| 1 |
<?php |
| 2 |
/** |
| 3 |
* Tokenizes PHP code. |
| 4 |
* |
| 5 |
* @author Greg Sherwood <[email protected]> |
| 6 |
* @copyright 2006-2015 Squiz Pty Ltd (ABN 77 084 670 600) |
| 7 |
* @license https://github.com/squizlabs/PHP_CodeSniffer/blob/master/licence.txt BSD Licence |
| 8 |
*/ |
| 9 |
|
| 10 |
namespace PHP_CodeSniffer\Tokenizers; |
| 11 |
|
| 12 |
use PHP_CodeSniffer\Util; |
| 13 |
|
| 14 |
class PHP extends Tokenizer |
| 15 |
{ |
| 16 |
|
| 17 |
/** |
| 18 |
* A list of tokens that are allowed to open a scope. |
| 19 |
* |
| 20 |
* This array also contains information about what kind of token the scope |
| 21 |
* opener uses to open and close the scope, if the token strictly requires |
| 22 |
* an opener, if the token can share a scope closer, and who it can be shared |
| 23 |
* with. An example of a token that shares a scope closer is a CASE scope. |
| 24 |
* |
| 25 |
* @var array |
| 26 |
*/ |
| 27 |
public $scopeOpeners = [ |
| 28 |
T_IF => [ |
| 29 |
'start' => [ |
| 30 |
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
| 31 |
T_COLON => T_COLON, |
| 32 |
], |
| 33 |
'end' => [ |
| 34 |
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
| 35 |
T_ENDIF => T_ENDIF, |
| 36 |
T_ELSE => T_ELSE, |
| 37 |
T_ELSEIF => T_ELSEIF, |
| 38 |
], |
| 39 |
'strict' => false, |
| 40 |
'shared' => false, |
| 41 |
'with' => [ |
| 42 |
T_ELSE => T_ELSE, |
| 43 |
T_ELSEIF => T_ELSEIF, |
| 44 |
], |
| 45 |
], |
| 46 |
T_TRY => [ |
| 47 |
'start' => [T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET], |
| 48 |
'end' => [T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET], |
| 49 |
'strict' => true, |
| 50 |
'shared' => false, |
| 51 |
'with' => [], |
| 52 |
], |
| 53 |
T_CATCH => [ |
| 54 |
'start' => [T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET], |
| 55 |
'end' => [T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET], |
| 56 |
'strict' => true, |
| 57 |
'shared' => false, |
| 58 |
'with' => [], |
| 59 |
], |
| 60 |
T_FINALLY => [ |
| 61 |
'start' => [T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET], |
| 62 |
'end' => [T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET], |
| 63 |
'strict' => true, |
| 64 |
'shared' => false, |
| 65 |
'with' => [], |
| 66 |
], |
| 67 |
T_ELSE => [ |
| 68 |
'start' => [ |
| 69 |
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
| 70 |
T_COLON => T_COLON, |
| 71 |
], |
| 72 |
'end' => [ |
| 73 |
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
| 74 |
T_ENDIF => T_ENDIF, |
| 75 |
], |
| 76 |
'strict' => false, |
| 77 |
'shared' => false, |
| 78 |
'with' => [ |
| 79 |
T_IF => T_IF, |
| 80 |
T_ELSEIF => T_ELSEIF, |
| 81 |
], |
| 82 |
], |
| 83 |
T_ELSEIF => [ |
| 84 |
'start' => [ |
| 85 |
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
| 86 |
T_COLON => T_COLON, |
| 87 |
], |
| 88 |
'end' => [ |
| 89 |
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
| 90 |
T_ENDIF => T_ENDIF, |
| 91 |
T_ELSE => T_ELSE, |
| 92 |
T_ELSEIF => T_ELSEIF, |
| 93 |
], |
| 94 |
'strict' => false, |
| 95 |
'shared' => false, |
| 96 |
'with' => [ |
| 97 |
T_IF => T_IF, |
| 98 |
T_ELSE => T_ELSE, |
| 99 |
], |
| 100 |
], |
| 101 |
T_FOR => [ |
| 102 |
'start' => [ |
| 103 |
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
| 104 |
T_COLON => T_COLON, |
| 105 |
], |
| 106 |
'end' => [ |
| 107 |
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
| 108 |
T_ENDFOR => T_ENDFOR, |
| 109 |
], |
| 110 |
'strict' => false, |
| 111 |
'shared' => false, |
| 112 |
'with' => [], |
| 113 |
], |
| 114 |
T_FOREACH => [ |
| 115 |
'start' => [ |
| 116 |
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
| 117 |
T_COLON => T_COLON, |
| 118 |
], |
| 119 |
'end' => [ |
| 120 |
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
| 121 |
T_ENDFOREACH => T_ENDFOREACH, |
| 122 |
], |
| 123 |
'strict' => false, |
| 124 |
'shared' => false, |
| 125 |
'with' => [], |
| 126 |
], |
| 127 |
T_INTERFACE => [ |
| 128 |
'start' => [T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET], |
| 129 |
'end' => [T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET], |
| 130 |
'strict' => true, |
| 131 |
'shared' => false, |
| 132 |
'with' => [], |
| 133 |
], |
| 134 |
T_FUNCTION => [ |
| 135 |
'start' => [T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET], |
| 136 |
'end' => [T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET], |
| 137 |
'strict' => true, |
| 138 |
'shared' => false, |
| 139 |
'with' => [], |
| 140 |
], |
| 141 |
T_CLASS => [ |
| 142 |
'start' => [T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET], |
| 143 |
'end' => [T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET], |
| 144 |
'strict' => true, |
| 145 |
'shared' => false, |
| 146 |
'with' => [], |
| 147 |
], |
| 148 |
T_TRAIT => [ |
| 149 |
'start' => [T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET], |
| 150 |
'end' => [T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET], |
| 151 |
'strict' => true, |
| 152 |
'shared' => false, |
| 153 |
'with' => [], |
| 154 |
], |
| 155 |
T_USE => [ |
| 156 |
'start' => [T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET], |
| 157 |
'end' => [T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET], |
| 158 |
'strict' => false, |
| 159 |
'shared' => false, |
| 160 |
'with' => [], |
| 161 |
], |
| 162 |
T_DECLARE => [ |
| 163 |
'start' => [ |
| 164 |
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
| 165 |
T_COLON => T_COLON, |
| 166 |
], |
| 167 |
'end' => [ |
| 168 |
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
| 169 |
T_ENDDECLARE => T_ENDDECLARE, |
| 170 |
], |
| 171 |
'strict' => false, |
| 172 |
'shared' => false, |
| 173 |
'with' => [], |
| 174 |
], |
| 175 |
T_NAMESPACE => [ |
| 176 |
'start' => [T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET], |
| 177 |
'end' => [T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET], |
| 178 |
'strict' => false, |
| 179 |
'shared' => false, |
| 180 |
'with' => [], |
| 181 |
], |
| 182 |
T_WHILE => [ |
| 183 |
'start' => [ |
| 184 |
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
| 185 |
T_COLON => T_COLON, |
| 186 |
], |
| 187 |
'end' => [ |
| 188 |
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
| 189 |
T_ENDWHILE => T_ENDWHILE, |
| 190 |
], |
| 191 |
'strict' => false, |
| 192 |
'shared' => false, |
| 193 |
'with' => [], |
| 194 |
], |
| 195 |
T_DO => [ |
| 196 |
'start' => [T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET], |
| 197 |
'end' => [T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET], |
| 198 |
'strict' => true, |
| 199 |
'shared' => false, |
| 200 |
'with' => [], |
| 201 |
], |
| 202 |
T_SWITCH => [ |
| 203 |
'start' => [ |
| 204 |
T_OPEN_CURLY_BRACKET => T_OPEN_CURLY_BRACKET, |
| 205 |
T_COLON => T_COLON, |
| 206 |
], |
| 207 |
'end' => [ |
| 208 |
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
| 209 |
T_ENDSWITCH => T_ENDSWITCH, |
| 210 |
], |
| 211 |
'strict' => true, |
| 212 |
'shared' => false, |
| 213 |
'with' => [], |
| 214 |
], |
| 215 |
T_CASE => [ |
| 216 |
'start' => [ |
| 217 |
T_COLON => T_COLON, |
| 218 |
T_SEMICOLON => T_SEMICOLON, |
| 219 |
], |
| 220 |
'end' => [ |
| 221 |
T_BREAK => T_BREAK, |
| 222 |
T_RETURN => T_RETURN, |
| 223 |
T_CONTINUE => T_CONTINUE, |
| 224 |
T_THROW => T_THROW, |
| 225 |
T_EXIT => T_EXIT, |
| 226 |
], |
| 227 |
'strict' => true, |
| 228 |
'shared' => true, |
| 229 |
'with' => [ |
| 230 |
T_DEFAULT => T_DEFAULT, |
| 231 |
T_CASE => T_CASE, |
| 232 |
T_SWITCH => T_SWITCH, |
| 233 |
], |
| 234 |
], |
| 235 |
T_DEFAULT => [ |
| 236 |
'start' => [ |
| 237 |
T_COLON => T_COLON, |
| 238 |
T_SEMICOLON => T_SEMICOLON, |
| 239 |
], |
| 240 |
'end' => [ |
| 241 |
T_BREAK => T_BREAK, |
| 242 |
T_RETURN => T_RETURN, |
| 243 |
T_CONTINUE => T_CONTINUE, |
| 244 |
T_THROW => T_THROW, |
| 245 |
T_EXIT => T_EXIT, |
| 246 |
], |
| 247 |
'strict' => true, |
| 248 |
'shared' => true, |
| 249 |
'with' => [ |
| 250 |
T_CASE => T_CASE, |
| 251 |
T_SWITCH => T_SWITCH, |
| 252 |
], |
| 253 |
], |
| 254 |
T_START_HEREDOC => [ |
| 255 |
'start' => [T_START_HEREDOC => T_START_HEREDOC], |
| 256 |
'end' => [T_END_HEREDOC => T_END_HEREDOC], |
| 257 |
'strict' => true, |
| 258 |
'shared' => false, |
| 259 |
'with' => [], |
| 260 |
], |
| 261 |
T_START_NOWDOC => [ |
| 262 |
'start' => [T_START_NOWDOC => T_START_NOWDOC], |
| 263 |
'end' => [T_END_NOWDOC => T_END_NOWDOC], |
| 264 |
'strict' => true, |
| 265 |
'shared' => false, |
| 266 |
'with' => [], |
| 267 |
], |
| 268 |
]; |
| 269 |
|
| 270 |
/** |
| 271 |
* A list of tokens that end the scope. |
| 272 |
* |
| 273 |
* This array is just a unique collection of the end tokens |
| 274 |
* from the scopeOpeners array. The data is duplicated here to |
| 275 |
* save time during parsing of the file. |
| 276 |
* |
| 277 |
* @var array |
| 278 |
*/ |
| 279 |
public $endScopeTokens = [ |
| 280 |
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
| 281 |
T_ENDIF => T_ENDIF, |
| 282 |
T_ENDFOR => T_ENDFOR, |
| 283 |
T_ENDFOREACH => T_ENDFOREACH, |
| 284 |
T_ENDWHILE => T_ENDWHILE, |
| 285 |
T_ENDSWITCH => T_ENDSWITCH, |
| 286 |
T_BREAK => T_BREAK, |
| 287 |
T_END_HEREDOC => T_END_HEREDOC, |
| 288 |
]; |
| 289 |
|
| 290 |
/** |
| 291 |
* Known lengths of tokens. |
| 292 |
* |
| 293 |
* @var array<int, int> |
| 294 |
*/ |
| 295 |
public $knownLengths = [ |
| 296 |
T_ABSTRACT => 8, |
| 297 |
T_AND_EQUAL => 2, |
| 298 |
T_ARRAY => 5, |
| 299 |
T_AS => 2, |
| 300 |
T_BOOLEAN_AND => 2, |
| 301 |
T_BOOLEAN_OR => 2, |
| 302 |
T_BREAK => 5, |
| 303 |
T_CALLABLE => 8, |
| 304 |
T_CASE => 4, |
| 305 |
T_CATCH => 5, |
| 306 |
T_CLASS => 5, |
| 307 |
T_CLASS_C => 9, |
| 308 |
T_CLONE => 5, |
| 309 |
T_CONCAT_EQUAL => 2, |
| 310 |
T_CONST => 5, |
| 311 |
T_CONTINUE => 8, |
| 312 |
T_CURLY_OPEN => 2, |
| 313 |
T_DEC => 2, |
| 314 |
T_DECLARE => 7, |
| 315 |
T_DEFAULT => 7, |
| 316 |
T_DIR => 7, |
| 317 |
T_DIV_EQUAL => 2, |
| 318 |
T_DO => 2, |
| 319 |
T_DOLLAR_OPEN_CURLY_BRACES => 2, |
| 320 |
T_DOUBLE_ARROW => 2, |
| 321 |
T_DOUBLE_COLON => 2, |
| 322 |
T_ECHO => 4, |
| 323 |
T_ELSE => 4, |
| 324 |
T_ELSEIF => 6, |
| 325 |
T_EMPTY => 5, |
| 326 |
T_ENDDECLARE => 10, |
| 327 |
T_ENDFOR => 6, |
| 328 |
T_ENDFOREACH => 10, |
| 329 |
T_ENDIF => 5, |
| 330 |
T_ENDSWITCH => 9, |
| 331 |
T_ENDWHILE => 8, |
| 332 |
T_EVAL => 4, |
| 333 |
T_EXTENDS => 7, |
| 334 |
T_FILE => 8, |
| 335 |
T_FINAL => 5, |
| 336 |
T_FINALLY => 7, |
| 337 |
T_FOR => 3, |
| 338 |
T_FOREACH => 7, |
| 339 |
T_FUNCTION => 8, |
| 340 |
T_FUNC_C => 12, |
| 341 |
T_GLOBAL => 6, |
| 342 |
T_GOTO => 4, |
| 343 |
T_HALT_COMPILER => 15, |
| 344 |
T_IF => 2, |
| 345 |
T_IMPLEMENTS => 10, |
| 346 |
T_INC => 2, |
| 347 |
T_INCLUDE => 7, |
| 348 |
T_INCLUDE_ONCE => 12, |
| 349 |
T_INSTANCEOF => 10, |
| 350 |
T_INSTEADOF => 9, |
| 351 |
T_INTERFACE => 9, |
| 352 |
T_ISSET => 5, |
| 353 |
T_IS_EQUAL => 2, |
| 354 |
T_IS_GREATER_OR_EQUAL => 2, |
| 355 |
T_IS_IDENTICAL => 3, |
| 356 |
T_IS_NOT_EQUAL => 2, |
| 357 |
T_IS_NOT_IDENTICAL => 3, |
| 358 |
T_IS_SMALLER_OR_EQUAL => 2, |
| 359 |
T_LINE => 8, |
| 360 |
T_LIST => 4, |
| 361 |
T_LOGICAL_AND => 3, |
| 362 |
T_LOGICAL_OR => 2, |
| 363 |
T_LOGICAL_XOR => 3, |
| 364 |
T_METHOD_C => 10, |
| 365 |
T_MINUS_EQUAL => 2, |
| 366 |
T_POW_EQUAL => 3, |
| 367 |
T_MOD_EQUAL => 2, |
| 368 |
T_MUL_EQUAL => 2, |
| 369 |
T_NAMESPACE => 9, |
| 370 |
T_NS_C => 13, |
| 371 |
T_NS_SEPARATOR => 1, |
| 372 |
T_NEW => 3, |
| 373 |
T_OBJECT_OPERATOR => 2, |
| 374 |
T_OPEN_TAG_WITH_ECHO => 3, |
| 375 |
T_OR_EQUAL => 2, |
| 376 |
T_PLUS_EQUAL => 2, |
| 377 |
T_PRINT => 5, |
| 378 |
T_PRIVATE => 7, |
| 379 |
T_PUBLIC => 6, |
| 380 |
T_PROTECTED => 9, |
| 381 |
T_REQUIRE => 7, |
| 382 |
T_REQUIRE_ONCE => 12, |
| 383 |
T_RETURN => 6, |
| 384 |
T_STATIC => 6, |
| 385 |
T_SWITCH => 6, |
| 386 |
T_THROW => 5, |
| 387 |
T_TRAIT => 5, |
| 388 |
T_TRAIT_C => 9, |
| 389 |
T_TRY => 3, |
| 390 |
T_UNSET => 5, |
| 391 |
T_USE => 3, |
| 392 |
T_VAR => 3, |
| 393 |
T_WHILE => 5, |
| 394 |
T_XOR_EQUAL => 2, |
| 395 |
T_YIELD => 5, |
| 396 |
T_OPEN_CURLY_BRACKET => 1, |
| 397 |
T_CLOSE_CURLY_BRACKET => 1, |
| 398 |
T_OPEN_SQUARE_BRACKET => 1, |
| 399 |
T_CLOSE_SQUARE_BRACKET => 1, |
| 400 |
T_OPEN_PARENTHESIS => 1, |
| 401 |
T_CLOSE_PARENTHESIS => 1, |
| 402 |
T_COLON => 1, |
| 403 |
T_STRING_CONCAT => 1, |
| 404 |
T_INLINE_THEN => 1, |
| 405 |
T_INLINE_ELSE => 1, |
| 406 |
T_NULLABLE => 1, |
| 407 |
T_NULL => 4, |
| 408 |
T_FALSE => 5, |
| 409 |
T_TRUE => 4, |
| 410 |
T_SEMICOLON => 1, |
| 411 |
T_EQUAL => 1, |
| 412 |
T_MULTIPLY => 1, |
| 413 |
T_DIVIDE => 1, |
| 414 |
T_PLUS => 1, |
| 415 |
T_MINUS => 1, |
| 416 |
T_MODULUS => 1, |
| 417 |
T_POW => 2, |
| 418 |
T_SPACESHIP => 3, |
| 419 |
T_COALESCE => 2, |
| 420 |
T_COALESCE_EQUAL => 3, |
| 421 |
T_BITWISE_AND => 1, |
| 422 |
T_BITWISE_OR => 1, |
| 423 |
T_BITWISE_XOR => 1, |
| 424 |
T_SL => 2, |
| 425 |
T_SR => 2, |
| 426 |
T_SL_EQUAL => 3, |
| 427 |
T_SR_EQUAL => 3, |
| 428 |
T_GREATER_THAN => 1, |
| 429 |
T_LESS_THAN => 1, |
| 430 |
T_BOOLEAN_NOT => 1, |
| 431 |
T_SELF => 4, |
| 432 |
T_PARENT => 6, |
| 433 |
T_COMMA => 1, |
| 434 |
T_THIS => 4, |
| 435 |
T_CLOSURE => 8, |
| 436 |
T_BACKTICK => 1, |
| 437 |
T_OPEN_SHORT_ARRAY => 1, |
| 438 |
T_CLOSE_SHORT_ARRAY => 1, |
| 439 |
]; |
| 440 |
|
| 441 |
/** |
| 442 |
* A cache of different token types, resolved into arrays. |
| 443 |
* |
| 444 |
* @var array |
| 445 |
* @see standardiseToken() |
| 446 |
*/ |
| 447 |
private static $resolveTokenCache = []; |
| 448 |
|
| 449 |
|
| 450 |
/** |
| 451 |
* Creates an array of tokens when given some PHP code. |
| 452 |
* |
| 453 |
* Starts by using token_get_all() but does a lot of extra processing |
| 454 |
* to insert information about the context of the token. |
| 455 |
* |
| 456 |
* @param string $string The string to tokenize. |
| 457 |
* |
| 458 |
* @return array |
| 459 |
*/ |
| 460 |
protected function tokenize($string) |
| 461 |
{ |
| 462 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 463 |
echo "\t*** START PHP TOKENIZING ***".PHP_EOL; |
| 464 |
$isWin = false; |
| 465 |
if (strtoupper(substr(PHP_OS, 0, 3)) === 'WIN') { |
| 466 |
$isWin = true; |
| 467 |
} |
| 468 |
} |
| 469 |
|
| 470 |
$tokens = @token_get_all($string); |
| 471 |
$finalTokens = []; |
| 472 |
|
| 473 |
$newStackPtr = 0; |
| 474 |
$numTokens = count($tokens); |
| 475 |
$lastNotEmptyToken = 0; |
| 476 |
|
| 477 |
$insideInlineIf = []; |
| 478 |
$insideUseGroup = false; |
| 479 |
|
| 480 |
$commentTokenizer = new Comment(); |
| 481 |
|
| 482 |
for ($stackPtr = 0; $stackPtr < $numTokens; $stackPtr++) { |
| 483 |
// Special case for tokens we have needed to blank out. |
| 484 |
if ($tokens[$stackPtr] === null) { |
| 485 |
continue; |
| 486 |
} |
| 487 |
|
| 488 |
$token = (array) $tokens[$stackPtr]; |
| 489 |
$tokenIsArray = isset($token[1]); |
| 490 |
|
| 491 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 492 |
if ($tokenIsArray === true) { |
| 493 |
$type = Util\Tokens::tokenName($token[0]); |
| 494 |
$content = Util\Common::prepareForOutput($token[1]); |
| 495 |
} else { |
| 496 |
$newToken = self::resolveSimpleToken($token[0]); |
| 497 |
$type = $newToken['type']; |
| 498 |
$content = Util\Common::prepareForOutput($token[0]); |
| 499 |
} |
| 500 |
|
| 501 |
echo "\tProcess token "; |
| 502 |
if ($tokenIsArray === true) { |
| 503 |
echo "[$stackPtr]"; |
| 504 |
} else { |
| 505 |
echo " $stackPtr "; |
| 506 |
} |
| 507 |
|
| 508 |
echo ": $type => $content"; |
| 509 |
}//end if |
| 510 |
|
| 511 |
if ($newStackPtr > 0 |
| 512 |
&& isset(Util\Tokens::$emptyTokens[$finalTokens[($newStackPtr - 1)]['code']]) === false |
| 513 |
) { |
| 514 |
$lastNotEmptyToken = ($newStackPtr - 1); |
| 515 |
} |
| 516 |
|
| 517 |
/* |
| 518 |
If we are using \r\n newline characters, the \r and \n are sometimes |
| 519 |
split over two tokens. This normally occurs after comments. We need |
| 520 |
to merge these two characters together so that our line endings are |
| 521 |
consistent for all lines. |
| 522 |
*/ |
| 523 |
|
| 524 |
if ($tokenIsArray === true && substr($token[1], -1) === "\r") { |
| 525 |
if (isset($tokens[($stackPtr + 1)]) === true |
| 526 |
&& is_array($tokens[($stackPtr + 1)]) === true |
| 527 |
&& $tokens[($stackPtr + 1)][1][0] === "\n" |
| 528 |
) { |
| 529 |
$token[1] .= "\n"; |
| 530 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 531 |
if ($isWin === true) { |
| 532 |
echo '\n'; |
| 533 |
} else { |
| 534 |
echo "\033[30;1m\\n\033[0m"; |
| 535 |
} |
| 536 |
} |
| 537 |
|
| 538 |
if ($tokens[($stackPtr + 1)][1] === "\n") { |
| 539 |
// This token's content has been merged into the previous, |
| 540 |
// so we can skip it. |
| 541 |
$tokens[($stackPtr + 1)] = ''; |
| 542 |
} else { |
| 543 |
$tokens[($stackPtr + 1)][1] = substr($tokens[($stackPtr + 1)][1], 1); |
| 544 |
} |
| 545 |
} |
| 546 |
}//end if |
| 547 |
|
| 548 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 549 |
echo PHP_EOL; |
| 550 |
} |
| 551 |
|
| 552 |
/* |
| 553 |
Parse doc blocks into something that can be easily iterated over. |
| 554 |
*/ |
| 555 |
|
| 556 |
if ($tokenIsArray === true |
| 557 |
&& ($token[0] === T_DOC_COMMENT |
| 558 |
|| ($token[0] === T_COMMENT && strpos($token[1], '/**') === 0)) |
| 559 |
) { |
| 560 |
$commentTokens = $commentTokenizer->tokenizeString($token[1], $this->eolChar, $newStackPtr); |
| 561 |
foreach ($commentTokens as $commentToken) { |
| 562 |
$finalTokens[$newStackPtr] = $commentToken; |
| 563 |
$newStackPtr++; |
| 564 |
} |
| 565 |
|
| 566 |
continue; |
| 567 |
} |
| 568 |
|
| 569 |
/* |
| 570 |
If this is a double quoted string, PHP will tokenize the whole |
| 571 |
thing which causes problems with the scope map when braces are |
| 572 |
within the string. So we need to merge the tokens together to |
| 573 |
provide a single string. |
| 574 |
*/ |
| 575 |
|
| 576 |
if ($tokenIsArray === false && ($token[0] === '"' || $token[0] === 'b"')) { |
| 577 |
// Binary casts need a special token. |
| 578 |
if ($token[0] === 'b"') { |
| 579 |
$finalTokens[$newStackPtr] = [ |
| 580 |
'code' => T_BINARY_CAST, |
| 581 |
'type' => 'T_BINARY_CAST', |
| 582 |
'content' => 'b', |
| 583 |
]; |
| 584 |
$newStackPtr++; |
| 585 |
} |
| 586 |
|
| 587 |
$tokenContent = '"'; |
| 588 |
$nestedVars = []; |
| 589 |
for ($i = ($stackPtr + 1); $i < $numTokens; $i++) { |
| 590 |
$subToken = (array) $tokens[$i]; |
| 591 |
$subTokenIsArray = isset($subToken[1]); |
| 592 |
|
| 593 |
if ($subTokenIsArray === true) { |
| 594 |
$tokenContent .= $subToken[1]; |
| 595 |
if ($subToken[1] === '{' |
| 596 |
&& $subToken[0] !== T_ENCAPSED_AND_WHITESPACE |
| 597 |
) { |
| 598 |
$nestedVars[] = $i; |
| 599 |
} |
| 600 |
} else { |
| 601 |
$tokenContent .= $subToken[0]; |
| 602 |
if ($subToken[0] === '}') { |
| 603 |
array_pop($nestedVars); |
| 604 |
} |
| 605 |
} |
| 606 |
|
| 607 |
if ($subTokenIsArray === false |
| 608 |
&& $subToken[0] === '"' |
| 609 |
&& empty($nestedVars) === true |
| 610 |
) { |
| 611 |
// We found the other end of the double quoted string. |
| 612 |
break; |
| 613 |
} |
| 614 |
}//end for |
| 615 |
|
| 616 |
$stackPtr = $i; |
| 617 |
|
| 618 |
// Convert each line within the double quoted string to a |
| 619 |
// new token, so it conforms with other multiple line tokens. |
| 620 |
$tokenLines = explode($this->eolChar, $tokenContent); |
| 621 |
$numLines = count($tokenLines); |
| 622 |
$newToken = []; |
| 623 |
|
| 624 |
for ($j = 0; $j < $numLines; $j++) { |
| 625 |
$newToken['content'] = $tokenLines[$j]; |
| 626 |
if ($j === ($numLines - 1)) { |
| 627 |
if ($tokenLines[$j] === '') { |
| 628 |
break; |
| 629 |
} |
| 630 |
} else { |
| 631 |
$newToken['content'] .= $this->eolChar; |
| 632 |
} |
| 633 |
|
| 634 |
$newToken['code'] = T_DOUBLE_QUOTED_STRING; |
| 635 |
$newToken['type'] = 'T_DOUBLE_QUOTED_STRING'; |
| 636 |
$finalTokens[$newStackPtr] = $newToken; |
| 637 |
$newStackPtr++; |
| 638 |
} |
| 639 |
|
| 640 |
// Continue, as we're done with this token. |
| 641 |
continue; |
| 642 |
}//end if |
| 643 |
|
| 644 |
/* |
| 645 |
Detect binary casting and assign the casts their own token. |
| 646 |
*/ |
| 647 |
|
| 648 |
if ($tokenIsArray === true |
| 649 |
&& $token[0] === T_CONSTANT_ENCAPSED_STRING |
| 650 |
&& (substr($token[1], 0, 2) === 'b"' |
| 651 |
|| substr($token[1], 0, 2) === "b'") |
| 652 |
) { |
| 653 |
$finalTokens[$newStackPtr] = [ |
| 654 |
'code' => T_BINARY_CAST, |
| 655 |
'type' => 'T_BINARY_CAST', |
| 656 |
'content' => 'b', |
| 657 |
]; |
| 658 |
$newStackPtr++; |
| 659 |
$token[1] = substr($token[1], 1); |
| 660 |
} |
| 661 |
|
| 662 |
if ($tokenIsArray === true |
| 663 |
&& $token[0] === T_STRING_CAST |
| 664 |
&& preg_match('`^\(\s*binary\s*\)$`i', $token[1]) === 1 |
| 665 |
) { |
| 666 |
$finalTokens[$newStackPtr] = [ |
| 667 |
'code' => T_BINARY_CAST, |
| 668 |
'type' => 'T_BINARY_CAST', |
| 669 |
'content' => $token[1], |
| 670 |
]; |
| 671 |
$newStackPtr++; |
| 672 |
continue; |
| 673 |
} |
| 674 |
|
| 675 |
/* |
| 676 |
If this is a heredoc, PHP will tokenize the whole |
| 677 |
thing which causes problems when heredocs don't |
| 678 |
contain real PHP code, which is almost never. |
| 679 |
We want to leave the start and end heredoc tokens |
| 680 |
alone though. |
| 681 |
*/ |
| 682 |
|
| 683 |
if ($tokenIsArray === true && $token[0] === T_START_HEREDOC) { |
| 684 |
// Add the start heredoc token to the final array. |
| 685 |
$finalTokens[$newStackPtr] = self::standardiseToken($token); |
| 686 |
|
| 687 |
// Check if this is actually a nowdoc and use a different token |
| 688 |
// to help the sniffs. |
| 689 |
$nowdoc = false; |
| 690 |
if (strpos($token[1], "'") !== false) { |
| 691 |
$finalTokens[$newStackPtr]['code'] = T_START_NOWDOC; |
| 692 |
$finalTokens[$newStackPtr]['type'] = 'T_START_NOWDOC'; |
| 693 |
$nowdoc = true; |
| 694 |
} |
| 695 |
|
| 696 |
$tokenContent = ''; |
| 697 |
for ($i = ($stackPtr + 1); $i < $numTokens; $i++) { |
| 698 |
$subTokenIsArray = is_array($tokens[$i]); |
| 699 |
if ($subTokenIsArray === true |
| 700 |
&& $tokens[$i][0] === T_END_HEREDOC |
| 701 |
) { |
| 702 |
// We found the other end of the heredoc. |
| 703 |
break; |
| 704 |
} |
| 705 |
|
| 706 |
if ($subTokenIsArray === true) { |
| 707 |
$tokenContent .= $tokens[$i][1]; |
| 708 |
} else { |
| 709 |
$tokenContent .= $tokens[$i]; |
| 710 |
} |
| 711 |
} |
| 712 |
|
| 713 |
if ($i === $numTokens) { |
| 714 |
// We got to the end of the file and never |
| 715 |
// found the closing token, so this probably wasn't |
| 716 |
// a heredoc. |
| 717 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 718 |
$type = $finalTokens[$newStackPtr]['type']; |
| 719 |
echo "\t\t* failed to find the end of the here/nowdoc".PHP_EOL; |
| 720 |
echo "\t\t* token $stackPtr changed from $type to T_STRING".PHP_EOL; |
| 721 |
} |
| 722 |
|
| 723 |
$finalTokens[$newStackPtr]['code'] = T_STRING; |
| 724 |
$finalTokens[$newStackPtr]['type'] = 'T_STRING'; |
| 725 |
$newStackPtr++; |
| 726 |
continue; |
| 727 |
} |
| 728 |
|
| 729 |
$stackPtr = $i; |
| 730 |
$newStackPtr++; |
| 731 |
|
| 732 |
// Convert each line within the heredoc to a |
| 733 |
// new token, so it conforms with other multiple line tokens. |
| 734 |
$tokenLines = explode($this->eolChar, $tokenContent); |
| 735 |
$numLines = count($tokenLines); |
| 736 |
$newToken = []; |
| 737 |
|
| 738 |
for ($j = 0; $j < $numLines; $j++) { |
| 739 |
$newToken['content'] = $tokenLines[$j]; |
| 740 |
if ($j === ($numLines - 1)) { |
| 741 |
if ($tokenLines[$j] === '') { |
| 742 |
break; |
| 743 |
} |
| 744 |
} else { |
| 745 |
$newToken['content'] .= $this->eolChar; |
| 746 |
} |
| 747 |
|
| 748 |
if ($nowdoc === true) { |
| 749 |
$newToken['code'] = T_NOWDOC; |
| 750 |
$newToken['type'] = 'T_NOWDOC'; |
| 751 |
} else { |
| 752 |
$newToken['code'] = T_HEREDOC; |
| 753 |
$newToken['type'] = 'T_HEREDOC'; |
| 754 |
} |
| 755 |
|
| 756 |
$finalTokens[$newStackPtr] = $newToken; |
| 757 |
$newStackPtr++; |
| 758 |
}//end for |
| 759 |
|
| 760 |
// Add the end heredoc token to the final array. |
| 761 |
$finalTokens[$newStackPtr] = self::standardiseToken($tokens[$stackPtr]); |
| 762 |
|
| 763 |
if ($nowdoc === true) { |
| 764 |
$finalTokens[$newStackPtr]['code'] = T_END_NOWDOC; |
| 765 |
$finalTokens[$newStackPtr]['type'] = 'T_END_NOWDOC'; |
| 766 |
} |
| 767 |
|
| 768 |
$newStackPtr++; |
| 769 |
|
| 770 |
// Continue, as we're done with this token. |
| 771 |
continue; |
| 772 |
}//end if |
| 773 |
|
| 774 |
/* |
| 775 |
Before PHP 7.0, the "yield from" was tokenized as |
| 776 |
T_YIELD, T_WHITESPACE and T_STRING. So look for |
| 777 |
and change this token in earlier versions. |
| 778 |
*/ |
| 779 |
|
| 780 |
if (PHP_VERSION_ID < 70000 |
| 781 |
&& PHP_VERSION_ID >= 50500 |
| 782 |
&& $tokenIsArray === true |
| 783 |
&& $token[0] === T_YIELD |
| 784 |
&& isset($tokens[($stackPtr + 1)]) === true |
| 785 |
&& isset($tokens[($stackPtr + 2)]) === true |
| 786 |
&& $tokens[($stackPtr + 1)][0] === T_WHITESPACE |
| 787 |
&& $tokens[($stackPtr + 2)][0] === T_STRING |
| 788 |
&& strtolower($tokens[($stackPtr + 2)][1]) === 'from' |
| 789 |
) { |
| 790 |
// Could be multi-line, so just the token stack. |
| 791 |
$token[0] = T_YIELD_FROM; |
| 792 |
$token[1] .= $tokens[($stackPtr + 1)][1].$tokens[($stackPtr + 2)][1]; |
| 793 |
|
| 794 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 795 |
for ($i = ($stackPtr + 1); $i <= ($stackPtr + 2); $i++) { |
| 796 |
$type = Util\Tokens::tokenName($tokens[$i][0]); |
| 797 |
$content = Util\Common::prepareForOutput($tokens[$i][1]); |
| 798 |
echo "\t\t* token $i merged into T_YIELD_FROM; was: $type => $content".PHP_EOL; |
| 799 |
} |
| 800 |
} |
| 801 |
|
| 802 |
$tokens[($stackPtr + 1)] = null; |
| 803 |
$tokens[($stackPtr + 2)] = null; |
| 804 |
} |
| 805 |
|
| 806 |
/* |
| 807 |
Before PHP 5.5, the yield keyword was tokenized as |
| 808 |
T_STRING. So look for and change this token in |
| 809 |
earlier versions. |
| 810 |
Checks also if it is just "yield" or "yield from". |
| 811 |
*/ |
| 812 |
|
| 813 |
if (PHP_VERSION_ID < 50500 |
| 814 |
&& $tokenIsArray === true |
| 815 |
&& $token[0] === T_STRING |
| 816 |
&& strtolower($token[1]) === 'yield' |
| 817 |
) { |
| 818 |
if (isset($tokens[($stackPtr + 1)]) === true |
| 819 |
&& isset($tokens[($stackPtr + 2)]) === true |
| 820 |
&& $tokens[($stackPtr + 1)][0] === T_WHITESPACE |
| 821 |
&& $tokens[($stackPtr + 2)][0] === T_STRING |
| 822 |
&& strtolower($tokens[($stackPtr + 2)][1]) === 'from' |
| 823 |
) { |
| 824 |
// Could be multi-line, so just just the token stack. |
| 825 |
$token[0] = T_YIELD_FROM; |
| 826 |
$token[1] .= $tokens[($stackPtr + 1)][1].$tokens[($stackPtr + 2)][1]; |
| 827 |
|
| 828 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 829 |
for ($i = ($stackPtr + 1); $i <= ($stackPtr + 2); $i++) { |
| 830 |
$type = Util\Tokens::tokenName($tokens[$i][0]); |
| 831 |
$content = Util\Common::prepareForOutput($tokens[$i][1]); |
| 832 |
echo "\t\t* token $i merged into T_YIELD_FROM; was: $type => $content".PHP_EOL; |
| 833 |
} |
| 834 |
} |
| 835 |
|
| 836 |
$tokens[($stackPtr + 1)] = null; |
| 837 |
$tokens[($stackPtr + 2)] = null; |
| 838 |
} else { |
| 839 |
$newToken = []; |
| 840 |
$newToken['code'] = T_YIELD; |
| 841 |
$newToken['type'] = 'T_YIELD'; |
| 842 |
$newToken['content'] = $token[1]; |
| 843 |
$finalTokens[$newStackPtr] = $newToken; |
| 844 |
|
| 845 |
$newStackPtr++; |
| 846 |
continue; |
| 847 |
}//end if |
| 848 |
}//end if |
| 849 |
|
| 850 |
/* |
| 851 |
Before PHP 5.6, the ... operator was tokenized as three |
| 852 |
T_STRING_CONCAT tokens in a row. So look for and combine |
| 853 |
these tokens in earlier versions. |
| 854 |
*/ |
| 855 |
|
| 856 |
if ($tokenIsArray === false |
| 857 |
&& $token[0] === '.' |
| 858 |
&& isset($tokens[($stackPtr + 1)]) === true |
| 859 |
&& isset($tokens[($stackPtr + 2)]) === true |
| 860 |
&& $tokens[($stackPtr + 1)] === '.' |
| 861 |
&& $tokens[($stackPtr + 2)] === '.' |
| 862 |
) { |
| 863 |
$newToken = []; |
| 864 |
$newToken['code'] = T_ELLIPSIS; |
| 865 |
$newToken['type'] = 'T_ELLIPSIS'; |
| 866 |
$newToken['content'] = '...'; |
| 867 |
$finalTokens[$newStackPtr] = $newToken; |
| 868 |
|
| 869 |
$newStackPtr++; |
| 870 |
$stackPtr += 2; |
| 871 |
continue; |
| 872 |
} |
| 873 |
|
| 874 |
/* |
| 875 |
Before PHP 5.6, the ** operator was tokenized as two |
| 876 |
T_MULTIPLY tokens in a row. So look for and combine |
| 877 |
these tokens in earlier versions. |
| 878 |
*/ |
| 879 |
|
| 880 |
if ($tokenIsArray === false |
| 881 |
&& $token[0] === '*' |
| 882 |
&& isset($tokens[($stackPtr + 1)]) === true |
| 883 |
&& $tokens[($stackPtr + 1)] === '*' |
| 884 |
) { |
| 885 |
$newToken = []; |
| 886 |
$newToken['code'] = T_POW; |
| 887 |
$newToken['type'] = 'T_POW'; |
| 888 |
$newToken['content'] = '**'; |
| 889 |
$finalTokens[$newStackPtr] = $newToken; |
| 890 |
|
| 891 |
$newStackPtr++; |
| 892 |
$stackPtr++; |
| 893 |
continue; |
| 894 |
} |
| 895 |
|
| 896 |
/* |
| 897 |
Before PHP 5.6, the **= operator was tokenized as |
| 898 |
T_MULTIPLY followed by T_MUL_EQUAL. So look for and combine |
| 899 |
these tokens in earlier versions. |
| 900 |
*/ |
| 901 |
|
| 902 |
if ($tokenIsArray === false |
| 903 |
&& $token[0] === '*' |
| 904 |
&& isset($tokens[($stackPtr + 1)]) === true |
| 905 |
&& is_array($tokens[($stackPtr + 1)]) === true |
| 906 |
&& $tokens[($stackPtr + 1)][1] === '*=' |
| 907 |
) { |
| 908 |
$newToken = []; |
| 909 |
$newToken['code'] = T_POW_EQUAL; |
| 910 |
$newToken['type'] = 'T_POW_EQUAL'; |
| 911 |
$newToken['content'] = '**='; |
| 912 |
$finalTokens[$newStackPtr] = $newToken; |
| 913 |
|
| 914 |
$newStackPtr++; |
| 915 |
$stackPtr++; |
| 916 |
continue; |
| 917 |
} |
| 918 |
|
| 919 |
/* |
| 920 |
Before PHP 7, the ??= operator was tokenized as |
| 921 |
T_INLINE_THEN, T_INLINE_THEN, T_EQUAL. |
| 922 |
Between PHP 7.0 and 7.2, the ??= operator was tokenized as |
| 923 |
T_COALESCE, T_EQUAL. |
| 924 |
So look for and combine these tokens in earlier versions. |
| 925 |
*/ |
| 926 |
|
| 927 |
if (($tokenIsArray === false |
| 928 |
&& $token[0] === '?' |
| 929 |
&& isset($tokens[($stackPtr + 1)]) === true |
| 930 |
&& $tokens[($stackPtr + 1)][0] === '?' |
| 931 |
&& isset($tokens[($stackPtr + 2)]) === true |
| 932 |
&& $tokens[($stackPtr + 2)][0] === '=') |
| 933 |
|| ($tokenIsArray === true |
| 934 |
&& $token[0] === T_COALESCE |
| 935 |
&& isset($tokens[($stackPtr + 1)]) === true |
| 936 |
&& $tokens[($stackPtr + 1)][0] === '=') |
| 937 |
) { |
| 938 |
$newToken = []; |
| 939 |
$newToken['code'] = T_COALESCE_EQUAL; |
| 940 |
$newToken['type'] = 'T_COALESCE_EQUAL'; |
| 941 |
$newToken['content'] = '??='; |
| 942 |
$finalTokens[$newStackPtr] = $newToken; |
| 943 |
|
| 944 |
$newStackPtr++; |
| 945 |
$stackPtr++; |
| 946 |
|
| 947 |
if ($tokenIsArray === false) { |
| 948 |
// Pre PHP 7. |
| 949 |
$stackPtr++; |
| 950 |
} |
| 951 |
|
| 952 |
continue; |
| 953 |
} |
| 954 |
|
| 955 |
/* |
| 956 |
Before PHP 7, the ?? operator was tokenized as |
| 957 |
T_INLINE_THEN followed by T_INLINE_THEN. |
| 958 |
So look for and combine these tokens in earlier versions. |
| 959 |
*/ |
| 960 |
|
| 961 |
if ($tokenIsArray === false |
| 962 |
&& $token[0] === '?' |
| 963 |
&& isset($tokens[($stackPtr + 1)]) === true |
| 964 |
&& $tokens[($stackPtr + 1)][0] === '?' |
| 965 |
) { |
| 966 |
$newToken = []; |
| 967 |
$newToken['code'] = T_COALESCE; |
| 968 |
$newToken['type'] = 'T_COALESCE'; |
| 969 |
$newToken['content'] = '??'; |
| 970 |
$finalTokens[$newStackPtr] = $newToken; |
| 971 |
|
| 972 |
$newStackPtr++; |
| 973 |
$stackPtr++; |
| 974 |
continue; |
| 975 |
} |
| 976 |
|
| 977 |
/* |
| 978 |
Before PHP 7.4, underscores inside T_LNUMBER and T_DNUMBER |
| 979 |
tokens split the token with a T_STRING. So look for |
| 980 |
and change these tokens in earlier versions. |
| 981 |
*/ |
| 982 |
|
| 983 |
if (PHP_VERSION_ID < 70400 |
| 984 |
&& ($tokenIsArray === true |
| 985 |
&& ($token[0] === T_LNUMBER |
| 986 |
|| $token[0] === T_DNUMBER) |
| 987 |
&& isset($tokens[($stackPtr + 1)]) === true |
| 988 |
&& is_array($tokens[($stackPtr + 1)]) === true |
| 989 |
&& $tokens[($stackPtr + 1)][0] === T_STRING |
| 990 |
&& $tokens[($stackPtr + 1)][1][0] === '_') |
| 991 |
) { |
| 992 |
$newContent = $token[1]; |
| 993 |
$newType = $token[0]; |
| 994 |
for ($i = ($stackPtr + 1); $i < $numTokens; $i++) { |
| 995 |
if (is_array($tokens[$i]) === false) { |
| 996 |
break; |
| 997 |
} |
| 998 |
|
| 999 |
if ($tokens[$i][0] === T_LNUMBER |
| 1000 |
|| $tokens[$i][0] === T_DNUMBER |
| 1001 |
) { |
| 1002 |
$newContent .= $tokens[$i][1]; |
| 1003 |
continue; |
| 1004 |
} |
| 1005 |
|
| 1006 |
if ($tokens[$i][0] === T_STRING |
| 1007 |
&& $tokens[$i][1][0] === '_' |
| 1008 |
&& ((strpos($newContent, '0x') === 0 |
| 1009 |
&& preg_match('`^((?<!\.)_[0-9A-F][0-9A-F\.]*)+$`iD', $tokens[$i][1]) === 1) |
| 1010 |
|| (strpos($newContent, '0x') !== 0 |
| 1011 |
&& substr($newContent, -1) !== '.' |
| 1012 |
&& substr(strtolower($newContent), -1) !== 'e' |
| 1013 |
&& preg_match('`^(?:(?<![\.e])_[0-9][0-9e\.]*)+$`iD', $tokens[$i][1]) === 1)) |
| 1014 |
) { |
| 1015 |
$newContent .= $tokens[$i][1]; |
| 1016 |
|
| 1017 |
// Support floats. |
| 1018 |
if (substr(strtolower($tokens[$i][1]), -1) === 'e' |
| 1019 |
&& ($tokens[($i + 1)] === '-' |
| 1020 |
|| $tokens[($i + 1)] === '+') |
| 1021 |
) { |
| 1022 |
$newContent .= $tokens[($i + 1)]; |
| 1023 |
$i++; |
| 1024 |
} |
| 1025 |
|
| 1026 |
continue; |
| 1027 |
}//end if |
| 1028 |
|
| 1029 |
break; |
| 1030 |
}//end for |
| 1031 |
|
| 1032 |
if ($newType === T_LNUMBER |
| 1033 |
&& ((stripos($newContent, '0x') === 0 && hexdec(str_replace('_', '', $newContent)) > PHP_INT_MAX) |
| 1034 |
|| (stripos($newContent, '0b') === 0 && bindec(str_replace('_', '', $newContent)) > PHP_INT_MAX) |
| 1035 |
|| (stripos($newContent, '0x') !== 0 |
| 1036 |
&& stripos($newContent, 'e') !== false || strpos($newContent, '.') !== false) |
| 1037 |
|| (strpos($newContent, '0') === 0 && stripos($newContent, '0x') !== 0 |
| 1038 |
&& stripos($newContent, '0b') !== 0 && octdec(str_replace('_', '', $newContent)) > PHP_INT_MAX) |
| 1039 |
|| (strpos($newContent, '0') !== 0 && str_replace('_', '', $newContent) > PHP_INT_MAX)) |
| 1040 |
) { |
| 1041 |
$newType = T_DNUMBER; |
| 1042 |
} |
| 1043 |
|
| 1044 |
$newToken = []; |
| 1045 |
$newToken['code'] = $newType; |
| 1046 |
$newToken['type'] = Util\Tokens::tokenName($newType); |
| 1047 |
$newToken['content'] = $newContent; |
| 1048 |
$finalTokens[$newStackPtr] = $newToken; |
| 1049 |
|
| 1050 |
$newStackPtr++; |
| 1051 |
$stackPtr = ($i - 1); |
| 1052 |
continue; |
| 1053 |
}//end if |
| 1054 |
|
| 1055 |
/* |
| 1056 |
Convert ? to T_NULLABLE OR T_INLINE_THEN |
| 1057 |
*/ |
| 1058 |
|
| 1059 |
if ($tokenIsArray === false && $token[0] === '?') { |
| 1060 |
$newToken = []; |
| 1061 |
$newToken['content'] = '?'; |
| 1062 |
|
| 1063 |
/* |
| 1064 |
* Check if the next non-empty token is one of the tokens which can be used |
| 1065 |
* in type declarations. If not, it's definitely a ternary. |
| 1066 |
* At this point, the only token types which need to be taken into consideration |
| 1067 |
* as potential type declarations are T_STRING, T_ARRAY, T_CALLABLE and T_NS_SEPARATOR. |
| 1068 |
*/ |
| 1069 |
|
| 1070 |
$lastRelevantNonEmpty = null; |
| 1071 |
|
| 1072 |
for ($i = ($stackPtr + 1); $i < $numTokens; $i++) { |
| 1073 |
if (is_array($tokens[$i]) === true) { |
| 1074 |
$tokenType = $tokens[$i][0]; |
| 1075 |
} else { |
| 1076 |
$tokenType = $tokens[$i]; |
| 1077 |
} |
| 1078 |
|
| 1079 |
if (isset(Util\Tokens::$emptyTokens[$tokenType]) === true) { |
| 1080 |
continue; |
| 1081 |
} |
| 1082 |
|
| 1083 |
if ($tokenType === T_STRING |
| 1084 |
|| $tokenType === T_ARRAY |
| 1085 |
|| $tokenType === T_NS_SEPARATOR |
| 1086 |
) { |
| 1087 |
$lastRelevantNonEmpty = $tokenType; |
| 1088 |
continue; |
| 1089 |
} |
| 1090 |
|
| 1091 |
if (($tokenType !== T_CALLABLE |
| 1092 |
&& isset($lastRelevantNonEmpty) === false) |
| 1093 |
|| ($lastRelevantNonEmpty === T_ARRAY |
| 1094 |
&& $tokenType === '(') |
| 1095 |
|| ($lastRelevantNonEmpty === T_STRING |
| 1096 |
&& ($tokenType === T_DOUBLE_COLON |
| 1097 |
|| $tokenType === '(' |
| 1098 |
|| $tokenType === ':')) |
| 1099 |
) { |
| 1100 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1101 |
echo "\t\t* token $stackPtr changed from ? to T_INLINE_THEN".PHP_EOL; |
| 1102 |
} |
| 1103 |
|
| 1104 |
$newToken['code'] = T_INLINE_THEN; |
| 1105 |
$newToken['type'] = 'T_INLINE_THEN'; |
| 1106 |
|
| 1107 |
$insideInlineIf[] = $stackPtr; |
| 1108 |
|
| 1109 |
$finalTokens[$newStackPtr] = $newToken; |
| 1110 |
$newStackPtr++; |
| 1111 |
continue 2; |
| 1112 |
} |
| 1113 |
|
| 1114 |
break; |
| 1115 |
}//end for |
| 1116 |
|
| 1117 |
/* |
| 1118 |
* This can still be a nullable type or a ternary. |
| 1119 |
* Do additional checking. |
| 1120 |
*/ |
| 1121 |
|
| 1122 |
$prevNonEmpty = null; |
| 1123 |
$lastSeenNonEmpty = null; |
| 1124 |
|
| 1125 |
for ($i = ($stackPtr - 1); $i >= 0; $i--) { |
| 1126 |
if (is_array($tokens[$i]) === true) { |
| 1127 |
$tokenType = $tokens[$i][0]; |
| 1128 |
} else { |
| 1129 |
$tokenType = $tokens[$i]; |
| 1130 |
} |
| 1131 |
|
| 1132 |
if ($tokenType === T_STATIC |
| 1133 |
&& ($lastSeenNonEmpty === T_DOUBLE_COLON |
| 1134 |
|| $lastSeenNonEmpty === '(') |
| 1135 |
) { |
| 1136 |
$lastSeenNonEmpty = $tokenType; |
| 1137 |
continue; |
| 1138 |
} |
| 1139 |
|
| 1140 |
if ($prevNonEmpty === null |
| 1141 |
&& isset(Util\Tokens::$emptyTokens[$tokenType]) === false |
| 1142 |
) { |
| 1143 |
// Found the previous non-empty token. |
| 1144 |
if ($tokenType === ':' || $tokenType === ',') { |
| 1145 |
$newToken['code'] = T_NULLABLE; |
| 1146 |
$newToken['type'] = 'T_NULLABLE'; |
| 1147 |
|
| 1148 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1149 |
echo "\t\t* token $stackPtr changed from ? to T_NULLABLE".PHP_EOL; |
| 1150 |
} |
| 1151 |
|
| 1152 |
break; |
| 1153 |
} |
| 1154 |
|
| 1155 |
$prevNonEmpty = $tokenType; |
| 1156 |
} |
| 1157 |
|
| 1158 |
if ($tokenType === T_FUNCTION |
| 1159 |
|| $tokenType === T_FN |
| 1160 |
|| isset(Util\Tokens::$methodPrefixes[$tokenType]) === true |
| 1161 |
|| $tokenType === T_VAR |
| 1162 |
) { |
| 1163 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1164 |
echo "\t\t* token $stackPtr changed from ? to T_NULLABLE".PHP_EOL; |
| 1165 |
} |
| 1166 |
|
| 1167 |
$newToken['code'] = T_NULLABLE; |
| 1168 |
$newToken['type'] = 'T_NULLABLE'; |
| 1169 |
break; |
| 1170 |
} else if (in_array($tokenType, [T_DOUBLE_ARROW, T_OPEN_TAG, T_OPEN_TAG_WITH_ECHO, '=', '{', ';'], true) === true) { |
| 1171 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1172 |
echo "\t\t* token $stackPtr changed from ? to T_INLINE_THEN".PHP_EOL; |
| 1173 |
} |
| 1174 |
|
| 1175 |
$newToken['code'] = T_INLINE_THEN; |
| 1176 |
$newToken['type'] = 'T_INLINE_THEN'; |
| 1177 |
|
| 1178 |
$insideInlineIf[] = $stackPtr; |
| 1179 |
break; |
| 1180 |
} |
| 1181 |
|
| 1182 |
if (isset(Util\Tokens::$emptyTokens[$tokenType]) === false) { |
| 1183 |
$lastSeenNonEmpty = $tokenType; |
| 1184 |
} |
| 1185 |
}//end for |
| 1186 |
|
| 1187 |
$finalTokens[$newStackPtr] = $newToken; |
| 1188 |
$newStackPtr++; |
| 1189 |
continue; |
| 1190 |
}//end if |
| 1191 |
|
| 1192 |
/* |
| 1193 |
Tokens after a double colon may look like scope openers, |
| 1194 |
such as when writing code like Foo::NAMESPACE, but they are |
| 1195 |
only ever variables or strings. |
| 1196 |
*/ |
| 1197 |
|
| 1198 |
if ($stackPtr > 1 |
| 1199 |
&& (is_array($tokens[($stackPtr - 1)]) === true |
| 1200 |
&& $tokens[($stackPtr - 1)][0] === T_PAAMAYIM_NEKUDOTAYIM) |
| 1201 |
&& $tokenIsArray === true |
| 1202 |
&& $token[0] !== T_STRING |
| 1203 |
&& $token[0] !== T_VARIABLE |
| 1204 |
&& $token[0] !== T_DOLLAR |
| 1205 |
&& isset(Util\Tokens::$emptyTokens[$token[0]]) === false |
| 1206 |
) { |
| 1207 |
$newToken = []; |
| 1208 |
$newToken['code'] = T_STRING; |
| 1209 |
$newToken['type'] = 'T_STRING'; |
| 1210 |
$newToken['content'] = $token[1]; |
| 1211 |
$finalTokens[$newStackPtr] = $newToken; |
| 1212 |
|
| 1213 |
$newStackPtr++; |
| 1214 |
continue; |
| 1215 |
} |
| 1216 |
|
| 1217 |
/* |
| 1218 |
Backfill the T_FN token for PHP versions < 7.4. |
| 1219 |
*/ |
| 1220 |
|
| 1221 |
if ($tokenIsArray === true |
| 1222 |
&& $token[0] === T_STRING |
| 1223 |
&& strtolower($token[1]) === 'fn' |
| 1224 |
) { |
| 1225 |
// Modify the original token stack so that |
| 1226 |
// future checks (like looking for T_NULLABLE) can |
| 1227 |
// detect the T_FN token more easily. |
| 1228 |
$tokens[$stackPtr][0] = T_FN; |
| 1229 |
$token[0] = T_FN; |
| 1230 |
} |
| 1231 |
|
| 1232 |
/* |
| 1233 |
The string-like token after a function keyword should always be |
| 1234 |
tokenized as T_STRING even if it appears to be a different token, |
| 1235 |
such as when writing code like: function default(): foo |
| 1236 |
so go forward and change the token type before it is processed. |
| 1237 |
*/ |
| 1238 |
|
| 1239 |
if ($tokenIsArray === true |
| 1240 |
&& ($token[0] === T_FUNCTION |
| 1241 |
|| $token[0] === T_FN) |
| 1242 |
&& $finalTokens[$lastNotEmptyToken]['code'] !== T_USE |
| 1243 |
) { |
| 1244 |
for ($x = ($stackPtr + 1); $x < $numTokens; $x++) { |
| 1245 |
if (is_array($tokens[$x]) === false |
| 1246 |
|| isset(Util\Tokens::$emptyTokens[$tokens[$x][0]]) === false |
| 1247 |
) { |
| 1248 |
// Non-empty content. |
| 1249 |
break; |
| 1250 |
} |
| 1251 |
} |
| 1252 |
|
| 1253 |
if ($x < $numTokens && is_array($tokens[$x]) === true) { |
| 1254 |
$tokens[$x][0] = T_STRING; |
| 1255 |
} |
| 1256 |
|
| 1257 |
/* |
| 1258 |
This is a special condition for T_ARRAY tokens used for |
| 1259 |
function return types. We want to keep the parenthesis map clean, |
| 1260 |
so let's tag these tokens as T_STRING. |
| 1261 |
*/ |
| 1262 |
|
| 1263 |
// Go looking for the colon to start the return type hint. |
| 1264 |
// Start by finding the closing parenthesis of the function. |
| 1265 |
$parenthesisStack = []; |
| 1266 |
$parenthesisCloser = false; |
| 1267 |
for ($x = ($stackPtr + 1); $x < $numTokens; $x++) { |
| 1268 |
if (is_array($tokens[$x]) === false && $tokens[$x] === '(') { |
| 1269 |
$parenthesisStack[] = $x; |
| 1270 |
} else if (is_array($tokens[$x]) === false && $tokens[$x] === ')') { |
| 1271 |
array_pop($parenthesisStack); |
| 1272 |
if (empty($parenthesisStack) === true) { |
| 1273 |
$parenthesisCloser = $x; |
| 1274 |
break; |
| 1275 |
} |
| 1276 |
} |
| 1277 |
} |
| 1278 |
|
| 1279 |
if ($parenthesisCloser !== false) { |
| 1280 |
for ($x = ($parenthesisCloser + 1); $x < $numTokens; $x++) { |
| 1281 |
if (is_array($tokens[$x]) === false |
| 1282 |
|| isset(Util\Tokens::$emptyTokens[$tokens[$x][0]]) === false |
| 1283 |
) { |
| 1284 |
// Non-empty content. |
| 1285 |
if (is_array($tokens[$x]) === true && $tokens[$x][0] === T_USE) { |
| 1286 |
// Found a use statements, so search ahead for the closing parenthesis. |
| 1287 |
for ($x += 1; $x < $numTokens; $x++) { |
| 1288 |
if (is_array($tokens[$x]) === false && $tokens[$x] === ')') { |
| 1289 |
continue(2); |
| 1290 |
} |
| 1291 |
} |
| 1292 |
} |
| 1293 |
|
| 1294 |
break; |
| 1295 |
} |
| 1296 |
} |
| 1297 |
|
| 1298 |
if (isset($tokens[$x]) === true |
| 1299 |
&& is_array($tokens[$x]) === false |
| 1300 |
&& $tokens[$x] === ':' |
| 1301 |
) { |
| 1302 |
$allowed = [ |
| 1303 |
T_STRING => T_STRING, |
| 1304 |
T_ARRAY => T_ARRAY, |
| 1305 |
T_CALLABLE => T_CALLABLE, |
| 1306 |
T_SELF => T_SELF, |
| 1307 |
T_PARENT => T_PARENT, |
| 1308 |
T_NS_SEPARATOR => T_NS_SEPARATOR, |
| 1309 |
]; |
| 1310 |
|
| 1311 |
$allowed += Util\Tokens::$emptyTokens; |
| 1312 |
|
| 1313 |
// Find the start of the return type. |
| 1314 |
for ($x += 1; $x < $numTokens; $x++) { |
| 1315 |
if (is_array($tokens[$x]) === true |
| 1316 |
&& isset(Util\Tokens::$emptyTokens[$tokens[$x][0]]) === true |
| 1317 |
) { |
| 1318 |
// Whitespace or comments before the return type. |
| 1319 |
continue; |
| 1320 |
} |
| 1321 |
|
| 1322 |
if (is_array($tokens[$x]) === false && $tokens[$x] === '?') { |
| 1323 |
// Found a nullable operator, so skip it. |
| 1324 |
// But also covert the token to save the tokenizer |
| 1325 |
// a bit of time later on. |
| 1326 |
$tokens[$x] = [ |
| 1327 |
T_NULLABLE, |
| 1328 |
'?', |
| 1329 |
]; |
| 1330 |
|
| 1331 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1332 |
echo "\t\t* token $x changed from ? to T_NULLABLE".PHP_EOL; |
| 1333 |
} |
| 1334 |
|
| 1335 |
continue; |
| 1336 |
} |
| 1337 |
|
| 1338 |
break; |
| 1339 |
}//end for |
| 1340 |
|
| 1341 |
// Any T_ARRAY tokens we find between here and the next |
| 1342 |
// token that can't be part of the return type need to be |
| 1343 |
// converted to T_STRING tokens. |
| 1344 |
for ($x; $x < $numTokens; $x++) { |
| 1345 |
if (is_array($tokens[$x]) === false || isset($allowed[$tokens[$x][0]]) === false) { |
| 1346 |
break; |
| 1347 |
} else if ($tokens[$x][0] === T_ARRAY) { |
| 1348 |
$tokens[$x][0] = T_STRING; |
| 1349 |
|
| 1350 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1351 |
echo "\t\t* token $x changed from T_ARRAY to T_STRING".PHP_EOL; |
| 1352 |
} |
| 1353 |
} |
| 1354 |
} |
| 1355 |
}//end if |
| 1356 |
}//end if |
| 1357 |
}//end if |
| 1358 |
|
| 1359 |
/* |
| 1360 |
Before PHP 7, the <=> operator was tokenized as |
| 1361 |
T_IS_SMALLER_OR_EQUAL followed by T_GREATER_THAN. |
| 1362 |
So look for and combine these tokens in earlier versions. |
| 1363 |
*/ |
| 1364 |
|
| 1365 |
if ($tokenIsArray === true |
| 1366 |
&& $token[0] === T_IS_SMALLER_OR_EQUAL |
| 1367 |
&& isset($tokens[($stackPtr + 1)]) === true |
| 1368 |
&& $tokens[($stackPtr + 1)][0] === '>' |
| 1369 |
) { |
| 1370 |
$newToken = []; |
| 1371 |
$newToken['code'] = T_SPACESHIP; |
| 1372 |
$newToken['type'] = 'T_SPACESHIP'; |
| 1373 |
$newToken['content'] = '<=>'; |
| 1374 |
$finalTokens[$newStackPtr] = $newToken; |
| 1375 |
|
| 1376 |
$newStackPtr++; |
| 1377 |
$stackPtr++; |
| 1378 |
continue; |
| 1379 |
} |
| 1380 |
|
| 1381 |
/* |
| 1382 |
PHP doesn't assign a token to goto labels, so we have to. |
| 1383 |
These are just string tokens with a single colon after them. Double |
| 1384 |
colons are already tokenized and so don't interfere with this check. |
| 1385 |
But we do have to account for CASE statements, that look just like |
| 1386 |
goto labels. |
| 1387 |
*/ |
| 1388 |
|
| 1389 |
if ($tokenIsArray === true |
| 1390 |
&& $token[0] === T_STRING |
| 1391 |
&& isset($tokens[($stackPtr + 1)]) === true |
| 1392 |
&& $tokens[($stackPtr + 1)] === ':' |
| 1393 |
&& $tokens[($stackPtr - 1)][0] !== T_PAAMAYIM_NEKUDOTAYIM |
| 1394 |
) { |
| 1395 |
$stopTokens = [ |
| 1396 |
T_CASE => true, |
| 1397 |
T_SEMICOLON => true, |
| 1398 |
T_OPEN_CURLY_BRACKET => true, |
| 1399 |
T_INLINE_THEN => true, |
| 1400 |
]; |
| 1401 |
|
| 1402 |
for ($x = ($newStackPtr - 1); $x > 0; $x--) { |
| 1403 |
if (isset($stopTokens[$finalTokens[$x]['code']]) === true) { |
| 1404 |
break; |
| 1405 |
} |
| 1406 |
} |
| 1407 |
|
| 1408 |
if ($finalTokens[$x]['code'] !== T_CASE |
| 1409 |
&& $finalTokens[$x]['code'] !== T_INLINE_THEN |
| 1410 |
) { |
| 1411 |
$finalTokens[$newStackPtr] = [ |
| 1412 |
'content' => $token[1].':', |
| 1413 |
'code' => T_GOTO_LABEL, |
| 1414 |
'type' => 'T_GOTO_LABEL', |
| 1415 |
]; |
| 1416 |
|
| 1417 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1418 |
echo "\t\t* token $stackPtr changed from T_STRING to T_GOTO_LABEL".PHP_EOL; |
| 1419 |
echo "\t\t* skipping T_COLON token ".($stackPtr + 1).PHP_EOL; |
| 1420 |
} |
| 1421 |
|
| 1422 |
$newStackPtr++; |
| 1423 |
$stackPtr++; |
| 1424 |
continue; |
| 1425 |
} |
| 1426 |
}//end if |
| 1427 |
|
| 1428 |
/* |
| 1429 |
If this token has newlines in its content, split each line up |
| 1430 |
and create a new token for each line. We do this so it's easier |
| 1431 |
to ascertain where errors occur on a line. |
| 1432 |
Note that $token[1] is the token's content. |
| 1433 |
*/ |
| 1434 |
|
| 1435 |
if ($tokenIsArray === true && strpos($token[1], $this->eolChar) !== false) { |
| 1436 |
$tokenLines = explode($this->eolChar, $token[1]); |
| 1437 |
$numLines = count($tokenLines); |
| 1438 |
$newToken = [ |
| 1439 |
'type' => Util\Tokens::tokenName($token[0]), |
| 1440 |
'code' => $token[0], |
| 1441 |
'content' => '', |
| 1442 |
]; |
| 1443 |
|
| 1444 |
for ($i = 0; $i < $numLines; $i++) { |
| 1445 |
$newToken['content'] = $tokenLines[$i]; |
| 1446 |
if ($i === ($numLines - 1)) { |
| 1447 |
if ($tokenLines[$i] === '') { |
| 1448 |
break; |
| 1449 |
} |
| 1450 |
} else { |
| 1451 |
$newToken['content'] .= $this->eolChar; |
| 1452 |
} |
| 1453 |
|
| 1454 |
$finalTokens[$newStackPtr] = $newToken; |
| 1455 |
$newStackPtr++; |
| 1456 |
} |
| 1457 |
} else { |
| 1458 |
if ($tokenIsArray === true && $token[0] === T_STRING) { |
| 1459 |
// Some T_STRING tokens should remain that way |
| 1460 |
// due to their context. |
| 1461 |
$context = [ |
| 1462 |
T_OBJECT_OPERATOR => true, |
| 1463 |
T_FUNCTION => true, |
| 1464 |
T_CLASS => true, |
| 1465 |
T_EXTENDS => true, |
| 1466 |
T_IMPLEMENTS => true, |
| 1467 |
T_NEW => true, |
| 1468 |
T_CONST => true, |
| 1469 |
T_NS_SEPARATOR => true, |
| 1470 |
T_USE => true, |
| 1471 |
T_NAMESPACE => true, |
| 1472 |
T_PAAMAYIM_NEKUDOTAYIM => true, |
| 1473 |
]; |
| 1474 |
|
| 1475 |
if (isset($context[$finalTokens[$lastNotEmptyToken]['code']]) === true) { |
| 1476 |
// Special case for syntax like: return new self |
| 1477 |
// where self should not be a string. |
| 1478 |
if ($finalTokens[$lastNotEmptyToken]['code'] === T_NEW |
| 1479 |
&& strtolower($token[1]) === 'self' |
| 1480 |
) { |
| 1481 |
$finalTokens[$newStackPtr] = [ |
| 1482 |
'content' => $token[1], |
| 1483 |
'code' => T_SELF, |
| 1484 |
'type' => 'T_SELF', |
| 1485 |
]; |
| 1486 |
} else { |
| 1487 |
$finalTokens[$newStackPtr] = [ |
| 1488 |
'content' => $token[1], |
| 1489 |
'code' => T_STRING, |
| 1490 |
'type' => 'T_STRING', |
| 1491 |
]; |
| 1492 |
} |
| 1493 |
|
| 1494 |
$newStackPtr++; |
| 1495 |
continue; |
| 1496 |
}//end if |
| 1497 |
}//end if |
| 1498 |
|
| 1499 |
$newToken = null; |
| 1500 |
if ($tokenIsArray === false) { |
| 1501 |
if (isset(self::$resolveTokenCache[$token[0]]) === true) { |
| 1502 |
$newToken = self::$resolveTokenCache[$token[0]]; |
| 1503 |
} |
| 1504 |
} else { |
| 1505 |
$cacheKey = null; |
| 1506 |
if ($token[0] === T_STRING) { |
| 1507 |
$cacheKey = strtolower($token[1]); |
| 1508 |
} else if ($token[0] !== T_CURLY_OPEN) { |
| 1509 |
$cacheKey = $token[0]; |
| 1510 |
} |
| 1511 |
|
| 1512 |
if ($cacheKey !== null && isset(self::$resolveTokenCache[$cacheKey]) === true) { |
| 1513 |
$newToken = self::$resolveTokenCache[$cacheKey]; |
| 1514 |
$newToken['content'] = $token[1]; |
| 1515 |
} |
| 1516 |
} |
| 1517 |
|
| 1518 |
if ($newToken === null) { |
| 1519 |
$newToken = self::standardiseToken($token); |
| 1520 |
} |
| 1521 |
|
| 1522 |
// Convert colons that are actually the ELSE component of an |
| 1523 |
// inline IF statement. |
| 1524 |
if (empty($insideInlineIf) === false && $newToken['code'] === T_COLON) { |
| 1525 |
// Make sure this isn't the return type separator of a closure. |
| 1526 |
$isInlineIf = true; |
| 1527 |
for ($i = ($stackPtr - 1); $i > 0; $i--) { |
| 1528 |
if (is_array($tokens[$i]) === false |
| 1529 |
|| ($tokens[$i][0] !== T_DOC_COMMENT |
| 1530 |
&& $tokens[$i][0] !== T_COMMENT |
| 1531 |
&& $tokens[$i][0] !== T_WHITESPACE) |
| 1532 |
) { |
| 1533 |
break; |
| 1534 |
} |
| 1535 |
} |
| 1536 |
|
| 1537 |
if ($tokens[$i] === ')') { |
| 1538 |
$parenCount = 1; |
| 1539 |
for ($i--; $i > 0; $i--) { |
| 1540 |
if ($tokens[$i] === '(') { |
| 1541 |
$parenCount--; |
| 1542 |
if ($parenCount === 0) { |
| 1543 |
break; |
| 1544 |
} |
| 1545 |
} else if ($tokens[$i] === ')') { |
| 1546 |
$parenCount++; |
| 1547 |
} |
| 1548 |
} |
| 1549 |
|
| 1550 |
// We've found the open parenthesis, so if the previous |
| 1551 |
// non-empty token is FUNCTION or USE, this is a closure. |
| 1552 |
for ($i--; $i > 0; $i--) { |
| 1553 |
if (is_array($tokens[$i]) === false |
| 1554 |
|| ($tokens[$i][0] !== T_DOC_COMMENT |
| 1555 |
&& $tokens[$i][0] !== T_COMMENT |
| 1556 |
&& $tokens[$i][0] !== T_WHITESPACE) |
| 1557 |
) { |
| 1558 |
break; |
| 1559 |
} |
| 1560 |
} |
| 1561 |
|
| 1562 |
if ($tokens[$i][0] === T_FUNCTION || $tokens[$i][0] === T_FN || $tokens[$i][0] === T_USE) { |
| 1563 |
$isInlineIf = false; |
| 1564 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1565 |
echo "\t\t* token is function return type, not T_INLINE_ELSE".PHP_EOL; |
| 1566 |
} |
| 1567 |
} |
| 1568 |
}//end if |
| 1569 |
|
| 1570 |
// Check to see if this is a CASE or DEFAULT opener. |
| 1571 |
$inlineIfToken = $insideInlineIf[(count($insideInlineIf) - 1)]; |
| 1572 |
for ($i = $stackPtr; $i > $inlineIfToken; $i--) { |
| 1573 |
if (is_array($tokens[$i]) === true |
| 1574 |
&& ($tokens[$i][0] === T_CASE |
| 1575 |
|| $tokens[$i][0] === T_DEFAULT) |
| 1576 |
) { |
| 1577 |
$isInlineIf = false; |
| 1578 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1579 |
echo "\t\t* token is T_CASE or T_DEFAULT opener, not T_INLINE_ELSE".PHP_EOL; |
| 1580 |
} |
| 1581 |
|
| 1582 |
break; |
| 1583 |
} |
| 1584 |
|
| 1585 |
if (is_array($tokens[$i]) === false |
| 1586 |
&& ($tokens[$i] === ';' |
| 1587 |
|| $tokens[$i] === '{') |
| 1588 |
) { |
| 1589 |
break; |
| 1590 |
} |
| 1591 |
} |
| 1592 |
|
| 1593 |
if ($isInlineIf === true) { |
| 1594 |
array_pop($insideInlineIf); |
| 1595 |
$newToken['code'] = T_INLINE_ELSE; |
| 1596 |
$newToken['type'] = 'T_INLINE_ELSE'; |
| 1597 |
|
| 1598 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1599 |
echo "\t\t* token changed from T_COLON to T_INLINE_ELSE".PHP_EOL; |
| 1600 |
} |
| 1601 |
} |
| 1602 |
}//end if |
| 1603 |
|
| 1604 |
// This is a special condition for T_ARRAY tokens used for |
| 1605 |
// type hinting function arguments as being arrays. We want to keep |
| 1606 |
// the parenthesis map clean, so let's tag these tokens as |
| 1607 |
// T_STRING. |
| 1608 |
if ($newToken['code'] === T_ARRAY) { |
| 1609 |
for ($i = $stackPtr; $i < $numTokens; $i++) { |
| 1610 |
if ($tokens[$i] === '(') { |
| 1611 |
break; |
| 1612 |
} else if ($tokens[$i][0] === T_VARIABLE) { |
| 1613 |
$newToken['code'] = T_STRING; |
| 1614 |
$newToken['type'] = 'T_STRING'; |
| 1615 |
break; |
| 1616 |
} |
| 1617 |
} |
| 1618 |
} |
| 1619 |
|
| 1620 |
// This is a special case when checking PHP 5.5+ code in PHP < 5.5 |
| 1621 |
// where "finally" should be T_FINALLY instead of T_STRING. |
| 1622 |
if ($newToken['code'] === T_STRING |
| 1623 |
&& strtolower($newToken['content']) === 'finally' |
| 1624 |
) { |
| 1625 |
$newToken['code'] = T_FINALLY; |
| 1626 |
$newToken['type'] = 'T_FINALLY'; |
| 1627 |
} |
| 1628 |
|
| 1629 |
// This is a special case for the PHP 5.5 classname::class syntax |
| 1630 |
// where "class" should be T_STRING instead of T_CLASS. |
| 1631 |
if (($newToken['code'] === T_CLASS |
| 1632 |
|| $newToken['code'] === T_FUNCTION) |
| 1633 |
&& $finalTokens[$lastNotEmptyToken]['code'] === T_DOUBLE_COLON |
| 1634 |
) { |
| 1635 |
$newToken['code'] = T_STRING; |
| 1636 |
$newToken['type'] = 'T_STRING'; |
| 1637 |
} |
| 1638 |
|
| 1639 |
// This is a special case for PHP 5.6 use function and use const |
| 1640 |
// where "function" and "const" should be T_STRING instead of T_FUNCTION |
| 1641 |
// and T_CONST. |
| 1642 |
if (($newToken['code'] === T_FUNCTION |
| 1643 |
|| $newToken['code'] === T_CONST) |
| 1644 |
&& ($finalTokens[$lastNotEmptyToken]['code'] === T_USE || $insideUseGroup === true) |
| 1645 |
) { |
| 1646 |
$newToken['code'] = T_STRING; |
| 1647 |
$newToken['type'] = 'T_STRING'; |
| 1648 |
} |
| 1649 |
|
| 1650 |
// This is a special case for use groups in PHP 7+ where leaving |
| 1651 |
// the curly braces as their normal tokens would confuse |
| 1652 |
// the scope map and sniffs. |
| 1653 |
if ($newToken['code'] === T_OPEN_CURLY_BRACKET |
| 1654 |
&& $finalTokens[$lastNotEmptyToken]['code'] === T_NS_SEPARATOR |
| 1655 |
) { |
| 1656 |
$newToken['code'] = T_OPEN_USE_GROUP; |
| 1657 |
$newToken['type'] = 'T_OPEN_USE_GROUP'; |
| 1658 |
$insideUseGroup = true; |
| 1659 |
} |
| 1660 |
|
| 1661 |
if ($insideUseGroup === true && $newToken['code'] === T_CLOSE_CURLY_BRACKET) { |
| 1662 |
$newToken['code'] = T_CLOSE_USE_GROUP; |
| 1663 |
$newToken['type'] = 'T_CLOSE_USE_GROUP'; |
| 1664 |
$insideUseGroup = false; |
| 1665 |
} |
| 1666 |
|
| 1667 |
$finalTokens[$newStackPtr] = $newToken; |
| 1668 |
$newStackPtr++; |
| 1669 |
}//end if |
| 1670 |
}//end for |
| 1671 |
|
| 1672 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1673 |
echo "\t*** END PHP TOKENIZING ***".PHP_EOL; |
| 1674 |
} |
| 1675 |
|
| 1676 |
return $finalTokens; |
| 1677 |
|
| 1678 |
}//end tokenize() |
| 1679 |
|
| 1680 |
|
| 1681 |
/** |
| 1682 |
* Performs additional processing after main tokenizing. |
| 1683 |
* |
| 1684 |
* This additional processing checks for CASE statements that are using curly |
| 1685 |
* braces for scope openers and closers. It also turns some T_FUNCTION tokens |
| 1686 |
* into T_CLOSURE when they are not standard function definitions. It also |
| 1687 |
* detects short array syntax and converts those square brackets into new tokens. |
| 1688 |
* It also corrects some usage of the static and class keywords. It also |
| 1689 |
* assigns tokens to function return types. |
| 1690 |
* |
| 1691 |
* @return void |
| 1692 |
*/ |
| 1693 |
protected function processAdditional() |
| 1694 |
{ |
| 1695 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1696 |
echo "\t*** START ADDITIONAL PHP PROCESSING ***".PHP_EOL; |
| 1697 |
} |
| 1698 |
|
| 1699 |
$numTokens = count($this->tokens); |
| 1700 |
for ($i = ($numTokens - 1); $i >= 0; $i--) { |
| 1701 |
// Check for any unset scope conditions due to alternate IF/ENDIF syntax. |
| 1702 |
if (isset($this->tokens[$i]['scope_opener']) === true |
| 1703 |
&& isset($this->tokens[$i]['scope_condition']) === false |
| 1704 |
) { |
| 1705 |
$this->tokens[$i]['scope_condition'] = $this->tokens[$this->tokens[$i]['scope_opener']]['scope_condition']; |
| 1706 |
} |
| 1707 |
|
| 1708 |
if ($this->tokens[$i]['code'] === T_FUNCTION) { |
| 1709 |
/* |
| 1710 |
Detect functions that are actually closures and |
| 1711 |
assign them a different token. |
| 1712 |
*/ |
| 1713 |
|
| 1714 |
if (isset($this->tokens[$i]['scope_opener']) === true) { |
| 1715 |
for ($x = ($i + 1); $x < $numTokens; $x++) { |
| 1716 |
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false |
| 1717 |
&& $this->tokens[$x]['code'] !== T_BITWISE_AND |
| 1718 |
) { |
| 1719 |
break; |
| 1720 |
} |
| 1721 |
} |
| 1722 |
|
| 1723 |
if ($this->tokens[$x]['code'] === T_OPEN_PARENTHESIS) { |
| 1724 |
$this->tokens[$i]['code'] = T_CLOSURE; |
| 1725 |
$this->tokens[$i]['type'] = 'T_CLOSURE'; |
| 1726 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1727 |
$line = $this->tokens[$i]['line']; |
| 1728 |
echo "\t* token $i on line $line changed from T_FUNCTION to T_CLOSURE".PHP_EOL; |
| 1729 |
} |
| 1730 |
|
| 1731 |
for ($x = ($this->tokens[$i]['scope_opener'] + 1); $x < $this->tokens[$i]['scope_closer']; $x++) { |
| 1732 |
if (isset($this->tokens[$x]['conditions'][$i]) === false) { |
| 1733 |
continue; |
| 1734 |
} |
| 1735 |
|
| 1736 |
$this->tokens[$x]['conditions'][$i] = T_CLOSURE; |
| 1737 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1738 |
$type = $this->tokens[$x]['type']; |
| 1739 |
echo "\t\t* cleaned $x ($type) *".PHP_EOL; |
| 1740 |
} |
| 1741 |
} |
| 1742 |
} |
| 1743 |
}//end if |
| 1744 |
|
| 1745 |
continue; |
| 1746 |
} else if ($this->tokens[$i]['code'] === T_CLASS && isset($this->tokens[$i]['scope_opener']) === true) { |
| 1747 |
/* |
| 1748 |
Detect anonymous classes and assign them a different token. |
| 1749 |
*/ |
| 1750 |
|
| 1751 |
for ($x = ($i + 1); $x < $numTokens; $x++) { |
| 1752 |
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
| 1753 |
break; |
| 1754 |
} |
| 1755 |
} |
| 1756 |
|
| 1757 |
if ($this->tokens[$x]['code'] === T_OPEN_PARENTHESIS |
| 1758 |
|| $this->tokens[$x]['code'] === T_OPEN_CURLY_BRACKET |
| 1759 |
|| $this->tokens[$x]['code'] === T_EXTENDS |
| 1760 |
|| $this->tokens[$x]['code'] === T_IMPLEMENTS |
| 1761 |
) { |
| 1762 |
$this->tokens[$i]['code'] = T_ANON_CLASS; |
| 1763 |
$this->tokens[$i]['type'] = 'T_ANON_CLASS'; |
| 1764 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1765 |
$line = $this->tokens[$i]['line']; |
| 1766 |
echo "\t* token $i on line $line changed from T_CLASS to T_ANON_CLASS".PHP_EOL; |
| 1767 |
} |
| 1768 |
|
| 1769 |
if ($this->tokens[$x]['code'] === T_OPEN_PARENTHESIS |
| 1770 |
&& isset($this->tokens[$x]['parenthesis_closer']) === true |
| 1771 |
) { |
| 1772 |
$closer = $this->tokens[$x]['parenthesis_closer']; |
| 1773 |
|
| 1774 |
$this->tokens[$i]['parenthesis_opener'] = $x; |
| 1775 |
$this->tokens[$i]['parenthesis_closer'] = $closer; |
| 1776 |
$this->tokens[$i]['parenthesis_owner'] = $i; |
| 1777 |
$this->tokens[$x]['parenthesis_owner'] = $i; |
| 1778 |
$this->tokens[$closer]['parenthesis_owner'] = $i; |
| 1779 |
|
| 1780 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1781 |
$line = $this->tokens[$i]['line']; |
| 1782 |
echo "\t\t* added parenthesis keys to T_ANON_CLASS token $i on line $line".PHP_EOL; |
| 1783 |
} |
| 1784 |
} |
| 1785 |
|
| 1786 |
for ($x = ($this->tokens[$i]['scope_opener'] + 1); $x < $this->tokens[$i]['scope_closer']; $x++) { |
| 1787 |
if (isset($this->tokens[$x]['conditions'][$i]) === false) { |
| 1788 |
continue; |
| 1789 |
} |
| 1790 |
|
| 1791 |
$this->tokens[$x]['conditions'][$i] = T_ANON_CLASS; |
| 1792 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1793 |
$type = $this->tokens[$x]['type']; |
| 1794 |
echo "\t\t* cleaned $x ($type) *".PHP_EOL; |
| 1795 |
} |
| 1796 |
} |
| 1797 |
}//end if |
| 1798 |
|
| 1799 |
continue; |
| 1800 |
} else if ($this->tokens[$i]['code'] === T_FN && isset($this->tokens[($i + 1)]) === true) { |
| 1801 |
// Possible arrow function. |
| 1802 |
for ($x = ($i + 1); $x < $numTokens; $x++) { |
| 1803 |
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false |
| 1804 |
&& $this->tokens[$x]['code'] !== T_BITWISE_AND |
| 1805 |
) { |
| 1806 |
// Non-whitespace content. |
| 1807 |
break; |
| 1808 |
} |
| 1809 |
} |
| 1810 |
|
| 1811 |
if (isset($this->tokens[$x]) === true && $this->tokens[$x]['code'] === T_OPEN_PARENTHESIS) { |
| 1812 |
$ignore = Util\Tokens::$emptyTokens; |
| 1813 |
$ignore += [ |
| 1814 |
T_STRING => T_STRING, |
| 1815 |
T_ARRAY => T_ARRAY, |
| 1816 |
T_COLON => T_COLON, |
| 1817 |
T_NS_SEPARATOR => T_NS_SEPARATOR, |
| 1818 |
T_NULLABLE => T_NULLABLE, |
| 1819 |
T_CALLABLE => T_CALLABLE, |
| 1820 |
T_PARENT => T_PARENT, |
| 1821 |
T_SELF => T_SELF, |
| 1822 |
]; |
| 1823 |
|
| 1824 |
$closer = $this->tokens[$x]['parenthesis_closer']; |
| 1825 |
for ($arrow = ($closer + 1); $arrow < $numTokens; $arrow++) { |
| 1826 |
if (isset($ignore[$this->tokens[$arrow]['code']]) === false) { |
| 1827 |
break; |
| 1828 |
} |
| 1829 |
} |
| 1830 |
|
| 1831 |
if ($this->tokens[$arrow]['code'] === T_DOUBLE_ARROW) { |
| 1832 |
$endTokens = [ |
| 1833 |
T_COLON => true, |
| 1834 |
T_COMMA => true, |
| 1835 |
T_SEMICOLON => true, |
| 1836 |
T_CLOSE_PARENTHESIS => true, |
| 1837 |
T_CLOSE_SQUARE_BRACKET => true, |
| 1838 |
T_CLOSE_CURLY_BRACKET => true, |
| 1839 |
T_CLOSE_SHORT_ARRAY => true, |
| 1840 |
T_OPEN_TAG => true, |
| 1841 |
T_CLOSE_TAG => true, |
| 1842 |
]; |
| 1843 |
|
| 1844 |
$inTernary = false; |
| 1845 |
|
| 1846 |
for ($scopeCloser = ($arrow + 1); $scopeCloser < $numTokens; $scopeCloser++) { |
| 1847 |
if (isset($endTokens[$this->tokens[$scopeCloser]['code']]) === true) { |
| 1848 |
break; |
| 1849 |
} |
| 1850 |
|
| 1851 |
if (isset($this->tokens[$scopeCloser]['scope_closer']) === true |
| 1852 |
&& $this->tokens[$scopeCloser]['code'] !== T_INLINE_ELSE |
| 1853 |
) { |
| 1854 |
// We minus 1 here in case the closer can be shared with us. |
| 1855 |
$scopeCloser = ($this->tokens[$scopeCloser]['scope_closer'] - 1); |
| 1856 |
continue; |
| 1857 |
} |
| 1858 |
|
| 1859 |
if (isset($this->tokens[$scopeCloser]['parenthesis_closer']) === true) { |
| 1860 |
$scopeCloser = $this->tokens[$scopeCloser]['parenthesis_closer']; |
| 1861 |
continue; |
| 1862 |
} |
| 1863 |
|
| 1864 |
if (isset($this->tokens[$scopeCloser]['bracket_closer']) === true) { |
| 1865 |
$scopeCloser = $this->tokens[$scopeCloser]['bracket_closer']; |
| 1866 |
continue; |
| 1867 |
} |
| 1868 |
|
| 1869 |
if ($this->tokens[$scopeCloser]['code'] === T_INLINE_THEN) { |
| 1870 |
$inTernary = true; |
| 1871 |
continue; |
| 1872 |
} |
| 1873 |
|
| 1874 |
if ($this->tokens[$scopeCloser]['code'] === T_INLINE_ELSE) { |
| 1875 |
if ($inTernary === false) { |
| 1876 |
break; |
| 1877 |
} |
| 1878 |
|
| 1879 |
$inTernary = false; |
| 1880 |
continue; |
| 1881 |
} |
| 1882 |
}//end for |
| 1883 |
|
| 1884 |
if ($scopeCloser !== $numTokens) { |
| 1885 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1886 |
$line = $this->tokens[$i]['line']; |
| 1887 |
echo "\t=> token $i on line $line processed as arrow function".PHP_EOL; |
| 1888 |
} |
| 1889 |
|
| 1890 |
$this->tokens[$i]['code'] = T_FN; |
| 1891 |
$this->tokens[$i]['type'] = 'T_FN'; |
| 1892 |
$this->tokens[$i]['scope_condition'] = $i; |
| 1893 |
$this->tokens[$i]['scope_opener'] = $arrow; |
| 1894 |
$this->tokens[$i]['scope_closer'] = $scopeCloser; |
| 1895 |
$this->tokens[$i]['parenthesis_owner'] = $i; |
| 1896 |
$this->tokens[$i]['parenthesis_opener'] = $x; |
| 1897 |
$this->tokens[$i]['parenthesis_closer'] = $this->tokens[$x]['parenthesis_closer']; |
| 1898 |
|
| 1899 |
$this->tokens[$arrow]['code'] = T_FN_ARROW; |
| 1900 |
$this->tokens[$arrow]['type'] = 'T_FN_ARROW'; |
| 1901 |
|
| 1902 |
$this->tokens[$arrow]['scope_condition'] = $i; |
| 1903 |
$this->tokens[$arrow]['scope_opener'] = $arrow; |
| 1904 |
$this->tokens[$arrow]['scope_closer'] = $scopeCloser; |
| 1905 |
$this->tokens[$scopeCloser]['scope_condition'] = $i; |
| 1906 |
$this->tokens[$scopeCloser]['scope_opener'] = $arrow; |
| 1907 |
$this->tokens[$scopeCloser]['scope_closer'] = $scopeCloser; |
| 1908 |
|
| 1909 |
$opener = $this->tokens[$i]['parenthesis_opener']; |
| 1910 |
$closer = $this->tokens[$i]['parenthesis_closer']; |
| 1911 |
$this->tokens[$opener]['parenthesis_owner'] = $i; |
| 1912 |
$this->tokens[$closer]['parenthesis_owner'] = $i; |
| 1913 |
|
| 1914 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1915 |
$line = $this->tokens[$arrow]['line']; |
| 1916 |
echo "\t\t* token $arrow on line $line changed from T_DOUBLE_ARROW to T_FN_ARROW".PHP_EOL; |
| 1917 |
} |
| 1918 |
}//end if |
| 1919 |
}//end if |
| 1920 |
}//end if |
| 1921 |
} else if ($this->tokens[$i]['code'] === T_OPEN_SQUARE_BRACKET) { |
| 1922 |
if (isset($this->tokens[$i]['bracket_closer']) === false) { |
| 1923 |
continue; |
| 1924 |
} |
| 1925 |
|
| 1926 |
// Unless there is a variable or a bracket before this token, |
| 1927 |
// it is the start of an array being defined using the short syntax. |
| 1928 |
$isShortArray = false; |
| 1929 |
$allowed = [ |
| 1930 |
T_CLOSE_SQUARE_BRACKET => T_CLOSE_SQUARE_BRACKET, |
| 1931 |
T_CLOSE_CURLY_BRACKET => T_CLOSE_CURLY_BRACKET, |
| 1932 |
T_CLOSE_PARENTHESIS => T_CLOSE_PARENTHESIS, |
| 1933 |
T_VARIABLE => T_VARIABLE, |
| 1934 |
T_OBJECT_OPERATOR => T_OBJECT_OPERATOR, |
| 1935 |
T_STRING => T_STRING, |
| 1936 |
T_CONSTANT_ENCAPSED_STRING => T_CONSTANT_ENCAPSED_STRING, |
| 1937 |
]; |
| 1938 |
|
| 1939 |
for ($x = ($i - 1); $x >= 0; $x--) { |
| 1940 |
// If we hit a scope opener, the statement has ended |
| 1941 |
// without finding anything, so it's probably an array |
| 1942 |
// using PHP 7.1 short list syntax. |
| 1943 |
if (isset($this->tokens[$x]['scope_opener']) === true) { |
| 1944 |
$isShortArray = true; |
| 1945 |
break; |
| 1946 |
} |
| 1947 |
|
| 1948 |
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
| 1949 |
if (isset($allowed[$this->tokens[$x]['code']]) === false) { |
| 1950 |
$isShortArray = true; |
| 1951 |
} |
| 1952 |
|
| 1953 |
break; |
| 1954 |
} |
| 1955 |
} |
| 1956 |
|
| 1957 |
if ($isShortArray === true) { |
| 1958 |
$this->tokens[$i]['code'] = T_OPEN_SHORT_ARRAY; |
| 1959 |
$this->tokens[$i]['type'] = 'T_OPEN_SHORT_ARRAY'; |
| 1960 |
|
| 1961 |
$closer = $this->tokens[$i]['bracket_closer']; |
| 1962 |
$this->tokens[$closer]['code'] = T_CLOSE_SHORT_ARRAY; |
| 1963 |
$this->tokens[$closer]['type'] = 'T_CLOSE_SHORT_ARRAY'; |
| 1964 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1965 |
$line = $this->tokens[$i]['line']; |
| 1966 |
echo "\t* token $i on line $line changed from T_OPEN_SQUARE_BRACKET to T_OPEN_SHORT_ARRAY".PHP_EOL; |
| 1967 |
$line = $this->tokens[$closer]['line']; |
| 1968 |
echo "\t* token $closer on line $line changed from T_CLOSE_SQUARE_BRACKET to T_CLOSE_SHORT_ARRAY".PHP_EOL; |
| 1969 |
} |
| 1970 |
} |
| 1971 |
|
| 1972 |
continue; |
| 1973 |
} else if ($this->tokens[$i]['code'] === T_STATIC) { |
| 1974 |
for ($x = ($i - 1); $x > 0; $x--) { |
| 1975 |
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
| 1976 |
break; |
| 1977 |
} |
| 1978 |
} |
| 1979 |
|
| 1980 |
if ($this->tokens[$x]['code'] === T_INSTANCEOF) { |
| 1981 |
$this->tokens[$i]['code'] = T_STRING; |
| 1982 |
$this->tokens[$i]['type'] = 'T_STRING'; |
| 1983 |
|
| 1984 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 1985 |
$line = $this->tokens[$i]['line']; |
| 1986 |
echo "\t* token $i on line $line changed from T_STATIC to T_STRING".PHP_EOL; |
| 1987 |
} |
| 1988 |
} |
| 1989 |
|
| 1990 |
continue; |
| 1991 |
} else if ($this->tokens[$i]['code'] === T_TRUE |
| 1992 |
|| $this->tokens[$i]['code'] === T_FALSE |
| 1993 |
|| $this->tokens[$i]['code'] === T_NULL |
| 1994 |
) { |
| 1995 |
for ($x = ($i + 1); $i < $numTokens; $x++) { |
| 1996 |
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
| 1997 |
// Non-whitespace content. |
| 1998 |
break; |
| 1999 |
} |
| 2000 |
} |
| 2001 |
|
| 2002 |
$context = [ |
| 2003 |
T_OBJECT_OPERATOR => true, |
| 2004 |
T_NS_SEPARATOR => true, |
| 2005 |
T_PAAMAYIM_NEKUDOTAYIM => true, |
| 2006 |
]; |
| 2007 |
if (isset($context[$this->tokens[$x]['code']]) === true) { |
| 2008 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 2009 |
$line = $this->tokens[$i]['line']; |
| 2010 |
$type = $this->tokens[$i]['type']; |
| 2011 |
echo "\t* token $i on line $line changed from $type to T_STRING".PHP_EOL; |
| 2012 |
} |
| 2013 |
|
| 2014 |
$this->tokens[$i]['code'] = T_STRING; |
| 2015 |
$this->tokens[$i]['type'] = 'T_STRING'; |
| 2016 |
} |
| 2017 |
} else if ($this->tokens[$i]['code'] === T_CONST) { |
| 2018 |
// Context sensitive keywords support. |
| 2019 |
for ($x = ($i + 1); $i < $numTokens; $x++) { |
| 2020 |
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
| 2021 |
// Non-whitespace content. |
| 2022 |
break; |
| 2023 |
} |
| 2024 |
} |
| 2025 |
|
| 2026 |
if ($this->tokens[$x]['code'] !== T_STRING) { |
| 2027 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 2028 |
$line = $this->tokens[$x]['line']; |
| 2029 |
$type = $this->tokens[$x]['type']; |
| 2030 |
echo "\t* token $x on line $line changed from $type to T_STRING".PHP_EOL; |
| 2031 |
} |
| 2032 |
|
| 2033 |
$this->tokens[$x]['code'] = T_STRING; |
| 2034 |
$this->tokens[$x]['type'] = 'T_STRING'; |
| 2035 |
} |
| 2036 |
}//end if |
| 2037 |
|
| 2038 |
if (($this->tokens[$i]['code'] !== T_CASE |
| 2039 |
&& $this->tokens[$i]['code'] !== T_DEFAULT) |
| 2040 |
|| isset($this->tokens[$i]['scope_opener']) === false |
| 2041 |
) { |
| 2042 |
// Only interested in CASE and DEFAULT statements from here on in. |
| 2043 |
continue; |
| 2044 |
} |
| 2045 |
|
| 2046 |
$scopeOpener = $this->tokens[$i]['scope_opener']; |
| 2047 |
$scopeCloser = $this->tokens[$i]['scope_closer']; |
| 2048 |
|
| 2049 |
// If the first char after the opener is a curly brace |
| 2050 |
// and that brace has been ignored, it is actually |
| 2051 |
// opening this case statement and the opener and closer are |
| 2052 |
// probably set incorrectly. |
| 2053 |
for ($x = ($scopeOpener + 1); $x < $numTokens; $x++) { |
| 2054 |
if (isset(Util\Tokens::$emptyTokens[$this->tokens[$x]['code']]) === false) { |
| 2055 |
// Non-whitespace content. |
| 2056 |
break; |
| 2057 |
} |
| 2058 |
} |
| 2059 |
|
| 2060 |
if ($this->tokens[$x]['code'] === T_CASE || $this->tokens[$x]['code'] === T_DEFAULT) { |
| 2061 |
// Special case for multiple CASE statements that share the same |
| 2062 |
// closer. Because we are going backwards through the file, this next |
| 2063 |
// CASE statement is already fixed, so just use its closer and don't |
| 2064 |
// worry about fixing anything. |
| 2065 |
$newCloser = $this->tokens[$x]['scope_closer']; |
| 2066 |
$this->tokens[$i]['scope_closer'] = $newCloser; |
| 2067 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 2068 |
$oldType = $this->tokens[$scopeCloser]['type']; |
| 2069 |
$newType = $this->tokens[$newCloser]['type']; |
| 2070 |
$line = $this->tokens[$i]['line']; |
| 2071 |
echo "\t* token $i (T_CASE) on line $line closer changed from $scopeCloser ($oldType) to $newCloser ($newType)".PHP_EOL; |
| 2072 |
} |
| 2073 |
|
| 2074 |
continue; |
| 2075 |
} |
| 2076 |
|
| 2077 |
if ($this->tokens[$x]['code'] !== T_OPEN_CURLY_BRACKET |
| 2078 |
|| isset($this->tokens[$x]['scope_condition']) === true |
| 2079 |
) { |
| 2080 |
// Not a CASE/DEFAULT with a curly brace opener. |
| 2081 |
continue; |
| 2082 |
} |
| 2083 |
|
| 2084 |
// The closer for this CASE/DEFAULT should be the closing curly brace and |
| 2085 |
// not whatever it already is. The opener needs to be the opening curly |
| 2086 |
// brace so everything matches up. |
| 2087 |
$newCloser = $this->tokens[$x]['bracket_closer']; |
| 2088 |
foreach ([$i, $x, $newCloser] as $index) { |
| 2089 |
$this->tokens[$index]['scope_condition'] = $i; |
| 2090 |
$this->tokens[$index]['scope_opener'] = $x; |
| 2091 |
$this->tokens[$index]['scope_closer'] = $newCloser; |
| 2092 |
} |
| 2093 |
|
| 2094 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 2095 |
$line = $this->tokens[$i]['line']; |
| 2096 |
$tokenType = $this->tokens[$i]['type']; |
| 2097 |
|
| 2098 |
$oldType = $this->tokens[$scopeOpener]['type']; |
| 2099 |
$newType = $this->tokens[$x]['type']; |
| 2100 |
echo "\t* token $i ($tokenType) on line $line opener changed from $scopeOpener ($oldType) to $x ($newType)".PHP_EOL; |
| 2101 |
|
| 2102 |
$oldType = $this->tokens[$scopeCloser]['type']; |
| 2103 |
$newType = $this->tokens[$newCloser]['type']; |
| 2104 |
echo "\t* token $i ($tokenType) on line $line closer changed from $scopeCloser ($oldType) to $newCloser ($newType)".PHP_EOL; |
| 2105 |
} |
| 2106 |
|
| 2107 |
if ($this->tokens[$scopeOpener]['scope_condition'] === $i) { |
| 2108 |
unset($this->tokens[$scopeOpener]['scope_condition']); |
| 2109 |
unset($this->tokens[$scopeOpener]['scope_opener']); |
| 2110 |
unset($this->tokens[$scopeOpener]['scope_closer']); |
| 2111 |
} |
| 2112 |
|
| 2113 |
if ($this->tokens[$scopeCloser]['scope_condition'] === $i) { |
| 2114 |
unset($this->tokens[$scopeCloser]['scope_condition']); |
| 2115 |
unset($this->tokens[$scopeCloser]['scope_opener']); |
| 2116 |
unset($this->tokens[$scopeCloser]['scope_closer']); |
| 2117 |
} else { |
| 2118 |
// We were using a shared closer. All tokens that were |
| 2119 |
// sharing this closer with us, except for the scope condition |
| 2120 |
// and it's opener, need to now point to the new closer. |
| 2121 |
$condition = $this->tokens[$scopeCloser]['scope_condition']; |
| 2122 |
$start = ($this->tokens[$condition]['scope_opener'] + 1); |
| 2123 |
for ($y = $start; $y < $scopeCloser; $y++) { |
| 2124 |
if (isset($this->tokens[$y]['scope_closer']) === true |
| 2125 |
&& $this->tokens[$y]['scope_closer'] === $scopeCloser |
| 2126 |
) { |
| 2127 |
$this->tokens[$y]['scope_closer'] = $newCloser; |
| 2128 |
|
| 2129 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 2130 |
$line = $this->tokens[$y]['line']; |
| 2131 |
$tokenType = $this->tokens[$y]['type']; |
| 2132 |
$oldType = $this->tokens[$scopeCloser]['type']; |
| 2133 |
$newType = $this->tokens[$newCloser]['type']; |
| 2134 |
echo "\t\t* token $y ($tokenType) on line $line closer changed from $scopeCloser ($oldType) to $newCloser ($newType)".PHP_EOL; |
| 2135 |
} |
| 2136 |
} |
| 2137 |
} |
| 2138 |
}//end if |
| 2139 |
|
| 2140 |
unset($this->tokens[$x]['bracket_opener']); |
| 2141 |
unset($this->tokens[$x]['bracket_closer']); |
| 2142 |
unset($this->tokens[$newCloser]['bracket_opener']); |
| 2143 |
unset($this->tokens[$newCloser]['bracket_closer']); |
| 2144 |
$this->tokens[$scopeCloser]['conditions'][] = $i; |
| 2145 |
|
| 2146 |
// Now fix up all the tokens that think they are |
| 2147 |
// inside the CASE/DEFAULT statement when they are really outside. |
| 2148 |
for ($x = $newCloser; $x < $scopeCloser; $x++) { |
| 2149 |
foreach ($this->tokens[$x]['conditions'] as $num => $oldCond) { |
| 2150 |
if ($oldCond === $this->tokens[$i]['code']) { |
| 2151 |
$oldConditions = $this->tokens[$x]['conditions']; |
| 2152 |
unset($this->tokens[$x]['conditions'][$num]); |
| 2153 |
|
| 2154 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 2155 |
$type = $this->tokens[$x]['type']; |
| 2156 |
$oldConds = ''; |
| 2157 |
foreach ($oldConditions as $condition) { |
| 2158 |
$oldConds .= Util\Tokens::tokenName($condition).','; |
| 2159 |
} |
| 2160 |
|
| 2161 |
$oldConds = rtrim($oldConds, ','); |
| 2162 |
|
| 2163 |
$newConds = ''; |
| 2164 |
foreach ($this->tokens[$x]['conditions'] as $condition) { |
| 2165 |
$newConds .= Util\Tokens::tokenName($condition).','; |
| 2166 |
} |
| 2167 |
|
| 2168 |
$newConds = rtrim($newConds, ','); |
| 2169 |
|
| 2170 |
echo "\t\t* cleaned $x ($type) *".PHP_EOL; |
| 2171 |
echo "\t\t\t=> conditions changed from $oldConds to $newConds".PHP_EOL; |
| 2172 |
} |
| 2173 |
|
| 2174 |
break; |
| 2175 |
}//end if |
| 2176 |
}//end foreach |
| 2177 |
}//end for |
| 2178 |
}//end for |
| 2179 |
|
| 2180 |
if (PHP_CODESNIFFER_VERBOSITY > 1) { |
| 2181 |
echo "\t*** END ADDITIONAL PHP PROCESSING ***".PHP_EOL; |
| 2182 |
} |
| 2183 |
|
| 2184 |
}//end processAdditional() |
| 2185 |
|
| 2186 |
|
| 2187 |
/** |
| 2188 |
* Takes a token produced from <code>token_get_all()</code> and produces a |
| 2189 |
* more uniform token. |
| 2190 |
* |
| 2191 |
* @param string|array $token The token to convert. |
| 2192 |
* |
| 2193 |
* @return array The new token. |
| 2194 |
*/ |
| 2195 |
public static function standardiseToken($token) |
| 2196 |
{ |
| 2197 |
if (isset($token[1]) === false) { |
| 2198 |
if (isset(self::$resolveTokenCache[$token[0]]) === true) { |
| 2199 |
return self::$resolveTokenCache[$token[0]]; |
| 2200 |
} |
| 2201 |
} else { |
| 2202 |
$cacheKey = null; |
| 2203 |
if ($token[0] === T_STRING) { |
| 2204 |
$cacheKey = strtolower($token[1]); |
| 2205 |
} else if ($token[0] !== T_CURLY_OPEN) { |
| 2206 |
$cacheKey = $token[0]; |
| 2207 |
} |
| 2208 |
|
| 2209 |
if ($cacheKey !== null && isset(self::$resolveTokenCache[$cacheKey]) === true) { |
| 2210 |
$newToken = self::$resolveTokenCache[$cacheKey]; |
| 2211 |
$newToken['content'] = $token[1]; |
| 2212 |
return $newToken; |
| 2213 |
} |
| 2214 |
} |
| 2215 |
|
| 2216 |
if (isset($token[1]) === false) { |
| 2217 |
return self::resolveSimpleToken($token[0]); |
| 2218 |
} |
| 2219 |
|
| 2220 |
if ($token[0] === T_STRING) { |
| 2221 |
switch ($cacheKey) { |
| 2222 |
case 'false': |
| 2223 |
$newToken['type'] = 'T_FALSE'; |
| 2224 |
break; |
| 2225 |
case 'true': |
| 2226 |
$newToken['type'] = 'T_TRUE'; |
| 2227 |
break; |
| 2228 |
case 'null': |
| 2229 |
$newToken['type'] = 'T_NULL'; |
| 2230 |
break; |
| 2231 |
case 'self': |
| 2232 |
$newToken['type'] = 'T_SELF'; |
| 2233 |
break; |
| 2234 |
case 'parent': |
| 2235 |
$newToken['type'] = 'T_PARENT'; |
| 2236 |
break; |
| 2237 |
default: |
| 2238 |
$newToken['type'] = 'T_STRING'; |
| 2239 |
break; |
| 2240 |
} |
| 2241 |
|
| 2242 |
$newToken['code'] = constant($newToken['type']); |
| 2243 |
|
| 2244 |
self::$resolveTokenCache[$cacheKey] = $newToken; |
| 2245 |
} else if ($token[0] === T_CURLY_OPEN) { |
| 2246 |
$newToken = [ |
| 2247 |
'code' => T_OPEN_CURLY_BRACKET, |
| 2248 |
'type' => 'T_OPEN_CURLY_BRACKET', |
| 2249 |
]; |
| 2250 |
} else { |
| 2251 |
$newToken = [ |
| 2252 |
'code' => $token[0], |
| 2253 |
'type' => Util\Tokens::tokenName($token[0]), |
| 2254 |
]; |
| 2255 |
|
| 2256 |
self::$resolveTokenCache[$token[0]] = $newToken; |
| 2257 |
}//end if |
| 2258 |
|
| 2259 |
$newToken['content'] = $token[1]; |
| 2260 |
return $newToken; |
| 2261 |
|
| 2262 |
}//end standardiseToken() |
| 2263 |
|
| 2264 |
|
| 2265 |
/** |
| 2266 |
* Converts simple tokens into a format that conforms to complex tokens |
| 2267 |
* produced by token_get_all(). |
| 2268 |
* |
| 2269 |
* Simple tokens are tokens that are not in array form when produced from |
| 2270 |
* token_get_all(). |
| 2271 |
* |
| 2272 |
* @param string $token The simple token to convert. |
| 2273 |
* |
| 2274 |
* @return array The new token in array format. |
| 2275 |
*/ |
| 2276 |
public static function resolveSimpleToken($token) |
| 2277 |
{ |
| 2278 |
$newToken = []; |
| 2279 |
|
| 2280 |
switch ($token) { |
| 2281 |
case '{': |
| 2282 |
$newToken['type'] = 'T_OPEN_CURLY_BRACKET'; |
| 2283 |
break; |
| 2284 |
case '}': |
| 2285 |
$newToken['type'] = 'T_CLOSE_CURLY_BRACKET'; |
| 2286 |
break; |
| 2287 |
case '[': |
| 2288 |
$newToken['type'] = 'T_OPEN_SQUARE_BRACKET'; |
| 2289 |
break; |
| 2290 |
case ']': |
| 2291 |
$newToken['type'] = 'T_CLOSE_SQUARE_BRACKET'; |
| 2292 |
break; |
| 2293 |
case '(': |
| 2294 |
$newToken['type'] = 'T_OPEN_PARENTHESIS'; |
| 2295 |
break; |
| 2296 |
case ')': |
| 2297 |
$newToken['type'] = 'T_CLOSE_PARENTHESIS'; |
| 2298 |
break; |
| 2299 |
case ':': |
| 2300 |
$newToken['type'] = 'T_COLON'; |
| 2301 |
break; |
| 2302 |
case '.': |
| 2303 |
$newToken['type'] = 'T_STRING_CONCAT'; |
| 2304 |
break; |
| 2305 |
case ';': |
| 2306 |
$newToken['type'] = 'T_SEMICOLON'; |
| 2307 |
break; |
| 2308 |
case '=': |
| 2309 |
$newToken['type'] = 'T_EQUAL'; |
| 2310 |
break; |
| 2311 |
case '*': |
| 2312 |
$newToken['type'] = 'T_MULTIPLY'; |
| 2313 |
break; |
| 2314 |
case '/': |
| 2315 |
$newToken['type'] = 'T_DIVIDE'; |
| 2316 |
break; |
| 2317 |
case '+': |
| 2318 |
$newToken['type'] = 'T_PLUS'; |
| 2319 |
break; |
| 2320 |
case '-': |
| 2321 |
$newToken['type'] = 'T_MINUS'; |
| 2322 |
break; |
| 2323 |
case '%': |
| 2324 |
$newToken['type'] = 'T_MODULUS'; |
| 2325 |
break; |
| 2326 |
case '^': |
| 2327 |
$newToken['type'] = 'T_BITWISE_XOR'; |
| 2328 |
break; |
| 2329 |
case '&': |
| 2330 |
$newToken['type'] = 'T_BITWISE_AND'; |
| 2331 |
break; |
| 2332 |
case '|': |
| 2333 |
$newToken['type'] = 'T_BITWISE_OR'; |
| 2334 |
break; |
| 2335 |
case '~': |
| 2336 |
$newToken['type'] = 'T_BITWISE_NOT'; |
| 2337 |
break; |
| 2338 |
case '<': |
| 2339 |
$newToken['type'] = 'T_LESS_THAN'; |
| 2340 |
break; |
| 2341 |
case '>': |
| 2342 |
$newToken['type'] = 'T_GREATER_THAN'; |
| 2343 |
break; |
| 2344 |
case '!': |
| 2345 |
$newToken['type'] = 'T_BOOLEAN_NOT'; |
| 2346 |
break; |
| 2347 |
case ',': |
| 2348 |
$newToken['type'] = 'T_COMMA'; |
| 2349 |
break; |
| 2350 |
case '@': |
| 2351 |
$newToken['type'] = 'T_ASPERAND'; |
| 2352 |
break; |
| 2353 |
case '$': |
| 2354 |
$newToken['type'] = 'T_DOLLAR'; |
| 2355 |
break; |
| 2356 |
case '`': |
| 2357 |
$newToken['type'] = 'T_BACKTICK'; |
| 2358 |
break; |
| 2359 |
default: |
| 2360 |
$newToken['type'] = 'T_NONE'; |
| 2361 |
break; |
| 2362 |
}//end switch |
| 2363 |
|
| 2364 |
$newToken['code'] = constant($newToken['type']); |
| 2365 |
$newToken['content'] = $token; |
| 2366 |
|
| 2367 |
self::$resolveTokenCache[$token] = $newToken; |
| 2368 |
return $newToken; |
| 2369 |
|
| 2370 |
}//end resolveSimpleToken() |
| 2371 |
|
| 2372 |
|
| 2373 |
}//end class |
| 2374 |
|
