sanitize utf8mb4 characters

gistfile1.php

/**
 * sanitize utf8mb4 characters
 */
function _utf8_4byte_to_3byte($input) {
  if (!empty($input)) {
    $utf8_2byte = 0xC0 /*1100 0000*/; $utf8_2byte_bmask = 0xE0 /*1110 0000*/;
    $utf8_3byte = 0xE0 /*1110 0000*/; $utf8_3byte_bmask = 0XF0 /*1111 0000*/;
    $utf8_4byte = 0xF0 /*1111 0000*/; $utf8_4byte_bmask = 0xF8 /*1111 1000*/;

    $sanitized = "";
    $len = strlen($input);
    for ($i = 0; $i < $len; ++$i) {
      $mb_char = $input[$i]; // Potentially a multibyte sequence
      $byte = ord($mb_char);
      if (($byte & $utf8_2byte_bmask) == $utf8_2byte) {
        $mb_char .= $input[++$i];
      }
      else if (($byte & $utf8_3byte_bmask) == $utf8_3byte) {
        $mb_char .= $input[++$i];
        $mb_char .= $input[++$i];
      }
      else if (($byte & $utf8_4byte_bmask) == $utf8_4byte) {
        // Replace with ? to avoid MySQL exception
        //$mb_char = '?';
        $mb_char = ' ';
        $i += 3;
      }
      $sanitized .=  $mb_char;
    }
    $input= $sanitized;
  }
  return $input;
}