Total Complexity | 112 |
Total Lines | 708 |
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 | * @inheritdoc |
||
198 | */ |
||
199 | public function createSyntaxErrorException(?string $description = null): SyntaxErrorException |
||
200 | { |
||
201 | return new SyntaxErrorException( |
||
202 | $this->source, |
||
203 | $this->pos, |
||
204 | $description ?? $this->unexpectedCharacterMessage($this->readCharCode($this->pos)) |
||
205 | ); |
||
206 | } |
||
207 | |||
208 | /** |
||
209 | * Reads the token after the given token. |
||
210 | * |
||
211 | * @param Token $prev |
||
212 | * @return Token |
||
213 | * @throws SyntaxErrorException |
||
214 | */ |
||
215 | protected function readToken(Token $prev): Token |
||
216 | { |
||
217 | $this->pos = $prev->getEnd(); |
||
218 | |||
219 | $this->skipWhitespace(); |
||
220 | |||
221 | $line = $this->line; |
||
222 | $col = (1 + $this->pos) - $this->lineStart; |
||
223 | |||
224 | if ($this->pos >= $this->bodyLength) { |
||
225 | return $this->createEndOfFileToken($line, $col, $prev); |
||
226 | } |
||
227 | |||
228 | $code = $this->readCharCode($this->pos); |
||
229 | |||
230 | // Punctuation: [!$&:=@|()\[\]{}]{1} |
||
231 | if (33 === $code || 36 === $code || 38 === $code || 58 === $code || 61 === $code || 64 === $code || 124 === $code || |
||
232 | 40 === $code || 41 === $code || 91 === $code || 93 === $code || 123 === $code || 125 === $code) { |
||
233 | return $this->lexPunctuation($code, $line, $col, $prev); |
||
234 | } |
||
235 | |||
236 | // Comment: #[\u0009\u0020-\uFFFF]* |
||
237 | if (35 === $code) { |
||
238 | return $this->lexComment($line, $col, $prev); |
||
239 | } |
||
240 | |||
241 | // Int: -?(0|[1-9][0-9]*) |
||
242 | // Float: -?(0|[1-9][0-9]*)(\.[0-9]+)?((E|e)(+|-)?[0-9]+)? |
||
243 | if (45 === $code || isNumber($code)) { |
||
244 | return $this->lexNumber($code, $line, $col, $prev); |
||
245 | } |
||
246 | |||
247 | // Name: [_A-Za-z][_0-9A-Za-z]* |
||
248 | if (isAlphaNumeric($code)) { |
||
249 | return $this->lexName($line, $col, $prev); |
||
250 | } |
||
251 | |||
252 | // Spread: ... |
||
253 | if ($this->bodyLength >= 3 && $this->isSpread($code)) { |
||
254 | return $this->lexSpread($line, $col, $prev); |
||
255 | } |
||
256 | |||
257 | // String: "([^"\\\u000A\u000D]|(\\(u[0-9a-fA-F]{4}|["\\/bfnrt])))*" |
||
258 | if ($this->isString($code)) { |
||
259 | return $this->lexString($line, $col, $prev); |
||
260 | } |
||
261 | |||
262 | // Block String: """("?"?(\\"""|\\(?!=""")|[^"\\]))*""" |
||
263 | if ($this->bodyLength >= 3 && $this->isTripleQuote($code)) { |
||
264 | return $this->lexBlockString($line, $col, $prev); |
||
265 | } |
||
266 | |||
267 | throw $this->createSyntaxErrorException(); |
||
268 | } |
||
269 | |||
270 | /** |
||
271 | * @return Token |
||
272 | */ |
||
273 | protected function createStartOfFileToken(): Token |
||
274 | { |
||
275 | return new Token(TokenKindEnum::SOF); |
||
276 | } |
||
277 | |||
278 | /** |
||
279 | * Creates an End Of File (EOF) token. |
||
280 | * |
||
281 | * @param int $line |
||
282 | * @param int $col |
||
283 | * @param Token $prev |
||
284 | * @return Token |
||
285 | */ |
||
286 | protected function createEndOfFileToken(int $line, int $col, Token $prev): Token |
||
287 | { |
||
288 | return new Token(TokenKindEnum::EOF, $this->bodyLength, $this->bodyLength, $line, $col, $prev); |
||
289 | } |
||
290 | |||
291 | /** |
||
292 | * Reads a punctuation token from the source file. |
||
293 | * |
||
294 | * @param int $code |
||
295 | * @param int $line |
||
296 | * @param int $col |
||
297 | * @param Token $prev |
||
298 | * @return Token |
||
299 | * @throws SyntaxErrorException |
||
300 | */ |
||
301 | protected function lexPunctuation(int $code, int $line, int $col, Token $prev): ?Token |
||
308 | } |
||
309 | |||
310 | /** |
||
311 | * Reads a name token from the source file. |
||
312 | * |
||
313 | * @param int $line |
||
314 | * @param int $col |
||
315 | * @param Token $prev |
||
316 | * @return Token |
||
317 | */ |
||
318 | protected function lexName(int $line, int $col, Token $prev): Token |
||
319 | { |
||
320 | $start = $this->pos; |
||
321 | |||
322 | ++$this->pos; |
||
323 | |||
324 | while ($this->pos !== $this->bodyLength && |
||
325 | ($code = $this->readCharCode($this->pos)) !== null && |
||
326 | isAlphaNumeric($code)) { |
||
327 | ++$this->pos; |
||
328 | } |
||
329 | |||
330 | $value = sliceString($this->body, $start, $this->pos); |
||
331 | |||
332 | return new Token(TokenKindEnum::NAME, $start, $this->pos, $line, $col, $prev, $value); |
||
333 | } |
||
334 | |||
335 | /** |
||
336 | * Reads a number (int or float) token from the source file. |
||
337 | * |
||
338 | * @param int $code |
||
339 | * @param int $line |
||
340 | * @param int $col |
||
341 | * @param Token $prev |
||
342 | * @return Token |
||
343 | * @throws SyntaxErrorException |
||
344 | */ |
||
345 | protected function lexNumber(int $code, int $line, int $col, Token $prev): Token |
||
400 | ); |
||
401 | } |
||
402 | |||
403 | /** |
||
404 | * Skips digits at the current position. |
||
405 | * |
||
406 | * @param int $code |
||
407 | * @throws SyntaxErrorException |
||
408 | */ |
||
409 | protected function skipDigits(int $code): void |
||
410 | { |
||
411 | if (isNumber($code)) { |
||
412 | do { |
||
413 | $code = $this->readCharCode(++$this->pos); |
||
414 | } while (isNumber($code)); |
||
415 | |||
416 | return; |
||
417 | } |
||
418 | |||
419 | throw $this->createSyntaxErrorException( |
||
420 | \sprintf('Invalid number, expected digit but got: %s.', printCharCode($code)) |
||
421 | ); |
||
422 | } |
||
423 | |||
424 | /** |
||
425 | * Reads a comment token from the source file. |
||
426 | * |
||
427 | * @param int $line |
||
428 | * @param int $col |
||
429 | * @param Token $prev |
||
430 | * @return Token |
||
431 | */ |
||
432 | protected function lexComment(int $line, int $col, Token $prev): Token |
||
433 | { |
||
434 | $start = $this->pos; |
||
435 | |||
436 | do { |
||
437 | $code = $this->readCharCode(++$this->pos); |
||
438 | } while ($code !== null && ($code > 0x001f || 0x0009 === $code)); // SourceCharacter but not LineTerminator |
||
439 | |||
440 | return new Token( |
||
441 | TokenKindEnum::COMMENT, |
||
442 | $start, |
||
443 | $this->pos, |
||
444 | $line, |
||
445 | $col, |
||
446 | $prev, |
||
447 | sliceString($this->body, $start + 1, $this->pos) |
||
448 | ); |
||
449 | } |
||
450 | |||
451 | /** |
||
452 | * Reads a spread token from the source. |
||
453 | * |
||
454 | * @param int $line |
||
455 | * @param int $col |
||
456 | * @param Token $prev |
||
457 | * @return Token |
||
458 | */ |
||
459 | protected function lexSpread(int $line, int $col, Token $prev): Token |
||
460 | { |
||
461 | return new Token(TokenKindEnum::SPREAD, $this->pos, $this->pos + 3, $line, $col, $prev); |
||
462 | } |
||
463 | |||
464 | /** |
||
465 | * Reads a string token from the source. |
||
466 | * |
||
467 | * @param int $line |
||
468 | * @param int $col |
||
469 | * @param Token $prev |
||
470 | * @return Token |
||
471 | * @throws SyntaxErrorException |
||
472 | */ |
||
473 | protected function lexString(int $line, int $col, Token $prev): Token |
||
474 | { |
||
475 | $start = $this->pos; |
||
476 | $chunkStart = ++$this->pos; // skip the quote |
||
477 | $value = ''; |
||
478 | |||
479 | while ($this->pos < $this->bodyLength && |
||
480 | ($code = $this->readCharCode($this->pos)) !== null && !isLineTerminator($code)) { |
||
481 | // Closing Quote (") |
||
482 | if (34 === $code) { |
||
483 | $value .= sliceString($this->body, $chunkStart, $this->pos); |
||
484 | return new Token(TokenKindEnum::STRING, $start, $this->pos + 1, $line, $col, $prev, $value); |
||
485 | } |
||
486 | |||
487 | if (isSourceCharacter($code)) { |
||
488 | throw $this->createSyntaxErrorException( |
||
489 | \sprintf('Invalid character within String: %s.', printCharCode($code)) |
||
490 | ); |
||
491 | } |
||
492 | |||
493 | ++$this->pos; |
||
494 | |||
495 | if (92 === $code) { |
||
496 | // \ |
||
497 | $value .= sliceString($this->body, $chunkStart, $this->pos - 1); |
||
498 | |||
499 | $code = $this->readCharCode($this->pos); |
||
500 | |||
501 | switch ($code) { |
||
502 | case 34: // " |
||
503 | $value .= '"'; |
||
504 | break; |
||
505 | case 47: // / |
||
506 | $value .= '/'; |
||
507 | break; |
||
508 | case 92: // \ |
||
509 | $value .= '\\'; |
||
510 | break; |
||
511 | case 98: // b |
||
512 | $value .= '\b'; |
||
513 | break; |
||
514 | case 102: // f |
||
515 | $value .= '\f'; |
||
516 | break; |
||
517 | case 110: // n |
||
518 | $value .= '\n'; |
||
519 | break; |
||
520 | case 114: // r |
||
521 | $value .= '\r'; |
||
522 | break; |
||
523 | case 116: // t |
||
524 | $value .= '\t'; |
||
525 | break; |
||
526 | case 117: // u |
||
527 | $unicodeString = sliceString($this->body, $this->pos + 1, $this->pos + 5); |
||
528 | |||
529 | if (!\preg_match('/[0-9A-Fa-f]{4}/', $unicodeString)) { |
||
530 | throw $this->createSyntaxErrorException( |
||
531 | \sprintf('Invalid character escape sequence: \\u%s.', $unicodeString) |
||
532 | ); |
||
533 | } |
||
534 | |||
535 | $value .= '\\u' . $unicodeString; |
||
536 | $this->pos += 4; |
||
537 | |||
538 | break; |
||
539 | default: |
||
540 | throw $this->createSyntaxErrorException( |
||
541 | \sprintf('Invalid character escape sequence: \\%s.', \chr($code)) |
||
542 | ); |
||
543 | } |
||
544 | |||
545 | ++$this->pos; |
||
546 | |||
547 | $chunkStart = $this->pos; |
||
548 | } |
||
549 | } |
||
550 | |||
551 | throw $this->createSyntaxErrorException('Unterminated string.'); |
||
552 | } |
||
553 | |||
554 | /** |
||
555 | * Reads a block string token from the source file. |
||
556 | * |
||
557 | * @param int $line |
||
558 | * @param int $col |
||
559 | * @param Token $prev |
||
560 | * @return Token |
||
561 | * @throws SyntaxErrorException |
||
562 | */ |
||
563 | protected function lexBlockString(int $line, int $col, Token $prev): Token |
||
601 | } |
||
602 | |||
603 | /** |
||
604 | * Skips whitespace at the current position. |
||
605 | */ |
||
606 | protected function skipWhitespace(): void |
||
607 | { |
||
608 | while ($this->pos < $this->bodyLength) { |
||
609 | $code = $this->readCharCode($this->pos); |
||
610 | |||
611 | if (9 === $code || 32 === $code || 44 === $code || 0xfeff === $code) { |
||
612 | // tab | space | comma | BOM |
||
613 | ++$this->pos; |
||
614 | } elseif (10 === $code) { |
||
615 | // new line (\n) |
||
616 | ++$this->pos; |
||
617 | ++$this->line; |
||
618 | $this->lineStart = $this->pos; |
||
619 | } elseif (13 === $code) { |
||
620 | // carriage return (\r) |
||
621 | if (10 === $this->readCharCode($this->pos + 1)) { |
||
622 | // carriage return and new line (\r\n) |
||
623 | $this->pos += 2; |
||
624 | } else { |
||
625 | ++$this->pos; |
||
626 | } |
||
627 | ++$this->line; |
||
628 | $this->lineStart = $this->pos; |
||
629 | } else { |
||
630 | break; |
||
631 | } |
||
632 | } |
||
633 | } |
||
634 | |||
635 | /** |
||
636 | * @param int $pos |
||
637 | * @return int |
||
638 | */ |
||
639 | protected function readCharCode(int $pos): int |
||
652 | } |
||
653 | |||
654 | /** |
||
655 | * Report a message that an unexpected character was encountered. |
||
656 | * |
||
657 | * @param int $code |
||
658 | * @return string |
||
659 | */ |
||
660 | protected function unexpectedCharacterMessage(int $code): string |
||
661 | { |
||
662 | if (isSourceCharacter($code) && !isLineTerminator($code)) { |
||
663 | return \sprintf('Cannot contain the invalid character %s.', printCharCode($code)); |
||
664 | } |
||
665 | |||
666 | if ($code === 39) { |
||
667 | // ' |
||
668 | return 'Unexpected single quote character (\'), did you mean to use a double quote (")?'; |
||
669 | } |
||
670 | |||
671 | return \sprintf('Cannot parse the unexpected character %s.', printCharCode($code)); |
||
672 | } |
||
673 | |||
674 | /** |
||
675 | * @param int $code |
||
676 | * @return bool |
||
677 | */ |
||
678 | protected function isSpread(int $code): bool |
||
679 | { |
||
680 | return 46 === $code && |
||
681 | $this->readCharCode($this->pos + 1) === 46 && |
||
682 | $this->readCharCode($this->pos + 2) === 46; // ... |
||
683 | } |
||
684 | |||
685 | /** |
||
686 | * @param int $code |
||
687 | * @return bool |
||
688 | */ |
||
689 | protected function isString(int $code): bool |
||
690 | { |
||
691 | return 34 === $code && $this->readCharCode($this->pos + 1) !== 34; |
||
692 | } |
||
693 | |||
694 | /** |
||
695 | * @param int $code |
||
696 | * @return bool |
||
697 | */ |
||
698 | protected function isTripleQuote(int $code): bool |
||
699 | { |
||
700 | return 34 === $code && |
||
701 | 34 === $this->readCharCode($this->pos + 1) && |
||
702 | 34 === $this->readCharCode($this->pos + 2); // """ |
||
703 | } |
||
704 | |||
705 | /** |
||
706 | * @param int $code |
||
707 | * @return bool |
||
708 | */ |
||
709 | protected function isEscapedTripleQuote(int $code): bool |
||
715 | } |
||
716 | } |
||
717 |