4 * Tweeper - a Twitter to RSS web scraper.
6 * Copyright (C) 2013-2015 Antonio Ospite <ao2@ao2.it>
8 * This program is free software: you can redistribute it and/or modify
9 * it under the terms of the GNU General Public License as published by
10 * the Free Software Foundation, either version 3 of the License, or
11 * (at your option) any later version.
13 * This program is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 * GNU General Public License for more details.
18 * You should have received a copy of the GNU General Public License
19 * along with this program. If not, see <http://www.gnu.org/licenses/>.
22 require_once 'Symfony/Component/Serializer/autoload.php';
24 use Symfony\Component\Serializer\Serializer;
25 use Symfony\Component\Serializer\Encoder\XmlEncoder;
26 use Symfony\Component\Serializer\Normalizer\ObjectNormalizer;
28 date_default_timezone_set('UTC');
31 * Scrape supported websites and perform conversion to RSS.
35 private static $userAgent = "Mozilla/5.0 (Windows NT 6.1; rv:22.0) Gecko/20130405 Firefox/22.0";
38 * Constructor sets up {@link $generate_enclosure}.
40 public function __construct($generate_enclosure = FALSE) {
41 $this->generate_enclosure = $generate_enclosure;
45 * Convert numeric Epoch to the date format expected in a RSS document.
47 public static function epochToRssDate($timestamp) {
48 if (!is_numeric($timestamp) || is_nan($timestamp)) {
52 return gmdate(DATE_RSS, $timestamp);
56 * Convert generic date string to the date format expected in a RSS document.
58 public static function strToRssDate($date) {
59 $timestamp = strtotime($date);
60 if (FALSE === $timestamp) {
64 return Tweeper::epochToRssDate($timestamp);
68 * Convert string to UpperCamelCase.
70 public static function toUpperCamelCase($str, $delim = ' ') {
71 $str_upper = ucwords($str, $delim);
72 $str_camel_case = str_replace($delim, '', $str_upper);
73 return $str_camel_case;
77 * Get the contents from a URL.
79 private static function getUrlContents($url) {
80 $ch = curl_init($url);
81 curl_setopt_array($ch, array(
82 CURLOPT_HEADER => FALSE,
83 // Follow http redirects to get the real URL.
84 CURLOPT_FOLLOWLOCATION => TRUE,
85 CURLOPT_RETURNTRANSFER => TRUE,
86 CURLOPT_SSL_VERIFYHOST => FALSE,
87 CURLOPT_SSL_VERIFYPEER => FALSE,
88 CURLOPT_HTTPHEADER => array('Accept-language: en'),
89 CURLOPT_USERAGENT => Tweeper::$userAgent,
91 $contents = curl_exec($ch);
92 if (FALSE === $contents) {
93 trigger_error(curl_error($ch));
101 * Get the headers from a URL.
103 private static function getUrlInfo($url) {
104 $ch = curl_init($url);
105 curl_setopt_array($ch, array(
106 CURLOPT_HEADER => TRUE,
107 CURLOPT_NOBODY => TRUE,
108 // Follow http redirects to get the real URL.
109 CURLOPT_FOLLOWLOCATION => TRUE,
110 CURLOPT_RETURNTRANSFER => TRUE,
111 CURLOPT_SSL_VERIFYHOST => FALSE,
112 CURLOPT_SSL_VERIFYPEER => FALSE,
113 CURLOPT_USERAGENT => Tweeper::$userAgent,
116 $url_info = curl_getinfo($ch);
117 if (FALSE === $url_info) {
118 trigger_error(curl_error($ch));
126 * Generate an RSS <enclosure/> element.
128 public static function generateEnclosure($url) {
129 $supported_content_types = array(
130 "application/octet-stream",
150 $url_info = Tweeper::getUrlInfo($url);
152 $supported = in_array($url_info['content_type'], $supported_content_types);
154 error_log("Unsupported enclosure content type \"" . $url_info['content_type'] . "\" for URL: " . $url_info['url']);
158 // The RSS specification says that the enclosure element URL must be http.
159 // See http://sourceforge.net/p/feedvalidator/bugs/72/
160 $http_url = preg_replace("/^https/", "http", $url_info['url']);
162 $dom = new DOMDocument();
163 $enc = $dom->createElement('enclosure');
164 $enc->setAttribute('url', $http_url);
165 $enc->setAttribute('length', $url_info['download_content_length']);
166 $enc->setAttribute('type', $url_info['content_type']);
172 * Mimic the message from libxml.c::php_libxml_ctx_error_level()
174 private static function logXmlError($error) {
177 switch ($error->level) {
178 case LIBXML_ERR_WARNING:
179 $output .= "Warning $error->code: ";
182 case LIBXML_ERR_ERROR:
183 $output .= "Error $error->code: ";
186 case LIBXML_ERR_FATAL:
187 $output .= "Fatal Error $error->code: ";
191 $output .= trim($error->message);
194 $output .= " in $error->file";
197 $output .= " in Entity,";
200 $output .= " line $error->line";
206 * Convert json to XML.
208 private static function jsonToXml($json, $root_node_name) {
209 // Apparently the ObjectNormalizer used afterwards is not able to handle
210 // the stdClass object created by json_decode() with the default setting
211 // $assoc = false; so use $assoc = true.
212 $data = json_decode($json, $assoc = TRUE);
217 $encoder = new XmlEncoder();
218 $normalizer = new ObjectNormalizer();
219 $serializer = new Serializer(array($normalizer), array($encoder));
221 $serializer_options = array(
222 'xml_encoding' => "UTF-8",
223 'xml_format_output' => TRUE,
224 'xml_root_node_name' => $root_node_name,
227 $xml_data = $serializer->serialize($data, 'xml', $serializer_options);
229 trigger_error("Cannot serialize data", E_USER_ERROR);
237 * Convert the Instagram content to XML.
239 private function getXmlInstagramCom($html) {
240 // Extract the json data from the html code.
241 $json_match_expr = '/window._sharedData = (.*);/';
242 $ret = preg_match($json_match_expr, $html, $matches);
244 trigger_error("Cannot match expression: $json_match_expr\n", E_USER_ERROR);
248 return Tweeper::jsonToXml($matches[1], 'instagram');
252 * Make the Facebook HTML processable.
254 private function preprocessHtmlFacebookCom($html) {
255 $html = str_replace('<!--', '', $html);
256 $html = str_replace('-->', '', $html);
261 * Convert the HTML retrieved from the site to XML.
263 private function htmlToXml($html, $host) {
264 $xmlDoc = new DOMDocument();
266 // Handle warnings and errors when loading invalid HTML.
267 $xml_errors_value = libxml_use_internal_errors(TRUE);
269 // If there is a host-specific method to get the XML data, use it!
270 $get_xml_host_method = 'getXml' . Tweeper::toUpperCamelCase($host, '.');
271 if (method_exists($this, $get_xml_host_method)) {
272 $xml_data = call_user_func_array(array($this, $get_xml_host_method), array($html));
273 $xmlDoc->loadXML($xml_data);
276 $xmlDoc->loadHTML($html);
279 foreach (libxml_get_errors() as $xml_error) {
280 Tweeper::logXmlError($xml_error);
282 libxml_clear_errors();
283 libxml_use_internal_errors($xml_errors_value);
289 * Load a stylesheet if the web site is supported.
291 private function loadStylesheet($host) {
292 $stylesheet = "file://" . __DIR__ . "/rss_converter_" . $host . ".xsl";
293 if (FALSE === file_exists($stylesheet)) {
294 trigger_error("Conversion to RSS not supported for $host ($stylesheet not found)", E_USER_ERROR);
298 $stylesheet_contents = Tweeper::getUrlContents($stylesheet);
300 $xslDoc = new DOMDocument();
301 $xslDoc->loadXML($stylesheet_contents);
303 $xsltProcessor = new XSLTProcessor();
304 $xsltProcessor->registerPHPFunctions();
305 $xsltProcessor->setParameter('', 'generate-enclosure', $this->generate_enclosure);
306 $xsltProcessor->importStylesheet($xslDoc);
308 return $xsltProcessor;
312 * Convert the site content to RSS.
314 public function tweep($src_url) {
315 $url = parse_url($src_url);
316 if (FALSE === $url || empty($url["host"])) {
317 trigger_error("Invalid URL: $src_url", E_USER_ERROR);
321 // Strip the leading www. to be more forgiving on input URLs.
322 $host = preg_replace('/^www\./', '', $url["host"]);
324 $xsltProcessor = $this->loadStylesheet($host);
325 if (NULL === $xsltProcessor) {
329 $html = Tweeper::getUrlContents($src_url);
330 if (FALSE === $html) {
334 $preprocess_html_host_method = 'preprocessHtml' . Tweeper::toUpperCamelCase($host, '.');
335 if (method_exists($this, $preprocess_html_host_method)) {
336 $html = call_user_func_array(array($this, $preprocess_html_host_method), array($html));
339 $xmlDoc = $this->htmlToXml($html, $host);
340 if (NULL === $xmlDoc) {
344 $output = $xsltProcessor->transformToXML($xmlDoc);
346 if (FALSE === $output) {
347 trigger_error('XSL transformation failed.', E_USER_ERROR);
356 * Check if the script is being run from the command line.
359 return (php_sapi_name() === "cli");
363 * Show the script usage.
365 function usage($argv) {
367 $usage = "{$argv[0]} [-e|-h|--help] <src_url>\n";
370 $usage = htmlentities("{$_SERVER['SCRIPT_NAME']}?src_url=<src_url>&generate_enclosure=<0|1>");
373 return "usage: $usage";
377 * Parse command line options.
379 function parse_options_cli($argv, $argc) {
381 'generate_enclosure' => FALSE,
388 $cli_options = getopt("eh", array("help"));
389 foreach ($cli_options as $opt => $val) {
392 $options['generate_enclosure'] = TRUE;
401 fwrite(STDERR, usage($argv));
406 $options['src_url'] = $argv[count($cli_options) + 1];
412 * Parse options passed from a query string.
414 function parse_options_query_string() {
416 'generate_enclosure' => FALSE,
419 if (isset($_GET['src_url'])) {
420 $options['src_url'] = $_GET['src_url'];
423 if (isset($_GET['generate_enclosure'])) {
424 $options['generate_enclosure'] = $_GET['generate_enclosure'] == 1;
431 $options = parse_options_cli($argv, $argc);
432 $error_stream = fopen('php://stderr', 'w');
435 $options = parse_options_query_string();
436 $error_stream = fopen('php://output', 'w');
439 if (!isset($options['src_url'])) {
440 fwrite($error_stream, usage(is_cli() ? $argv : NULL));
444 $tweeper = new Tweeper($options['generate_enclosure']);
445 $output = $tweeper->tweep($options['src_url']);
446 if (is_null($output)) {