3 * tweeper - a Twitter to RSS web scraper
5 * Copyright (C) 2013 Antonio Ospite <ospite@studenti.unina.it>
7 * This program is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU General Public License as published by
9 * the Free Software Foundation, either version 3 of the License, or
10 * (at your option) any later version.
12 * This program is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU General Public License for more details.
17 * You should have received a copy of the GNU General Public License
18 * along with this program. If not, see <http://www.gnu.org/licenses/>.
21 date_default_timezone_set('UTC');
25 private static $USER_AGENT = "Mozilla/5.0 (Windows NT 6.1; rv:22.0) Gecko/20130405 Firefox/22.0";
27 public function __construct($stylesheet, $generate_enclosure = FALSE) {
28 $stylesheet_contents = $this->get_contents($stylesheet);
30 $xslDoc = new DOMDocument();
31 $xslDoc->loadXML($stylesheet_contents);
33 $this->xsltProcessor = new XSLTProcessor();
34 $this->xsltProcessor->registerPHPFunctions();
35 $this->xsltProcessor->setParameter('', 'generateEnclosure', $generate_enclosure);
36 $this->xsltProcessor->importStylesheet($xslDoc);
39 public static function epoch_to_gmdate($timestamp)
41 return gmdate('D, d M Y H:i:s', $timestamp) . ' GMT';
44 public static function str_to_gmdate($date)
46 $timestamp = strtotime($date);
47 return Tweeper::epoch_to_gmdate($timestamp);
50 private static function get_contents($url)
52 $ch = curl_init($url);
53 curl_setopt_array($ch, array(
54 CURLOPT_HEADER => FALSE,
55 CURLOPT_RETURNTRANSFER => TRUE,
56 CURLOPT_SSL_VERIFYHOST => FALSE,
57 CURLOPT_SSL_VERIFYPEER => FALSE,
58 CURLOPT_HTTPHEADER => array('Accept-language: en'),
59 CURLOPT_USERAGENT => Tweeper::$USER_AGENT,
61 $contents = curl_exec($ch);
67 private static function get_info($url)
69 $ch = curl_init($url);
70 curl_setopt_array($ch, array(
71 CURLOPT_HEADER => TRUE,
72 CURLOPT_NOBODY => TRUE,
73 CURLOPT_FOLLOWLOCATION => TRUE, // follow http redirects to get the real URL
74 CURLOPT_RETURNTRANSFER => TRUE,
75 CURLOPT_SSL_VERIFYHOST => FALSE,
76 CURLOPT_SSL_VERIFYPEER => FALSE,
77 CURLOPT_USERAGENT => Tweeper::$USER_AGENT,
80 $url_info = curl_getinfo($ch);
86 public static function generate_enclosure($url)
88 $supported_content_types = array(
105 $url_info = Tweeper::get_info($url);
107 $supported = in_array($url_info['content_type'], $supported_content_types);
109 error_log("Unsupported enclosure content type \"" . $url_info['content_type'] . "\" for URL: " . $url_info['url']);
113 $dom = new DomDocument();
114 $enc = $dom->createElement('enclosure');
115 $enc->setAttribute('url', $url_info['url']);
116 $enc->setAttribute('length', $url_info['download_content_length']);
117 $enc->setAttribute('type', $url_info['content_type']);
119 $dom->appendChild($enc);
121 return $dom->saveXML($enc);
124 public function tweep($uri) {
125 $html = Tweeper::get_contents($uri);
127 $xmlDoc = new DOMDocument();
128 $xmlDoc->loadHTML($html);
130 $output = $this->xsltProcessor->transformToXML($xmlDoc);
132 if (FALSE === $output) {
133 trigger_error('XSL transformation failed.', E_USER_ERROR);
140 function usage($argv)
142 if (php_sapi_name() != 'cli')
143 $usage = htmlentities("{$_SERVER['SCRIPT_NAME']}?src_url=<src_url>&generate_enclosure=<0|1>");
145 $usage = "{$argv[0]} [-e|-h|--help] <src_url>\n";
147 return "usage: $usage";
150 function parse_options_cli($argv, $argc)
153 'generate_enclosure' => FALSE
159 $cli_options = getopt("eh", array("help"));
160 foreach ($cli_options as $opt => $val) {
163 $options['generate_enclosure'] = TRUE;
174 $options['src_url'] = $argv[count($cli_options) + 1];
179 function parse_options_query_string()
182 'generate_enclosure' => FALSE
185 if (isset($_GET['src_url']))
186 $options['src_url'] = $_GET['src_url'];
188 if (isset($_GET['generate_enclosure']))
189 $options['generate_enclosure'] = $_GET['generate_enclosure'] == 1;
195 if (php_sapi_name() != 'cli')
196 $options = parse_options_query_string();
198 $options = parse_options_cli($argv, $argc);
200 if (!isset($options['src_url']))
203 $url = parse_url($options['src_url']);
204 if (FALSE === $url || empty($url["host"]))
205 die("Invalid url: ${options['src_url']}\n");
207 $stylesheet = "file://" . __DIR__ . "/rss_converter_" . $url["host"] . ".xsl";
208 if (FALSE === file_exists($stylesheet))
209 die("Conversion to RSS not supported: {$url["host"]}\n");
211 $tweeper = new Tweeper($stylesheet, $options['generate_enclosure']);
212 echo $tweeper->tweep($options['src_url']);