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