[Summary view]

Differences Between: [Versions 310 and 311] [Versions 310 and 400] [Versions 310 and 401] [Versions 310 and 402] [Versions 310 and 403] [Versions 39 and 310]

   1  <?php
   2  
   3  // This file is part of Moodle - http://moodle.org/
   4  //
   5  // Moodle is free software: you can redistribute it and/or modify
   6  // it under the terms of the GNU General Public License as published by
   7  // the Free Software Foundation, either version 3 of the License, or
   8  // (at your option) any later version.
   9  //
  10  // Moodle is distributed in the hope that it will be useful,
  11  // but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  13  // GNU General Public License for more details.
  14  //
  15  // You should have received a copy of the GNU General Public License
  16  // along with Moodle.  If not, see <http://www.gnu.org/licenses/>.
  17  
  18  /**
  19   * This plugin is used to access files by providing an url
  20   *
  21   * @since Moodle 2.0
  22   * @package    repository_url
  23   * @copyright  2010 Dongsheng Cai {@link http://dongsheng.org}
  24   * @license    http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
  25   */
  26  require_once($CFG->dirroot . '/repository/lib.php');
  27  require_once (__DIR__.'/locallib.php');
  28  
  29  /**
  30   * repository_url class
  31   * A subclass of repository, which is used to download a file from a specific url
  32   *
  33   * @since Moodle 2.0
  34   * @package    repository_url
  35   * @copyright  2009 Dongsheng Cai {@link http://dongsheng.org}
  36   * @license    http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
  37   */
  38  class repository_url extends repository {
  39      /** @var int Maximum time of recursion. */
  40      const MAX_RECURSION_TIME = 5;
  41      var $processedfiles = array();
  42      /** @var int Recursion counter. */
  43      var $recursioncounter = 0;
  44  
  45      /**
  46       * @param int $repositoryid
  47       * @param object $context
  48       * @param array $options
  49       */
  50      public function __construct($repositoryid, $context = SYSCONTEXTID, $options = array()){
  51          global $CFG;
  52          parent::__construct($repositoryid, $context, $options);
  53          $this->file_url = optional_param('file', '', PARAM_RAW);
  54          $this->file_url = $this->escape_url($this->file_url);
  55      }
  56  
  57      public function check_login() {
  58          if (!empty($this->file_url)) {
  59              return true;
  60          } else {
  61              return false;
  62          }
  63      }
  64      /**
  65       * @return mixed
  66       */
  67      public function print_login() {
  68          $strdownload = get_string('download', 'repository');
  69          $strname     = get_string('rename', 'repository_url');
  70          $strurl      = get_string('url', 'repository_url');
  71          if ($this->options['ajax']) {
  72              $url = new stdClass();
  73              $url->label = $strurl.': ';
  74              $url->id   = 'fileurl';
  75              $url->type = 'text';
  76              $url->name = 'file';
  77  
  78              $ret['login'] = array($url);
  79              $ret['login_btn_label'] = get_string('download', 'repository_url');
  80              $ret['allowcaching'] = true; // indicates that login form can be cached in filepicker.js
  81              return $ret;
  82          } else {
  83              echo <<<EOD
  84  <table>
  85  <tr>
  86  <td>{$strurl}: </td><td><input name="file" type="text" /></td>
  87  </tr>
  88  </table>
  89  <input type="submit" value="{$strdownload}" />
  90  EOD;
  91  
  92          }
  93      }
  94  
  95      /**
  96       * @param mixed $path
  97       * @param string $search
  98       * @return array
  99       */
 100      public function get_listing($path='', $page='') {
 101          $ret = array();
 102          $ret['list'] = array();
 103          $ret['nosearch'] = true;
 104          $ret['norefresh'] = true;
 105          $ret['nologin'] = true;
 106  
 107          $this->file_url = clean_param($this->file_url, PARAM_URL);
 108          if (empty($this->file_url)) {
 109              throw new repository_exception('validfiletype', 'repository_url');
 110          }
 111  
 112          $this->parse_file(null, $this->file_url, $ret, true);
 113          return $ret;
 114      }
 115  
 116      /**
 117       * Parses one file (either html or css)
 118       *
 119       * @param string $baseurl (optional) URL of the file where link to this file was found
 120       * @param string $relativeurl relative or absolute link to the file
 121       * @param array $list
 122       * @param bool $mainfile true only for main HTML false and false for all embedded/linked files
 123       */
 124      protected function parse_file($baseurl, $relativeurl, &$list, $mainfile = false) {
 125          if (preg_match('/([\'"])(.*)\1/', $relativeurl, $matches)) {
 126              $relativeurl = $matches[2];
 127          }
 128          if (empty($baseurl)) {
 129              $url = $relativeurl;
 130          } else {
 131              $url = htmlspecialchars_decode(url_to_absolute($baseurl, $relativeurl));
 132          }
 133          if (in_array($url, $this->processedfiles)) {
 134              // Avoid endless recursion for the same URL with same parameters.
 135              return;
 136          }
 137          // Remove the query string before check.
 138          $recursioncheckurl = preg_replace('/\?.*/', '', $url);
 139          if (in_array($recursioncheckurl, $this->processedfiles)) {
 140              $this->recursioncounter++;
 141          }
 142          if ($this->recursioncounter >= self::MAX_RECURSION_TIME) {
 143              // Avoid endless recursion for the same URL with different parameters.
 144              return;
 145          }
 146          $this->processedfiles[] = $url;
 147          $curl = new curl;
 148          $curl->setopt(array('CURLOPT_FOLLOWLOCATION' => true, 'CURLOPT_MAXREDIRS' => 3));
 149          $msg = $curl->head($url);
 150          $info = $curl->get_info();
 151          if ($info['http_code'] != 200) {
 152              if ($mainfile) {
 153                  $list['error'] = $msg;
 154              }
 155          } else {
 156              $csstoanalyze = '';
 157              if ($mainfile && (strstr($info['content_type'], 'text/html') || empty($info['content_type']))) {
 158                  // parse as html
 159                  $htmlcontent = $curl->get($info['url']);
 160                  $ddoc = new DOMDocument();
 161                  @$ddoc->loadHTML($htmlcontent);
 162                  // extract <img>
 163                  $tags = $ddoc->getElementsByTagName('img');
 164                  foreach ($tags as $tag) {
 165                      $url = $tag->getAttribute('src');
 166                      $this->add_image_to_list($info['url'], $url, $list);
 167                  }
 168                  // analyse embedded css (<style>)
 169                  $tags = $ddoc->getElementsByTagName('style');
 170                  foreach ($tags as $tag) {
 171                      if ($tag->getAttribute('type') == 'text/css') {
 172                          $csstoanalyze .= $tag->textContent."\n";
 173                      }
 174                  }
 175                  // analyse links to css (<link type='text/css' href='...'>)
 176                  $tags = $ddoc->getElementsByTagName('link');
 177                  foreach ($tags as $tag) {
 178                      if ($tag->getAttribute('type') == 'text/css' && strlen($tag->getAttribute('href'))) {
 179                          $this->parse_file($info['url'], $tag->getAttribute('href'), $list);
 180                      }
 181                  }
 182              } else if (strstr($info['content_type'], 'css')) {
 183                  // parse as css
 184                  $csscontent = $curl->get($info['url']);
 185                  $csstoanalyze .= $csscontent."\n";
 186              } else if (strstr($info['content_type'], 'image/')) {
 187                  // download this file
 188                  $this->add_image_to_list($info['url'], $info['url'], $list);
 189              } else {
 190                  $list['error'] = get_string('validfiletype', 'repository_url');
 191              }
 192  
 193              // parse all found css styles
 194              if (strlen($csstoanalyze)) {
 195                  $urls = extract_css_urls($csstoanalyze);
 196                  if (!empty($urls['property'])) {
 197                      foreach ($urls['property'] as $url) {
 198                          $this->add_image_to_list($info['url'], $url, $list);
 199                      }
 200                  }
 201                  if (!empty($urls['import'])) {
 202                      foreach ($urls['import'] as $cssurl) {
 203                          $this->parse_file($info['url'], $cssurl, $list);
 204                      }
 205                  }
 206              }
 207          }
 208      }
 209      protected function add_image_to_list($baseurl, $url, &$list) {
 210          if (empty($list['list'])) {
 211              $list['list'] = array();
 212          }
 213          $src = url_to_absolute($baseurl, htmlspecialchars_decode($url));
 214          foreach ($list['list'] as $image) {
 215              if ($image['source'] == $src) {
 216                  return;
 217              }
 218          }
 219          $list['list'][] = array(
 220              'title'=>$this->guess_filename($url, ''),
 221              'source'=>$src,
 222              'thumbnail'=>$src,
 223              'thumbnail_height'=>84,
 224              'thumbnail_width'=>84
 225          );
 226      }
 227      public function guess_filename($url, $type) {
 228          $pattern = '#\/([\w_\?\-.]+)$#';
 229          $matches = null;
 230          preg_match($pattern, $url, $matches);
 231          if (empty($matches[1])) {
 232              return $url;
 233          } else {
 234              return $matches[1];
 235          }
 236      }
 237  
 238      /**
 239       * Escapes a url by replacing spaces with %20.
 240       *
 241       * Note: In general moodle does not automatically escape urls, but for the purposes of making this plugin more user friendly
 242       * and make it consistent with some other areas in moodle (such as mod_url), urls will automatically be escaped.
 243       *
 244       * If moodle_url or PARAM_URL is changed to clean characters that need to be escaped, then this function can be removed
 245       *
 246       * @param string $url An unescaped url.
 247       * @return string The escaped url
 248       */
 249      protected function escape_url($url) {
 250          $url = str_replace('"', '%22', $url);
 251          $url = str_replace('\'', '%27', $url);
 252          $url = str_replace(' ', '%20', $url);
 253          $url = str_replace('<', '%3C', $url);
 254          $url = str_replace('>', '%3E', $url);
 255          return $url;
 256      }
 257  
 258      public function supported_returntypes() {
 259          return (FILE_INTERNAL | FILE_EXTERNAL);
 260      }
 261  
 262      /**
 263       * Return the source information
 264       *
 265       * @param stdClass $url
 266       * @return string|null
 267       */
 268      public function get_file_source_info($url) {
 269          return $url;
 270      }
 271  
 272      /**
 273       * file types supported by url downloader plugin
 274       *
 275       * @return array
 276       */
 277      public function supported_filetypes() {
 278          return array('web_image');
 279      }
 280  
 281      /**
 282       * Is this repository accessing private data?
 283       *
 284       * @return bool
 285       */
 286      public function contains_private_data() {
 287          return false;
 288      }
 289  }