* * This program is free software: you can redistribute it and/or modify * it under the terms of the GNU 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 General Public License for more details. * * You should have received a copy of the GNU General Public License * along with this program. If not, see . */ use DOMDocument; use XSLTProcessor; use Symfony\Component\Serializer\Serializer; use Symfony\Component\Serializer\Encoder\XmlEncoder; use Symfony\Component\Serializer\Normalizer\ObjectNormalizer; date_default_timezone_set('UTC'); /** * Scrape supported websites and perform conversion to RSS. */ class Tweeper { private static $userAgent = "APIs-Google (+https://developers.google.com/webmasters/APIs-Google.html)"; private static $maxConnectionTimeout = 5; private static $maxConnectionRetries = 5; /** * Create a new Tweeper object controlling optional settings. * * @param bool $generate_enclosure * Enables the creation of elements (disabled by default). * @param bool $show_usernames * Enables showing the username in front of the content for multi-user * sites (enabled by default). Only some stylesheets supports this * functionality (twitter, instagram, pump.io). * @param bool $show_multimedia * Enables showing multimedia content (images, videos) directly in the * item description (enabled by default). Only some stylesheets supports * this functionality (twitter, instagram, dilbert). * @param bool $verbose_output * Enables showing non-fatal errors like XML parsing errors. */ public function __construct($generate_enclosure = FALSE, $show_usernames = TRUE, $show_multimedia = TRUE, $verbose_output = TRUE) { $this->generate_enclosure = $generate_enclosure; $this->show_usernames = $show_usernames; $this->show_multimedia = $show_multimedia; $this->verbose_output = $verbose_output; } /** * Convert numeric Epoch to the date format expected in a RSS document. */ public static function epochToRssDate($timestamp) { if (!is_numeric($timestamp) || is_nan($timestamp)) { $timestamp = 0; } return gmdate(DATE_RSS, $timestamp); } /** * Convert generic date string to the date format expected in a RSS document. */ public static function strToRssDate($date) { $timestamp = strtotime($date); if (FALSE === $timestamp) { $timestamp = 0; } return Tweeper::epochToRssDate($timestamp); } /** * Convert string to UpperCamelCase. */ public static function toUpperCamelCase($str, $delim = ' ') { $str_upper = ucwords($str, $delim); $str_camel_case = str_replace($delim, '', $str_upper); return $str_camel_case; } /** * Perform a cURL session multiple times when it fails with a timeout. * * @param resource $ch * a cURL session handle. */ private static function curlExec($ch) { $ret = FALSE; $attempt = 0; do { $ret = curl_exec($ch); if (FALSE === $ret) { trigger_error(curl_error($ch), E_USER_WARNING); } } while (curl_errno($ch) == CURLE_OPERATION_TIMEDOUT && ++$attempt < Tweeper::$maxConnectionRetries); return $ret; } /** * Get the contents from a URL. */ private static function getUrlContents($url, $user_agent = NULL) { $ch = curl_init($url); curl_setopt_array($ch, [ CURLOPT_HEADER => FALSE, CURLOPT_CONNECTTIMEOUT => Tweeper::$maxConnectionTimeout, // Follow http redirects to get the real URL. CURLOPT_FOLLOWLOCATION => TRUE, CURLOPT_COOKIEFILE => "", CURLOPT_RETURNTRANSFER => TRUE, CURLOPT_HTTPHEADER => ['Accept-language: en'], CURLOPT_USERAGENT => isset($user_agent) ? $user_agent : Tweeper::$userAgent, ]); $contents = Tweeper::curlExec($ch); curl_close($ch); return $contents; } /** * Get the headers from a URL. */ private static function getUrlInfo($url, $user_agent = NULL) { $ch = curl_init($url); curl_setopt_array($ch, [ CURLOPT_HEADER => TRUE, CURLOPT_NOBODY => TRUE, CURLOPT_CONNECTTIMEOUT => Tweeper::$maxConnectionTimeout, // Follow http redirects to get the real URL. CURLOPT_FOLLOWLOCATION => TRUE, CURLOPT_RETURNTRANSFER => TRUE, CURLOPT_USERAGENT => isset($user_agent) ? $user_agent : Tweeper::$userAgent, ]); $ret = Tweeper::curlExec($ch); if (FALSE === $ret) { curl_close($ch); return FALSE; } $url_info = curl_getinfo($ch); if (FALSE === $url_info) { trigger_error(curl_error($ch), E_USER_WARNING); } curl_close($ch); return $url_info; } /** * Generate an RSS element. */ public static function generateEnclosure($url) { $supported_content_types = [ "application/octet-stream", "application/ogg", "application/pdf", "audio/aac", "audio/mp4", "audio/mpeg", "audio/ogg", "audio/vorbis", "audio/wav", "audio/webm", "audio/x-midi", "image/gif", "image/jpeg", "image/png", "video/avi", "video/mp4", "video/mpeg", "video/ogg", ]; $url_info = Tweeper::getUrlInfo($url); if (FALSE === $url_info) { trigger_error("Failed to retrieve info for URL: " . $url, E_USER_WARNING); return ''; } $supported = in_array($url_info['content_type'], $supported_content_types); if (!$supported) { trigger_error("Unsupported enclosure content type \"" . $url_info['content_type'] . "\" for URL: " . $url_info['url'], E_USER_WARNING); return ''; } // The RSS specification says that the enclosure element URL must be http. // See http://sourceforge.net/p/feedvalidator/bugs/72/ $http_url = preg_replace("/^https/", "http", $url_info['url']); // When the server does not provide a Content-Length header, // curl_getinfo() would return a negative value for // "download_content_length", however RSS recommends to use 0 when the // enclosure's size cannot be determined. // See: https://www.feedvalidator.org/docs/error/UseZeroForUnknown.html $length = max($url_info['download_content_length'], 0); $dom = new DOMDocument(); $enc = $dom->createElement('enclosure'); $enc->setAttribute('url', $http_url); $enc->setAttribute('length', $length); $enc->setAttribute('type', $url_info['content_type']); return $enc; } /** * Mimic the message from libxml.c::php_libxml_ctx_error_level() */ private static function logXmlError($error) { $output = ""; switch ($error->level) { case LIBXML_ERR_WARNING: $output .= "Warning $error->code: "; break; case LIBXML_ERR_ERROR: $output .= "Error $error->code: "; break; case LIBXML_ERR_FATAL: $output .= "Fatal Error $error->code: "; break; } $output .= trim($error->message); if ($error->file) { $output .= " in $error->file"; } else { $output .= " in Entity,"; } $output .= " line $error->line"; trigger_error($output, E_USER_WARNING); } /** * Convert json to XML. */ private static function jsonToXml($json, $root_node_name) { // Apparently the ObjectNormalizer used afterwards is not able to handle // the stdClass object created by json_decode() with the default setting // $assoc = false; so use $assoc = true. $data = json_decode($json, $assoc = TRUE); if (!$data) { return NULL; } $encoder = new XmlEncoder(); $normalizer = new ObjectNormalizer(); $serializer = new Serializer([$normalizer], [$encoder]); $serializer_options = [ 'xml_encoding' => "UTF-8", 'xml_format_output' => TRUE, 'xml_root_node_name' => $root_node_name, ]; $xml_data = $serializer->serialize($data, 'xml', $serializer_options); if (!$xml_data) { trigger_error("Cannot serialize data", E_USER_WARNING); return NULL; } return $xml_data; } /** * Convert the Instagram content to XML. */ private function getXmlInstagramCom($html) { // Extract the json data from the html code. $json_match_expr = '/window._sharedData = (.*);/'; $ret = preg_match($json_match_expr, $html, $matches); if ($ret !== 1) { trigger_error("Cannot match expression: $json_match_expr\n", E_USER_WARNING); return NULL; } $data = json_decode($matches[1], $assoc = TRUE); // The "qe" object contains elements which will result in invalid XML // element names, so remove it. unset($data["qe"]); // The "knobs" object contains elements with undefined namespaces, so // remove it to silence an error message. unset($data["knobs"]); $json = json_encode($data); return Tweeper::jsonToXml($json, 'instagram'); } /** * Make the Facebook HTML processable. */ private function preprocessHtmlFacebookCom($html) { $html = str_replace('', '', $html); return $html; } /** * Convert the HTML retrieved from the site to XML. */ private function htmlToXml($html, $host) { $xmlDoc = new DOMDocument(); // Handle warnings and errors when loading invalid HTML. $xml_errors_value = libxml_use_internal_errors(TRUE); // If there is a host-specific method to get the XML data, use it! $get_xml_host_method = 'getXml' . Tweeper::toUpperCamelCase($host, '.'); if (method_exists($this, $get_xml_host_method)) { $xml_data = call_user_func_array([$this, $get_xml_host_method], [$html]); $xmlDoc->loadXML($xml_data); } else { $xmlDoc->loadHTML($html); } if ($this->verbose_output) { foreach (libxml_get_errors() as $xml_error) { Tweeper::logXmlError($xml_error); } } libxml_clear_errors(); libxml_use_internal_errors($xml_errors_value); return $xmlDoc; } /** * Load a stylesheet if the web site is supported. */ private function loadStylesheet($host) { $stylesheet = "file://" . __DIR__ . "/rss_converter_" . $host . ".xsl"; if (FALSE === file_exists($stylesheet)) { trigger_error("Conversion to RSS not supported for $host ($stylesheet not found)", E_USER_WARNING); return NULL; } $stylesheet_contents = file_get_contents($stylesheet); if (FALSE === $stylesheet_contents) { trigger_error("Cannot open $stylesheet", E_USER_WARNING); return NULL; } $xslDoc = new DOMDocument(); $xslDoc->loadXML($stylesheet_contents); $xsltProcessor = new XSLTProcessor(); $xsltProcessor->registerPHPFunctions(); $xsltProcessor->setParameter('', 'generate-enclosure', $this->generate_enclosure); $xsltProcessor->setParameter('', 'show-usernames', $this->show_usernames); $xsltProcessor->setParameter('', 'show-multimedia', $this->show_multimedia); $xsltProcessor->importStylesheet($xslDoc); return $xsltProcessor; } /** * Convert the site content to RSS. */ public function tweep($src_url, $host = NULL, $validate_scheme = TRUE) { $url = parse_url($src_url); if (FALSE === $url) { trigger_error("Invalid URL: $src_url", E_USER_WARNING); return NULL; } if (TRUE === $validate_scheme) { $scheme = $url["scheme"]; if (!in_array($scheme, ["http", "https"])) { trigger_error("unsupported scheme: $scheme", E_USER_WARNING); return NULL; } } // If the host is not given derive it from the URL. if (NULL === $host) { if (empty($url["host"])) { trigger_error("Invalid host in URL: $src_url", E_USER_WARNING); return NULL; } // Strip the leading www. to be more forgiving on input URLs. $host = preg_replace('/^www\./', '', $url["host"]); } $xsltProcessor = $this->loadStylesheet($host); if (NULL === $xsltProcessor) { return NULL; } $html = Tweeper::getUrlContents($src_url); if (FALSE === $html) { trigger_error("Failed to retrieve $src_url", E_USER_WARNING); return NULL; } $preprocess_html_host_method = 'preprocessHtml' . Tweeper::toUpperCamelCase($host, '.'); if (method_exists($this, $preprocess_html_host_method)) { $html = call_user_func_array([$this, $preprocess_html_host_method], [$html]); } $xmlDoc = $this->htmlToXml($html, $host); if (NULL === $xmlDoc) { return NULL; } $output = $xsltProcessor->transformToXML($xmlDoc); if (FALSE === $output) { trigger_error('XSL transformation failed.', E_USER_WARNING); return NULL; } return $output; } }