[Summary view] [Print] [Text view]

   1  <?php
   2  # MantisBT - A PHP based bugtracking system
   3  
   4  # MantisBT is free software: you can redistribute it and/or modify
   5  # it under the terms of the GNU General Public License as published by
   6  # the Free Software Foundation, either version 2 of the License, or
   7  # (at your option) any later version.
   8  #
   9  # MantisBT is distributed in the hope that it will be useful,
  10  # but WITHOUT ANY WARRANTY; without even the implied warranty of
  11  # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12  # GNU General Public License for more details.
  13  #
  14  # You should have received a copy of the GNU General Public License
  15  # along with MantisBT.  If not, see <http://www.gnu.org/licenses/>.
  16  
  17  /**
  18   * @package Tests
  19   * @subpackage String
  20   * @copyright Copyright (C) 2002 - 2011  MantisBT Team - mantisbt-dev@lists.sourceforge.net
  21   * @link http://www.mantisbt.org
  22   */
  23  
  24  require_once 'PHPUnit/Framework.php';
  25  
  26  //require_once 'string_api.php';
  27  
  28  class Mantis_StringTest extends PHPUnit_Framework_TestCase {
  29  
  30      /**
  31        * Tests string_sanitize_url()
  32        *
  33        * @dataProvider provider
  34        */
  35      public function testStringSanitize( $in, $out )
  36      {
  37          $a = string_do_sanitize_url($in, false);
  38          $this->assertEquals( $out, $a );
  39      }
  40  
  41      public function provider()
  42      {
  43          $testStrings = array(
  44              array( '', 'index.php' ),
  45              array( 'abc.php', 'abc.php' ),
  46              array( 'abc.php?','abc.php'),
  47              array( 'abc.php#a','abc.php#a'),
  48              array( 'abc.php?abc=def','abc.php?abc=def'),
  49              array( 'abc.php?abc=def#a','abc.php?abc=def#a'),
  50              array( 'abc.php?abc=def&z=xyz','abc.php?abc=def&amp;z=xyz'),
  51              array( 'abc.php?abc=def&amp;z=xyz','abc.php?abc=def&amp;z=xyz'),
  52              array( 'abc.php?abc=def&z=xyz#a','abc.php?abc=def&amp;z=xyz#a'),
  53              array( 'abc.php?abc=def&amp;z=xyz#a','abc.php?abc=def&amp;z=xyz#a'),
  54  /*    FIXME    array( 'abc.php?abc=def&z=<script>alert("foo")</script>z#a','abc.php?abc=def&amp;z=alert%28%22foo%29%22%3cz#a'), */
  55  /* FIXME    array( 'abc.php?abc=def&z=z#<script>alert("foo")</script>a','abc.php?abc=def&amp;z=z#alert%28%22foo%22%3ca'), */
  56              array( 'plugin.php?page=Source/index','plugin.php?page=Source%2Findex'),
  57              array( 'plugin.php?page=Source/list&id=1','plugin.php?page=Source%2Flist&amp;id=1'),
  58              array( 'plugin.php?page=Source/list&id=1#abc','plugin.php?page=Source%2Flist&amp;id=1#abc'),
  59             );
  60  
  61          /*
  62             FIXME
  63              array( $my_path.'abc.php',
  64              array( $my_path.'abc.php?',
  65              array( $my_path.'abc.php#a',
  66              array( $my_path.'abc.php?abc=def',
  67              array( $my_path.'abc.php?abc=def#a',
  68              array( $my_path.'abc.php?abc=def&z=xyz',
  69              array( $my_path.'abc.php?abc=def&amp;z=xyz',
  70              array( $my_path.'abc.php?abc=def&z=xyz#a',
  71              array( $my_path.'abc.php?abc=def&amp;z=xyz#a',
  72              array( $my_path.'abc.php?abc=def&z=<script>alert("foo")</script>z#a',
  73              array( $my_path.'abc.php?abc=def&z=z#<script>alert("foo")</script>a',
  74              array( $my_path.'plugin.php?page=Source/index',
  75              array( $my_path.'plugin.php?page=Source/list&id=1',
  76              array( $my_path.'plugin.php?page=Source/list&id=1#abc',
  77              array( 'http://www.test.my.url/'),
  78          */
  79             return $testStrings;
  80   }
  81  
  82  }
  83  
  84  
  85  // FIXME: hardcoced here to avoid external dependencies, should use code in string_api.php
  86  function string_do_sanitize_url( $p_url, $p_return_absolute = false ) {
  87      $t_url = strip_tags( urldecode( $p_url ) );
  88  
  89      $t_path = '/';
  90      $t_short_path = '/';
  91  
  92      $t_pattern = '(?:/*(?P<script>[^\?#]*))(?:\?(?P<query>[^#]*))?(?:#(?P<anchor>[^#]*))?';
  93  
  94      # Break the given URL into pieces for path, script, query, and anchor
  95      $t_type = 0;
  96      if ( preg_match( "@^(?P<path>$t_path)$t_pattern\$@", $t_url, $t_matches ) ) {
  97          $t_type = 1;
  98      } else if ( preg_match( "@^(?P<path>$t_short_path)$t_pattern\$@", $t_url, $t_matches ) ) {
  99          $t_type = 2;
 100      } else if ( preg_match( "@^(?P<path>)$t_pattern\$@", $t_url, $t_matches ) ) {
 101          $t_type = 3;
 102      }
 103  
 104      # Check for URL's pointing to other domains
 105      if ( 0 == $t_type || empty( $t_matches['script'] ) ||
 106          3 == $t_type && preg_match( '@(?:[^:]*)?://@', $t_url ) > 0 ) {
 107  
 108          return ( $p_return_absolute ? $t_path . '/' : '' ) . 'index.php';
 109      }
 110  
 111      # Start extracting regex matches
 112      $t_script = $t_matches['script'];
 113      $t_script_path = $t_matches['path'];
 114  
 115      # Clean/encode query params
 116      $t_query = '';
 117      if ( isset( $t_matches['query'] ) ) {
 118          $t_pairs = array();
 119          parse_str( html_entity_decode( $t_matches['query'] ), $t_pairs );
 120  
 121          $t_clean_pairs = array();
 122          foreach( $t_pairs as $t_key => $t_value ) {
 123              $t_clean_pairs[] = rawurlencode( $t_key ) . '=' . rawurlencode( $t_value );
 124          }
 125  
 126          if ( !empty( $t_clean_pairs ) ) {
 127              $t_query = '?' . join( '&amp;', $t_clean_pairs );
 128          }
 129      }
 130  
 131      # encode link anchor
 132      $t_anchor = '';
 133      if ( isset( $t_matches['anchor'] ) ) {
 134          $t_anchor = '#' . rawurlencode( $t_matches['anchor'] );
 135      }
 136  
 137      # Return an appropriate re-combined URL string
 138      if ( $p_return_absolute ) {
 139          return $t_path . '/' . $t_script . $t_query . $t_anchor;
 140      } else {
 141          return ( !empty( $t_script_path ) ? $t_script_path . '/' : '' ) . $t_script . $t_query . $t_anchor;
 142      }
 143  }
 144