Chris@0: <?php
Chris@0: 
Chris@0: namespace Drupal\Component\Utility;
Chris@0: 
Chris@0: /**
Chris@0:  * Provides Unicode-related conversions and operations.
Chris@0:  *
Chris@0:  * @ingroup utility
Chris@0:  */
Chris@0: class Unicode {
Chris@0: 
Chris@0:   /**
Chris@0:    * Matches Unicode characters that are word boundaries.
Chris@0:    *
Chris@0:    * Characters with the following General_category (gc) property values are used
Chris@0:    * as word boundaries. While this does not fully conform to the Word Boundaries
Chris@0:    * algorithm described in http://unicode.org/reports/tr29, as PCRE does not
Chris@0:    * contain the Word_Break property table, this simpler algorithm has to do.
Chris@0:    * - Cc, Cf, Cn, Co, Cs: Other.
Chris@0:    * - Pc, Pd, Pe, Pf, Pi, Po, Ps: Punctuation.
Chris@0:    * - Sc, Sk, Sm, So: Symbols.
Chris@0:    * - Zl, Zp, Zs: Separators.
Chris@0:    *
Chris@0:    * Non-boundary characters include the following General_category (gc) property
Chris@0:    * values:
Chris@0:    * - Ll, Lm, Lo, Lt, Lu: Letters.
Chris@0:    * - Mc, Me, Mn: Combining Marks.
Chris@0:    * - Nd, Nl, No: Numbers.
Chris@0:    *
Chris@0:    * Note that the PCRE property matcher is not used because we wanted to be
Chris@0:    * compatible with Unicode 5.2.0 regardless of the PCRE version used (and any
Chris@0:    * bugs in PCRE property tables).
Chris@0:    *
Chris@0:    * @see http://unicode.org/glossary
Chris@0:    */
Chris@0:   const PREG_CLASS_WORD_BOUNDARY = <<<'EOD'
Chris@0: \x{0}-\x{2F}\x{3A}-\x{40}\x{5B}-\x{60}\x{7B}-\x{A9}\x{AB}-\x{B1}\x{B4}
Chris@0: \x{B6}-\x{B8}\x{BB}\x{BF}\x{D7}\x{F7}\x{2C2}-\x{2C5}\x{2D2}-\x{2DF}
Chris@0: \x{2E5}-\x{2EB}\x{2ED}\x{2EF}-\x{2FF}\x{375}\x{37E}-\x{385}\x{387}\x{3F6}
Chris@0: \x{482}\x{55A}-\x{55F}\x{589}-\x{58A}\x{5BE}\x{5C0}\x{5C3}\x{5C6}
Chris@0: \x{5F3}-\x{60F}\x{61B}-\x{61F}\x{66A}-\x{66D}\x{6D4}\x{6DD}\x{6E9}
Chris@0: \x{6FD}-\x{6FE}\x{700}-\x{70F}\x{7F6}-\x{7F9}\x{830}-\x{83E}
Chris@0: \x{964}-\x{965}\x{970}\x{9F2}-\x{9F3}\x{9FA}-\x{9FB}\x{AF1}\x{B70}
Chris@0: \x{BF3}-\x{BFA}\x{C7F}\x{CF1}-\x{CF2}\x{D79}\x{DF4}\x{E3F}\x{E4F}
Chris@0: \x{E5A}-\x{E5B}\x{F01}-\x{F17}\x{F1A}-\x{F1F}\x{F34}\x{F36}\x{F38}
Chris@0: \x{F3A}-\x{F3D}\x{F85}\x{FBE}-\x{FC5}\x{FC7}-\x{FD8}\x{104A}-\x{104F}
Chris@0: \x{109E}-\x{109F}\x{10FB}\x{1360}-\x{1368}\x{1390}-\x{1399}\x{1400}
Chris@0: \x{166D}-\x{166E}\x{1680}\x{169B}-\x{169C}\x{16EB}-\x{16ED}
Chris@0: \x{1735}-\x{1736}\x{17B4}-\x{17B5}\x{17D4}-\x{17D6}\x{17D8}-\x{17DB}
Chris@0: \x{1800}-\x{180A}\x{180E}\x{1940}-\x{1945}\x{19DE}-\x{19FF}
Chris@0: \x{1A1E}-\x{1A1F}\x{1AA0}-\x{1AA6}\x{1AA8}-\x{1AAD}\x{1B5A}-\x{1B6A}
Chris@0: \x{1B74}-\x{1B7C}\x{1C3B}-\x{1C3F}\x{1C7E}-\x{1C7F}\x{1CD3}\x{1FBD}
Chris@0: \x{1FBF}-\x{1FC1}\x{1FCD}-\x{1FCF}\x{1FDD}-\x{1FDF}\x{1FED}-\x{1FEF}
Chris@0: \x{1FFD}-\x{206F}\x{207A}-\x{207E}\x{208A}-\x{208E}\x{20A0}-\x{20B8}
Chris@0: \x{2100}-\x{2101}\x{2103}-\x{2106}\x{2108}-\x{2109}\x{2114}
Chris@0: \x{2116}-\x{2118}\x{211E}-\x{2123}\x{2125}\x{2127}\x{2129}\x{212E}
Chris@0: \x{213A}-\x{213B}\x{2140}-\x{2144}\x{214A}-\x{214D}\x{214F}
Chris@0: \x{2190}-\x{244A}\x{249C}-\x{24E9}\x{2500}-\x{2775}\x{2794}-\x{2B59}
Chris@0: \x{2CE5}-\x{2CEA}\x{2CF9}-\x{2CFC}\x{2CFE}-\x{2CFF}\x{2E00}-\x{2E2E}
Chris@0: \x{2E30}-\x{3004}\x{3008}-\x{3020}\x{3030}\x{3036}-\x{3037}
Chris@0: \x{303D}-\x{303F}\x{309B}-\x{309C}\x{30A0}\x{30FB}\x{3190}-\x{3191}
Chris@0: \x{3196}-\x{319F}\x{31C0}-\x{31E3}\x{3200}-\x{321E}\x{322A}-\x{3250}
Chris@0: \x{3260}-\x{327F}\x{328A}-\x{32B0}\x{32C0}-\x{33FF}\x{4DC0}-\x{4DFF}
Chris@0: \x{A490}-\x{A4C6}\x{A4FE}-\x{A4FF}\x{A60D}-\x{A60F}\x{A673}\x{A67E}
Chris@0: \x{A6F2}-\x{A716}\x{A720}-\x{A721}\x{A789}-\x{A78A}\x{A828}-\x{A82B}
Chris@0: \x{A836}-\x{A839}\x{A874}-\x{A877}\x{A8CE}-\x{A8CF}\x{A8F8}-\x{A8FA}
Chris@0: \x{A92E}-\x{A92F}\x{A95F}\x{A9C1}-\x{A9CD}\x{A9DE}-\x{A9DF}
Chris@0: \x{AA5C}-\x{AA5F}\x{AA77}-\x{AA79}\x{AADE}-\x{AADF}\x{ABEB}
Chris@0: \x{E000}-\x{F8FF}\x{FB29}\x{FD3E}-\x{FD3F}\x{FDFC}-\x{FDFD}
Chris@0: \x{FE10}-\x{FE19}\x{FE30}-\x{FE6B}\x{FEFF}-\x{FF0F}\x{FF1A}-\x{FF20}
Chris@0: \x{FF3B}-\x{FF40}\x{FF5B}-\x{FF65}\x{FFE0}-\x{FFFD}
Chris@0: EOD;
Chris@0: 
Chris@0:   /**
Chris@0:    * Indicates that standard PHP (emulated) unicode support is being used.
Chris@0:    */
Chris@0:   const STATUS_SINGLEBYTE = 0;
Chris@0: 
Chris@0:   /**
Chris@0:    * Indicates that full unicode support with the PHP mbstring extension is
Chris@0:    * being used.
Chris@0:    */
Chris@0:   const STATUS_MULTIBYTE = 1;
Chris@0: 
Chris@0:   /**
Chris@0:    * Indicates an error during check for PHP unicode support.
Chris@0:    */
Chris@0:   const STATUS_ERROR = -1;
Chris@0: 
Chris@0:   /**
Chris@0:    * Holds the multibyte capabilities of the current environment.
Chris@0:    *
Chris@0:    * @var int
Chris@0:    */
Chris@0:   protected static $status = 0;
Chris@0: 
Chris@0:   /**
Chris@0:    * Gets the current status of unicode/multibyte support on this environment.
Chris@0:    *
Chris@0:    * @return int
Chris@0:    *   The status of multibyte support. It can be one of:
Chris@0:    *   - \Drupal\Component\Utility\Unicode::STATUS_MULTIBYTE
Chris@0:    *     Full unicode support using an extension.
Chris@0:    *   - \Drupal\Component\Utility\Unicode::STATUS_SINGLEBYTE
Chris@0:    *     Standard PHP (emulated) unicode support.
Chris@0:    *   - \Drupal\Component\Utility\Unicode::STATUS_ERROR
Chris@0:    *     An error occurred. No unicode support.
Chris@0:    */
Chris@0:   public static function getStatus() {
Chris@0:     return static::$status;
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Sets the value for multibyte support status for the current environment.
Chris@0:    *
Chris@0:    * The following status keys are supported:
Chris@0:    *   - \Drupal\Component\Utility\Unicode::STATUS_MULTIBYTE
Chris@0:    *     Full unicode support using an extension.
Chris@0:    *   - \Drupal\Component\Utility\Unicode::STATUS_SINGLEBYTE
Chris@0:    *     Standard PHP (emulated) unicode support.
Chris@0:    *   - \Drupal\Component\Utility\Unicode::STATUS_ERROR
Chris@0:    *     An error occurred. No unicode support.
Chris@0:    *
Chris@0:    * @param int $status
Chris@0:    *   The new status of multibyte support.
Chris@0:    */
Chris@0:   public static function setStatus($status) {
Chris@0:     if (!in_array($status, [static::STATUS_SINGLEBYTE, static::STATUS_MULTIBYTE, static::STATUS_ERROR])) {
Chris@0:       throw new \InvalidArgumentException('Invalid status value for unicode support.');
Chris@0:     }
Chris@0:     static::$status = $status;
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Checks for Unicode support in PHP and sets the proper settings if possible.
Chris@0:    *
Chris@0:    * Because of the need to be able to handle text in various encodings, we do
Chris@0:    * not support mbstring function overloading. HTTP input/output conversion
Chris@0:    * must be disabled for similar reasons.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   A string identifier of a failed multibyte extension check, if any.
Chris@0:    *   Otherwise, an empty string.
Chris@0:    */
Chris@0:   public static function check() {
Chris@0:     // Check for mbstring extension.
Chris@0:     if (!function_exists('mb_strlen')) {
Chris@0:       static::$status = static::STATUS_SINGLEBYTE;
Chris@0:       return 'mb_strlen';
Chris@0:     }
Chris@0: 
Chris@0:     // Check mbstring configuration.
Chris@0:     if (ini_get('mbstring.func_overload') != 0) {
Chris@0:       static::$status = static::STATUS_ERROR;
Chris@0:       return 'mbstring.func_overload';
Chris@0:     }
Chris@0:     if (ini_get('mbstring.encoding_translation') != 0) {
Chris@0:       static::$status = static::STATUS_ERROR;
Chris@0:       return 'mbstring.encoding_translation';
Chris@0:     }
Chris@0:     // mbstring.http_input and mbstring.http_output are deprecated and empty by
Chris@0:     // default in PHP 5.6.
Chris@0:     if (version_compare(PHP_VERSION, '5.6.0') == -1) {
Chris@0:       if (ini_get('mbstring.http_input') != 'pass') {
Chris@0:         static::$status = static::STATUS_ERROR;
Chris@0:         return 'mbstring.http_input';
Chris@0:       }
Chris@0:       if (ini_get('mbstring.http_output') != 'pass') {
Chris@0:         static::$status = static::STATUS_ERROR;
Chris@0:         return 'mbstring.http_output';
Chris@0:       }
Chris@0:     }
Chris@0: 
Chris@0:     // Set appropriate configuration.
Chris@0:     mb_internal_encoding('utf-8');
Chris@0:     mb_language('uni');
Chris@0:     static::$status = static::STATUS_MULTIBYTE;
Chris@0:     return '';
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Decodes UTF byte-order mark (BOM) into the encoding's name.
Chris@0:    *
Chris@0:    * @param string $data
Chris@0:    *   The data possibly containing a BOM. This can be the entire contents of
Chris@0:    *   a file, or just a fragment containing at least the first five bytes.
Chris@0:    *
Chris@0:    * @return string|bool
Chris@0:    *   The name of the encoding, or FALSE if no byte order mark was present.
Chris@0:    */
Chris@0:   public static function encodingFromBOM($data) {
Chris@0:     static $bomMap = [
Chris@0:       "\xEF\xBB\xBF" => 'UTF-8',
Chris@0:       "\xFE\xFF" => 'UTF-16BE',
Chris@0:       "\xFF\xFE" => 'UTF-16LE',
Chris@0:       "\x00\x00\xFE\xFF" => 'UTF-32BE',
Chris@0:       "\xFF\xFE\x00\x00" => 'UTF-32LE',
Chris@0:       "\x2B\x2F\x76\x38" => 'UTF-7',
Chris@0:       "\x2B\x2F\x76\x39" => 'UTF-7',
Chris@0:       "\x2B\x2F\x76\x2B" => 'UTF-7',
Chris@0:       "\x2B\x2F\x76\x2F" => 'UTF-7',
Chris@0:       "\x2B\x2F\x76\x38\x2D" => 'UTF-7',
Chris@0:     ];
Chris@0: 
Chris@0:     foreach ($bomMap as $bom => $encoding) {
Chris@0:       if (strpos($data, $bom) === 0) {
Chris@0:         return $encoding;
Chris@0:       }
Chris@0:     }
Chris@0:     return FALSE;
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Converts data to UTF-8.
Chris@0:    *
Chris@0:    * Requires the iconv, GNU recode or mbstring PHP extension.
Chris@0:    *
Chris@0:    * @param string $data
Chris@0:    *   The data to be converted.
Chris@0:    * @param string $encoding
Chris@0:    *   The encoding that the data is in.
Chris@0:    *
Chris@0:    * @return string|bool
Chris@0:    *   Converted data or FALSE.
Chris@0:    */
Chris@0:   public static function convertToUtf8($data, $encoding) {
Chris@0:     if (function_exists('iconv')) {
Chris@0:       return @iconv($encoding, 'utf-8', $data);
Chris@0:     }
Chris@0:     elseif (function_exists('mb_convert_encoding')) {
Chris@0:       return @mb_convert_encoding($data, 'utf-8', $encoding);
Chris@0:     }
Chris@0:     elseif (function_exists('recode_string')) {
Chris@0:       return @recode_string($encoding . '..utf-8', $data);
Chris@0:     }
Chris@0:     // Cannot convert.
Chris@0:     return FALSE;
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Truncates a UTF-8-encoded string safely to a number of bytes.
Chris@0:    *
Chris@0:    * If the end position is in the middle of a UTF-8 sequence, it scans backwards
Chris@0:    * until the beginning of the byte sequence.
Chris@0:    *
Chris@0:    * Use this function whenever you want to chop off a string at an unsure
Chris@0:    * location. On the other hand, if you're sure that you're splitting on a
Chris@0:    * character boundary (e.g. after using strpos() or similar), you can safely
Chris@0:    * use substr() instead.
Chris@0:    *
Chris@0:    * @param string $string
Chris@0:    *   The string to truncate.
Chris@0:    * @param int $len
Chris@0:    *   An upper limit on the returned string length.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The truncated string.
Chris@0:    */
Chris@0:   public static function truncateBytes($string, $len) {
Chris@0:     if (strlen($string) <= $len) {
Chris@0:       return $string;
Chris@0:     }
Chris@0:     if ((ord($string[$len]) < 0x80) || (ord($string[$len]) >= 0xC0)) {
Chris@0:       return substr($string, 0, $len);
Chris@0:     }
Chris@0:     // Scan backwards to beginning of the byte sequence.
Chris@0:     // @todo Make the code more readable in https://www.drupal.org/node/2911497.
Chris@0:     while (--$len >= 0 && ord($string[$len]) >= 0x80 && ord($string[$len]) < 0xC0) {
Chris@0:     }
Chris@0: 
Chris@0:     return substr($string, 0, $len);
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Counts the number of characters in a UTF-8 string.
Chris@0:    *
Chris@0:    * This is less than or equal to the byte count.
Chris@0:    *
Chris@0:    * @param string $text
Chris@0:    *   The string to run the operation on.
Chris@0:    *
Chris@0:    * @return int
Chris@0:    *   The length of the string.
Chris@0:    */
Chris@0:   public static function strlen($text) {
Chris@0:     if (static::getStatus() == static::STATUS_MULTIBYTE) {
Chris@0:       return mb_strlen($text);
Chris@0:     }
Chris@0:     else {
Chris@0:       // Do not count UTF-8 continuation bytes.
Chris@0:       return strlen(preg_replace("/[\x80-\xBF]/", '', $text));
Chris@0:     }
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Converts a UTF-8 string to uppercase.
Chris@0:    *
Chris@0:    * @param string $text
Chris@0:    *   The string to run the operation on.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The string in uppercase.
Chris@0:    */
Chris@0:   public static function strtoupper($text) {
Chris@0:     if (static::getStatus() == static::STATUS_MULTIBYTE) {
Chris@0:       return mb_strtoupper($text);
Chris@0:     }
Chris@0:     else {
Chris@0:       // Use C-locale for ASCII-only uppercase.
Chris@0:       $text = strtoupper($text);
Chris@0:       // Case flip Latin-1 accented letters.
Chris@0:       $text = preg_replace_callback('/\xC3[\xA0-\xB6\xB8-\xBE]/', '\Drupal\Component\Utility\Unicode::caseFlip', $text);
Chris@0:       return $text;
Chris@0:     }
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Converts a UTF-8 string to lowercase.
Chris@0:    *
Chris@0:    * @param string $text
Chris@0:    *   The string to run the operation on.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The string in lowercase.
Chris@0:    */
Chris@0:   public static function strtolower($text) {
Chris@0:     if (static::getStatus() == static::STATUS_MULTIBYTE) {
Chris@0:       return mb_strtolower($text);
Chris@0:     }
Chris@0:     else {
Chris@0:       // Use C-locale for ASCII-only lowercase.
Chris@0:       $text = strtolower($text);
Chris@0:       // Case flip Latin-1 accented letters.
Chris@0:       $text = preg_replace_callback('/\xC3[\x80-\x96\x98-\x9E]/', '\Drupal\Component\Utility\Unicode::caseFlip', $text);
Chris@0:       return $text;
Chris@0:     }
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Capitalizes the first character of a UTF-8 string.
Chris@0:    *
Chris@0:    * @param string $text
Chris@0:    *   The string to convert.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The string with the first character as uppercase.
Chris@0:    */
Chris@0:   public static function ucfirst($text) {
Chris@0:     return static::strtoupper(static::substr($text, 0, 1)) . static::substr($text, 1);
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Converts the first character of a UTF-8 string to lowercase.
Chris@0:    *
Chris@0:    * @param string $text
Chris@0:    *   The string that will be converted.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The string with the first character as lowercase.
Chris@0:    *
Chris@0:    * @ingroup php_wrappers
Chris@0:    */
Chris@0:   public static function lcfirst($text) {
Chris@0:     // Note: no mbstring equivalent!
Chris@0:     return static::strtolower(static::substr($text, 0, 1)) . static::substr($text, 1);
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Capitalizes the first character of each word in a UTF-8 string.
Chris@0:    *
Chris@0:    * @param string $text
Chris@0:    *   The text that will be converted.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The input $text with each word capitalized.
Chris@0:    *
Chris@0:    * @ingroup php_wrappers
Chris@0:    */
Chris@0:   public static function ucwords($text) {
Chris@0:     $regex = '/(^|[' . static::PREG_CLASS_WORD_BOUNDARY . '])([^' . static::PREG_CLASS_WORD_BOUNDARY . '])/u';
Chris@0:     return preg_replace_callback($regex, function (array $matches) {
Chris@0:       return $matches[1] . Unicode::strtoupper($matches[2]);
Chris@0:     }, $text);
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Cuts off a piece of a string based on character indices and counts.
Chris@0:    *
Chris@0:    * Follows the same behavior as PHP's own substr() function. Note that for
Chris@0:    * cutting off a string at a known character/substring location, the usage of
Chris@0:    * PHP's normal strpos/substr is safe and much faster.
Chris@0:    *
Chris@0:    * @param string $text
Chris@0:    *   The input string.
Chris@0:    * @param int $start
Chris@0:    *   The position at which to start reading.
Chris@0:    * @param int $length
Chris@0:    *   The number of characters to read.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The shortened string.
Chris@0:    */
Chris@0:   public static function substr($text, $start, $length = NULL) {
Chris@0:     if (static::getStatus() == static::STATUS_MULTIBYTE) {
Chris@0:       return $length === NULL ? mb_substr($text, $start) : mb_substr($text, $start, $length);
Chris@0:     }
Chris@0:     else {
Chris@0:       $strlen = strlen($text);
Chris@0:       // Find the starting byte offset.
Chris@0:       $bytes = 0;
Chris@0:       if ($start > 0) {
Chris@0:         // Count all the characters except continuation bytes from the start
Chris@0:         // until we have found $start characters or the end of the string.
Chris@0:         $bytes = -1; $chars = -1;
Chris@0:         while ($bytes < $strlen - 1 && $chars < $start) {
Chris@0:           $bytes++;
Chris@0:           $c = ord($text[$bytes]);
Chris@0:           if ($c < 0x80 || $c >= 0xC0) {
Chris@0:             $chars++;
Chris@0:           }
Chris@0:         }
Chris@0:       }
Chris@0:       elseif ($start < 0) {
Chris@0:         // Count all the characters except continuation bytes from the end
Chris@0:         // until we have found abs($start) characters.
Chris@0:         $start = abs($start);
Chris@0:         $bytes = $strlen; $chars = 0;
Chris@0:         while ($bytes > 0 && $chars < $start) {
Chris@0:           $bytes--;
Chris@0:           $c = ord($text[$bytes]);
Chris@0:           if ($c < 0x80 || $c >= 0xC0) {
Chris@0:             $chars++;
Chris@0:           }
Chris@0:         }
Chris@0:       }
Chris@0:       $istart = $bytes;
Chris@0: 
Chris@0:       // Find the ending byte offset.
Chris@0:       if ($length === NULL) {
Chris@0:         $iend = $strlen;
Chris@0:       }
Chris@0:       elseif ($length > 0) {
Chris@0:         // Count all the characters except continuation bytes from the starting
Chris@0:         // index until we have found $length characters or reached the end of
Chris@0:         // the string, then backtrace one byte.
Chris@0:         $iend = $istart - 1;
Chris@0:         $chars = -1;
Chris@0:         $last_real = FALSE;
Chris@0:         while ($iend < $strlen - 1 && $chars < $length) {
Chris@0:           $iend++;
Chris@0:           $c = ord($text[$iend]);
Chris@0:           $last_real = FALSE;
Chris@0:           if ($c < 0x80 || $c >= 0xC0) {
Chris@0:             $chars++;
Chris@0:             $last_real = TRUE;
Chris@0:           }
Chris@0:         }
Chris@0:         // Backtrace one byte if the last character we found was a real
Chris@0:         // character and we don't need it.
Chris@0:         if ($last_real && $chars >= $length) {
Chris@0:           $iend--;
Chris@0:         }
Chris@0:       }
Chris@0:       elseif ($length < 0) {
Chris@0:         // Count all the characters except continuation bytes from the end
Chris@0:         // until we have found abs($start) characters, then backtrace one byte.
Chris@0:         $length = abs($length);
Chris@0:         $iend = $strlen; $chars = 0;
Chris@0:         while ($iend > 0 && $chars < $length) {
Chris@0:           $iend--;
Chris@0:           $c = ord($text[$iend]);
Chris@0:           if ($c < 0x80 || $c >= 0xC0) {
Chris@0:             $chars++;
Chris@0:           }
Chris@0:         }
Chris@0:         // Backtrace one byte if we are not at the beginning of the string.
Chris@0:         if ($iend > 0) {
Chris@0:           $iend--;
Chris@0:         }
Chris@0:       }
Chris@0:       else {
Chris@0:         // $length == 0, return an empty string.
Chris@0:         return '';
Chris@0:       }
Chris@0: 
Chris@0:       return substr($text, $istart, max(0, $iend - $istart + 1));
Chris@0:     }
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Truncates a UTF-8-encoded string safely to a number of characters.
Chris@0:    *
Chris@0:    * @param string $string
Chris@0:    *   The string to truncate.
Chris@0:    * @param int $max_length
Chris@0:    *   An upper limit on the returned string length, including trailing ellipsis
Chris@0:    *   if $add_ellipsis is TRUE.
Chris@0:    * @param bool $wordsafe
Chris@0:    *   If TRUE, attempt to truncate on a word boundary. Word boundaries are
Chris@0:    *   spaces, punctuation, and Unicode characters used as word boundaries in
Chris@0:    *   non-Latin languages; see Unicode::PREG_CLASS_WORD_BOUNDARY for more
Chris@0:    *   information. If a word boundary cannot be found that would make the length
Chris@0:    *   of the returned string fall within length guidelines (see parameters
Chris@0:    *   $max_length and $min_wordsafe_length), word boundaries are ignored.
Chris@0:    * @param bool $add_ellipsis
Chris@0:    *   If TRUE, add '...' to the end of the truncated string (defaults to
Chris@0:    *   FALSE). The string length will still fall within $max_length.
Chris@0:    * @param int $min_wordsafe_length
Chris@0:    *   If $wordsafe is TRUE, the minimum acceptable length for truncation (before
Chris@0:    *   adding an ellipsis, if $add_ellipsis is TRUE). Has no effect if $wordsafe
Chris@0:    *   is FALSE. This can be used to prevent having a very short resulting string
Chris@0:    *   that will not be understandable. For instance, if you are truncating the
Chris@0:    *   string "See myverylongurlexample.com for more information" to a word-safe
Chris@0:    *   return length of 20, the only available word boundary within 20 characters
Chris@0:    *   is after the word "See", which wouldn't leave a very informative string. If
Chris@0:    *   you had set $min_wordsafe_length to 10, though, the function would realise
Chris@0:    *   that "See" alone is too short, and would then just truncate ignoring word
Chris@0:    *   boundaries, giving you "See myverylongurl..." (assuming you had set
Chris@0:    *   $add_ellipses to TRUE).
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The truncated string.
Chris@0:    */
Chris@0:   public static function truncate($string, $max_length, $wordsafe = FALSE, $add_ellipsis = FALSE, $min_wordsafe_length = 1) {
Chris@0:     $ellipsis = '';
Chris@0:     $max_length = max($max_length, 0);
Chris@0:     $min_wordsafe_length = max($min_wordsafe_length, 0);
Chris@0: 
Chris@0:     if (static::strlen($string) <= $max_length) {
Chris@0:       // No truncation needed, so don't add ellipsis, just return.
Chris@0:       return $string;
Chris@0:     }
Chris@0: 
Chris@0:     if ($add_ellipsis) {
Chris@0:       // Truncate ellipsis in case $max_length is small.
Chris@0:       $ellipsis = static::substr('…', 0, $max_length);
Chris@0:       $max_length -= static::strlen($ellipsis);
Chris@0:       $max_length = max($max_length, 0);
Chris@0:     }
Chris@0: 
Chris@0:     if ($max_length <= $min_wordsafe_length) {
Chris@0:       // Do not attempt word-safe if lengths are bad.
Chris@0:       $wordsafe = FALSE;
Chris@0:     }
Chris@0: 
Chris@0:     if ($wordsafe) {
Chris@0:       $matches = [];
Chris@0:       // Find the last word boundary, if there is one within $min_wordsafe_length
Chris@0:       // to $max_length characters. preg_match() is always greedy, so it will
Chris@0:       // find the longest string possible.
Chris@0:       $found = preg_match('/^(.{' . $min_wordsafe_length . ',' . $max_length . '})[' . Unicode::PREG_CLASS_WORD_BOUNDARY . ']/u', $string, $matches);
Chris@0:       if ($found) {
Chris@0:         $string = $matches[1];
Chris@0:       }
Chris@0:       else {
Chris@0:         $string = static::substr($string, 0, $max_length);
Chris@0:       }
Chris@0:     }
Chris@0:     else {
Chris@0:       $string = static::substr($string, 0, $max_length);
Chris@0:     }
Chris@0: 
Chris@0:     if ($add_ellipsis) {
Chris@0:       // If we're adding an ellipsis, remove any trailing periods.
Chris@0:       $string = rtrim($string, '.');
Chris@0: 
Chris@0:       $string .= $ellipsis;
Chris@0:     }
Chris@0: 
Chris@0:     return $string;
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Compares UTF-8-encoded strings in a binary safe case-insensitive manner.
Chris@0:    *
Chris@0:    * @param string $str1
Chris@0:    *   The first string.
Chris@0:    * @param string $str2
Chris@0:    *   The second string.
Chris@0:    *
Chris@0:    * @return int
Chris@0:    *   Returns < 0 if $str1 is less than $str2; > 0 if $str1 is greater than
Chris@0:    *   $str2, and 0 if they are equal.
Chris@0:    */
Chris@0:   public static function strcasecmp($str1, $str2) {
Chris@0:     return strcmp(static::strtoupper($str1), static::strtoupper($str2));
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Encodes MIME/HTTP headers that contain incorrectly encoded characters.
Chris@0:    *
Chris@0:    * For example, Unicode::mimeHeaderEncode('tést.txt') returns
Chris@0:    * "=?UTF-8?B?dMOpc3QudHh0?=".
Chris@0:    *
Chris@0:    * See http://www.rfc-editor.org/rfc/rfc2047.txt for more information.
Chris@0:    *
Chris@0:    * Notes:
Chris@0:    * - Only encode strings that contain non-ASCII characters.
Chris@0:    * - We progressively cut-off a chunk with self::truncateBytes(). This ensures
Chris@0:    *   each chunk starts and ends on a character boundary.
Chris@0:    * - Using \n as the chunk separator may cause problems on some systems and
Chris@0:    *   may have to be changed to \r\n or \r.
Chris@0:    *
Chris@0:    * @param string $string
Chris@0:    *   The header to encode.
Chris@12:    * @param bool $shorten
Chris@12:    *   If TRUE, only return the first chunk of a multi-chunk encoded string.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The mime-encoded header.
Chris@0:    */
Chris@12:   public static function mimeHeaderEncode($string, $shorten = FALSE) {
Chris@0:     if (preg_match('/[^\x20-\x7E]/', $string)) {
Chris@0:       // floor((75 - strlen("=?UTF-8?B??=")) * 0.75);
Chris@0:       $chunk_size = 47;
Chris@0:       $len = strlen($string);
Chris@0:       $output = '';
Chris@0:       while ($len > 0) {
Chris@0:         $chunk = static::truncateBytes($string, $chunk_size);
Chris@0:         $output .= ' =?UTF-8?B?' . base64_encode($chunk) . "?=\n";
Chris@12:         if ($shorten) {
Chris@12:           break;
Chris@12:         }
Chris@0:         $c = strlen($chunk);
Chris@0:         $string = substr($string, $c);
Chris@0:         $len -= $c;
Chris@0:       }
Chris@0:       return trim($output);
Chris@0:     }
Chris@0:     return $string;
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Decodes MIME/HTTP encoded header values.
Chris@0:    *
Chris@0:    * @param string $header
Chris@0:    *   The header to decode.
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The mime-decoded header.
Chris@0:    */
Chris@0:   public static function mimeHeaderDecode($header) {
Chris@0:     $callback = function ($matches) {
Chris@0:       $data = ($matches[2] == 'B') ? base64_decode($matches[3]) : str_replace('_', ' ', quoted_printable_decode($matches[3]));
Chris@0:       if (strtolower($matches[1]) != 'utf-8') {
Chris@0:         $data = static::convertToUtf8($data, $matches[1]);
Chris@0:       }
Chris@0:       return $data;
Chris@0:     };
Chris@0:     // First step: encoded chunks followed by other encoded chunks (need to collapse whitespace)
Chris@0:     $header = preg_replace_callback('/=\?([^?]+)\?(Q|B)\?([^?]+|\?(?!=))\?=\s+(?==\?)/', $callback, $header);
Chris@0:     // Second step: remaining chunks (do not collapse whitespace)
Chris@0:     return preg_replace_callback('/=\?([^?]+)\?(Q|B)\?([^?]+|\?(?!=))\?=/', $callback, $header);
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Flip U+C0-U+DE to U+E0-U+FD and back. Can be used as preg_replace callback.
Chris@0:    *
Chris@0:    * @param array $matches
Chris@0:    *   An array of matches by preg_replace_callback().
Chris@0:    *
Chris@0:    * @return string
Chris@0:    *   The flipped text.
Chris@0:    */
Chris@0:   public static function caseFlip($matches) {
Chris@0:     return $matches[0][0] . chr(ord($matches[0][1]) ^ 32);
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Checks whether a string is valid UTF-8.
Chris@0:    *
Chris@0:    * All functions designed to filter input should use drupal_validate_utf8
Chris@0:    * to ensure they operate on valid UTF-8 strings to prevent bypass of the
Chris@0:    * filter.
Chris@0:    *
Chris@0:    * When text containing an invalid UTF-8 lead byte (0xC0 - 0xFF) is presented
Chris@0:    * as UTF-8 to Internet Explorer 6, the program may misinterpret subsequent
Chris@0:    * bytes. When these subsequent bytes are HTML control characters such as
Chris@0:    * quotes or angle brackets, parts of the text that were deemed safe by filters
Chris@0:    * end up in locations that are potentially unsafe; An onerror attribute that
Chris@0:    * is outside of a tag, and thus deemed safe by a filter, can be interpreted
Chris@0:    * by the browser as if it were inside the tag.
Chris@0:    *
Chris@0:    * The function does not return FALSE for strings containing character codes
Chris@0:    * above U+10FFFF, even though these are prohibited by RFC 3629.
Chris@0:    *
Chris@0:    * @param string $text
Chris@0:    *   The text to check.
Chris@0:    *
Chris@0:    * @return bool
Chris@0:    *   TRUE if the text is valid UTF-8, FALSE if not.
Chris@0:    */
Chris@0:   public static function validateUtf8($text) {
Chris@0:     if (strlen($text) == 0) {
Chris@0:       return TRUE;
Chris@0:     }
Chris@0:     // With the PCRE_UTF8 modifier 'u', preg_match() fails silently on strings
Chris@0:     // containing invalid UTF-8 byte sequences. It does not reject character
Chris@0:     // codes above U+10FFFF (represented by 4 or more octets), though.
Chris@0:     return (preg_match('/^./us', $text) == 1);
Chris@0:   }
Chris@0: 
Chris@0:   /**
Chris@0:    * Finds the position of the first occurrence of a string in another string.
Chris@0:    *
Chris@0:    * @param string $haystack
Chris@0:    *   The string to search in.
Chris@0:    * @param string $needle
Chris@0:    *   The string to find in $haystack.
Chris@0:    * @param int $offset
Chris@0:    *   If specified, start the search at this number of characters from the
Chris@0:    *   beginning (default 0).
Chris@0:    *
Chris@0:    * @return int|false
Chris@0:    *   The position where $needle occurs in $haystack, always relative to the
Chris@0:    *   beginning (independent of $offset), or FALSE if not found. Note that
Chris@0:    *   a return value of 0 is not the same as FALSE.
Chris@0:    */
Chris@0:   public static function strpos($haystack, $needle, $offset = 0) {
Chris@0:     if (static::getStatus() == static::STATUS_MULTIBYTE) {
Chris@0:       return mb_strpos($haystack, $needle, $offset);
Chris@0:     }
Chris@0:     else {
Chris@0:       // Remove Unicode continuation characters, to be compatible with
Chris@0:       // Unicode::strlen() and Unicode::substr().
Chris@0:       $haystack = preg_replace("/[\x80-\xBF]/", '', $haystack);
Chris@0:       $needle = preg_replace("/[\x80-\xBF]/", '', $needle);
Chris@0:       return strpos($haystack, $needle, $offset);
Chris@0:     }
Chris@0:   }
Chris@0: 
Chris@0: }