| 1 |
<?php |
| 2 |
|
| 3 |
/* |
| 4 |
* This file is part of the Symfony package. |
| 5 |
* |
| 6 |
* (c) Fabien Potencier <fabien@symfony.com> |
| 7 |
* |
| 8 |
* For the full copyright and license information, please view the LICENSE |
| 9 |
* file that was distributed with this source code. |
| 10 |
*/ |
| 11 |
|
| 12 |
namespace Symfony\Component\String; |
| 13 |
|
| 14 |
use Symfony\Component\String\Exception\ExceptionInterface; |
| 15 |
use Symfony\Component\String\Exception\InvalidArgumentException; |
| 16 |
|
| 17 |
/** |
| 18 |
* Represents a string of Unicode grapheme clusters encoded as UTF-8. |
| 19 |
* |
| 20 |
* A letter followed by combining characters (accents typically) form what Unicode defines |
| 21 |
* as a grapheme cluster: a character as humans mean it in written texts. This class knows |
| 22 |
* about the concept and won't split a letter apart from its combining accents. It also |
| 23 |
* ensures all string comparisons happen on their canonically-composed representation, |
| 24 |
* ignoring e.g. the order in which accents are listed when a letter has many of them. |
| 25 |
* |
| 26 |
* @see https://unicode.org/reports/tr15/ |
| 27 |
* |
| 28 |
* @author Nicolas Grekas <p@tchwork.com> |
| 29 |
* @author Hugo Hamon <hugohamon@neuf.fr> |
| 30 |
* |
| 31 |
* @throws ExceptionInterface |
| 32 |
*/ |
| 33 |
class UnicodeString extends AbstractUnicodeString |
| 34 |
{ |
| 35 |
public function __construct(string $string = '') |
| 36 |
{ |
| 37 |
$this->string = normalizer_is_normalized($string) ? $string : normalizer_normalize($string); |
| 38 |
|
| 39 |
if (false === $this->string) { |
| 40 |
throw new InvalidArgumentException('Invalid UTF-8 string.'); |
| 41 |
} |
| 42 |
} |
| 43 |
|
| 44 |
public function append(string ...$suffix): static |
| 45 |
{ |
| 46 |
$str = clone $this; |
| 47 |
$str->string = $this->string.(1 >= \count($suffix) ? ($suffix[0] ?? '') : implode('', $suffix)); |
| 48 |
normalizer_is_normalized($str->string) ?: $str->string = normalizer_normalize($str->string); |
| 49 |
|
| 50 |
if (false === $str->string) { |
| 51 |
throw new InvalidArgumentException('Invalid UTF-8 string.'); |
| 52 |
} |
| 53 |
|
| 54 |
return $str; |
| 55 |
} |
| 56 |
|
| 57 |
public function chunk(int $length = 1): array |
| 58 |
{ |
| 59 |
if (1 > $length) { |
| 60 |
throw new InvalidArgumentException('The chunk length must be greater than zero.'); |
| 61 |
} |
| 62 |
|
| 63 |
if ('' === $this->string) { |
| 64 |
return []; |
| 65 |
} |
| 66 |
|
| 67 |
$rx = '/('; |
| 68 |
while (65535 < $length) { |
| 69 |
$rx .= '\X{65535}'; |
| 70 |
$length -= 65535; |
| 71 |
} |
| 72 |
$rx .= '\X{'.$length.'})/u'; |
| 73 |
|
| 74 |
$str = clone $this; |
| 75 |
$chunks = []; |
| 76 |
|
| 77 |
foreach (preg_split($rx, $this->string, -1, \PREG_SPLIT_DELIM_CAPTURE | \PREG_SPLIT_NO_EMPTY) as $chunk) { |
| 78 |
$str->string = $chunk; |
| 79 |
$chunks[] = clone $str; |
| 80 |
} |
| 81 |
|
| 82 |
return $chunks; |
| 83 |
} |
| 84 |
|
| 85 |
public function endsWith(string|iterable|AbstractString $suffix): bool |
| 86 |
{ |
| 87 |
if ($suffix instanceof AbstractString) { |
| 88 |
$suffix = $suffix->string; |
| 89 |
} elseif (!\is_string($suffix)) { |
| 90 |
return parent::endsWith($suffix); |
| 91 |
} |
| 92 |
|
| 93 |
$form = null === $this->ignoreCase ? \Normalizer::NFD : \Normalizer::NFC; |
| 94 |
normalizer_is_normalized($suffix, $form) ?: $suffix = normalizer_normalize($suffix, $form); |
| 95 |
|
| 96 |
if ('' === $suffix || false === $suffix) { |
| 97 |
return false; |
| 98 |
} |
| 99 |
|
| 100 |
if ($this->ignoreCase) { |
| 101 |
return 0 === mb_stripos(grapheme_extract($this->string, \strlen($suffix), \GRAPHEME_EXTR_MAXBYTES, \strlen($this->string) - \strlen($suffix)), $suffix, 0, 'UTF-8'); |
| 102 |
} |
| 103 |
|
| 104 |
return $suffix === grapheme_extract($this->string, \strlen($suffix), \GRAPHEME_EXTR_MAXBYTES, \strlen($this->string) - \strlen($suffix)); |
| 105 |
} |
| 106 |
|
| 107 |
public function equalsTo(string|iterable|AbstractString $string): bool |
| 108 |
{ |
| 109 |
if ($string instanceof AbstractString) { |
| 110 |
$string = $string->string; |
| 111 |
} elseif (!\is_string($string)) { |
| 112 |
return parent::equalsTo($string); |
| 113 |
} |
| 114 |
|
| 115 |
$form = null === $this->ignoreCase ? \Normalizer::NFD : \Normalizer::NFC; |
| 116 |
normalizer_is_normalized($string, $form) ?: $string = normalizer_normalize($string, $form); |
| 117 |
|
| 118 |
if ('' !== $string && false !== $string && $this->ignoreCase) { |
| 119 |
return \strlen($string) === \strlen($this->string) && 0 === mb_stripos($this->string, $string, 0, 'UTF-8'); |
| 120 |
} |
| 121 |
|
| 122 |
return $string === $this->string; |
| 123 |
} |
| 124 |
|
| 125 |
public function indexOf(string|iterable|AbstractString $needle, int $offset = 0): ?int |
| 126 |
{ |
| 127 |
if ($needle instanceof AbstractString) { |
| 128 |
$needle = $needle->string; |
| 129 |
} elseif (!\is_string($needle)) { |
| 130 |
return parent::indexOf($needle, $offset); |
| 131 |
} |
| 132 |
|
| 133 |
$form = null === $this->ignoreCase ? \Normalizer::NFD : \Normalizer::NFC; |
| 134 |
normalizer_is_normalized($needle, $form) ?: $needle = normalizer_normalize($needle, $form); |
| 135 |
|
| 136 |
if ('' === $needle || false === $needle) { |
| 137 |
return null; |
| 138 |
} |
| 139 |
|
| 140 |
try { |
| 141 |
$i = $this->ignoreCase ? grapheme_stripos($this->string, $needle, $offset) : grapheme_strpos($this->string, $needle, $offset); |
| 142 |
} catch (\ValueError) { |
| 143 |
return null; |
| 144 |
} |
| 145 |
|
| 146 |
return false === $i ? null : $i; |
| 147 |
} |
| 148 |
|
| 149 |
public function indexOfLast(string|iterable|AbstractString $needle, int $offset = 0): ?int |
| 150 |
{ |
| 151 |
if ($needle instanceof AbstractString) { |
| 152 |
$needle = $needle->string; |
| 153 |
} elseif (!\is_string($needle)) { |
| 154 |
return parent::indexOfLast($needle, $offset); |
| 155 |
} |
| 156 |
|
| 157 |
$form = null === $this->ignoreCase ? \Normalizer::NFD : \Normalizer::NFC; |
| 158 |
normalizer_is_normalized($needle, $form) ?: $needle = normalizer_normalize($needle, $form); |
| 159 |
|
| 160 |
if ('' === $needle || false === $needle) { |
| 161 |
return null; |
| 162 |
} |
| 163 |
|
| 164 |
$string = $this->string; |
| 165 |
|
| 166 |
if (0 > $offset) { |
| 167 |
// workaround https://bugs.php.net/74264 |
| 168 |
if (0 > $offset += grapheme_strlen($needle)) { |
| 169 |
$string = grapheme_substr($string, 0, $offset); |
| 170 |
} |
| 171 |
$offset = 0; |
| 172 |
} |
| 173 |
|
| 174 |
$i = $this->ignoreCase ? grapheme_strripos($string, $needle, $offset) : grapheme_strrpos($string, $needle, $offset); |
| 175 |
|
| 176 |
return false === $i ? null : $i; |
| 177 |
} |
| 178 |
|
| 179 |
public function join(array $strings, string $lastGlue = null): static |
| 180 |
{ |
| 181 |
$str = parent::join($strings, $lastGlue); |
| 182 |
normalizer_is_normalized($str->string) ?: $str->string = normalizer_normalize($str->string); |
| 183 |
|
| 184 |
return $str; |
| 185 |
} |
| 186 |
|
| 187 |
public function length(): int |
| 188 |
{ |
| 189 |
return grapheme_strlen($this->string); |
| 190 |
} |
| 191 |
|
| 192 |
public function normalize(int $form = self::NFC): static |
| 193 |
{ |
| 194 |
$str = clone $this; |
| 195 |
|
| 196 |
if (\in_array($form, [self::NFC, self::NFKC], true)) { |
| 197 |
normalizer_is_normalized($str->string, $form) ?: $str->string = normalizer_normalize($str->string, $form); |
| 198 |
} elseif (!\in_array($form, [self::NFD, self::NFKD], true)) { |
| 199 |
throw new InvalidArgumentException('Unsupported normalization form.'); |
| 200 |
} elseif (!normalizer_is_normalized($str->string, $form)) { |
| 201 |
$str->string = normalizer_normalize($str->string, $form); |
| 202 |
$str->ignoreCase = null; |
| 203 |
} |
| 204 |
|
| 205 |
return $str; |
| 206 |
} |
| 207 |
|
| 208 |
public function prepend(string ...$prefix): static |
| 209 |
{ |
| 210 |
$str = clone $this; |
| 211 |
$str->string = (1 >= \count($prefix) ? ($prefix[0] ?? '') : implode('', $prefix)).$this->string; |
| 212 |
normalizer_is_normalized($str->string) ?: $str->string = normalizer_normalize($str->string); |
| 213 |
|
| 214 |
if (false === $str->string) { |
| 215 |
throw new InvalidArgumentException('Invalid UTF-8 string.'); |
| 216 |
} |
| 217 |
|
| 218 |
return $str; |
| 219 |
} |
| 220 |
|
| 221 |
public function replace(string $from, string $to): static |
| 222 |
{ |
| 223 |
$str = clone $this; |
| 224 |
normalizer_is_normalized($from) ?: $from = normalizer_normalize($from); |
| 225 |
|
| 226 |
if ('' !== $from && false !== $from) { |
| 227 |
$tail = $str->string; |
| 228 |
$result = ''; |
| 229 |
$indexOf = $this->ignoreCase ? 'grapheme_stripos' : 'grapheme_strpos'; |
| 230 |
|
| 231 |
while ('' !== $tail && false !== $i = $indexOf($tail, $from)) { |
| 232 |
$slice = grapheme_substr($tail, 0, $i); |
| 233 |
$result .= $slice.$to; |
| 234 |
$tail = substr($tail, \strlen($slice) + \strlen($from)); |
| 235 |
} |
| 236 |
|
| 237 |
$str->string = $result.$tail; |
| 238 |
normalizer_is_normalized($str->string) ?: $str->string = normalizer_normalize($str->string); |
| 239 |
|
| 240 |
if (false === $str->string) { |
| 241 |
throw new InvalidArgumentException('Invalid UTF-8 string.'); |
| 242 |
} |
| 243 |
} |
| 244 |
|
| 245 |
return $str; |
| 246 |
} |
| 247 |
|
| 248 |
public function replaceMatches(string $fromRegexp, string|callable $to): static |
| 249 |
{ |
| 250 |
$str = parent::replaceMatches($fromRegexp, $to); |
| 251 |
normalizer_is_normalized($str->string) ?: $str->string = normalizer_normalize($str->string); |
| 252 |
|
| 253 |
return $str; |
| 254 |
} |
| 255 |
|
| 256 |
public function slice(int $start = 0, int $length = null): static |
| 257 |
{ |
| 258 |
$str = clone $this; |
| 259 |
|
| 260 |
$str->string = (string) grapheme_substr($this->string, $start, $length ?? 2147483647); |
| 261 |
|
| 262 |
return $str; |
| 263 |
} |
| 264 |
|
| 265 |
public function splice(string $replacement, int $start = 0, int $length = null): static |
| 266 |
{ |
| 267 |
$str = clone $this; |
| 268 |
|
| 269 |
$start = $start ? \strlen(grapheme_substr($this->string, 0, $start)) : 0; |
| 270 |
$length = $length ? \strlen(grapheme_substr($this->string, $start, $length ?? 2147483647)) : $length; |
| 271 |
$str->string = substr_replace($this->string, $replacement, $start, $length ?? 2147483647); |
| 272 |
normalizer_is_normalized($str->string) ?: $str->string = normalizer_normalize($str->string); |
| 273 |
|
| 274 |
if (false === $str->string) { |
| 275 |
throw new InvalidArgumentException('Invalid UTF-8 string.'); |
| 276 |
} |
| 277 |
|
| 278 |
return $str; |
| 279 |
} |
| 280 |
|
| 281 |
public function split(string $delimiter, int $limit = null, int $flags = null): array |
| 282 |
{ |
| 283 |
if (1 > $limit ??= 2147483647) { |
| 284 |
throw new InvalidArgumentException('Split limit must be a positive integer.'); |
| 285 |
} |
| 286 |
|
| 287 |
if ('' === $delimiter) { |
| 288 |
throw new InvalidArgumentException('Split delimiter is empty.'); |
| 289 |
} |
| 290 |
|
| 291 |
if (null !== $flags) { |
| 292 |
return parent::split($delimiter.'u', $limit, $flags); |
| 293 |
} |
| 294 |
|
| 295 |
normalizer_is_normalized($delimiter) ?: $delimiter = normalizer_normalize($delimiter); |
| 296 |
|
| 297 |
if (false === $delimiter) { |
| 298 |
throw new InvalidArgumentException('Split delimiter is not a valid UTF-8 string.'); |
| 299 |
} |
| 300 |
|
| 301 |
$str = clone $this; |
| 302 |
$tail = $this->string; |
| 303 |
$chunks = []; |
| 304 |
$indexOf = $this->ignoreCase ? 'grapheme_stripos' : 'grapheme_strpos'; |
| 305 |
|
| 306 |
while (1 < $limit && false !== $i = $indexOf($tail, $delimiter)) { |
| 307 |
$str->string = grapheme_substr($tail, 0, $i); |
| 308 |
$chunks[] = clone $str; |
| 309 |
$tail = substr($tail, \strlen($str->string) + \strlen($delimiter)); |
| 310 |
--$limit; |
| 311 |
} |
| 312 |
|
| 313 |
$str->string = $tail; |
| 314 |
$chunks[] = clone $str; |
| 315 |
|
| 316 |
return $chunks; |
| 317 |
} |
| 318 |
|
| 319 |
public function startsWith(string|iterable|AbstractString $prefix): bool |
| 320 |
{ |
| 321 |
if ($prefix instanceof AbstractString) { |
| 322 |
$prefix = $prefix->string; |
| 323 |
} elseif (!\is_string($prefix)) { |
| 324 |
return parent::startsWith($prefix); |
| 325 |
} |
| 326 |
|
| 327 |
$form = null === $this->ignoreCase ? \Normalizer::NFD : \Normalizer::NFC; |
| 328 |
normalizer_is_normalized($prefix, $form) ?: $prefix = normalizer_normalize($prefix, $form); |
| 329 |
|
| 330 |
if ('' === $prefix || false === $prefix) { |
| 331 |
return false; |
| 332 |
} |
| 333 |
|
| 334 |
if ($this->ignoreCase) { |
| 335 |
return 0 === mb_stripos(grapheme_extract($this->string, \strlen($prefix), \GRAPHEME_EXTR_MAXBYTES), $prefix, 0, 'UTF-8'); |
| 336 |
} |
| 337 |
|
| 338 |
return $prefix === grapheme_extract($this->string, \strlen($prefix), \GRAPHEME_EXTR_MAXBYTES); |
| 339 |
} |
| 340 |
|
| 341 |
public function __wakeup() |
| 342 |
{ |
| 343 |
if (!\is_string($this->string)) { |
| 344 |
throw new \BadMethodCallException('Cannot unserialize '.__CLASS__); |
| 345 |
} |
| 346 |
|
| 347 |
normalizer_is_normalized($this->string) ?: $this->string = normalizer_normalize($this->string); |
| 348 |
} |
| 349 |
|
| 350 |
public function __clone() |
| 351 |
{ |
| 352 |
if (null === $this->ignoreCase) { |
| 353 |
normalizer_is_normalized($this->string) ?: $this->string = normalizer_normalize($this->string); |
| 354 |
} |
| 355 |
|
| 356 |
$this->ignoreCase = false; |
| 357 |
} |
| 358 |
} |
| 359 |
|