1: | <?php declare(strict_types = 1); |
2: | |
3: | namespace PHPStan\PhpDocParser\Parser; |
4: | |
5: | use LogicException; |
6: | use PHPStan\PhpDocParser\Lexer\Lexer; |
7: | use function array_pop; |
8: | use function assert; |
9: | use function count; |
10: | use function in_array; |
11: | use function strlen; |
12: | use function substr; |
13: | |
14: | class TokenIterator |
15: | { |
16: | |
17: | |
18: | private array $tokens; |
19: | |
20: | private int $index; |
21: | |
22: | |
23: | private array $savePoints = []; |
24: | |
25: | |
26: | private array $skippedTokenTypes = [Lexer::TOKEN_HORIZONTAL_WS]; |
27: | |
28: | private ?string $newline = null; |
29: | |
30: | |
31: | |
32: | |
33: | public function __construct(array $tokens, int $index = 0) |
34: | { |
35: | $this->tokens = $tokens; |
36: | $this->index = $index; |
37: | |
38: | $this->skipIrrelevantTokens(); |
39: | } |
40: | |
41: | |
42: | |
43: | |
44: | |
45: | public function getTokens(): array |
46: | { |
47: | return $this->tokens; |
48: | } |
49: | |
50: | |
51: | public function getContentBetween(int $startPos, int $endPos): string |
52: | { |
53: | if ($startPos < 0 || $endPos > count($this->tokens)) { |
54: | throw new LogicException(); |
55: | } |
56: | |
57: | $content = ''; |
58: | for ($i = $startPos; $i < $endPos; $i++) { |
59: | $content .= $this->tokens[$i][Lexer::VALUE_OFFSET]; |
60: | } |
61: | |
62: | return $content; |
63: | } |
64: | |
65: | |
66: | public function getTokenCount(): int |
67: | { |
68: | return count($this->tokens); |
69: | } |
70: | |
71: | |
72: | public function currentTokenValue(): string |
73: | { |
74: | return $this->tokens[$this->index][Lexer::VALUE_OFFSET]; |
75: | } |
76: | |
77: | |
78: | public function currentTokenType(): int |
79: | { |
80: | return $this->tokens[$this->index][Lexer::TYPE_OFFSET]; |
81: | } |
82: | |
83: | |
84: | public function currentTokenOffset(): int |
85: | { |
86: | $offset = 0; |
87: | for ($i = 0; $i < $this->index; $i++) { |
88: | $offset += strlen($this->tokens[$i][Lexer::VALUE_OFFSET]); |
89: | } |
90: | |
91: | return $offset; |
92: | } |
93: | |
94: | |
95: | public function currentTokenLine(): int |
96: | { |
97: | return $this->tokens[$this->index][Lexer::LINE_OFFSET]; |
98: | } |
99: | |
100: | |
101: | public function currentTokenIndex(): int |
102: | { |
103: | return $this->index; |
104: | } |
105: | |
106: | |
107: | public function endIndexOfLastRelevantToken(): int |
108: | { |
109: | $endIndex = $this->currentTokenIndex(); |
110: | $endIndex--; |
111: | while (in_array($this->tokens[$endIndex][Lexer::TYPE_OFFSET], $this->skippedTokenTypes, true)) { |
112: | if (!isset($this->tokens[$endIndex - 1])) { |
113: | break; |
114: | } |
115: | $endIndex--; |
116: | } |
117: | |
118: | return $endIndex; |
119: | } |
120: | |
121: | |
122: | public function isCurrentTokenValue(string $tokenValue): bool |
123: | { |
124: | return $this->tokens[$this->index][Lexer::VALUE_OFFSET] === $tokenValue; |
125: | } |
126: | |
127: | |
128: | public function isCurrentTokenType(int ...$tokenType): bool |
129: | { |
130: | return in_array($this->tokens[$this->index][Lexer::TYPE_OFFSET], $tokenType, true); |
131: | } |
132: | |
133: | |
134: | public function isPrecededByHorizontalWhitespace(): bool |
135: | { |
136: | return ($this->tokens[$this->index - 1][Lexer::TYPE_OFFSET] ?? -1) === Lexer::TOKEN_HORIZONTAL_WS; |
137: | } |
138: | |
139: | |
140: | |
141: | |
142: | |
143: | public function consumeTokenType(int $tokenType): void |
144: | { |
145: | if ($this->tokens[$this->index][Lexer::TYPE_OFFSET] !== $tokenType) { |
146: | $this->throwError($tokenType); |
147: | } |
148: | |
149: | if ($tokenType === Lexer::TOKEN_PHPDOC_EOL) { |
150: | if ($this->newline === null) { |
151: | $this->detectNewline(); |
152: | } |
153: | } |
154: | |
155: | $this->index++; |
156: | $this->skipIrrelevantTokens(); |
157: | } |
158: | |
159: | |
160: | |
161: | |
162: | |
163: | public function consumeTokenValue(int $tokenType, string $tokenValue): void |
164: | { |
165: | if ($this->tokens[$this->index][Lexer::TYPE_OFFSET] !== $tokenType || $this->tokens[$this->index][Lexer::VALUE_OFFSET] !== $tokenValue) { |
166: | $this->throwError($tokenType, $tokenValue); |
167: | } |
168: | |
169: | $this->index++; |
170: | $this->skipIrrelevantTokens(); |
171: | } |
172: | |
173: | |
174: | |
175: | public function tryConsumeTokenValue(string $tokenValue): bool |
176: | { |
177: | if ($this->tokens[$this->index][Lexer::VALUE_OFFSET] !== $tokenValue) { |
178: | return false; |
179: | } |
180: | |
181: | $this->index++; |
182: | $this->skipIrrelevantTokens(); |
183: | |
184: | return true; |
185: | } |
186: | |
187: | |
188: | |
189: | public function tryConsumeTokenType(int $tokenType): bool |
190: | { |
191: | if ($this->tokens[$this->index][Lexer::TYPE_OFFSET] !== $tokenType) { |
192: | return false; |
193: | } |
194: | |
195: | if ($tokenType === Lexer::TOKEN_PHPDOC_EOL) { |
196: | if ($this->newline === null) { |
197: | $this->detectNewline(); |
198: | } |
199: | } |
200: | |
201: | $this->index++; |
202: | $this->skipIrrelevantTokens(); |
203: | |
204: | return true; |
205: | } |
206: | |
207: | |
208: | |
209: | public function skipNewLineTokens(): void |
210: | { |
211: | if (!$this->isCurrentTokenType(Lexer::TOKEN_PHPDOC_EOL)) { |
212: | return; |
213: | } |
214: | |
215: | do { |
216: | $foundNewLine = $this->tryConsumeTokenType(Lexer::TOKEN_PHPDOC_EOL); |
217: | } while ($foundNewLine === true); |
218: | } |
219: | |
220: | |
221: | private function detectNewline(): void |
222: | { |
223: | $value = $this->currentTokenValue(); |
224: | if (substr($value, 0, 2) === "\r\n") { |
225: | $this->newline = "\r\n"; |
226: | } elseif (substr($value, 0, 1) === "\n") { |
227: | $this->newline = "\n"; |
228: | } |
229: | } |
230: | |
231: | |
232: | public function getSkippedHorizontalWhiteSpaceIfAny(): string |
233: | { |
234: | if ($this->index > 0 && $this->tokens[$this->index - 1][Lexer::TYPE_OFFSET] === Lexer::TOKEN_HORIZONTAL_WS) { |
235: | return $this->tokens[$this->index - 1][Lexer::VALUE_OFFSET]; |
236: | } |
237: | |
238: | return ''; |
239: | } |
240: | |
241: | |
242: | |
243: | public function joinUntil(int ...$tokenType): string |
244: | { |
245: | $s = ''; |
246: | while (!in_array($this->tokens[$this->index][Lexer::TYPE_OFFSET], $tokenType, true)) { |
247: | $s .= $this->tokens[$this->index++][Lexer::VALUE_OFFSET]; |
248: | } |
249: | return $s; |
250: | } |
251: | |
252: | |
253: | public function next(): void |
254: | { |
255: | $this->index++; |
256: | $this->skipIrrelevantTokens(); |
257: | } |
258: | |
259: | |
260: | private function skipIrrelevantTokens(): void |
261: | { |
262: | if (!isset($this->tokens[$this->index])) { |
263: | return; |
264: | } |
265: | |
266: | while (in_array($this->tokens[$this->index][Lexer::TYPE_OFFSET], $this->skippedTokenTypes, true)) { |
267: | if (!isset($this->tokens[$this->index + 1])) { |
268: | break; |
269: | } |
270: | $this->index++; |
271: | } |
272: | } |
273: | |
274: | |
275: | public function addEndOfLineToSkippedTokens(): void |
276: | { |
277: | $this->skippedTokenTypes = [Lexer::TOKEN_HORIZONTAL_WS, Lexer::TOKEN_PHPDOC_EOL]; |
278: | } |
279: | |
280: | |
281: | public function removeEndOfLineFromSkippedTokens(): void |
282: | { |
283: | $this->skippedTokenTypes = [Lexer::TOKEN_HORIZONTAL_WS]; |
284: | } |
285: | |
286: | |
287: | public function forwardToTheEnd(): void |
288: | { |
289: | $lastToken = count($this->tokens) - 1; |
290: | $this->index = $lastToken; |
291: | } |
292: | |
293: | |
294: | public function pushSavePoint(): void |
295: | { |
296: | $this->savePoints[] = $this->index; |
297: | } |
298: | |
299: | |
300: | public function dropSavePoint(): void |
301: | { |
302: | array_pop($this->savePoints); |
303: | } |
304: | |
305: | |
306: | public function rollback(): void |
307: | { |
308: | $index = array_pop($this->savePoints); |
309: | assert($index !== null); |
310: | $this->index = $index; |
311: | } |
312: | |
313: | |
314: | |
315: | |
316: | |
317: | private function throwError(int $expectedTokenType, ?string $expectedTokenValue = null): void |
318: | { |
319: | throw new ParserException( |
320: | $this->currentTokenValue(), |
321: | $this->currentTokenType(), |
322: | $this->currentTokenOffset(), |
323: | $expectedTokenType, |
324: | $expectedTokenValue, |
325: | $this->currentTokenLine(), |
326: | ); |
327: | } |
328: | |
329: | |
330: | |
331: | |
332: | |
333: | |
334: | public function hasTokenImmediatelyBefore(int $pos, int $expectedTokenType): bool |
335: | { |
336: | $tokens = $this->tokens; |
337: | $pos--; |
338: | for (; $pos >= 0; $pos--) { |
339: | $token = $tokens[$pos]; |
340: | $type = $token[Lexer::TYPE_OFFSET]; |
341: | if ($type === $expectedTokenType) { |
342: | return true; |
343: | } |
344: | if (!in_array($type, [ |
345: | Lexer::TOKEN_HORIZONTAL_WS, |
346: | Lexer::TOKEN_PHPDOC_EOL, |
347: | ], true)) { |
348: | break; |
349: | } |
350: | } |
351: | return false; |
352: | } |
353: | |
354: | |
355: | |
356: | |
357: | |
358: | |
359: | public function hasTokenImmediatelyAfter(int $pos, int $expectedTokenType): bool |
360: | { |
361: | $tokens = $this->tokens; |
362: | $pos++; |
363: | for ($c = count($tokens); $pos < $c; $pos++) { |
364: | $token = $tokens[$pos]; |
365: | $type = $token[Lexer::TYPE_OFFSET]; |
366: | if ($type === $expectedTokenType) { |
367: | return true; |
368: | } |
369: | if (!in_array($type, [ |
370: | Lexer::TOKEN_HORIZONTAL_WS, |
371: | Lexer::TOKEN_PHPDOC_EOL, |
372: | ], true)) { |
373: | break; |
374: | } |
375: | } |
376: | |
377: | return false; |
378: | } |
379: | |
380: | public function getDetectedNewline(): ?string |
381: | { |
382: | return $this->newline; |
383: | } |
384: | |
385: | |
386: | |
387: | |
388: | public function hasParentheses(int $startPos, int $endPos): bool |
389: | { |
390: | return $this->hasTokenImmediatelyBefore($startPos, Lexer::TOKEN_OPEN_PARENTHESES) |
391: | && $this->hasTokenImmediatelyAfter($endPos, Lexer::TOKEN_CLOSE_PARENTHESES); |
392: | } |
393: | |
394: | } |
395: | |