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');
23 define('USER_AGENT', 'Mozilla/5.0 (Windows NT 6.1; rv:22.0) Gecko/20130405 Firefox/22.0');
25 function str_to_gmdate($date)
27 $timestamp = strtotime($date);
28 return epoch_to_gmdate($timestamp);
31 function epoch_to_gmdate($timestamp)
33 return gmdate('D, d M Y H:i:s', $timestamp) . ' GMT';
36 function get_url_info($url)
38 $ch = curl_init($url);
39 curl_setopt_array($ch, array(
40 CURLOPT_HEADER => TRUE,
41 CURLOPT_NOBODY => TRUE,
42 CURLOPT_FOLLOWLOCATION => TRUE, // follow http redirects to get the real URL
43 CURLOPT_RETURNTRANSFER => TRUE,
44 CURLOPT_SSL_VERIFYHOST => FALSE,
45 CURLOPT_SSL_VERIFYPEER => FALSE,
46 CURLOPT_USERAGENT => USER_AGENT,
49 $url_info = curl_getinfo($ch);;
55 function generate_enclosure($url)
57 $supported_content_types = array(
74 $url_info = get_url_info($url);
76 $supported = in_array($url_info['content_type'], $supported_content_types);
78 error_log("Unsupported enclosure content type \"" . $url_info['content_type'] . "\" for URL: " . $url_info['url']);
82 $dom = new DomDocument();
83 $enc = $dom->createElement('enclosure');
84 $enc->setAttribute('url', $url_info['url']);
85 $enc->setAttribute('length', $url_info['download_content_length']);
86 $enc->setAttribute('type', $url_info['content_type']);
88 $dom->appendChild($enc);
90 return $dom->saveXML($enc);
96 public function __construct($stylesheet, $generate_enclosure = FALSE) {
97 $stylesheet_contents = $this->get_contents($stylesheet);
99 $xslDoc = new DOMDocument();
100 $xslDoc->loadXML($stylesheet_contents);
102 $this->xsltProcessor = new XSLTProcessor();
103 $this->xsltProcessor->registerPHPFunctions();
104 $this->xsltProcessor->setParameter('', 'generateEnclosure', $generate_enclosure);
105 $this->xsltProcessor->importStylesheet($xslDoc);
108 private function get_contents($uri) {
109 # https://www.wjsams.com/c/docs/Wiki/Php-HowToSetUserAgentOnFileGetContents
113 'header' => join(array(
114 "Accept-language: en\r\n",
115 "User-Agent: {" + USER_AGENT + "}\r\n"
120 $context = stream_context_create($opts);
121 $contents = file_get_contents($uri, false, $context);
125 public function tweep($uri) {
126 $html = $this->get_contents($uri);
128 $xmlDoc = new DOMDocument();
129 $xmlDoc->loadHTML($html);
131 $output = $this->xsltProcessor->transformToXML($xmlDoc);
133 if (FALSE === $output) {
134 trigger_error('XSL transformation failed.', E_USER_ERROR);
141 function usage($argv)
143 if (php_sapi_name() != 'cli')
144 $usage = htmlentities("{$_SERVER['SCRIPT_NAME']}?src_url=<src_url>&generate_enclosure=<0|1>");
146 $usage = "{$argv[0]} [-e|-h|--help] <src_url>\n";
148 return "usage: $usage";
151 function parse_options_cli($argv, $argc)
154 'generate_enclosure' => FALSE
160 $cli_options = getopt("eh", array("help"));
161 foreach ($cli_options as $opt => $val) {
164 $options['generate_enclosure'] = TRUE;
175 $options['src_url'] = $argv[count($cli_options) + 1];
180 function parse_options_query_string()
183 'generate_enclosure' => FALSE
186 if (isset($_GET['src_url']))
187 $options['src_url'] = $_GET['src_url'];
189 if (isset($_GET['generate_enclosure']))
190 $options['generate_enclosure'] = $_GET['generate_enclosure'] == 1;
196 if (php_sapi_name() != 'cli')
197 $options = parse_options_query_string();
199 $options = parse_options_cli($argv, $argc);
201 if (!isset($options['src_url']))
204 $url = parse_url($options['src_url']);
205 if (FALSE === $url || empty($url["host"]))
206 die("Invalid url: ${options['src_url']}\n");
208 $stylesheet = __DIR__ . "/rss_converter_" . $url["host"] . ".xsl";
209 if (FALSE === file_exists($stylesheet))
210 die("Conversion to RSS not supported: {$url["host"]}\n");
212 $tweeper = new Tweeper($stylesheet, $options['generate_enclosure']);
213 echo $tweeper->tweep($options['src_url']);