[Summary view] [Print] [Text view]

   1  <?php
   2  /**
   3  * Tools for validing a UTF-8 string is well formed.
   4  * The Original Code is Mozilla Communicator client code.
   5  * The Initial Developer of the Original Code is
   6  * Netscape Communications Corporation.
   7  * Portions created by the Initial Developer are Copyright (C) 1998
   8  * the Initial Developer. All Rights Reserved.
   9  * Ported to PHP by Henri Sivonen (http://hsivonen.iki.fi)
  10  * Slight modifications to fit with phputf8 library by Harry Fuecks (hfuecks gmail com)
  11  * @see http://lxr.mozilla.org/seamonkey/source/intl/uconv/src/nsUTF8ToUnicode.cpp
  12  * @see http://lxr.mozilla.org/seamonkey/source/intl/uconv/src/nsUnicodeToUTF8.cpp
  13  * @see http://hsivonen.iki.fi/php-utf8/
  14  * @package utf8
  15  */
  16  
  17  //--------------------------------------------------------------------
  18  /**
  19  * Tests a string as to whether it's valid UTF-8 and supported by the
  20  * Unicode standard
  21  * Note: this function has been modified to simple return true or false
  22  * @author <[email protected]>
  23  * @param string UTF-8 encoded string
  24  * @return boolean true if valid
  25  * @see http://hsivonen.iki.fi/php-utf8/
  26  * @see utf8_compliant
  27  * @package utf8
  28  */
  29  function utf8_is_valid($str) {
  30  
  31      $mState = 0;     // cached expected number of octets after the current octet
  32                       // until the beginning of the next UTF8 character sequence
  33      $mUcs4  = 0;     // cached Unicode character
  34      $mBytes = 1;     // cached expected number of octets in the current sequence
  35  
  36      $len = strlen($str);
  37  
  38      for($i = 0; $i < $len; $i++) {
  39  
  40          /*
  41           * Joomla modification - As of PHP 7.4, curly brace access has been deprecated. As a result the line below has
  42           * been modified to use square brace syntax
  43           * See https://github.com/php/php-src/commit/d574df63dc375f5fc9202ce5afde23f866b6450a
  44           * for additional references
  45           */
  46          $in = ord($str[$i]);
  47  
  48          if ( $mState == 0) {
  49  
  50              // When mState is zero we expect either a US-ASCII character or a
  51              // multi-octet sequence.
  52              if (0 == (0x80 & ($in))) {
  53                  // US-ASCII, pass straight through.
  54                  $mBytes = 1;
  55  
  56              } else if (0xC0 == (0xE0 & ($in))) {
  57                  // First octet of 2 octet sequence
  58                  $mUcs4 = ($in);
  59                  $mUcs4 = ($mUcs4 & 0x1F) << 6;
  60                  $mState = 1;
  61                  $mBytes = 2;
  62  
  63              } else if (0xE0 == (0xF0 & ($in))) {
  64                  // First octet of 3 octet sequence
  65                  $mUcs4 = ($in);
  66                  $mUcs4 = ($mUcs4 & 0x0F) << 12;
  67                  $mState = 2;
  68                  $mBytes = 3;
  69  
  70              } else if (0xF0 == (0xF8 & ($in))) {
  71                  // First octet of 4 octet sequence
  72                  $mUcs4 = ($in);
  73                  $mUcs4 = ($mUcs4 & 0x07) << 18;
  74                  $mState = 3;
  75                  $mBytes = 4;
  76  
  77              } else if (0xF8 == (0xFC & ($in))) {
  78                  /* First octet of 5 octet sequence.
  79                  *
  80                  * This is illegal because the encoded codepoint must be either
  81                  * (a) not the shortest form or
  82                  * (b) outside the Unicode range of 0-0x10FFFF.
  83                  * Rather than trying to resynchronize, we will carry on until the end
  84                  * of the sequence and let the later error handling code catch it.
  85                  */
  86                  $mUcs4 = ($in);
  87                  $mUcs4 = ($mUcs4 & 0x03) << 24;
  88                  $mState = 4;
  89                  $mBytes = 5;
  90  
  91              } else if (0xFC == (0xFE & ($in))) {
  92                  // First octet of 6 octet sequence, see comments for 5 octet sequence.
  93                  $mUcs4 = ($in);
  94                  $mUcs4 = ($mUcs4 & 1) << 30;
  95                  $mState = 5;
  96                  $mBytes = 6;
  97  
  98              } else {
  99                  /* Current octet is neither in the US-ASCII range nor a legal first
 100                   * octet of a multi-octet sequence.
 101                   */
 102                  return FALSE;
 103  
 104              }
 105  
 106          } else {
 107  
 108              // When mState is non-zero, we expect a continuation of the multi-octet
 109              // sequence
 110              if (0x80 == (0xC0 & ($in))) {
 111  
 112                  // Legal continuation.
 113                  $shift = ($mState - 1) * 6;
 114                  $tmp = $in;
 115                  $tmp = ($tmp & 0x0000003F) << $shift;
 116                  $mUcs4 |= $tmp;
 117  
 118                  /**
 119                  * End of the multi-octet sequence. mUcs4 now contains the final
 120                  * Unicode codepoint to be output
 121                  */
 122                  if (0 == --$mState) {
 123  
 124                      /*
 125                      * Check for illegal sequences and codepoints.
 126                      */
 127                      // From Unicode 3.1, non-shortest form is illegal
 128                      if (((2 == $mBytes) && ($mUcs4 < 0x0080)) ||
 129                          ((3 == $mBytes) && ($mUcs4 < 0x0800)) ||
 130                          ((4 == $mBytes) && ($mUcs4 < 0x10000)) ||
 131                          (4 < $mBytes) ||
 132                          // From Unicode 3.2, surrogate characters are illegal
 133                          (($mUcs4 & 0xFFFFF800) == 0xD800) ||
 134                          // Codepoints outside the Unicode range are illegal
 135                          ($mUcs4 > 0x10FFFF)) {
 136  
 137                          return FALSE;
 138  
 139                      }
 140  
 141                      //initialize UTF8 cache
 142                      $mState = 0;
 143                      $mUcs4  = 0;
 144                      $mBytes = 1;
 145                  }
 146  
 147              } else {
 148                  /**
 149                  *((0xC0 & (*in) != 0x80) && (mState != 0))
 150                  * Incomplete multi-octet sequence.
 151                  */
 152  
 153                  return FALSE;
 154              }
 155          }
 156      }
 157      return TRUE;
 158  }
 159  
 160  //--------------------------------------------------------------------
 161  /**
 162  * Tests whether a string complies as UTF-8. This will be much
 163  * faster than utf8_is_valid but will pass five and six octet
 164  * UTF-8 sequences, which are not supported by Unicode and
 165  * so cannot be displayed correctly in a browser. In other words
 166  * it is not as strict as utf8_is_valid but it's faster. If you use
 167  * is to validate user input, you place yourself at the risk that
 168  * attackers will be able to inject 5 and 6 byte sequences (which
 169  * may or may not be a significant risk, depending on what you are
 170  * are doing)
 171  * @see utf8_is_valid
 172  * @see http://www.php.net/manual/en/reference.pcre.pattern.modifiers.php#54805
 173  * @param string UTF-8 string to check
 174  * @return boolean TRUE if string is valid UTF-8
 175  * @package utf8
 176  */
 177  function utf8_compliant($str) {
 178      if ( strlen($str) == 0 ) {
 179          return TRUE;
 180      }
 181      // If even just the first character can be matched, when the /u
 182      // modifier is used, then it's valid UTF-8. If the UTF-8 is somehow
 183      // invalid, nothing at all will match, even if the string contains
 184      // some valid sequences
 185      return (preg_match('/^.{1}/us',$str,$ar) == 1);
 186  }
 187