digiaonline /
graphql-php
| Total Complexity | 112 |
| Total Lines | 711 |
| Duplicated Lines | 0 % |
| Changes | 0 | ||
Complex classes like Lexer often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use Lexer, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 7 | class Lexer implements LexerInterface |
||
| 8 | { |
||
| 9 | protected const ENCODING = 'UTF-8'; |
||
| 10 | |||
| 11 | /** |
||
| 12 | * A map between punctuation character code and the corresponding token kind. |
||
| 13 | * |
||
| 14 | * @var array |
||
| 15 | */ |
||
| 16 | protected static $codeTokenKindMap = [ |
||
| 17 | 33 => TokenKindEnum::BANG, |
||
| 18 | 36 => TokenKindEnum::DOLLAR, |
||
| 19 | 38 => TokenKindEnum::AMP, |
||
| 20 | 40 => TokenKindEnum::PAREN_L, |
||
| 21 | 41 => TokenKindEnum::PAREN_R, |
||
| 22 | 58 => TokenKindEnum::COLON, |
||
| 23 | 61 => TokenKindEnum::EQUALS, |
||
| 24 | 64 => TokenKindEnum::AT, |
||
| 25 | 91 => TokenKindEnum::BRACKET_L, |
||
| 26 | 93 => TokenKindEnum::BRACKET_R, |
||
| 27 | 123 => TokenKindEnum::BRACE_L, |
||
| 28 | 124 => TokenKindEnum::PIPE, |
||
| 29 | 125 => TokenKindEnum::BRACE_R, |
||
| 30 | ]; |
||
| 31 | |||
| 32 | /** |
||
| 33 | * The source file for this lexer. |
||
| 34 | * |
||
| 35 | * @var Source |
||
| 36 | */ |
||
| 37 | protected $source; |
||
| 38 | |||
| 39 | /** |
||
| 40 | * The contents of the source file. |
||
| 41 | * |
||
| 42 | * @var string |
||
| 43 | */ |
||
| 44 | protected $body; |
||
| 45 | |||
| 46 | /** |
||
| 47 | * The total number of characters in the source file. |
||
| 48 | * |
||
| 49 | * @var int |
||
| 50 | */ |
||
| 51 | protected $bodyLength; |
||
| 52 | |||
| 53 | /** |
||
| 54 | * The options for this lexer. |
||
| 55 | * |
||
| 56 | * @var array |
||
| 57 | */ |
||
| 58 | protected $options = []; |
||
| 59 | |||
| 60 | /** |
||
| 61 | * The previously focused non-ignored token. |
||
| 62 | * |
||
| 63 | * @var Token |
||
| 64 | */ |
||
| 65 | protected $lastToken; |
||
| 66 | |||
| 67 | /** |
||
| 68 | * The currently focused non-ignored token. |
||
| 69 | * |
||
| 70 | * @var Token |
||
| 71 | */ |
||
| 72 | protected $token; |
||
| 73 | |||
| 74 | /** |
||
| 75 | * The current position. |
||
| 76 | * |
||
| 77 | * @var int |
||
| 78 | */ |
||
| 79 | protected $pos; |
||
| 80 | |||
| 81 | /** |
||
| 82 | * The (1-indexed) line containing the current token. |
||
| 83 | * |
||
| 84 | * @var int |
||
| 85 | */ |
||
| 86 | protected $line; |
||
| 87 | |||
| 88 | /** |
||
| 89 | * The character offset at which the current line begins. |
||
| 90 | * |
||
| 91 | * @var int |
||
| 92 | */ |
||
| 93 | protected $lineStart; |
||
| 94 | |||
| 95 | /** |
||
| 96 | * @var array |
||
| 97 | */ |
||
| 98 | protected static $charCodeCache = []; |
||
| 99 | |||
| 100 | /** |
||
| 101 | * Lexer constructor. |
||
| 102 | * @param Source|null $source |
||
| 103 | * @param array $options |
||
| 104 | */ |
||
| 105 | public function __construct(Source $source, array $options) |
||
| 106 | { |
||
| 107 | $startOfFileToken = $this->createStartOfFileToken(); |
||
| 108 | |||
| 109 | $this->lastToken = $startOfFileToken; |
||
| 110 | $this->token = $startOfFileToken; |
||
| 111 | $this->line = 1; |
||
| 112 | $this->lineStart = 0; |
||
| 113 | $this->body = $source->getBody(); |
||
| 114 | $this->bodyLength = \strlen($this->body); |
||
| 115 | $this->source = $source; |
||
| 116 | $this->options = $options; |
||
| 117 | } |
||
| 118 | |||
| 119 | /** |
||
| 120 | * @inheritdoc |
||
| 121 | * @throws SyntaxErrorException |
||
| 122 | */ |
||
| 123 | public function advance(): Token |
||
| 124 | { |
||
| 125 | $this->lastToken = $this->token; |
||
| 126 | return $this->token = $this->lookahead(); |
||
| 127 | } |
||
| 128 | |||
| 129 | /** |
||
| 130 | * @inheritdoc |
||
| 131 | * @throws SyntaxErrorException |
||
| 132 | */ |
||
| 133 | public function lookahead(): Token |
||
| 134 | { |
||
| 135 | $token = $this->token; |
||
| 136 | |||
| 137 | if (TokenKindEnum::EOF !== $token->getKind()) { |
||
| 138 | do { |
||
| 139 | $next = $this->readToken($token); |
||
| 140 | $token->setNext($next); |
||
| 141 | $token = $next; |
||
| 142 | } while (TokenKindEnum::COMMENT === $token->getKind()); |
||
| 143 | } |
||
| 144 | |||
| 145 | return $token; |
||
| 146 | } |
||
| 147 | |||
| 148 | /** |
||
| 149 | * @inheritdoc |
||
| 150 | */ |
||
| 151 | public function getOption(string $name, $default = null) |
||
| 152 | { |
||
| 153 | return $this->options[$name] ?? $default; |
||
| 154 | } |
||
| 155 | |||
| 156 | /** |
||
| 157 | * @inheritdoc |
||
| 158 | */ |
||
| 159 | public function getTokenKind(): string |
||
| 160 | { |
||
| 161 | return $this->token->getKind(); |
||
| 162 | } |
||
| 163 | |||
| 164 | /** |
||
| 165 | * @inheritdoc |
||
| 166 | */ |
||
| 167 | public function getTokenValue(): ?string |
||
| 168 | { |
||
| 169 | return $this->token->getValue(); |
||
| 170 | } |
||
| 171 | |||
| 172 | /** |
||
| 173 | * @inheritdoc |
||
| 174 | */ |
||
| 175 | public function getToken(): Token |
||
| 176 | { |
||
| 177 | return $this->token; |
||
| 178 | } |
||
| 179 | |||
| 180 | /** |
||
| 181 | * @inheritdoc |
||
| 182 | */ |
||
| 183 | public function getSource(): Source |
||
| 184 | { |
||
| 185 | return $this->source; |
||
| 186 | } |
||
| 187 | |||
| 188 | /** |
||
| 189 | * @inheritdoc |
||
| 190 | */ |
||
| 191 | public function getLastToken(): Token |
||
| 192 | { |
||
| 193 | return $this->lastToken; |
||
| 194 | } |
||
| 195 | |||
| 196 | /** |
||
| 197 | * Reads the token after the given token. |
||
| 198 | * |
||
| 199 | * @param Token $prev |
||
| 200 | * @return Token |
||
| 201 | * @throws SyntaxErrorException |
||
| 202 | */ |
||
| 203 | protected function readToken(Token $prev): Token |
||
| 204 | { |
||
| 205 | $this->pos = $prev->getEnd(); |
||
| 206 | |||
| 207 | $this->skipWhitespace(); |
||
| 208 | |||
| 209 | $line = $this->line; |
||
| 210 | $col = (1 + $this->pos) - $this->lineStart; |
||
| 211 | |||
| 212 | if ($this->pos >= $this->bodyLength) { |
||
| 213 | return $this->createEndOfFileToken($line, $col, $prev); |
||
| 214 | } |
||
| 215 | |||
| 216 | $code = $this->readCharCode($this->pos); |
||
| 217 | |||
| 218 | // Punctuation: [!$&:=@|()\[\]{}]{1} |
||
| 219 | if (33 === $code || 36 === $code || 38 === $code || 58 === $code || 61 === $code || 64 === $code || 124 === $code || |
||
| 220 | 40 === $code || 41 === $code || 91 === $code || 93 === $code || 123 === $code || 125 === $code) { |
||
| 221 | return $this->lexPunctuation($code, $line, $col, $prev); |
||
| 222 | } |
||
| 223 | |||
| 224 | // Comment: #[\u0009\u0020-\uFFFF]* |
||
| 225 | if (35 === $code) { |
||
| 226 | return $this->lexComment($line, $col, $prev); |
||
| 227 | } |
||
| 228 | |||
| 229 | // Int: -?(0|[1-9][0-9]*) |
||
| 230 | // Float: -?(0|[1-9][0-9]*)(\.[0-9]+)?((E|e)(+|-)?[0-9]+)? |
||
| 231 | if (45 === $code || isNumber($code)) { |
||
| 232 | return $this->lexNumber($code, $line, $col, $prev); |
||
| 233 | } |
||
| 234 | |||
| 235 | // Name: [_A-Za-z][_0-9A-Za-z]* |
||
| 236 | if (isAlphaNumeric($code)) { |
||
| 237 | return $this->lexName($line, $col, $prev); |
||
| 238 | } |
||
| 239 | |||
| 240 | // Spread: ... |
||
| 241 | if ($this->bodyLength >= 3 && $this->isSpread($code)) { |
||
| 242 | return $this->lexSpread($line, $col, $prev); |
||
| 243 | } |
||
| 244 | |||
| 245 | // String: "([^"\\\u000A\u000D]|(\\(u[0-9a-fA-F]{4}|["\\/bfnrt])))*" |
||
| 246 | if ($this->isString($code)) { |
||
| 247 | return $this->lexString($line, $col, $prev); |
||
| 248 | } |
||
| 249 | |||
| 250 | // Block String: """("?"?(\\"""|\\(?!=""")|[^"\\]))*""" |
||
| 251 | if ($this->bodyLength >= 3 && $this->isTripleQuote($code)) { |
||
| 252 | return $this->lexBlockString($line, $col, $prev); |
||
| 253 | } |
||
| 254 | |||
| 255 | throw $this->createSyntaxErrorException(); |
||
| 256 | } |
||
| 257 | |||
| 258 | /** |
||
| 259 | * @return Token |
||
| 260 | */ |
||
| 261 | protected function createStartOfFileToken(): Token |
||
| 264 | } |
||
| 265 | |||
| 266 | /** |
||
| 267 | * Creates an End Of File (EOF) token. |
||
| 268 | * |
||
| 269 | * @param int $line |
||
| 270 | * @param int $col |
||
| 271 | * @param Token $prev |
||
| 272 | * @return Token |
||
| 273 | */ |
||
| 274 | protected function createEndOfFileToken(int $line, int $col, Token $prev): Token |
||
| 275 | { |
||
| 276 | return new Token(TokenKindEnum::EOF, $this->bodyLength, $this->bodyLength, $line, $col, $prev); |
||
| 277 | } |
||
| 278 | |||
| 279 | /** |
||
| 280 | * Reads a punctuation token from the source file. |
||
| 281 | * |
||
| 282 | * @param int $code |
||
| 283 | * @param int $line |
||
| 284 | * @param int $col |
||
| 285 | * @param Token $prev |
||
| 286 | * @return Token |
||
| 287 | * @throws SyntaxErrorException |
||
| 288 | */ |
||
| 289 | protected function lexPunctuation(int $code, int $line, int $col, Token $prev): ?Token |
||
| 296 | } |
||
| 297 | |||
| 298 | /** |
||
| 299 | * Reads a name token from the source file. |
||
| 300 | * |
||
| 301 | * @param int $line |
||
| 302 | * @param int $col |
||
| 303 | * @param Token $prev |
||
| 304 | * @return Token |
||
| 305 | */ |
||
| 306 | protected function lexName(int $line, int $col, Token $prev): Token |
||
| 321 | } |
||
| 322 | |||
| 323 | /** |
||
| 324 | * Reads a number (int or float) token from the source file. |
||
| 325 | * |
||
| 326 | * @param int $code |
||
| 327 | * @param int $line |
||
| 328 | * @param int $col |
||
| 329 | * @param Token $prev |
||
| 330 | * @return Token |
||
| 331 | * @throws SyntaxErrorException |
||
| 332 | */ |
||
| 333 | protected function lexNumber(int $code, int $line, int $col, Token $prev): Token |
||
| 388 | ); |
||
| 389 | } |
||
| 390 | |||
| 391 | /** |
||
| 392 | * Skips digits at the current position. |
||
| 393 | * |
||
| 394 | * @param int $code |
||
| 395 | * @throws SyntaxErrorException |
||
| 396 | */ |
||
| 397 | protected function skipDigits(int $code): void |
||
| 398 | { |
||
| 399 | if (isNumber($code)) { |
||
| 400 | do { |
||
| 401 | $code = $this->readCharCode(++$this->pos); |
||
| 402 | } while (isNumber($code)); |
||
| 403 | |||
| 404 | return; |
||
| 405 | } |
||
| 406 | |||
| 407 | throw $this->createSyntaxErrorException( |
||
| 408 | \sprintf('Invalid number, expected digit but got: %s.', printCharCode($code)) |
||
| 409 | ); |
||
| 410 | } |
||
| 411 | |||
| 412 | /** |
||
| 413 | * Reads a comment token from the source file. |
||
| 414 | * |
||
| 415 | * @param int $line |
||
| 416 | * @param int $col |
||
| 417 | * @param Token $prev |
||
| 418 | * @return Token |
||
| 419 | */ |
||
| 420 | protected function lexComment(int $line, int $col, Token $prev): Token |
||
| 421 | { |
||
| 422 | $start = $this->pos; |
||
| 423 | |||
| 424 | do { |
||
| 425 | $code = $this->readCharCode(++$this->pos); |
||
| 426 | } while ($code !== null && ($code > 0x001f || 0x0009 === $code)); // SourceCharacter but not LineTerminator |
||
| 427 | |||
| 428 | return new Token( |
||
| 429 | TokenKindEnum::COMMENT, |
||
| 430 | $start, |
||
| 431 | $this->pos, |
||
| 432 | $line, |
||
| 433 | $col, |
||
| 434 | $prev, |
||
| 435 | sliceString($this->body, $start + 1, $this->pos) |
||
| 436 | ); |
||
| 437 | } |
||
| 438 | |||
| 439 | /** |
||
| 440 | * Reads a spread token from the source. |
||
| 441 | * |
||
| 442 | * @param int $line |
||
| 443 | * @param int $col |
||
| 444 | * @param Token $prev |
||
| 445 | * @return Token |
||
| 446 | */ |
||
| 447 | protected function lexSpread(int $line, int $col, Token $prev): Token |
||
| 448 | { |
||
| 449 | return new Token(TokenKindEnum::SPREAD, $this->pos, $this->pos + 3, $line, $col, $prev); |
||
| 450 | } |
||
| 451 | |||
| 452 | /** |
||
| 453 | * Reads a string token from the source. |
||
| 454 | * |
||
| 455 | * @param int $line |
||
| 456 | * @param int $col |
||
| 457 | * @param Token $prev |
||
| 458 | * @return Token |
||
| 459 | * @throws SyntaxErrorException |
||
| 460 | */ |
||
| 461 | protected function lexString(int $line, int $col, Token $prev): Token |
||
| 462 | { |
||
| 463 | $start = $this->pos; |
||
| 464 | $chunkStart = ++$this->pos; // skip the quote |
||
| 465 | $value = ''; |
||
| 466 | |||
| 467 | while ($this->pos < $this->bodyLength && |
||
| 468 | ($code = $this->readCharCode($this->pos)) !== null && !isLineTerminator($code)) { |
||
| 469 | // Closing Quote (") |
||
| 470 | if (34 === $code) { |
||
| 471 | $value .= sliceString($this->body, $chunkStart, $this->pos); |
||
| 472 | return new Token(TokenKindEnum::STRING, $start, $this->pos + 1, $line, $col, $prev, $value); |
||
| 473 | } |
||
| 474 | |||
| 475 | if (isSourceCharacter($code)) { |
||
| 476 | throw $this->createSyntaxErrorException( |
||
| 477 | \sprintf('Invalid character within String: %s.', printCharCode($code)) |
||
| 478 | ); |
||
| 479 | } |
||
| 480 | |||
| 481 | ++$this->pos; |
||
| 482 | |||
| 483 | if (92 === $code) { |
||
| 484 | // \ |
||
| 485 | $value .= sliceString($this->body, $chunkStart, $this->pos - 1); |
||
| 486 | |||
| 487 | $code = $this->readCharCode($this->pos); |
||
| 488 | |||
| 489 | switch ($code) { |
||
| 490 | case 34: // " |
||
| 491 | $value .= '"'; |
||
| 492 | break; |
||
| 493 | case 47: // / |
||
| 494 | $value .= '/'; |
||
| 495 | break; |
||
| 496 | case 92: // \ |
||
| 497 | $value .= '\\'; |
||
| 498 | break; |
||
| 499 | case 98: // b |
||
| 500 | $value .= '\b'; |
||
| 501 | break; |
||
| 502 | case 102: // f |
||
| 503 | $value .= '\f'; |
||
| 504 | break; |
||
| 505 | case 110: // n |
||
| 506 | $value .= '\n'; |
||
| 507 | break; |
||
| 508 | case 114: // r |
||
| 509 | $value .= '\r'; |
||
| 510 | break; |
||
| 511 | case 116: // t |
||
| 512 | $value .= '\t'; |
||
| 513 | break; |
||
| 514 | case 117: // u |
||
| 515 | $unicodeString = sliceString($this->body, $this->pos + 1, $this->pos + 5); |
||
| 516 | |||
| 517 | if (!\preg_match('/[0-9A-Fa-f]{4}/', $unicodeString)) { |
||
| 518 | throw $this->createSyntaxErrorException( |
||
| 519 | \sprintf('Invalid character escape sequence: \\u%s.', $unicodeString) |
||
| 520 | ); |
||
| 521 | } |
||
| 522 | |||
| 523 | $value .= '\\u' . $unicodeString; |
||
| 524 | $this->pos += 4; |
||
| 525 | |||
| 526 | break; |
||
| 527 | default: |
||
| 528 | throw $this->createSyntaxErrorException( |
||
| 529 | \sprintf('Invalid character escape sequence: \\%s.', \chr($code)) |
||
| 530 | ); |
||
| 531 | } |
||
| 532 | |||
| 533 | ++$this->pos; |
||
| 534 | |||
| 535 | $chunkStart = $this->pos; |
||
| 536 | } |
||
| 537 | } |
||
| 538 | |||
| 539 | throw $this->createSyntaxErrorException('Unterminated string.'); |
||
| 540 | } |
||
| 541 | |||
| 542 | /** |
||
| 543 | * Reads a block string token from the source file. |
||
| 544 | * |
||
| 545 | * @param int $line |
||
| 546 | * @param int $col |
||
| 547 | * @param Token $prev |
||
| 548 | * @return Token |
||
| 549 | * @throws SyntaxErrorException |
||
| 550 | */ |
||
| 551 | protected function lexBlockString(int $line, int $col, Token $prev): Token |
||
| 589 | } |
||
| 590 | |||
| 591 | /** |
||
| 592 | * Skips whitespace at the current position. |
||
| 593 | */ |
||
| 594 | protected function skipWhitespace(): void |
||
| 619 | } |
||
| 620 | } |
||
| 621 | } |
||
| 622 | |||
| 623 | /** |
||
| 624 | * @param int $pos |
||
| 625 | * @return int |
||
| 626 | */ |
||
| 627 | protected function readCharCode(int $pos): int |
||
| 628 | { |
||
| 629 | $char = \mb_substr($this->body, $pos, 1, self::ENCODING); |
||
| 630 | |||
| 631 | if ('' === $char) { |
||
| 632 | return 0; |
||
| 633 | } |
||
| 634 | |||
| 635 | if (!isset(self::$charCodeCache[$char])) { |
||
| 636 | self::$charCodeCache[$char] = \mb_ord($char, self::ENCODING); |
||
| 637 | } |
||
| 638 | |||
| 639 | return self::$charCodeCache[$char]; |
||
| 640 | } |
||
| 641 | |||
| 642 | /** |
||
| 643 | * Creates a `SyntaxErrorException` for the current position in the source. |
||
| 644 | * |
||
| 645 | * @param null|string $description |
||
| 646 | * @return SyntaxErrorException |
||
| 647 | */ |
||
| 648 | protected function createSyntaxErrorException(?string $description = null): SyntaxErrorException |
||
| 649 | { |
||
| 650 | return new SyntaxErrorException( |
||
| 651 | $this->source, |
||
| 652 | $this->pos, |
||
| 653 | $description ?? $this->unexpectedCharacterMessage($this->readCharCode($this->pos)) |
||
| 654 | ); |
||
| 655 | } |
||
| 656 | |||
| 657 | /** |
||
| 658 | * Report a message that an unexpected character was encountered. |
||
| 659 | * |
||
| 660 | * @param int $code |
||
| 661 | * @return string |
||
| 662 | */ |
||
| 663 | protected function unexpectedCharacterMessage(int $code): string |
||
| 664 | { |
||
| 665 | if (isSourceCharacter($code) && !isLineTerminator($code)) { |
||
| 666 | return \sprintf('Cannot contain the invalid character %s.', printCharCode($code)); |
||
| 667 | } |
||
| 668 | |||
| 669 | if ($code === 39) { |
||
| 670 | // ' |
||
| 671 | return 'Unexpected single quote character (\'), did you mean to use a double quote (")?'; |
||
| 672 | } |
||
| 673 | |||
| 674 | return \sprintf('Cannot parse the unexpected character %s.', printCharCode($code)); |
||
| 675 | } |
||
| 676 | |||
| 677 | /** |
||
| 678 | * @param int $code |
||
| 679 | * @return bool |
||
| 680 | */ |
||
| 681 | protected function isSpread(int $code): bool |
||
| 682 | { |
||
| 683 | return 46 === $code && |
||
| 684 | $this->readCharCode($this->pos + 1) === 46 && |
||
| 685 | $this->readCharCode($this->pos + 2) === 46; // ... |
||
| 686 | } |
||
| 687 | |||
| 688 | /** |
||
| 689 | * @param int $code |
||
| 690 | * @return bool |
||
| 691 | */ |
||
| 692 | protected function isString(int $code): bool |
||
| 693 | { |
||
| 694 | return 34 === $code && $this->readCharCode($this->pos + 1) !== 34; |
||
| 695 | } |
||
| 696 | |||
| 697 | /** |
||
| 698 | * @param int $code |
||
| 699 | * @return bool |
||
| 700 | */ |
||
| 701 | protected function isTripleQuote(int $code): bool |
||
| 702 | { |
||
| 703 | return 34 === $code && |
||
| 704 | 34 === $this->readCharCode($this->pos + 1) && |
||
| 705 | 34 === $this->readCharCode($this->pos + 2); // """ |
||
| 706 | } |
||
| 707 | |||
| 708 | /** |
||
| 709 | * @param int $code |
||
| 710 | * @return bool |
||
| 711 | */ |
||
| 712 | protected function isEscapedTripleQuote(int $code): bool |
||
| 718 | } |
||
| 719 | } |
||
| 720 |
