!Network::isRedirectBlocked($url) && $curlResponse->isRedirectUrl()) { $redirects++; $this->logger->info('Post redirect.', ['url' => $url, 'to' => $curlResponse->getRedirectUrl()]); curl_close($ch); @@ -302,7 +289,7 @@ class HTTPRequest curl_close($ch); - $this->profiler->saveTimestamp($stamp1, 'network', System::callstack()); + $this->profiler->saveTimestamp($stamp1, 'network'); // Very old versions of Lighttpd don't like the "Expect" header, so we remove it when needed if ($curlResponse->getReturnCode() == 417) { @@ -325,23 +312,20 @@ class HTTPRequest } /** - * 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 - * - * @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 + * {@inheritDoc} */ public function finalUrl(string $url, int $depth = 1, bool $fetchbody = false) { + if (Network::isUrlBlocked($url)) { + $this->logger->info('Domain is blocked.', ['url' => $url]); + return $url; + } + + if (Network::isRedirectBlocked($url)) { + $this->logger->info('Domain should not be redirected.', ['url' => $url]); + return $url; + } + $url = Network::stripTrackingQueryParams($url); if ($depth > 10) { @@ -356,6 +340,7 @@ class HTTPRequest curl_setopt($ch, CURLOPT_URL, $url); curl_setopt($ch, CURLOPT_HEADER, 1); curl_setopt($ch, CURLOPT_NOBODY, 1); + curl_setopt($ch, CURLOPT_CONNECTTIMEOUT, 10); curl_setopt($ch, CURLOPT_TIMEOUT, 10); curl_setopt($ch, CURLOPT_RETURNTRANSFER, true); curl_setopt($ch, CURLOPT_USERAGENT, $this->getUserAgent()); @@ -365,7 +350,7 @@ class HTTPRequest $http_code = $curl_info['http_code']; curl_close($ch); - $this->profiler->saveTimestamp($stamp1, "network", System::callstack()); + $this->profiler->saveTimestamp($stamp1, "network"); if ($http_code == 0) { return $url; @@ -400,6 +385,7 @@ class HTTPRequest curl_setopt($ch, CURLOPT_URL, $url); curl_setopt($ch, CURLOPT_HEADER, 0); curl_setopt($ch, CURLOPT_NOBODY, 0); + curl_setopt($ch, CURLOPT_CONNECTTIMEOUT, 10); curl_setopt($ch, CURLOPT_TIMEOUT, 10); curl_setopt($ch, CURLOPT_RETURNTRANSFER, true); curl_setopt($ch, CURLOPT_USERAGENT, $this->getUserAgent()); @@ -407,7 +393,7 @@ class HTTPRequest $body = curl_exec($ch); curl_close($ch); - $this->profiler->saveTimestamp($stamp1, "network", System::callstack()); + $this->profiler->saveTimestamp($stamp1, "network"); if (trim($body) == "") { return $url; @@ -443,48 +429,27 @@ class HTTPRequest } /** - * Curl wrapper - * - * If binary flag is true, return binary results. - * Set the cookiejar argument to a string (e.g. "/tmp/friendica-cookies.txt") - * to preserve cookies from one request to the next. + * {@inheritDoc} * - * @param string $url URL to fetch - * @param bool $binary default false - * TRUE if asked to return binary results (file download) - * @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 + * @param int $redirects The recursion counter for internal use - default 0 * - * @return string The fetched content * @throws \Friendica\Network\HTTPException\InternalServerErrorException */ - public function fetchUrl(string $url, bool $binary = false, int $timeout = 0, string $accept_content = '', string $cookiejar = '', int &$redirects = 0) + public function fetch(string $url, bool $binary = false, int $timeout = 0, string $accept_content = '', string $cookiejar = '', int &$redirects = 0) { - $ret = $this->fetchUrlFull($url, $binary, $timeout, $accept_content, $cookiejar, $redirects); + $ret = $this->fetchFull($url, $binary, $timeout, $accept_content, $cookiejar, $redirects); return $ret->getBody(); } /** - * Curl wrapper with array of return values. + * {@inheritDoc} * - * Inner workings and parameters are the same as @ref fetchUrl but returns an array with - * all the information collected during the fetch. + * @param int $redirects The recursion counter for internal use - default 0 * - * @param string $url URL to fetch - * @param bool $binary default false - * TRUE if asked to return binary results (file download) - * @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 function fetchUrlFull(string $url, bool $binary = false, int $timeout = 0, string $accept_content = '', string $cookiejar = '', int &$redirects = 0) + public function fetchFull(string $url, bool $binary = false, int $timeout = 0, string $accept_content = '', string $cookiejar = '', int &$redirects = 0) { return $this->get( $url, @@ -499,9 +464,7 @@ class HTTPRequest } /** - * Returns the current UserAgent as a String - * - * @return string the UserAgent as a String + * {@inheritDoc} */ public function getUserAgent() {