[Summary view] [Print] [Text view]

   1  <?php
   2  /**
   3   * Test feeds random 16-byte strings to both the pure PHP and ICU-based
   4   * UtfNormal::cleanUp() code paths, and checks to see if there's a
   5   * difference. Will run forever until it finds one or you kill it.
   6   *
   7   * Copyright (C) 2004 Brion Vibber <[email protected]>
   8   * https://www.mediawiki.org/
   9   *
  10   * This program is free software; you can redistribute it and/or modify
  11   * it under the terms of the GNU General Public License as published by
  12   * the Free Software Foundation; either version 2 of the License, or
  13   * (at your option) any later version.
  14   *
  15   * This program is distributed in the hope that it will be useful,
  16   * but WITHOUT ANY WARRANTY; without even the implied warranty of
  17   * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  18   * GNU General Public License for more details.
  19   *
  20   * You should have received a copy of the GNU General Public License along
  21   * with this program; if not, write to the Free Software Foundation, Inc.,
  22   * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  23   * http://www.gnu.org/copyleft/gpl.html
  24   *
  25   * @file
  26   * @ingroup UtfNormal
  27   */
  28  
  29  if ( PHP_SAPI != 'cli' ) {
  30      die( "Run me from the command line please.\n" );
  31  }
  32  
  33  /** */
  34  require_once  'UtfNormal.php';
  35  require_once  '../diff/DifferenceEngine.php';
  36  
  37  dl( 'php_utfnormal.so' );
  38  
  39  # mt_srand( 99999 );
  40  
  41  function randomString( $length, $nullOk, $ascii = false ) {
  42      $out = '';
  43      for ( $i = 0; $i < $length; $i++ )
  44          $out .= chr( mt_rand( $nullOk ? 0 : 1, $ascii ? 127 : 255 ) );
  45  
  46      return $out;
  47  }
  48  
  49  /* Duplicate of the cleanUp() path for ICU usage */
  50  function donorm( $str ) {
  51      # We exclude a few chars that ICU would not.
  52      $str = preg_replace( '/[\x00-\x08\x0b\x0c\x0e-\x1f]/', UTF8_REPLACEMENT, $str );
  53      $str = str_replace( UTF8_FFFE, UTF8_REPLACEMENT, $str );
  54      $str = str_replace( UTF8_FFFF, UTF8_REPLACEMENT, $str );
  55  
  56      # UnicodeString constructor fails if the string ends with a head byte.
  57      # Add a junk char at the end, we'll strip it off
  58      return rtrim( utf8_normalize( $str . "\x01", UtfNormal::UNORM_NFC ), "\x01" );
  59  }
  60  
  61  function showDiffs( $a, $b ) {
  62      $ota = explode( "\n", str_replace( "\r\n", "\n", $a ) );
  63      $nta = explode( "\n", str_replace( "\r\n", "\n", $b ) );
  64  
  65      $diffs = new Diff( $ota, $nta );
  66      $formatter = new TableDiffFormatter();
  67      $funky = $formatter->format( $diffs );
  68      $matches = array();
  69      preg_match_all( '/<(?:ins|del) class="diffchange">(.*?)<\/(?:ins|del)>/', $funky, $matches );
  70      foreach ( $matches[1] as $bit ) {
  71          $hex = bin2hex( $bit );
  72          echo "\t$hex\n";
  73      }
  74  }
  75  
  76  $size = 16;
  77  $n = 0;
  78  while ( true ) {
  79      $n++;
  80      echo "$n\n";
  81  
  82      $str = randomString( $size, true );
  83      $clean = UtfNormal::cleanUp( $str );
  84      $norm = donorm( $str );
  85  
  86      echo strlen( $clean ) . ", " . strlen( $norm );
  87      if ( $clean == $norm ) {
  88          echo " (match)\n";
  89      } else {
  90          echo " (FAIL)\n";
  91          echo "\traw: " . bin2hex( $str ) . "\n" .
  92              "\tphp: " . bin2hex( $clean ) . "\n" .
  93              "\ticu: " . bin2hex( $norm ) . "\n";
  94          echo "\n\tdiffs:\n";
  95          showDiffs( $clean, $norm );
  96          die();
  97      }
  98  
  99      $str = '';
 100      $clean = '';
 101      $norm = '';
 102  }