7 * Tweeper - a Twitter to RSS web scraper.
9 * Copyright (C) 2013-2018 Antonio Ospite <ao2@ao2.it>
11 * This program is free software: you can redistribute it and/or modify
12 * it under the terms of the GNU General Public License as published by
13 * the Free Software Foundation, either version 3 of the License, or
14 * (at your option) any later version.
16 * This program is distributed in the hope that it will be useful,
17 * but WITHOUT ANY WARRANTY; without even the implied warranty of
18 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19 * GNU General Public License for more details.
21 * You should have received a copy of the GNU General Public License
22 * along with this program. If not, see <http://www.gnu.org/licenses/>.
28 use Symfony\Component\Serializer\Serializer;
29 use Symfony\Component\Serializer\Encoder\XmlEncoder;
30 use Symfony\Component\Serializer\Normalizer\ObjectNormalizer;
32 date_default_timezone_set('UTC');
35 * Scrape supported websites and perform conversion to RSS.
39 private static $userAgent = "Mozilla/5.0";
40 private static $maxConnectionTimeout = 5;
41 private static $maxConnectionRetries = 5;
44 * Create a new Tweeper object controlling optional settings.
46 * @param bool $generate_enclosure
47 * Enables the creation of <enclosure/> elements (disabled by default).
48 * @param bool $show_usernames
49 * Enables showing the username in front of the content for multi-user
50 * sites (enabled by default). Only some stylesheets supports this
51 * functionality (twitter, instagram, pump.io).
52 * @param bool $show_multimedia
53 * Enables showing multimedia content (images, videos) directly in the
54 * item description (enabled by default). Only some stylesheets supports
55 * this functionality (twitter, instagram, dilbert).
57 public function __construct($generate_enclosure = FALSE, $show_usernames = TRUE, $show_multimedia = TRUE) {
58 $this->generate_enclosure = $generate_enclosure;
59 $this->show_usernames = $show_usernames;
60 $this->show_multimedia = $show_multimedia;
64 * Convert numeric Epoch to the date format expected in a RSS document.
66 public static function epochToRssDate($timestamp) {
67 if (!is_numeric($timestamp) || is_nan($timestamp)) {
71 return gmdate(DATE_RSS, $timestamp);
75 * Convert generic date string to the date format expected in a RSS document.
77 public static function strToRssDate($date) {
78 $timestamp = strtotime($date);
79 if (FALSE === $timestamp) {
83 return Tweeper::epochToRssDate($timestamp);
87 * Convert string to UpperCamelCase.
89 public static function toUpperCamelCase($str, $delim = ' ') {
90 $str_upper = ucwords($str, $delim);
91 $str_camel_case = str_replace($delim, '', $str_upper);
92 return $str_camel_case;
96 * Perform a cURL session multiple times when it fails with a timeout.
99 * a cURL session handle.
101 private static function curlExec($ch) {
105 $ret = curl_exec($ch);
106 if (FALSE === $ret) {
107 trigger_error(curl_error($ch), E_USER_WARNING);
109 } while (curl_errno($ch) == CURLE_OPERATION_TIMEDOUT && ++$attempt < Tweeper::$maxConnectionRetries);
115 * Get the contents from a URL.
117 private static function getUrlContents($url) {
118 $ch = curl_init($url);
119 curl_setopt_array($ch, array(
120 CURLOPT_HEADER => FALSE,
121 CURLOPT_CONNECTTIMEOUT => Tweeper::$maxConnectionTimeout,
122 // Follow http redirects to get the real URL.
123 CURLOPT_FOLLOWLOCATION => TRUE,
124 CURLOPT_COOKIEFILE => "",
125 CURLOPT_RETURNTRANSFER => TRUE,
126 CURLOPT_HTTPHEADER => array('Accept-language: en'),
127 CURLOPT_USERAGENT => Tweeper::$userAgent,
129 $contents = Tweeper::curlExec($ch);
136 * Get the headers from a URL.
138 private static function getUrlInfo($url) {
139 $ch = curl_init($url);
140 curl_setopt_array($ch, array(
141 CURLOPT_HEADER => TRUE,
142 CURLOPT_NOBODY => TRUE,
143 CURLOPT_CONNECTTIMEOUT => Tweeper::$maxConnectionTimeout,
144 // Follow http redirects to get the real URL.
145 CURLOPT_FOLLOWLOCATION => TRUE,
146 CURLOPT_RETURNTRANSFER => TRUE,
147 CURLOPT_USERAGENT => Tweeper::$userAgent,
150 $ret = Tweeper::curlExec($ch);
151 if (FALSE === $ret) {
156 $url_info = curl_getinfo($ch);
157 if (FALSE === $url_info) {
158 trigger_error(curl_error($ch), E_USER_WARNING);
166 * Generate an RSS <enclosure/> element.
168 public static function generateEnclosure($url) {
169 $supported_content_types = array(
170 "application/octet-stream",
190 $url_info = Tweeper::getUrlInfo($url);
191 if (FALSE === $url_info) {
192 trigger_error("Failed to retrieve info for URL: " . $url, E_USER_WARNING);
196 $supported = in_array($url_info['content_type'], $supported_content_types);
198 trigger_error("Unsupported enclosure content type \"" . $url_info['content_type'] . "\" for URL: " . $url_info['url'], E_USER_WARNING);
202 // The RSS specification says that the enclosure element URL must be http.
203 // See http://sourceforge.net/p/feedvalidator/bugs/72/
204 $http_url = preg_replace("/^https/", "http", $url_info['url']);
206 // When the server does not provide a Content-Length header,
207 // curl_getinfo() would return a negative value for
208 // "download_content_length", however RSS recommends to use 0 when the
209 // enclosure's size cannot be determined.
210 // See: https://www.feedvalidator.org/docs/error/UseZeroForUnknown.html
211 $length = max($url_info['download_content_length'], 0);
213 $dom = new DOMDocument();
214 $enc = $dom->createElement('enclosure');
215 $enc->setAttribute('url', $http_url);
216 $enc->setAttribute('length', $length);
217 $enc->setAttribute('type', $url_info['content_type']);
223 * Mimic the message from libxml.c::php_libxml_ctx_error_level()
225 private static function logXmlError($error) {
228 switch ($error->level) {
229 case LIBXML_ERR_WARNING:
230 $output .= "Warning $error->code: ";
233 case LIBXML_ERR_ERROR:
234 $output .= "Error $error->code: ";
237 case LIBXML_ERR_FATAL:
238 $output .= "Fatal Error $error->code: ";
242 $output .= trim($error->message);
245 $output .= " in $error->file";
248 $output .= " in Entity,";
251 $output .= " line $error->line";
253 trigger_error($output, E_USER_WARNING);
257 * Convert json to XML.
259 private static function jsonToXml($json, $root_node_name) {
260 // Apparently the ObjectNormalizer used afterwards is not able to handle
261 // the stdClass object created by json_decode() with the default setting
262 // $assoc = false; so use $assoc = true.
263 $data = json_decode($json, $assoc = TRUE);
268 $encoder = new XmlEncoder();
269 $normalizer = new ObjectNormalizer();
270 $serializer = new Serializer(array($normalizer), array($encoder));
272 $serializer_options = array(
273 'xml_encoding' => "UTF-8",
274 'xml_format_output' => TRUE,
275 'xml_root_node_name' => $root_node_name,
278 $xml_data = $serializer->serialize($data, 'xml', $serializer_options);
280 trigger_error("Cannot serialize data", E_USER_WARNING);
288 * Convert the Instagram content to XML.
290 private function getXmlInstagramCom($html) {
291 // Extract the json data from the html code.
292 $json_match_expr = '/window._sharedData = (.*);/';
293 $ret = preg_match($json_match_expr, $html, $matches);
295 trigger_error("Cannot match expression: $json_match_expr\n", E_USER_WARNING);
299 $data = json_decode($matches[1], $assoc = TRUE);
301 // The "qe" object contains elements which will result in invalid XML
302 // element names, so remove it.
305 // The "knobs" object contains elements with undefined namespaces, so
306 // remove it to silence an error message.
307 unset($data["knobs"]);
309 $json = json_encode($data);
311 return Tweeper::jsonToXml($json, 'instagram');
315 * Make the Facebook HTML processable.
317 private function preprocessHtmlFacebookCom($html) {
318 $html = str_replace('<!--', '', $html);
319 $html = str_replace('-->', '', $html);
324 * Convert the HTML retrieved from the site to XML.
326 private function htmlToXml($html, $host) {
327 $xmlDoc = new DOMDocument();
329 // Handle warnings and errors when loading invalid HTML.
330 $xml_errors_value = libxml_use_internal_errors(TRUE);
332 // If there is a host-specific method to get the XML data, use it!
333 $get_xml_host_method = 'getXml' . Tweeper::toUpperCamelCase($host, '.');
334 if (method_exists($this, $get_xml_host_method)) {
335 $xml_data = call_user_func_array(array($this, $get_xml_host_method), array($html));
336 $xmlDoc->loadXML($xml_data);
339 $xmlDoc->loadHTML($html);
342 foreach (libxml_get_errors() as $xml_error) {
343 Tweeper::logXmlError($xml_error);
345 libxml_clear_errors();
346 libxml_use_internal_errors($xml_errors_value);
352 * Load a stylesheet if the web site is supported.
354 private function loadStylesheet($host) {
355 $stylesheet = "file://" . __DIR__ . "/rss_converter_" . $host . ".xsl";
356 if (FALSE === file_exists($stylesheet)) {
357 trigger_error("Conversion to RSS not supported for $host ($stylesheet not found)", E_USER_WARNING);
361 $stylesheet_contents = Tweeper::getUrlContents($stylesheet);
362 if (FALSE === $stylesheet_contents) {
363 trigger_error("Cannot open $stylesheet", E_USER_WARNING);
367 $xslDoc = new DOMDocument();
368 $xslDoc->loadXML($stylesheet_contents);
370 $xsltProcessor = new XSLTProcessor();
371 $xsltProcessor->registerPHPFunctions();
372 $xsltProcessor->setParameter('', 'generate-enclosure', $this->generate_enclosure);
373 $xsltProcessor->setParameter('', 'show-usernames', $this->show_usernames);
374 $xsltProcessor->setParameter('', 'show-multimedia', $this->show_multimedia);
375 $xsltProcessor->importStylesheet($xslDoc);
377 return $xsltProcessor;
381 * Convert the site content to RSS.
383 public function tweep($src_url, $host = NULL, $validate_scheme = TRUE) {
384 $url = parse_url($src_url);
385 if (FALSE === $url) {
386 trigger_error("Invalid URL: $src_url", E_USER_WARNING);
390 if (TRUE === $validate_scheme) {
391 $scheme = $url["scheme"];
392 if (!in_array($scheme, array("http", "https"))) {
393 trigger_error("unsupported scheme: $scheme", E_USER_WARNING);
398 // If the host is not given derive it from the URL.
399 if (NULL === $host) {
400 if (empty($url["host"])) {
401 trigger_error("Invalid host in URL: $src_url", E_USER_WARNING);
404 // Strip the leading www. to be more forgiving on input URLs.
405 $host = preg_replace('/^www\./', '', $url["host"]);
408 $xsltProcessor = $this->loadStylesheet($host);
409 if (NULL === $xsltProcessor) {
413 $html = Tweeper::getUrlContents($src_url);
414 if (FALSE === $html) {
415 trigger_error("Failed to retrieve $src_url", E_USER_WARNING);
419 $preprocess_html_host_method = 'preprocessHtml' . Tweeper::toUpperCamelCase($host, '.');
420 if (method_exists($this, $preprocess_html_host_method)) {
421 $html = call_user_func_array(array($this, $preprocess_html_host_method), array($html));
424 $xmlDoc = $this->htmlToXml($html, $host);
425 if (NULL === $xmlDoc) {
429 $output = $xsltProcessor->transformToXML($xmlDoc);
430 if (FALSE === $output) {
431 trigger_error('XSL transformation failed.', E_USER_WARNING);