<?php

/**
 * Reads the språkbanken metadata repo to retrive the list of open metadata files
 * @param string $type lexicon or corpus
 * @return array list of resources of a specific type from språkbanken metadata
 */
function get_sb_resource_list($type) {
  
  $cache = cache_get('get_sb_resource_list'.$type);
  
  $resources = array();
  
  if (empty($cache->data)){  

    $base_url = 'https://svn.spraakdata.gu.se/sb-arkiv/pub/metadata/';
    
    if($type == 'lexicon' || $type == 'corpus'){
      $html = file_get_contents($base_url.$type);

      $dom = new DOMDocument;
      $dom->loadHTML($html);

      foreach ($dom->getElementsByTagName('a') as $node) {
        $info = pathinfo($node->getAttribute('href'));

        if($info['extension'] == 'xml'){
          $resources[] = $info['filename'];
        }
      } 
    }
    
    $expire = mktime(0, 0, 0, date("m"), date("d") + 1, date("Y"));

    cache_set('get_sb_resource_list'.$type, $resources, 'cache', $expire);    
  }else{
    $resources = $cache->data;
  }
  
  return $resources;
}