Complex classes like Docblock 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. You can also have a look at the cohesion graph to spot any un-connected, or weakly-connected components.
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 Docblock, and based on these observations, apply Extract Interface, too.
| 1 | <?php declare(strict_types=1); |
||
| 16 | use phootwork\collection\ArrayList; |
||
| 17 | use phootwork\collection\Map; |
||
| 18 | use phootwork\lang\Comparator; |
||
| 19 | use ReflectionClass; |
||
| 20 | use ReflectionFunctionAbstract; |
||
| 21 | use ReflectionProperty; |
||
| 22 | |||
| 23 | class Docblock implements \Stringable { |
||
| 24 | protected string $shortDescription; |
||
|
|
|||
| 25 | protected string $longDescription; |
||
| 26 | protected ArrayList $tags; |
||
| 27 | protected ?Comparator $comparator = null; |
||
| 28 | |||
| 29 | const REGEX_TAGNAME = '[\w\-\_\\\\]+'; |
||
| 30 | |||
| 31 | /** |
||
| 32 | * Static docblock factory |
||
| 33 | * |
||
| 34 | * @param ReflectionFunctionAbstract|ReflectionClass|ReflectionProperty|string $docblock a docblock to parse |
||
| 35 | * |
||
| 36 | * @return $this |
||
| 37 | */ |
||
| 38 | public static function create(ReflectionFunctionAbstract|ReflectionClass|ReflectionProperty|string $docblock = ''): self { |
||
| 39 | 1 | return new static($docblock); |
|
| 40 | 1 | } |
|
| 41 | |||
| 42 | /** |
||
| 43 | * Creates a new docblock instance and parses the initial string or reflector object if given |
||
| 44 | * |
||
| 45 | * @param ReflectionFunctionAbstract|ReflectionClass|ReflectionProperty|string $docblock a docblock to parse |
||
| 46 | */ |
||
| 47 | final public function __construct(ReflectionFunctionAbstract|ReflectionClass|ReflectionProperty|string $docblock = '') { |
||
| 48 | 13 | $this->tags = new ArrayList(); |
|
| 49 | 13 | $this->parse($docblock); |
|
| 50 | 13 | } |
|
| 51 | 11 | ||
| 52 | /** |
||
| 53 | * @see https://github.com/phpDocumentor/ReflectionDocBlock/blob/master/src/phpDocumentor/Reflection/DocBlock.php Original Method |
||
| 54 | * |
||
| 55 | * @param ReflectionFunctionAbstract|ReflectionClass|ReflectionProperty|string $docblock |
||
| 56 | * |
||
| 57 | * @throws InvalidArgumentException if there is no getDocComment() method available |
||
| 58 | */ |
||
| 59 | protected function parse(ReflectionFunctionAbstract|ReflectionClass|ReflectionProperty|string $docblock): void { |
||
| 60 | 13 | $docblock = is_object($docblock) ? $docblock->getDocComment() : $docblock; |
|
| 61 | 13 | $docblock = $this->cleanInput($docblock); |
|
| 62 | 2 | ||
| 63 | 1 | [$short, $long, $tags] = $this->splitDocBlock($docblock); |
|
| 64 | 1 | $this->shortDescription = $short; |
|
| 65 | $this->longDescription = $long; |
||
| 66 | $this->parseTags($tags); |
||
| 67 | 1 | } |
|
| 68 | |||
| 69 | /** |
||
| 70 | 12 | * Strips the asterisks from the DocBlock comment. |
|
| 71 | * |
||
| 72 | 12 | * @see https://github.com/phpDocumentor/ReflectionDocBlock/blob/master/src/phpDocumentor/Reflection/DocBlock.php Original Method |
|
| 73 | 12 | * |
|
| 74 | 12 | * @param string $comment String containing the comment text. |
|
| 75 | 12 | * |
|
| 76 | 11 | * @return string |
|
| 77 | */ |
||
| 78 | protected function cleanInput(string $comment): string { |
||
| 79 | $comment = trim(preg_replace('#[ \t]*(?:\/\*\*|\*\/|\*)?[ \t]{0,1}(.*)?#u', '$1', $comment)); |
||
| 80 | |||
| 81 | // reg ex above is not able to remove */ from a single line docblock |
||
| 82 | if (substr($comment, -2) == '*/') { |
||
| 83 | $comment = trim(substr($comment, 0, -2)); |
||
| 84 | } |
||
| 85 | |||
| 86 | // normalize strings |
||
| 87 | 12 | $comment = str_replace(["\r\n", "\r"], "\n", $comment); |
|
| 88 | 12 | ||
| 89 | return $comment; |
||
| 90 | } |
||
| 91 | 12 | ||
| 92 | 1 | /** |
|
| 93 | * Splits the Docblock into a short description, long description and |
||
| 94 | * block of tags. |
||
| 95 | * |
||
| 96 | 12 | * @see https://github.com/phpDocumentor/ReflectionDocBlock/blob/master/src/phpDocumentor/Reflection/DocBlock.php Original Method |
|
| 97 | * |
||
| 98 | 12 | * @param string $comment Comment to split into the sub-parts. |
|
| 99 | * |
||
| 100 | * @author RichardJ Special thanks to RichardJ for the regex responsible |
||
| 101 | * for the split. |
||
| 102 | * |
||
| 103 | * @return string[] containing the short-, long description and an element |
||
| 104 | * containing the tags. |
||
| 105 | */ |
||
| 106 | protected function splitDocBlock(string $comment): array { |
||
| 107 | $matches = []; |
||
| 108 | |||
| 109 | if (str_starts_with($comment, '@')) { |
||
| 110 | $matches = ['', '', $comment]; |
||
| 111 | } else { |
||
| 112 | // clears all extra horizontal whitespace from the line endings |
||
| 113 | // to prevent parsing issues |
||
| 114 | $comment = preg_replace('/\h*$/Sum', '', $comment); |
||
| 115 | 11 | ||
| 116 | 11 | /* |
|
| 117 | * Splits the docblock into a short description, long description and |
||
| 118 | 11 | * tags section |
|
| 119 | 2 | * - The short description is started from the first character until |
|
| 120 | * a dot is encountered followed by a newline OR |
||
| 121 | * two consecutive newlines (horizontal whitespace is taken into |
||
| 122 | * account to consider spacing errors) |
||
| 123 | 9 | * - The long description, any character until a new line is |
|
| 124 | * encountered followed by an @ and word characters (a tag). |
||
| 125 | * This is optional. |
||
| 126 | * - Tags; the remaining characters |
||
| 127 | * |
||
| 128 | * Big thanks to RichardJ for contributing this Regular Expression |
||
| 129 | */ |
||
| 130 | preg_match( |
||
| 131 | '/ |
||
| 132 | \A ( |
||
| 133 | [^\n.]+ |
||
| 134 | (?: |
||
| 135 | (?! \. \n | \n{2} ) # disallow the first seperator here |
||
| 136 | [\n.] (?! [ \t]* @\pL ) # disallow second seperator |
||
| 137 | [^\n.]+ |
||
| 138 | )* |
||
| 139 | 9 | \.? |
|
| 140 | 9 | ) |
|
| 141 | (?: |
||
| 142 | \s* # first seperator (actually newlines but it\'s all whitespace) |
||
| 143 | (?! @\pL ) # disallow the rest, to make sure this one doesn\'t match, |
||
| 144 | #if it doesn\'t exist |
||
| 145 | ( |
||
| 146 | [^\n]+ |
||
| 147 | (?: \n+ |
||
| 148 | (?! [ \t]* @\pL ) # disallow second seperator (@param) |
||
| 149 | [^\n]+ |
||
| 150 | )* |
||
| 151 | ) |
||
| 152 | )? |
||
| 153 | (\s+ [\s\S]*)? # everything that follows |
||
| 154 | /ux', |
||
| 155 | $comment, |
||
| 156 | $matches |
||
| 157 | ); |
||
| 158 | array_shift($matches); |
||
| 159 | } |
||
| 160 | |||
| 161 | while (count($matches) < 3) { |
||
| 162 | $matches[] = ''; |
||
| 163 | } |
||
| 164 | |||
| 165 | return $matches; |
||
| 166 | } |
||
| 167 | 9 | ||
| 168 | /** |
||
| 169 | * Parses the tags |
||
| 170 | 11 | * |
|
| 171 | 7 | * @see https://github.com/phpDocumentor/ReflectionDocBlock/blob/master/src/phpDocumentor/Reflection/DocBlock.php Original Method |
|
| 172 | * |
||
| 173 | * @param string $tags |
||
| 174 | 11 | * |
|
| 175 | * @throws LogicException |
||
| 176 | * @throws InvalidArgumentException |
||
| 177 | */ |
||
| 178 | protected function parseTags(string $tags): void { |
||
| 179 | $tags = trim($tags); |
||
| 180 | if ($tags !== '') { |
||
| 181 | |||
| 182 | // sanitize lines |
||
| 183 | $result = []; |
||
| 184 | foreach (explode("\n", $tags) as $line) { |
||
| 185 | if ($this->isTagLine($line) || count($result) == 0) { |
||
| 186 | $result[] = $line; |
||
| 187 | 12 | } else { |
|
| 188 | 12 | $result[count($result) - 1] .= "\n" . $line; |
|
| 189 | 12 | } |
|
| 190 | } |
||
| 191 | |||
| 192 | 5 | // create proper Tag objects |
|
| 193 | 5 | if (count($result)) { |
|
| 194 | 5 | $this->tags->clear(); |
|
| 195 | 5 | foreach ($result as $line) { |
|
| 196 | $this->tags->add($this->parseTag($line)); |
||
| 197 | 1 | } |
|
| 198 | } |
||
| 199 | } |
||
| 200 | } |
||
| 201 | |||
| 202 | 5 | /** |
|
| 203 | 5 | * Checks whether the given line is a tag line (= starts with @) or not |
|
| 204 | 5 | * |
|
| 205 | 5 | * @param string $line |
|
| 206 | * |
||
| 207 | * @return bool |
||
| 208 | */ |
||
| 209 | 11 | protected function isTagLine(string $line): bool { |
|
| 210 | return str_starts_with($line, '@'); |
||
| 211 | } |
||
| 212 | |||
| 213 | /** |
||
| 214 | * Parses an individual tag line |
||
| 215 | * |
||
| 216 | * @param string $line |
||
| 217 | * |
||
| 218 | 5 | * @throws InvalidArgumentException |
|
| 219 | 5 | * |
|
| 220 | * @return AbstractTag |
||
| 221 | */ |
||
| 222 | protected function parseTag(string $line): AbstractTag { |
||
| 223 | $matches = []; |
||
| 224 | if (!preg_match('/^@(' . self::REGEX_TAGNAME . ')(?:\s*([^\s].*)|$)?/us', $line, $matches)) { |
||
| 225 | throw new InvalidArgumentException('Invalid tag line detected: ' . $line); |
||
| 226 | } |
||
| 227 | |||
| 228 | $tagName = $matches[1]; |
||
| 229 | $content = $matches[2] ?? ''; |
||
| 230 | |||
| 231 | 5 | return TagFactory::create($tagName, $content); |
|
| 232 | 5 | } |
|
| 233 | 5 | ||
| 234 | 1 | /** |
|
| 235 | * Returns the short description |
||
| 236 | * |
||
| 237 | 4 | * @return string the short description |
|
| 238 | 4 | */ |
|
| 239 | public function getShortDescription(): string { |
||
| 240 | 4 | return $this->shortDescription; |
|
| 241 | } |
||
| 242 | |||
| 243 | /** |
||
| 244 | * Sets the short description |
||
| 245 | * |
||
| 246 | * @param string $description the new description |
||
| 247 | * |
||
| 248 | 3 | * @return $this |
|
| 249 | 3 | */ |
|
| 250 | public function setShortDescription(string $description = ''): self { |
||
| 251 | $this->shortDescription = $description; |
||
| 252 | |||
| 253 | return $this; |
||
| 254 | } |
||
| 255 | |||
| 256 | /** |
||
| 257 | * Returns the long description |
||
| 258 | * |
||
| 259 | 2 | * @return string the long description |
|
| 260 | 2 | */ |
|
| 261 | public function getLongDescription(): string { |
||
| 262 | 2 | return $this->longDescription; |
|
| 263 | } |
||
| 264 | |||
| 265 | /** |
||
| 266 | * Sets the long description |
||
| 267 | * |
||
| 268 | * @param string $description the new description |
||
| 269 | * |
||
| 270 | 2 | * @return $this |
|
| 271 | 2 | */ |
|
| 272 | public function setLongDescription(string $description = ''): self { |
||
| 273 | $this->longDescription = $description; |
||
| 274 | |||
| 275 | return $this; |
||
| 276 | } |
||
| 277 | |||
| 278 | /** |
||
| 279 | * Adds a tag to this docblock |
||
| 280 | * |
||
| 281 | 2 | * @param AbstractTag $tag |
|
| 282 | 2 | * |
|
| 283 | * @return $this |
||
| 284 | 2 | */ |
|
| 285 | public function appendTag(AbstractTag $tag): self { |
||
| 286 | $this->tags->add($tag); |
||
| 287 | |||
| 288 | return $this; |
||
| 289 | } |
||
| 290 | |||
| 291 | /** |
||
| 292 | * Removes tags (by tag name) |
||
| 293 | * |
||
| 294 | 4 | * @param string $tagName |
|
| 295 | 4 | */ |
|
| 296 | public function removeTags(string $tagName = ''): void { |
||
| 297 | 4 | $this->tags = $this->tags->filter(function (AbstractTag $tag) use ($tagName): bool { |
|
| 298 | return $tagName !== $tag->getTagName(); |
||
| 299 | }); |
||
| 300 | } |
||
| 301 | |||
| 302 | /** |
||
| 303 | * Checks whether a tag is present |
||
| 304 | * |
||
| 305 | 1 | * @param string $tagName |
|
| 306 | * |
||
| 307 | 1 | * @return bool |
|
| 308 | 1 | */ |
|
| 309 | 1 | public function hasTag(string $tagName): bool { |
|
| 310 | return $this->tags->search($tagName, |
||
| 311 | fn (AbstractTag $tag, string $query): bool => $tag->getTagName() === $query |
||
| 312 | ); |
||
| 313 | } |
||
| 314 | |||
| 315 | /** |
||
| 316 | * Gets tags (by tag name) |
||
| 317 | * |
||
| 318 | 2 | * @param string $tagName |
|
| 319 | * |
||
| 320 | 2 | * @return ArrayList the tags |
|
| 321 | 2 | */ |
|
| 322 | public function getTags(string $tagName = ''): ArrayList { |
||
| 323 | return $tagName === '' ? $this->tags : $this->tags->filter( |
||
| 324 | fn (AbstractTag $tag): bool => $tag->getTagName() === $tagName |
||
| 325 | ); |
||
| 326 | } |
||
| 327 | |||
| 328 | /** |
||
| 329 | * A list of tags sorted by tag-name |
||
| 330 | * |
||
| 331 | 2 | * @return ArrayList |
|
| 332 | */ |
||
| 333 | 2 | public function getSortedTags(): ArrayList { |
|
| 334 | 2 | $this->comparator = $this->comparator ?? new TagNameComparator(); |
|
| 335 | |||
| 336 | // 1) group by tag name |
||
| 337 | $group = new Map(); |
||
| 338 | /** @var AbstractTag $tag */ |
||
| 339 | foreach ($this->tags->toArray() as $tag) { |
||
| 340 | if (!$group->has($tag->getTagName())) { |
||
| 341 | $group->set($tag->getTagName(), new ArrayList()); |
||
| 342 | 6 | } |
|
| 343 | 6 | ||
| 344 | 6 | /** @var ArrayList $list */ |
|
| 345 | $list = $group->get($tag->getTagName()); |
||
| 346 | $list->add($tag); |
||
| 347 | } |
||
| 348 | 6 | ||
| 349 | 6 | // 2) Sort the group by tag name |
|
| 350 | 5 | $group->sortKeys(new TagNameComparator()); |
|
| 351 | 5 | ||
| 352 | // 3) flatten the group |
||
| 353 | $sorted = new ArrayList(); |
||
| 354 | 5 | /** @var array $tags */ |
|
| 355 | foreach ($group->values()->toArray() as $tags) { |
||
| 356 | $sorted->add(...$tags); |
||
| 357 | } |
||
| 358 | 6 | ||
| 359 | return $sorted; |
||
| 360 | } |
||
| 361 | 6 | ||
| 362 | 6 | /** |
|
| 363 | 5 | * Returns true when there is no content in the docblock |
|
| 364 | * |
||
| 365 | * @return bool |
||
| 366 | 6 | */ |
|
| 367 | public function isEmpty(): bool { |
||
| 368 | return $this->shortDescription === '' |
||
| 369 | && $this->longDescription === '' |
||
| 370 | && $this->tags->size() === 0; |
||
| 371 | } |
||
| 372 | |||
| 373 | /** |
||
| 374 | 1 | * Returns the string version of the docblock |
|
| 375 | 1 | * |
|
| 376 | 1 | * @return string |
|
| 377 | 1 | */ |
|
| 378 | public function toString(): string { |
||
| 379 | $docblock = "/**\n"; |
||
| 380 | |||
| 381 | // short description |
||
| 382 | $short = trim($this->shortDescription); |
||
| 383 | if ($short !== '') { |
||
| 384 | $docblock .= $this->writeLines(explode("\n", $short)); |
||
| 385 | 5 | } |
|
| 386 | 5 | ||
| 387 | // short description |
||
| 388 | $long = trim($this->longDescription); |
||
| 389 | 5 | if ($long !== '') { |
|
| 390 | 5 | $docblock .= $this->writeLines(explode("\n", $long), !empty($short)); |
|
| 391 | 3 | } |
|
| 392 | |||
| 393 | // tags |
||
| 394 | $tags = $this->getSortedTags()->map(function (AbstractTag $tag): string { |
||
| 395 | 5 | return (string) $tag; |
|
| 396 | 5 | }); |
|
| 397 | 2 | ||
| 398 | if (!$tags->isEmpty()) { |
||
| 399 | /** @psalm-suppress MixedArgumentTypeCoercion */ |
||
| 400 | $docblock .= $this->writeLines($tags->toArray(), $short !== '' || $long !== ''); |
||
| 401 | } |
||
| 402 | 4 | ||
| 403 | 5 | $docblock .= ' */'; |
|
| 404 | |||
| 405 | 5 | return $docblock; |
|
| 406 | 4 | } |
|
| 407 | |||
| 408 | /** |
||
| 409 | 5 | * Writes multiple lines with ' * ' prefixed for docblock |
|
| 410 | * |
||
| 411 | 5 | * @param string[] $lines the lines to be written |
|
| 412 | * @param bool $newline if a new line should be added before |
||
| 413 | * |
||
| 414 | * @return string the lines as string |
||
| 415 | */ |
||
| 416 | protected function writeLines(array $lines, bool $newline = false): string { |
||
| 417 | $docblock = ''; |
||
| 418 | if ($newline) { |
||
| 419 | $docblock .= " *\n"; |
||
| 420 | } |
||
| 421 | |||
| 422 | 5 | foreach ($lines as $line) { |
|
| 423 | 5 | if (str_contains($line, "\n")) { |
|
| 424 | 5 | $sublines = explode("\n", $line); |
|
| 425 | 3 | $line = array_shift($sublines); |
|
| 426 | $docblock .= " * $line\n"; |
||
| 427 | $docblock .= $this->writeLines($sublines); |
||
| 428 | 5 | } else { |
|
| 429 | 5 | $docblock .= " * $line\n"; |
|
| 430 | 1 | } |
|
| 431 | 1 | } |
|
| 432 | 1 | ||
| 433 | 1 | return $docblock; |
|
| 434 | } |
||
| 435 | 5 | ||
| 436 | /** |
||
| 437 | * Magic toString() method |
||
| 438 | * |
||
| 439 | 5 | * @return string |
|
| 440 | */ |
||
| 441 | public function __toString(): string { |
||
| 442 | return $this->toString(); |
||
| 443 | } |
||
| 444 | } |
||
| 445 |