Search moodle.org's
Developer Documentation

See Release Notes

  • Bug fixes for general core bugs in 4.3.x will end 7 October 2024 (12 months).
  • Bug fixes for security issues in 4.3.x will end 21 April 2025 (18 months).
  • PHP version: minimum PHP 8.0.0 Note: minimum PHP version has increased since Moodle 4.1. PHP 8.2.x is supported too.

Differences Between: [Versions 310 and 403] [Versions 311 and 403] [Versions 39 and 403] [Versions 400 and 403] [Versions 401 and 403] [Versions 402 and 403]

   1  <?php
   2  
   3  namespace PhpOffice\PhpSpreadsheet\Shared;
   4  
   5  class StringHelper
   6  {
   7      /**
   8       * Control characters array.
   9       *
  10       * @var string[]
  11       */
  12      private static $controlCharacters = [];
  13  
  14      /**
  15       * SYLK Characters array.
  16       *
  17       * @var array
  18       */
  19      private static $SYLKCharacters = [];
  20  
  21      /**
  22       * Decimal separator.
  23       *
  24       * @var ?string
  25       */
  26      private static $decimalSeparator;
  27  
  28      /**
  29       * Thousands separator.
  30       *
  31       * @var ?string
  32       */
  33      private static $thousandsSeparator;
  34  
  35      /**
  36       * Currency code.
  37       *
  38       * @var string
  39       */
  40      private static $currencyCode;
  41  
  42      /**
  43       * Is iconv extension avalable?
  44       *
  45       * @var ?bool
  46       */
  47      private static $isIconvEnabled;
  48  
  49      /**
  50       * iconv options.
  51       *
  52       * @var string
  53       */
  54      private static $iconvOptions = '//IGNORE//TRANSLIT';
  55  
  56      /**
  57       * Build control characters array.
  58       */
  59      private static function buildControlCharacters(): void
  60      {
  61          for ($i = 0; $i <= 31; ++$i) {
  62              if ($i != 9 && $i != 10 && $i != 13) {
  63                  $find = '_x' . sprintf('%04s', strtoupper(dechex($i))) . '_';
  64                  $replace = chr($i);
  65                  self::$controlCharacters[$find] = $replace;
  66              }
  67          }
  68      }
  69  
  70      /**
  71       * Build SYLK characters array.
  72       */
  73      private static function buildSYLKCharacters(): void
  74      {
  75          self::$SYLKCharacters = [
  76              "\x1B 0" => chr(0),
  77              "\x1B 1" => chr(1),
  78              "\x1B 2" => chr(2),
  79              "\x1B 3" => chr(3),
  80              "\x1B 4" => chr(4),
  81              "\x1B 5" => chr(5),
  82              "\x1B 6" => chr(6),
  83              "\x1B 7" => chr(7),
  84              "\x1B 8" => chr(8),
  85              "\x1B 9" => chr(9),
  86              "\x1B :" => chr(10),
  87              "\x1B ;" => chr(11),
  88              "\x1B <" => chr(12),
  89              "\x1B =" => chr(13),
  90              "\x1B >" => chr(14),
  91              "\x1B ?" => chr(15),
  92              "\x1B!0" => chr(16),
  93              "\x1B!1" => chr(17),
  94              "\x1B!2" => chr(18),
  95              "\x1B!3" => chr(19),
  96              "\x1B!4" => chr(20),
  97              "\x1B!5" => chr(21),
  98              "\x1B!6" => chr(22),
  99              "\x1B!7" => chr(23),
 100              "\x1B!8" => chr(24),
 101              "\x1B!9" => chr(25),
 102              "\x1B!:" => chr(26),
 103              "\x1B!;" => chr(27),
 104              "\x1B!<" => chr(28),
 105              "\x1B!=" => chr(29),
 106              "\x1B!>" => chr(30),
 107              "\x1B!?" => chr(31),
 108              "\x1B'?" => chr(127),
 109              "\x1B(0" => '€', // 128 in CP1252
 110              "\x1B(2" => '‚', // 130 in CP1252
 111              "\x1B(3" => 'ƒ', // 131 in CP1252
 112              "\x1B(4" => '„', // 132 in CP1252
 113              "\x1B(5" => '…', // 133 in CP1252
 114              "\x1B(6" => '†', // 134 in CP1252
 115              "\x1B(7" => '‡', // 135 in CP1252
 116              "\x1B(8" => 'ˆ', // 136 in CP1252
 117              "\x1B(9" => '‰', // 137 in CP1252
 118              "\x1B(:" => 'Š', // 138 in CP1252
 119              "\x1B(;" => '‹', // 139 in CP1252
 120              "\x1BNj" => 'Œ', // 140 in CP1252
 121              "\x1B(>" => 'Ž', // 142 in CP1252
 122              "\x1B)1" => '‘', // 145 in CP1252
 123              "\x1B)2" => '’', // 146 in CP1252
 124              "\x1B)3" => '“', // 147 in CP1252
 125              "\x1B)4" => '”', // 148 in CP1252
 126              "\x1B)5" => '•', // 149 in CP1252
 127              "\x1B)6" => '–', // 150 in CP1252
 128              "\x1B)7" => '—', // 151 in CP1252
 129              "\x1B)8" => '˜', // 152 in CP1252
 130              "\x1B)9" => '™', // 153 in CP1252
 131              "\x1B):" => 'š', // 154 in CP1252
 132              "\x1B);" => '›', // 155 in CP1252
 133              "\x1BNz" => 'œ', // 156 in CP1252
 134              "\x1B)>" => 'ž', // 158 in CP1252
 135              "\x1B)?" => 'Ÿ', // 159 in CP1252
 136              "\x1B*0" => ' ', // 160 in CP1252
 137              "\x1BN!" => '¡', // 161 in CP1252
 138              "\x1BN\"" => '¢', // 162 in CP1252
 139              "\x1BN#" => '£', // 163 in CP1252
 140              "\x1BN(" => '¤', // 164 in CP1252
 141              "\x1BN%" => '¥', // 165 in CP1252
 142              "\x1B*6" => '¦', // 166 in CP1252
 143              "\x1BN'" => '§', // 167 in CP1252
 144              "\x1BNH " => '¨', // 168 in CP1252
 145              "\x1BNS" => '©', // 169 in CP1252
 146              "\x1BNc" => 'ª', // 170 in CP1252
 147              "\x1BN+" => '«', // 171 in CP1252
 148              "\x1B*<" => '¬', // 172 in CP1252
 149              "\x1B*=" => '­', // 173 in CP1252
 150              "\x1BNR" => '®', // 174 in CP1252
 151              "\x1B*?" => '¯', // 175 in CP1252
 152              "\x1BN0" => '°', // 176 in CP1252
 153              "\x1BN1" => '±', // 177 in CP1252
 154              "\x1BN2" => '²', // 178 in CP1252
 155              "\x1BN3" => '³', // 179 in CP1252
 156              "\x1BNB " => '´', // 180 in CP1252
 157              "\x1BN5" => 'µ', // 181 in CP1252
 158              "\x1BN6" => '¶', // 182 in CP1252
 159              "\x1BN7" => '·', // 183 in CP1252
 160              "\x1B+8" => '¸', // 184 in CP1252
 161              "\x1BNQ" => '¹', // 185 in CP1252
 162              "\x1BNk" => 'º', // 186 in CP1252
 163              "\x1BN;" => '»', // 187 in CP1252
 164              "\x1BN<" => '¼', // 188 in CP1252
 165              "\x1BN=" => '½', // 189 in CP1252
 166              "\x1BN>" => '¾', // 190 in CP1252
 167              "\x1BN?" => '¿', // 191 in CP1252
 168              "\x1BNAA" => 'À', // 192 in CP1252
 169              "\x1BNBA" => 'Á', // 193 in CP1252
 170              "\x1BNCA" => 'Â', // 194 in CP1252
 171              "\x1BNDA" => 'Ã', // 195 in CP1252
 172              "\x1BNHA" => 'Ä', // 196 in CP1252
 173              "\x1BNJA" => 'Å', // 197 in CP1252
 174              "\x1BNa" => 'Æ', // 198 in CP1252
 175              "\x1BNKC" => 'Ç', // 199 in CP1252
 176              "\x1BNAE" => 'È', // 200 in CP1252
 177              "\x1BNBE" => 'É', // 201 in CP1252
 178              "\x1BNCE" => 'Ê', // 202 in CP1252
 179              "\x1BNHE" => 'Ë', // 203 in CP1252
 180              "\x1BNAI" => 'Ì', // 204 in CP1252
 181              "\x1BNBI" => 'Í', // 205 in CP1252
 182              "\x1BNCI" => 'Î', // 206 in CP1252
 183              "\x1BNHI" => 'Ï', // 207 in CP1252
 184              "\x1BNb" => 'Ð', // 208 in CP1252
 185              "\x1BNDN" => 'Ñ', // 209 in CP1252
 186              "\x1BNAO" => 'Ò', // 210 in CP1252
 187              "\x1BNBO" => 'Ó', // 211 in CP1252
 188              "\x1BNCO" => 'Ô', // 212 in CP1252
 189              "\x1BNDO" => 'Õ', // 213 in CP1252
 190              "\x1BNHO" => 'Ö', // 214 in CP1252
 191              "\x1B-7" => '×', // 215 in CP1252
 192              "\x1BNi" => 'Ø', // 216 in CP1252
 193              "\x1BNAU" => 'Ù', // 217 in CP1252
 194              "\x1BNBU" => 'Ú', // 218 in CP1252
 195              "\x1BNCU" => 'Û', // 219 in CP1252
 196              "\x1BNHU" => 'Ü', // 220 in CP1252
 197              "\x1B-=" => 'Ý', // 221 in CP1252
 198              "\x1BNl" => 'Þ', // 222 in CP1252
 199              "\x1BN{" => 'ß', // 223 in CP1252
 200              "\x1BNAa" => 'à', // 224 in CP1252
 201              "\x1BNBa" => 'á', // 225 in CP1252
 202              "\x1BNCa" => 'â', // 226 in CP1252
 203              "\x1BNDa" => 'ã', // 227 in CP1252
 204              "\x1BNHa" => 'ä', // 228 in CP1252
 205              "\x1BNJa" => 'å', // 229 in CP1252
 206              "\x1BNq" => 'æ', // 230 in CP1252
 207              "\x1BNKc" => 'ç', // 231 in CP1252
 208              "\x1BNAe" => 'è', // 232 in CP1252
 209              "\x1BNBe" => 'é', // 233 in CP1252
 210              "\x1BNCe" => 'ê', // 234 in CP1252
 211              "\x1BNHe" => 'ë', // 235 in CP1252
 212              "\x1BNAi" => 'ì', // 236 in CP1252
 213              "\x1BNBi" => 'í', // 237 in CP1252
 214              "\x1BNCi" => 'î', // 238 in CP1252
 215              "\x1BNHi" => 'ï', // 239 in CP1252
 216              "\x1BNs" => 'ð', // 240 in CP1252
 217              "\x1BNDn" => 'ñ', // 241 in CP1252
 218              "\x1BNAo" => 'ò', // 242 in CP1252
 219              "\x1BNBo" => 'ó', // 243 in CP1252
 220              "\x1BNCo" => 'ô', // 244 in CP1252
 221              "\x1BNDo" => 'õ', // 245 in CP1252
 222              "\x1BNHo" => 'ö', // 246 in CP1252
 223              "\x1B/7" => '÷', // 247 in CP1252
 224              "\x1BNy" => 'ø', // 248 in CP1252
 225              "\x1BNAu" => 'ù', // 249 in CP1252
 226              "\x1BNBu" => 'ú', // 250 in CP1252
 227              "\x1BNCu" => 'û', // 251 in CP1252
 228              "\x1BNHu" => 'ü', // 252 in CP1252
 229              "\x1B/=" => 'ý', // 253 in CP1252
 230              "\x1BN|" => 'þ', // 254 in CP1252
 231              "\x1BNHy" => 'ÿ', // 255 in CP1252
 232          ];
 233      }
 234  
 235      /**
 236       * Get whether iconv extension is available.
 237       *
 238       * @return bool
 239       */
 240      public static function getIsIconvEnabled()
 241      {
 242          if (isset(self::$isIconvEnabled)) {
 243              return self::$isIconvEnabled;
 244          }
 245  
 246          // Assume no problems with iconv
 247          self::$isIconvEnabled = true;
 248  
 249          // Fail if iconv doesn't exist
 250          if (!function_exists('iconv')) {
 251              self::$isIconvEnabled = false;
 252          } elseif (!@iconv('UTF-8', 'UTF-16LE', 'x')) {
 253              // Sometimes iconv is not working, and e.g. iconv('UTF-8', 'UTF-16LE', 'x') just returns false,
 254              self::$isIconvEnabled = false;
 255          } elseif (defined('PHP_OS') && @stristr(PHP_OS, 'AIX') && defined('ICONV_IMPL') && (@strcasecmp(ICONV_IMPL, 'unknown') == 0) && defined('ICONV_VERSION') && (@strcasecmp(ICONV_VERSION, 'unknown') == 0)) {
 256              // CUSTOM: IBM AIX iconv() does not work
 257              self::$isIconvEnabled = false;
 258          }
 259  
 260          // Deactivate iconv default options if they fail (as seen on IMB i)
 261          if (self::$isIconvEnabled && !@iconv('UTF-8', 'UTF-16LE' . self::$iconvOptions, 'x')) {
 262              self::$iconvOptions = '';
 263          }
 264  
 265          return self::$isIconvEnabled;
 266      }
 267  
 268      private static function buildCharacterSets(): void
 269      {
 270          if (empty(self::$controlCharacters)) {
 271              self::buildControlCharacters();
 272          }
 273  
 274          if (empty(self::$SYLKCharacters)) {
 275              self::buildSYLKCharacters();
 276          }
 277      }
 278  
 279      /**
 280       * Convert from OpenXML escaped control character to PHP control character.
 281       *
 282       * Excel 2007 team:
 283       * ----------------
 284       * That's correct, control characters are stored directly in the shared-strings table.
 285       * We do encode characters that cannot be represented in XML using the following escape sequence:
 286       * _xHHHH_ where H represents a hexadecimal character in the character's value...
 287       * So you could end up with something like _x0008_ in a string (either in a cell value (<v>)
 288       * element or in the shared string <t> element.
 289       *
 290       * @param string $textValue Value to unescape
 291       *
 292       * @return string
 293       */
 294      public static function controlCharacterOOXML2PHP($textValue)
 295      {
 296          self::buildCharacterSets();
 297  
 298          return str_replace(array_keys(self::$controlCharacters), array_values(self::$controlCharacters), $textValue);
 299      }
 300  
 301      /**
 302       * Convert from PHP control character to OpenXML escaped control character.
 303       *
 304       * Excel 2007 team:
 305       * ----------------
 306       * That's correct, control characters are stored directly in the shared-strings table.
 307       * We do encode characters that cannot be represented in XML using the following escape sequence:
 308       * _xHHHH_ where H represents a hexadecimal character in the character's value...
 309       * So you could end up with something like _x0008_ in a string (either in a cell value (<v>)
 310       * element or in the shared string <t> element.
 311       *
 312       * @param string $textValue Value to escape
 313       *
 314       * @return string
 315       */
 316      public static function controlCharacterPHP2OOXML($textValue)
 317      {
 318          self::buildCharacterSets();
 319  
 320          return str_replace(array_values(self::$controlCharacters), array_keys(self::$controlCharacters), $textValue);
 321      }
 322  
 323      /**
 324       * Try to sanitize UTF8, replacing invalid sequences with Unicode substitution characters.
 325       */
 326      public static function sanitizeUTF8(string $textValue): string
 327      {
 328          $textValue = str_replace(["\xef\xbf\xbe", "\xef\xbf\xbf"], "\xef\xbf\xbd", $textValue);
 329          $subst = mb_substitute_character(); // default is question mark
 330          mb_substitute_character(65533); // Unicode substitution character
 331          // Phpstan does not think this can return false.
 332          $returnValue = mb_convert_encoding($textValue, 'UTF-8', 'UTF-8');
 333          mb_substitute_character(/** @scrutinizer ignore-type */ $subst);
 334  
 335          return self::returnString($returnValue);
 336      }
 337  
 338      /**
 339       * Strictly to satisfy Scrutinizer.
 340       *
 341       * @param mixed $value
 342       */
 343      private static function returnString($value): string
 344      {
 345          return is_string($value) ? $value : '';
 346      }
 347  
 348      /**
 349       * Check if a string contains UTF8 data.
 350       */
 351      public static function isUTF8(string $textValue): bool
 352      {
 353          return $textValue === self::sanitizeUTF8($textValue);
 354      }
 355  
 356      /**
 357       * Formats a numeric value as a string for output in various output writers forcing
 358       * point as decimal separator in case locale is other than English.
 359       *
 360       * @param float|int|string $numericValue
 361       */
 362      public static function formatNumber($numericValue): string
 363      {
 364          if (is_float($numericValue)) {
 365              return str_replace(',', '.', (string) $numericValue);
 366          }
 367  
 368          return (string) $numericValue;
 369      }
 370  
 371      /**
 372       * Converts a UTF-8 string into BIFF8 Unicode string data (8-bit string length)
 373       * Writes the string using uncompressed notation, no rich text, no Asian phonetics
 374       * If mbstring extension is not available, ASCII is assumed, and compressed notation is used
 375       * although this will give wrong results for non-ASCII strings
 376       * see OpenOffice.org's Documentation of the Microsoft Excel File Format, sect. 2.5.3.
 377       *
 378       * @param string $textValue UTF-8 encoded string
 379       * @param mixed[] $arrcRuns Details of rich text runs in $value
 380       */
 381      public static function UTF8toBIFF8UnicodeShort(string $textValue, array $arrcRuns = []): string
 382      {
 383          // character count
 384          $ln = self::countCharacters($textValue, 'UTF-8');
 385          // option flags
 386          if (empty($arrcRuns)) {
 387              $data = pack('CC', $ln, 0x0001);
 388              // characters
 389              $data .= self::convertEncoding($textValue, 'UTF-16LE', 'UTF-8');
 390          } else {
 391              $data = pack('vC', $ln, 0x09);
 392              $data .= pack('v', count($arrcRuns));
 393              // characters
 394              $data .= self::convertEncoding($textValue, 'UTF-16LE', 'UTF-8');
 395              foreach ($arrcRuns as $cRun) {
 396                  $data .= pack('v', $cRun['strlen']);
 397                  $data .= pack('v', $cRun['fontidx']);
 398              }
 399          }
 400  
 401          return $data;
 402      }
 403  
 404      /**
 405       * Converts a UTF-8 string into BIFF8 Unicode string data (16-bit string length)
 406       * Writes the string using uncompressed notation, no rich text, no Asian phonetics
 407       * If mbstring extension is not available, ASCII is assumed, and compressed notation is used
 408       * although this will give wrong results for non-ASCII strings
 409       * see OpenOffice.org's Documentation of the Microsoft Excel File Format, sect. 2.5.3.
 410       *
 411       * @param string $textValue UTF-8 encoded string
 412       */
 413      public static function UTF8toBIFF8UnicodeLong(string $textValue): string
 414      {
 415          // character count
 416          $ln = self::countCharacters($textValue, 'UTF-8');
 417  
 418          // characters
 419          $chars = self::convertEncoding($textValue, 'UTF-16LE', 'UTF-8');
 420  
 421          return pack('vC', $ln, 0x0001) . $chars;
 422      }
 423  
 424      /**
 425       * Convert string from one encoding to another.
 426       *
 427       * @param string $to Encoding to convert to, e.g. 'UTF-8'
 428       * @param string $from Encoding to convert from, e.g. 'UTF-16LE'
 429       */
 430      public static function convertEncoding(string $textValue, string $to, string $from): string
 431      {
 432          if (self::getIsIconvEnabled()) {
 433              $result = iconv($from, $to . self::$iconvOptions, $textValue);
 434              if (false !== $result) {
 435                  return $result;
 436              }
 437          }
 438  
 439          return self::returnString(mb_convert_encoding($textValue, $to, $from));
 440      }
 441  
 442      /**
 443       * Get character count.
 444       *
 445       * @param string $encoding Encoding
 446       *
 447       * @return int Character count
 448       */
 449      public static function countCharacters(string $textValue, string $encoding = 'UTF-8'): int
 450      {
 451          return mb_strlen($textValue, $encoding);
 452      }
 453  
 454      /**
 455       * Get character count using mb_strwidth rather than mb_strlen.
 456       *
 457       * @param string $encoding Encoding
 458       *
 459       * @return int Character count
 460       */
 461      public static function countCharactersDbcs(string $textValue, string $encoding = 'UTF-8'): int
 462      {
 463          return mb_strwidth($textValue, $encoding);
 464      }
 465  
 466      /**
 467       * Get a substring of a UTF-8 encoded string.
 468       *
 469       * @param string $textValue UTF-8 encoded string
 470       * @param int $offset Start offset
 471       * @param ?int $length Maximum number of characters in substring
 472       */
 473      public static function substring(string $textValue, int $offset, ?int $length = 0): string
 474      {
 475          return mb_substr($textValue, $offset, $length, 'UTF-8');
 476      }
 477  
 478      /**
 479       * Convert a UTF-8 encoded string to upper case.
 480       *
 481       * @param string $textValue UTF-8 encoded string
 482       */
 483      public static function strToUpper(string $textValue): string
 484      {
 485          return mb_convert_case($textValue, MB_CASE_UPPER, 'UTF-8');
 486      }
 487  
 488      /**
 489       * Convert a UTF-8 encoded string to lower case.
 490       *
 491       * @param string $textValue UTF-8 encoded string
 492       */
 493      public static function strToLower(string $textValue): string
 494      {
 495          return mb_convert_case($textValue, MB_CASE_LOWER, 'UTF-8');
 496      }
 497  
 498      /**
 499       * Convert a UTF-8 encoded string to title/proper case
 500       * (uppercase every first character in each word, lower case all other characters).
 501       *
 502       * @param string $textValue UTF-8 encoded string
 503       */
 504      public static function strToTitle(string $textValue): string
 505      {
 506          return mb_convert_case($textValue, MB_CASE_TITLE, 'UTF-8');
 507      }
 508  
 509      public static function mbIsUpper(string $character): bool
 510      {
 511          return mb_strtolower($character, 'UTF-8') !== $character;
 512      }
 513  
 514      /**
 515       * Splits a UTF-8 string into an array of individual characters.
 516       */
 517      public static function mbStrSplit(string $string): array
 518      {
 519          // Split at all position not after the start: ^
 520          // and not before the end: $
 521          $split = preg_split('/(?<!^)(?!$)/u', $string);
 522  
 523          return ($split === false) ? [] : $split;
 524      }
 525  
 526      /**
 527       * Reverse the case of a string, so that all uppercase characters become lowercase
 528       * and all lowercase characters become uppercase.
 529       *
 530       * @param string $textValue UTF-8 encoded string
 531       */
 532      public static function strCaseReverse(string $textValue): string
 533      {
 534          $characters = self::mbStrSplit($textValue);
 535          foreach ($characters as &$character) {
 536              if (self::mbIsUpper($character)) {
 537                  $character = mb_strtolower($character, 'UTF-8');
 538              } else {
 539                  $character = mb_strtoupper($character, 'UTF-8');
 540              }
 541          }
 542  
 543          return implode('', $characters);
 544      }
 545  
 546      /**
 547       * Get the decimal separator. If it has not yet been set explicitly, try to obtain number
 548       * formatting information from locale.
 549       */
 550      public static function getDecimalSeparator(): string
 551      {
 552          if (!isset(self::$decimalSeparator)) {
 553              $localeconv = localeconv();
 554              self::$decimalSeparator = ($localeconv['decimal_point'] != '')
 555                  ? $localeconv['decimal_point'] : $localeconv['mon_decimal_point'];
 556  
 557              if (self::$decimalSeparator == '') {
 558                  // Default to .
 559                  self::$decimalSeparator = '.';
 560              }
 561          }
 562  
 563          return self::$decimalSeparator;
 564      }
 565  
 566      /**
 567       * Set the decimal separator. Only used by NumberFormat::toFormattedString()
 568       * to format output by \PhpOffice\PhpSpreadsheet\Writer\Html and \PhpOffice\PhpSpreadsheet\Writer\Pdf.
 569       *
 570       * @param string $separator Character for decimal separator
 571       */
 572      public static function setDecimalSeparator(string $separator): void
 573      {
 574          self::$decimalSeparator = $separator;
 575      }
 576  
 577      /**
 578       * Get the thousands separator. If it has not yet been set explicitly, try to obtain number
 579       * formatting information from locale.
 580       */
 581      public static function getThousandsSeparator(): string
 582      {
 583          if (!isset(self::$thousandsSeparator)) {
 584              $localeconv = localeconv();
 585              self::$thousandsSeparator = ($localeconv['thousands_sep'] != '')
 586                  ? $localeconv['thousands_sep'] : $localeconv['mon_thousands_sep'];
 587  
 588              if (self::$thousandsSeparator == '') {
 589                  // Default to .
 590                  self::$thousandsSeparator = ',';
 591              }
 592          }
 593  
 594          return self::$thousandsSeparator;
 595      }
 596  
 597      /**
 598       * Set the thousands separator. Only used by NumberFormat::toFormattedString()
 599       * to format output by \PhpOffice\PhpSpreadsheet\Writer\Html and \PhpOffice\PhpSpreadsheet\Writer\Pdf.
 600       *
 601       * @param string $separator Character for thousands separator
 602       */
 603      public static function setThousandsSeparator(string $separator): void
 604      {
 605          self::$thousandsSeparator = $separator;
 606      }
 607  
 608      /**
 609       *    Get the currency code. If it has not yet been set explicitly, try to obtain the
 610       *        symbol information from locale.
 611       */
 612      public static function getCurrencyCode(): string
 613      {
 614          if (!empty(self::$currencyCode)) {
 615              return self::$currencyCode;
 616          }
 617          self::$currencyCode = '$';
 618          $localeconv = localeconv();
 619          if (!empty($localeconv['currency_symbol'])) {
 620              self::$currencyCode = $localeconv['currency_symbol'];
 621  
 622              return self::$currencyCode;
 623          }
 624          if (!empty($localeconv['int_curr_symbol'])) {
 625              self::$currencyCode = $localeconv['int_curr_symbol'];
 626  
 627              return self::$currencyCode;
 628          }
 629  
 630          return self::$currencyCode;
 631      }
 632  
 633      /**
 634       * Set the currency code. Only used by NumberFormat::toFormattedString()
 635       *        to format output by \PhpOffice\PhpSpreadsheet\Writer\Html and \PhpOffice\PhpSpreadsheet\Writer\Pdf.
 636       *
 637       * @param string $currencyCode Character for currency code
 638       */
 639      public static function setCurrencyCode(string $currencyCode): void
 640      {
 641          self::$currencyCode = $currencyCode;
 642      }
 643  
 644      /**
 645       * Convert SYLK encoded string to UTF-8.
 646       *
 647       * @param string $textValue SYLK encoded string
 648       *
 649       * @return string UTF-8 encoded string
 650       */
 651      public static function SYLKtoUTF8(string $textValue): string
 652      {
 653          self::buildCharacterSets();
 654  
 655          // If there is no escape character in the string there is nothing to do
 656          if (strpos($textValue, '') === false) {
 657              return $textValue;
 658          }
 659  
 660          foreach (self::$SYLKCharacters as $k => $v) {
 661              $textValue = str_replace($k, $v, $textValue);
 662          }
 663  
 664          return $textValue;
 665      }
 666  
 667      /**
 668       * Retrieve any leading numeric part of a string, or return the full string if no leading numeric
 669       * (handles basic integer or float, but not exponent or non decimal).
 670       *
 671       * @param string $textValue
 672       *
 673       * @return mixed string or only the leading numeric part of the string
 674       */
 675      public static function testStringAsNumeric($textValue)
 676      {
 677          if (is_numeric($textValue)) {
 678              return $textValue;
 679          }
 680          $v = (float) $textValue;
 681  
 682          return (is_numeric(substr($textValue, 0, strlen((string) $v)))) ? $v : $textValue;
 683      }
 684  }