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 (Windows NT 6.1; WOW64; rv:60.0) Gecko/20100101 Firefox/60.0";
42 * Constructor sets up {@link $generate_enclosure}.
44 public function __construct($generate_enclosure = FALSE) {
45 $this->generate_enclosure = $generate_enclosure;
49 * Convert numeric Epoch to the date format expected in a RSS document.
51 public static function epochToRssDate($timestamp) {
52 if (!is_numeric($timestamp) || is_nan($timestamp)) {
56 return gmdate(DATE_RSS, $timestamp);
60 * Convert generic date string to the date format expected in a RSS document.
62 public static function strToRssDate($date) {
63 $timestamp = strtotime($date);
64 if (FALSE === $timestamp) {
68 return Tweeper::epochToRssDate($timestamp);
72 * Convert string to UpperCamelCase.
74 public static function toUpperCamelCase($str, $delim = ' ') {
75 $str_upper = ucwords($str, $delim);
76 $str_camel_case = str_replace($delim, '', $str_upper);
77 return $str_camel_case;
81 * Get the contents from a URL.
83 private static function getUrlContents($url) {
84 $ch = curl_init($url);
85 curl_setopt_array($ch, array(
86 CURLOPT_HEADER => FALSE,
87 // Follow http redirects to get the real URL.
88 CURLOPT_FOLLOWLOCATION => TRUE,
89 CURLOPT_RETURNTRANSFER => TRUE,
90 CURLOPT_SSL_VERIFYHOST => FALSE,
91 CURLOPT_SSL_VERIFYPEER => FALSE,
92 CURLOPT_HTTPHEADER => array('Accept-language: en'),
93 CURLOPT_USERAGENT => Tweeper::$userAgent,
95 $contents = curl_exec($ch);
96 if (FALSE === $contents) {
97 trigger_error(curl_error($ch));
105 * Get the headers from a URL.
107 private static function getUrlInfo($url) {
108 $ch = curl_init($url);
109 curl_setopt_array($ch, array(
110 CURLOPT_HEADER => TRUE,
111 CURLOPT_NOBODY => TRUE,
112 // Follow http redirects to get the real URL.
113 CURLOPT_FOLLOWLOCATION => TRUE,
114 CURLOPT_RETURNTRANSFER => TRUE,
115 CURLOPT_SSL_VERIFYHOST => FALSE,
116 CURLOPT_SSL_VERIFYPEER => FALSE,
117 CURLOPT_USERAGENT => Tweeper::$userAgent,
120 $url_info = curl_getinfo($ch);
121 if (FALSE === $url_info) {
122 trigger_error(curl_error($ch));
130 * Generate an RSS <enclosure/> element.
132 public static function generateEnclosure($url) {
133 $supported_content_types = array(
134 "application/octet-stream",
154 $url_info = Tweeper::getUrlInfo($url);
156 $supported = in_array($url_info['content_type'], $supported_content_types);
158 error_log("Unsupported enclosure content type \"" . $url_info['content_type'] . "\" for URL: " . $url_info['url']);
162 // The RSS specification says that the enclosure element URL must be http.
163 // See http://sourceforge.net/p/feedvalidator/bugs/72/
164 $http_url = preg_replace("/^https/", "http", $url_info['url']);
166 $dom = new DOMDocument();
167 $enc = $dom->createElement('enclosure');
168 $enc->setAttribute('url', $http_url);
169 $enc->setAttribute('length', $url_info['download_content_length']);
170 $enc->setAttribute('type', $url_info['content_type']);
176 * Mimic the message from libxml.c::php_libxml_ctx_error_level()
178 private static function logXmlError($error) {
181 switch ($error->level) {
182 case LIBXML_ERR_WARNING:
183 $output .= "Warning $error->code: ";
186 case LIBXML_ERR_ERROR:
187 $output .= "Error $error->code: ";
190 case LIBXML_ERR_FATAL:
191 $output .= "Fatal Error $error->code: ";
195 $output .= trim($error->message);
198 $output .= " in $error->file";
201 $output .= " in Entity,";
204 $output .= " line $error->line";
210 * Convert json to XML.
212 private static function jsonToXml($json, $root_node_name) {
213 // Apparently the ObjectNormalizer used afterwards is not able to handle
214 // the stdClass object created by json_decode() with the default setting
215 // $assoc = false; so use $assoc = true.
216 $data = json_decode($json, $assoc = TRUE);
221 $encoder = new XmlEncoder();
222 $normalizer = new ObjectNormalizer();
223 $serializer = new Serializer(array($normalizer), array($encoder));
225 $serializer_options = array(
226 'xml_encoding' => "UTF-8",
227 'xml_format_output' => TRUE,
228 'xml_root_node_name' => $root_node_name,
231 $xml_data = $serializer->serialize($data, 'xml', $serializer_options);
233 trigger_error("Cannot serialize data", E_USER_ERROR);
241 * Convert the Instagram content to XML.
243 private function getXmlInstagramCom($html) {
244 // Extract the json data from the html code.
245 $json_match_expr = '/window._sharedData = (.*);/';
246 $ret = preg_match($json_match_expr, $html, $matches);
248 trigger_error("Cannot match expression: $json_match_expr\n", E_USER_ERROR);
252 $data = json_decode($matches[1], $assoc = TRUE);
254 // The "qe" object contains elements which will result in invalid XML
255 // element names, so remove it.
258 // The "knobs" object contains elements with undefined namespaces, so
259 // remove it to silence an error message.
260 unset($data["knobs"]);
262 $json = json_encode($data);
264 return Tweeper::jsonToXml($json, 'instagram');
268 * Make the Facebook HTML processable.
270 private function preprocessHtmlFacebookCom($html) {
271 $html = str_replace('<!--', '', $html);
272 $html = str_replace('-->', '', $html);
277 * Convert the HTML retrieved from the site to XML.
279 private function htmlToXml($html, $host) {
280 $xmlDoc = new DOMDocument();
282 // Handle warnings and errors when loading invalid HTML.
283 $xml_errors_value = libxml_use_internal_errors(TRUE);
285 // If there is a host-specific method to get the XML data, use it!
286 $get_xml_host_method = 'getXml' . Tweeper::toUpperCamelCase($host, '.');
287 if (method_exists($this, $get_xml_host_method)) {
288 $xml_data = call_user_func_array(array($this, $get_xml_host_method), array($html));
289 $xmlDoc->loadXML($xml_data);
292 $xmlDoc->loadHTML($html);
295 foreach (libxml_get_errors() as $xml_error) {
296 Tweeper::logXmlError($xml_error);
298 libxml_clear_errors();
299 libxml_use_internal_errors($xml_errors_value);
305 * Load a stylesheet if the web site is supported.
307 private function loadStylesheet($host) {
308 $stylesheet = "file://" . __DIR__ . "/rss_converter_" . $host . ".xsl";
309 if (FALSE === file_exists($stylesheet)) {
310 trigger_error("Conversion to RSS not supported for $host ($stylesheet not found)", E_USER_ERROR);
314 $stylesheet_contents = Tweeper::getUrlContents($stylesheet);
316 $xslDoc = new DOMDocument();
317 $xslDoc->loadXML($stylesheet_contents);
319 $xsltProcessor = new XSLTProcessor();
320 $xsltProcessor->registerPHPFunctions();
321 $xsltProcessor->setParameter('', 'generate-enclosure', $this->generate_enclosure);
322 $xsltProcessor->importStylesheet($xslDoc);
324 return $xsltProcessor;
328 * Convert the site content to RSS.
330 public function tweep($src_url, $host=NULL, $validate_scheme=TRUE) {
331 $url = parse_url($src_url);
332 if (FALSE === $url) {
333 trigger_error("Invalid URL: $src_url", E_USER_ERROR);
337 if (TRUE === $validate_scheme) {
338 $scheme = $url["scheme"];
339 if (!in_array($scheme, array("http", "https"))) {
340 trigger_error("unsupported scheme: $scheme", E_USER_ERROR);
345 // if the host is not given derive it from the URL
346 if (NULL === $host) {
347 if (empty($url["host"])) {
348 trigger_error("Invalid host in URL: $src_url", E_USER_ERROR);
351 // Strip the leading www. to be more forgiving on input URLs.
352 $host = preg_replace('/^www\./', '', $url["host"]);
355 $xsltProcessor = $this->loadStylesheet($host);
356 if (NULL === $xsltProcessor) {
360 $html = Tweeper::getUrlContents($src_url);
361 if (FALSE === $html) {
365 $preprocess_html_host_method = 'preprocessHtml' . Tweeper::toUpperCamelCase($host, '.');
366 if (method_exists($this, $preprocess_html_host_method)) {
367 $html = call_user_func_array(array($this, $preprocess_html_host_method), array($html));
370 $xmlDoc = $this->htmlToXml($html, $host);
371 if (NULL === $xmlDoc) {
375 $output = $xsltProcessor->transformToXML($xmlDoc);
376 if (FALSE === $output) {
377 trigger_error('XSL transformation failed.', E_USER_ERROR);