| Total Complexity | 42 |
| Total Lines | 390 |
| Duplicated Lines | 0 % |
| Changes | 0 | ||
Complex classes like SqlToken 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 SqlToken, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 24 | final class SqlToken implements ArrayAccess |
||
| 25 | { |
||
| 26 | public const TYPE_CODE = 0; |
||
| 27 | public const TYPE_STATEMENT = 1; |
||
| 28 | public const TYPE_TOKEN = 2; |
||
| 29 | public const TYPE_PARENTHESIS = 3; |
||
| 30 | public const TYPE_KEYWORD = 4; |
||
| 31 | public const TYPE_OPERATOR = 5; |
||
| 32 | public const TYPE_IDENTIFIER = 6; |
||
| 33 | public const TYPE_STRING_LITERAL = 7; |
||
| 34 | |||
| 35 | private int $type = self::TYPE_TOKEN; |
||
| 36 | private ?string $content = null; |
||
| 37 | private ?int $startOffset = null; |
||
| 38 | private ?int $endOffset = null; |
||
| 39 | private ?SqlToken $parent = null; |
||
| 40 | private array $children = []; |
||
| 41 | |||
| 42 | /** |
||
| 43 | * Returns the SQL code representing the token. |
||
| 44 | * |
||
| 45 | * @return string SQL code. |
||
| 46 | */ |
||
| 47 | public function __toString(): string |
||
| 48 | { |
||
| 49 | return $this->getSql(); |
||
| 50 | } |
||
| 51 | |||
| 52 | /** |
||
| 53 | * Returns whether there is a child token at the specified offset. |
||
| 54 | * |
||
| 55 | * This method is required by the SPL {@see ArrayAccess} interface. It is implicitly called when you use something |
||
| 56 | * like `isset($token[$offset])`. |
||
| 57 | * |
||
| 58 | * @param int $offset child token offset. |
||
| 59 | * |
||
| 60 | * @return bool whether the token exists. |
||
| 61 | */ |
||
| 62 | public function offsetExists($offset): bool |
||
| 63 | { |
||
| 64 | return isset($this->children[$this->calculateOffset($offset)]); |
||
| 65 | } |
||
| 66 | |||
| 67 | /** |
||
| 68 | * Returns a child token at the specified offset. |
||
| 69 | * |
||
| 70 | * This method is required by the SPL {@see ArrayAccess} interface. It is implicitly called when you use something |
||
| 71 | * like `$child = $token[$offset];`. |
||
| 72 | * |
||
| 73 | * @param int $offset child token offset. |
||
| 74 | * |
||
| 75 | * @return SqlToken|null the child token at the specified offset, `null` if there's no token. |
||
| 76 | */ |
||
| 77 | public function offsetGet($offset): ?SqlToken |
||
| 82 | } |
||
| 83 | |||
| 84 | /** |
||
| 85 | * Adds a child token to the token. |
||
| 86 | * |
||
| 87 | * This method is required by the SPL {@see ArrayAccess} interface. It is implicitly called when you use something |
||
| 88 | * like `$token[$offset] = $child;`. |
||
| 89 | * |
||
| 90 | * @param int|null $offset child token offset. |
||
| 91 | * @param SqlToken $token token to be added. |
||
| 92 | */ |
||
| 93 | public function offsetSet($offset, $token): void |
||
| 104 | } |
||
| 105 | |||
| 106 | /** |
||
| 107 | * Removes a child token at the specified offset. |
||
| 108 | * |
||
| 109 | * This method is required by the SPL {@see ArrayAccess} interface. It is implicitly called when you use something |
||
| 110 | * like `unset($token[$offset])`. |
||
| 111 | * |
||
| 112 | * @param int $offset child token offset. |
||
| 113 | */ |
||
| 114 | public function offsetUnset($offset): void |
||
| 123 | } |
||
| 124 | |||
| 125 | /** |
||
| 126 | * Returns child tokens. |
||
| 127 | * |
||
| 128 | * @return SqlToken[] child tokens. |
||
| 129 | */ |
||
| 130 | public function getChildren(): array |
||
| 131 | { |
||
| 132 | return $this->children; |
||
| 133 | } |
||
| 134 | |||
| 135 | /** |
||
| 136 | * Sets a list of child tokens. |
||
| 137 | * |
||
| 138 | * @param SqlToken[] $children child tokens. |
||
| 139 | */ |
||
| 140 | public function setChildren(array $children): void |
||
| 141 | { |
||
| 142 | $this->children = []; |
||
| 143 | |||
| 144 | foreach ($children as $child) { |
||
| 145 | $child->parent = $this; |
||
| 146 | $this->children[] = $child; |
||
| 147 | } |
||
| 148 | |||
| 149 | $this->updateCollectionOffsets(); |
||
| 150 | } |
||
| 151 | |||
| 152 | /** |
||
| 153 | * Returns whether the token represents a collection of tokens. |
||
| 154 | * |
||
| 155 | * @return bool whether the token represents a collection of tokens. |
||
| 156 | */ |
||
| 157 | public function getIsCollection(): bool |
||
| 158 | { |
||
| 159 | return in_array($this->type, [ |
||
| 160 | self::TYPE_CODE, |
||
| 161 | self::TYPE_STATEMENT, |
||
| 162 | self::TYPE_PARENTHESIS, |
||
| 163 | ], true); |
||
| 164 | } |
||
| 165 | |||
| 166 | /** |
||
| 167 | * Returns whether the token represents a collection of tokens and has non-zero number of children. |
||
| 168 | * |
||
| 169 | * @return bool whether the token has children. |
||
| 170 | */ |
||
| 171 | public function getHasChildren(): bool |
||
| 172 | { |
||
| 173 | return $this->getIsCollection() && !empty($this->children); |
||
| 174 | } |
||
| 175 | |||
| 176 | /** |
||
| 177 | * Returns the SQL code representing the token. |
||
| 178 | * |
||
| 179 | * @return string SQL code. |
||
| 180 | */ |
||
| 181 | public function getSql(): string |
||
| 182 | { |
||
| 183 | $code = $this; |
||
| 184 | |||
| 185 | while ($code->parent !== null) { |
||
| 186 | $code = $code->parent; |
||
| 187 | } |
||
| 188 | |||
| 189 | return mb_substr($code->content, $this->startOffset, $this->endOffset - $this->startOffset, 'UTF-8'); |
||
| 190 | } |
||
| 191 | |||
| 192 | /** |
||
| 193 | * Returns whether this token (including its children) matches the specified "pattern" SQL code. |
||
| 194 | * |
||
| 195 | * Usage Example: |
||
| 196 | * |
||
| 197 | * ```php |
||
| 198 | * $patternToken = (new \Yiisoft\Db\Sqlite\SqlTokenizer('SELECT any FROM any'))->tokenize(); |
||
| 199 | * if ($sqlToken->matches($patternToken, 0, $firstMatchIndex, $lastMatchIndex)) { |
||
| 200 | * // ... |
||
| 201 | * } |
||
| 202 | * ``` |
||
| 203 | * |
||
| 204 | * @param SqlToken $patternToken tokenized SQL code to match against. In addition to normal SQL, the `any` keyword |
||
| 205 | * is supported which will match any number of keywords, identifiers, whitespaces. |
||
| 206 | * @param int $offset token children offset to start lookup with. |
||
| 207 | * @param int|null $firstMatchIndex token children offset where a successful match begins. |
||
| 208 | * @param int|null $lastMatchIndex token children offset where a successful match ends. |
||
| 209 | * |
||
| 210 | * @return bool whether this token matches the pattern SQL code. |
||
| 211 | */ |
||
| 212 | public function matches( |
||
| 213 | self $patternToken, |
||
| 214 | int $offset = 0, |
||
| 215 | ?int &$firstMatchIndex = null, |
||
| 216 | ?int &$lastMatchIndex = null |
||
| 217 | ): bool { |
||
| 218 | if (!$patternToken->getHasChildren()) { |
||
| 219 | return false; |
||
| 220 | } |
||
| 221 | |||
| 222 | $patternToken = $patternToken[0]; |
||
| 223 | |||
| 224 | return $this->tokensMatch($patternToken, $this, $offset, $firstMatchIndex, $lastMatchIndex); |
||
|
|
|||
| 225 | } |
||
| 226 | |||
| 227 | /** |
||
| 228 | * Tests the given token to match the specified pattern token. |
||
| 229 | * |
||
| 230 | * @param SqlToken $patternToken |
||
| 231 | * @param SqlToken $token |
||
| 232 | * @param int $offset |
||
| 233 | * @param int|null $firstMatchIndex |
||
| 234 | * @param int|null $lastMatchIndex |
||
| 235 | * |
||
| 236 | * @return bool |
||
| 237 | */ |
||
| 238 | private function tokensMatch( |
||
| 239 | self $patternToken, |
||
| 240 | self $token, |
||
| 241 | int $offset = 0, |
||
| 242 | ?int &$firstMatchIndex = null, |
||
| 243 | ?int &$lastMatchIndex = null |
||
| 244 | ): bool { |
||
| 245 | if ($patternToken->getIsCollection() !== $token->getIsCollection() || (!$patternToken->getIsCollection() && $patternToken->content !== $token->content)) { |
||
| 246 | return false; |
||
| 247 | } |
||
| 248 | |||
| 249 | if ($patternToken->children === $token->children) { |
||
| 250 | $firstMatchIndex = $lastMatchIndex = $offset; |
||
| 251 | |||
| 252 | return true; |
||
| 253 | } |
||
| 254 | |||
| 255 | $firstMatchIndex = $lastMatchIndex = null; |
||
| 256 | $wildcard = false; |
||
| 257 | |||
| 258 | for ($index = 0, $count = count($patternToken->children); $index < $count; $index++) { |
||
| 259 | /** |
||
| 260 | * Here we iterate token by token with an exception of "any" that toggles an iteration until we matched |
||
| 261 | * with a next pattern token or EOF. |
||
| 262 | */ |
||
| 263 | if ($patternToken[$index]->content === 'any') { |
||
| 264 | $wildcard = true; |
||
| 265 | continue; |
||
| 266 | } |
||
| 267 | |||
| 268 | for ($limit = $wildcard ? count($token->children) : $offset + 1; $offset < $limit; $offset++) { |
||
| 269 | if (!$wildcard && !isset($token[$offset])) { |
||
| 270 | break; |
||
| 271 | } |
||
| 272 | |||
| 273 | if (!$this->tokensMatch($patternToken[$index], $token[$offset])) { |
||
| 274 | continue; |
||
| 275 | } |
||
| 276 | |||
| 277 | if ($firstMatchIndex === null) { |
||
| 278 | $firstMatchIndex = $offset; |
||
| 279 | $lastMatchIndex = $offset; |
||
| 280 | } else { |
||
| 281 | $lastMatchIndex = $offset; |
||
| 282 | } |
||
| 283 | |||
| 284 | $wildcard = false; |
||
| 285 | $offset++; |
||
| 286 | |||
| 287 | continue 2; |
||
| 288 | } |
||
| 289 | |||
| 290 | return false; |
||
| 291 | } |
||
| 292 | |||
| 293 | return true; |
||
| 294 | } |
||
| 295 | |||
| 296 | /** |
||
| 297 | * Returns an absolute offset in the children array. |
||
| 298 | * |
||
| 299 | * @param int $offset |
||
| 300 | * |
||
| 301 | * @return int |
||
| 302 | */ |
||
| 303 | private function calculateOffset(int $offset): int |
||
| 304 | { |
||
| 305 | if ($offset >= 0) { |
||
| 306 | return $offset; |
||
| 307 | } |
||
| 308 | |||
| 309 | return count($this->children) + $offset; |
||
| 310 | } |
||
| 311 | |||
| 312 | /** |
||
| 313 | * Updates token SQL code start and end offsets based on its children. |
||
| 314 | */ |
||
| 315 | private function updateCollectionOffsets(): void |
||
| 316 | { |
||
| 317 | if (!empty($this->children)) { |
||
| 318 | $this->startOffset = reset($this->children)->startOffset; |
||
| 319 | $this->endOffset = end($this->children)->endOffset; |
||
| 320 | } |
||
| 321 | |||
| 322 | if ($this->parent !== null) { |
||
| 323 | $this->parent->updateCollectionOffsets(); |
||
| 324 | } |
||
| 325 | } |
||
| 326 | |||
| 327 | /** |
||
| 328 | * Set token type. It has to be one of the following constants: |
||
| 329 | * |
||
| 330 | * - {@see TYPE_CODE} |
||
| 331 | * - {@see TYPE_STATEMENT} |
||
| 332 | * - {@see TYPE_TOKEN} |
||
| 333 | * - {@see TYPE_PARENTHESIS} |
||
| 334 | * - {@see TYPE_KEYWORD} |
||
| 335 | * - {@see TYPE_OPERATOR} |
||
| 336 | * - {@see TYPE_IDENTIFIER} |
||
| 337 | * - {@see TYPE_STRING_LITERAL} |
||
| 338 | * |
||
| 339 | * @param int $value token type. It has to be one of the following constants: |
||
| 340 | * |
||
| 341 | * @return self |
||
| 342 | */ |
||
| 343 | public function type(int $value): self |
||
| 344 | { |
||
| 345 | $this->type = $value; |
||
| 346 | |||
| 347 | return $this; |
||
| 348 | } |
||
| 349 | |||
| 350 | /** |
||
| 351 | * Set token content. |
||
| 352 | * |
||
| 353 | * @param string|null $value |
||
| 354 | * |
||
| 355 | * @return self |
||
| 356 | */ |
||
| 357 | public function content(?string $value): self |
||
| 358 | { |
||
| 359 | $this->content = $value; |
||
| 360 | |||
| 361 | return $this; |
||
| 362 | } |
||
| 363 | |||
| 364 | /** |
||
| 365 | * Set original SQL token start position. |
||
| 366 | * |
||
| 367 | * @param int $value original SQL token start position. |
||
| 368 | * |
||
| 369 | * @return self |
||
| 370 | */ |
||
| 371 | public function startOffset(int $value): self |
||
| 372 | { |
||
| 373 | $this->startOffset = $value; |
||
| 374 | |||
| 375 | return $this; |
||
| 376 | } |
||
| 377 | |||
| 378 | /** |
||
| 379 | * Set original SQL token end position. |
||
| 380 | * |
||
| 381 | * @param int $value original SQL token end position. |
||
| 382 | * |
||
| 383 | * @return self |
||
| 384 | */ |
||
| 385 | public function endOffset(int $value): self |
||
| 386 | { |
||
| 387 | $this->endOffset = $value; |
||
| 388 | |||
| 389 | return $this; |
||
| 390 | } |
||
| 391 | |||
| 392 | /** |
||
| 393 | * Set parent token. |
||
| 394 | * |
||
| 395 | * @param SqlToken $value parent token. |
||
| 396 | * |
||
| 397 | * @return self |
||
| 398 | */ |
||
| 399 | public function parent(SqlToken $value): self |
||
| 400 | { |
||
| 401 | $this->parent = $value; |
||
| 402 | |||
| 403 | return $this; |
||
| 404 | } |
||
| 405 | |||
| 406 | public function getContent(): ?string |
||
| 409 | } |
||
| 410 | |||
| 411 | public function getType(): int |
||
| 412 | { |
||
| 413 | return $this->type; |
||
| 414 | } |
||
| 415 | } |
||
| 416 |