Complex classes like AbstractCsv 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 AbstractCsv, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 34 | abstract class AbstractCsv implements ByteSequence |
||
| 35 | { |
||
| 36 | /** |
||
| 37 | * The stream filter mode (read or write). |
||
| 38 | * |
||
| 39 | * @var int |
||
| 40 | */ |
||
| 41 | protected $stream_filter_mode; |
||
| 42 | |||
| 43 | /** |
||
| 44 | * collection of stream filters. |
||
| 45 | * |
||
| 46 | * @var bool[] |
||
| 47 | */ |
||
| 48 | protected $stream_filters = []; |
||
| 49 | |||
| 50 | /** |
||
| 51 | * The CSV document BOM sequence. |
||
| 52 | * |
||
| 53 | * @var string|null |
||
| 54 | */ |
||
| 55 | protected $input_bom = null; |
||
| 56 | |||
| 57 | /** |
||
| 58 | * The Output file BOM character. |
||
| 59 | * |
||
| 60 | * @var string |
||
| 61 | */ |
||
| 62 | protected $output_bom = ''; |
||
| 63 | |||
| 64 | /** |
||
| 65 | * the field delimiter (one character only). |
||
| 66 | * |
||
| 67 | * @var string |
||
| 68 | */ |
||
| 69 | protected $delimiter = ','; |
||
| 70 | |||
| 71 | /** |
||
| 72 | * the field enclosure character (one character only). |
||
| 73 | * |
||
| 74 | * @var string |
||
| 75 | */ |
||
| 76 | protected $enclosure = '"'; |
||
| 77 | |||
| 78 | /** |
||
| 79 | * the field escape character (one character only). |
||
| 80 | * |
||
| 81 | * @var string |
||
| 82 | */ |
||
| 83 | protected $escape = '\\'; |
||
| 84 | |||
| 85 | /** |
||
| 86 | * The CSV document. |
||
| 87 | * |
||
| 88 | * @var SplFileObject|Stream |
||
| 89 | */ |
||
| 90 | protected $document; |
||
| 91 | |||
| 92 | /** |
||
| 93 | * Tells whether the Input BOM must be included or skipped. |
||
| 94 | * |
||
| 95 | * @var bool |
||
| 96 | */ |
||
| 97 | protected $is_input_bom_included = false; |
||
| 98 | |||
| 99 | /** |
||
| 100 | * New instance. |
||
| 101 | * |
||
| 102 | * @param SplFileObject|Stream $document The CSV Object instance |
||
| 103 | */ |
||
| 104 | 42 | protected function __construct($document) |
|
| 105 | { |
||
| 106 | 42 | $this->document = $document; |
|
| 107 | 42 | [$this->delimiter, $this->enclosure, $this->escape] = $this->document->getCsvControl(); |
|
| 108 | 42 | $this->resetProperties(); |
|
| 109 | 42 | } |
|
| 110 | |||
| 111 | /** |
||
| 112 | * Reset dynamic object properties to improve performance. |
||
| 113 | */ |
||
| 114 | 39 | protected function resetProperties(): void |
|
| 115 | { |
||
| 116 | 39 | } |
|
| 117 | |||
| 118 | /** |
||
| 119 | * {@inheritdoc} |
||
| 120 | */ |
||
| 121 | 42 | public function __destruct() |
|
| 122 | { |
||
| 123 | 42 | unset($this->document); |
|
| 124 | 42 | } |
|
| 125 | |||
| 126 | /** |
||
| 127 | * {@inheritdoc} |
||
| 128 | */ |
||
| 129 | 3 | public function __clone() |
|
| 130 | { |
||
| 131 | 3 | throw new Exception(sprintf('An object of class %s cannot be cloned', static::class)); |
|
| 132 | } |
||
| 133 | |||
| 134 | /** |
||
| 135 | * Return a new instance from a SplFileObject. |
||
| 136 | * |
||
| 137 | * @return static |
||
| 138 | */ |
||
| 139 | 39 | public static function createFromFileObject(SplFileObject $file) |
|
| 140 | { |
||
| 141 | 39 | return new static($file); |
|
| 142 | } |
||
| 143 | |||
| 144 | /** |
||
| 145 | * Return a new instance from a PHP resource stream. |
||
| 146 | * |
||
| 147 | * @param resource $stream |
||
| 148 | * |
||
| 149 | * @return static |
||
| 150 | */ |
||
| 151 | 6 | public static function createFromStream($stream) |
|
| 152 | { |
||
| 153 | 6 | return new static(new Stream($stream)); |
|
| 154 | } |
||
| 155 | |||
| 156 | /** |
||
| 157 | * Return a new instance from a string. |
||
| 158 | * |
||
| 159 | * @return static |
||
| 160 | */ |
||
| 161 | 9 | public static function createFromString(string $content = '') |
|
| 162 | { |
||
| 163 | 9 | return new static(Stream::createFromString($content)); |
|
| 164 | } |
||
| 165 | |||
| 166 | /** |
||
| 167 | * Return a new instance from a file path. |
||
| 168 | * |
||
| 169 | * @param resource|null $context the resource context |
||
| 170 | * |
||
| 171 | * @return static |
||
| 172 | */ |
||
| 173 | 6 | public static function createFromPath(string $path, string $open_mode = 'r+', $context = null) |
|
| 174 | { |
||
| 175 | 6 | return new static(Stream::createFromPath($path, $open_mode, $context)); |
|
| 176 | } |
||
| 177 | |||
| 178 | /** |
||
| 179 | * Returns the current field delimiter. |
||
| 180 | */ |
||
| 181 | 21 | public function getDelimiter(): string |
|
| 182 | { |
||
| 183 | 21 | return $this->delimiter; |
|
| 184 | } |
||
| 185 | |||
| 186 | /** |
||
| 187 | * Returns the current field enclosure. |
||
| 188 | */ |
||
| 189 | 3 | public function getEnclosure(): string |
|
| 190 | { |
||
| 191 | 3 | return $this->enclosure; |
|
| 192 | } |
||
| 193 | |||
| 194 | /** |
||
| 195 | * Returns the pathname of the underlying document. |
||
| 196 | */ |
||
| 197 | 12 | public function getPathname(): string |
|
| 198 | { |
||
| 199 | 12 | return $this->document->getPathname(); |
|
| 200 | } |
||
| 201 | |||
| 202 | /** |
||
| 203 | * Returns the current field escape character. |
||
| 204 | */ |
||
| 205 | 3 | public function getEscape(): string |
|
| 206 | { |
||
| 207 | 3 | return $this->escape; |
|
| 208 | } |
||
| 209 | |||
| 210 | /** |
||
| 211 | * Returns the BOM sequence in use on Output methods. |
||
| 212 | */ |
||
| 213 | 3 | public function getOutputBOM(): string |
|
| 214 | { |
||
| 215 | 3 | return $this->output_bom; |
|
| 216 | } |
||
| 217 | |||
| 218 | /** |
||
| 219 | * Returns the BOM sequence of the given CSV. |
||
| 220 | */ |
||
| 221 | 60 | public function getInputBOM(): string |
|
| 222 | { |
||
| 223 | 60 | if (null !== $this->input_bom) { |
|
| 224 | 6 | return $this->input_bom; |
|
| 225 | } |
||
| 226 | |||
| 227 | 60 | $this->document->setFlags(SplFileObject::READ_CSV); |
|
| 228 | 60 | $this->document->rewind(); |
|
| 229 | 60 | $this->input_bom = bom_match((string) $this->document->fread(4)); |
|
| 230 | |||
| 231 | 60 | return $this->input_bom; |
|
| 232 | } |
||
| 233 | |||
| 234 | /** |
||
| 235 | * Returns the stream filter mode. |
||
| 236 | */ |
||
| 237 | 3 | public function getStreamFilterMode(): int |
|
| 238 | { |
||
| 239 | 3 | return $this->stream_filter_mode; |
|
| 240 | } |
||
| 241 | |||
| 242 | /** |
||
| 243 | * Tells whether the stream filter capabilities can be used. |
||
| 244 | */ |
||
| 245 | 6 | public function supportsStreamFilter(): bool |
|
| 246 | { |
||
| 247 | 6 | return $this->document instanceof Stream; |
|
| 248 | } |
||
| 249 | |||
| 250 | /** |
||
| 251 | * Tell whether the specify stream filter is attach to the current stream. |
||
| 252 | */ |
||
| 253 | 3 | public function hasStreamFilter(string $filtername): bool |
|
| 254 | { |
||
| 255 | 3 | return $this->stream_filters[$filtername] ?? false; |
|
| 256 | } |
||
| 257 | |||
| 258 | /** |
||
| 259 | * Tells whether the BOM can be stripped if presents. |
||
| 260 | */ |
||
| 261 | 3 | public function isInputBOMIncluded(): bool |
|
| 262 | { |
||
| 263 | 3 | return $this->is_input_bom_included; |
|
| 264 | } |
||
| 265 | |||
| 266 | /** |
||
| 267 | * Retuns the CSV document as a Generator of string chunk. |
||
| 268 | * |
||
| 269 | * @param int $length number of bytes read |
||
| 270 | * |
||
| 271 | * @throws Exception if the number of bytes is lesser than 1 |
||
| 272 | */ |
||
| 273 | 18 | public function chunk(int $length): Generator |
|
| 274 | { |
||
| 275 | 18 | if ($length < 1) { |
|
| 276 | 3 | throw new InvalidArgument(sprintf('%s() expects the length to be a positive integer %d given', __METHOD__, $length)); |
|
| 277 | } |
||
| 278 | |||
| 279 | 15 | $input_bom = $this->getInputBOM(); |
|
| 280 | 15 | $this->document->rewind(); |
|
| 281 | 15 | $this->document->setFlags(0); |
|
| 282 | 15 | $this->document->fseek(strlen($input_bom)); |
|
| 283 | /** @var array<int, string> $chunks */ |
||
| 284 | 15 | $chunks = str_split($this->output_bom.$this->document->fread($length), $length); |
|
| 285 | 15 | foreach ($chunks as $chunk) { |
|
| 286 | 15 | yield $chunk; |
|
| 287 | } |
||
| 288 | |||
| 289 | 15 | while ($this->document->valid()) { |
|
| 290 | 3 | yield $this->document->fread($length); |
|
| 291 | } |
||
| 292 | 15 | } |
|
| 293 | |||
| 294 | /** |
||
| 295 | * DEPRECATION WARNING! This method will be removed in the next major point release. |
||
| 296 | * |
||
| 297 | * @deprecated deprecated since version 9.1.0 |
||
| 298 | * @see AbstractCsv::getContent |
||
| 299 | * |
||
| 300 | * Retrieves the CSV content |
||
| 301 | */ |
||
| 302 | 3 | public function __toString(): string |
|
| 303 | { |
||
| 304 | 3 | return $this->getContent(); |
|
| 305 | } |
||
| 306 | |||
| 307 | /** |
||
| 308 | * Retrieves the CSV content. |
||
| 309 | */ |
||
| 310 | 21 | public function getContent(): string |
|
| 311 | { |
||
| 312 | 21 | $raw = ''; |
|
| 313 | 21 | foreach ($this->chunk(8192) as $chunk) { |
|
| 314 | 21 | $raw .= $chunk; |
|
| 315 | } |
||
| 316 | |||
| 317 | 21 | return $raw; |
|
| 318 | } |
||
| 319 | |||
| 320 | /** |
||
| 321 | * Outputs all data on the CSV file. |
||
| 322 | * |
||
| 323 | * @return int Returns the number of characters read from the handle |
||
| 324 | * and passed through to the output. |
||
| 325 | */ |
||
| 326 | 12 | public function output(string $filename = null): int |
|
| 327 | { |
||
| 328 | 12 | if (null !== $filename) { |
|
| 329 | 9 | $this->sendHeaders($filename); |
|
| 330 | } |
||
| 331 | |||
| 332 | 9 | $this->document->rewind(); |
|
| 333 | 9 | if (!$this->is_input_bom_included) { |
|
| 334 | 9 | $this->document->fseek(strlen($this->getInputBOM())); |
|
| 335 | } |
||
| 336 | |||
| 337 | 9 | echo $this->output_bom; |
|
| 338 | |||
| 339 | 9 | return strlen($this->output_bom) + (int) $this->document->fpassthru(); |
|
| 340 | } |
||
| 341 | |||
| 342 | /** |
||
| 343 | * Send the CSV headers. |
||
| 344 | * |
||
| 345 | * Adapted from Symfony\Component\HttpFoundation\ResponseHeaderBag::makeDisposition |
||
| 346 | * |
||
| 347 | * @throws Exception if the submitted header is invalid according to RFC 6266 |
||
| 348 | * |
||
| 349 | * @see https://tools.ietf.org/html/rfc6266#section-4.3 |
||
| 350 | */ |
||
| 351 | 9 | protected function sendHeaders(string $filename): void |
|
| 352 | { |
||
| 353 | 9 | if (strlen($filename) != strcspn($filename, '\\/')) { |
|
| 354 | 3 | throw new InvalidArgument('The filename cannot contain the "/" and "\\" characters.'); |
|
| 355 | } |
||
| 356 | |||
| 357 | 6 | $flag = FILTER_FLAG_STRIP_LOW; |
|
| 358 | 6 | if (strlen($filename) !== mb_strlen($filename)) { |
|
| 359 | 3 | $flag |= FILTER_FLAG_STRIP_HIGH; |
|
| 360 | } |
||
| 361 | |||
| 362 | /** @var string $filtered_name */ |
||
| 363 | 6 | $filtered_name = filter_var($filename, FILTER_SANITIZE_STRING, $flag); |
|
| 364 | 6 | $filename_fallback = str_replace('%', '', $filtered_name); |
|
| 365 | |||
| 366 | 6 | $disposition = sprintf('attachment; filename="%s"', str_replace('"', '\\"', $filename_fallback)); |
|
| 367 | 6 | if ($filename !== $filename_fallback) { |
|
| 368 | 3 | $disposition .= sprintf("; filename*=utf-8''%s", rawurlencode($filename)); |
|
| 369 | } |
||
| 370 | |||
| 371 | 6 | header('Content-Type: text/csv'); |
|
| 372 | 6 | header('Content-Transfer-Encoding: binary'); |
|
| 373 | 6 | header('Content-Description: File Transfer'); |
|
| 374 | 6 | header('Content-Disposition: '.$disposition); |
|
| 375 | 6 | } |
|
| 376 | |||
| 377 | public function detectDelimiter(): string |
||
| 378 | { |
||
| 379 | $this->document->rewind(); |
||
| 380 | $delimiters = array( |
||
| 381 | ';' => 0, |
||
| 382 | ',' => 0, |
||
| 383 | "\t" => 0, |
||
| 384 | 21 | "|" => 0 |
|
| 385 | ); |
||
| 386 | 21 | ||
| 387 | 12 | $firstLine = $this->document->fgets(); |
|
| 388 | foreach ($delimiters as $delimiter => &$count) { |
||
| 389 | $count = count(str_getcsv($firstLine, $delimiter)); |
||
| 390 | 18 | } |
|
| 391 | 18 | ||
| 392 | 18 | if(max($delimiters) == 0) |
|
| 393 | throw new DelimiterNotFound("Standard delimiter not found in file"); |
||
| 394 | 18 | ||
| 395 | $this->delimiter = array_search(max($delimiters), $delimiters); |
||
| 396 | return $this->delimiter |
||
| 397 | 3 | } |
|
|
|
|||
| 398 | |||
| 399 | /** |
||
| 400 | * Sets the field delimiter. |
||
| 401 | * |
||
| 402 | * @throws Exception If the Csv control character is not one character only. |
||
| 403 | * |
||
| 404 | * @return static |
||
| 405 | */ |
||
| 406 | public function setDelimiter(string $delimiter): self |
||
| 407 | 3 | { |
|
| 408 | if ($delimiter === $this->delimiter) { |
||
| 409 | 3 | return $this; |
|
| 410 | 3 | } |
|
| 411 | |||
| 412 | if (1 === strlen($delimiter)) { |
||
| 413 | 3 | $this->delimiter = $delimiter; |
|
| 414 | 3 | $this->resetProperties(); |
|
| 415 | 3 | ||
| 416 | return $this; |
||
| 417 | 3 | } |
|
| 418 | |||
| 419 | throw new InvalidArgument(sprintf('%s() expects delimiter to be a single character %s given', __METHOD__, $delimiter)); |
||
| 420 | 3 | } |
|
| 421 | |||
| 422 | /** |
||
| 423 | * Sets the field enclosure. |
||
| 424 | * |
||
| 425 | * @throws Exception If the Csv control character is not one character only. |
||
| 426 | * |
||
| 427 | * @return static |
||
| 428 | */ |
||
| 429 | public function setEnclosure(string $enclosure): self |
||
| 430 | 3 | { |
|
| 431 | if ($enclosure === $this->enclosure) { |
||
| 432 | 3 | return $this; |
|
| 433 | 3 | } |
|
| 434 | |||
| 435 | if (1 === strlen($enclosure)) { |
||
| 436 | 3 | $this->enclosure = $enclosure; |
|
| 437 | 3 | $this->resetProperties(); |
|
| 438 | 3 | ||
| 439 | return $this; |
||
| 440 | 3 | } |
|
| 441 | |||
| 442 | throw new InvalidArgument(sprintf('%s() expects enclosure to be a single character %s given', __METHOD__, $enclosure)); |
||
| 443 | 3 | } |
|
| 444 | |||
| 445 | /** |
||
| 446 | * Sets the field escape character. |
||
| 447 | * |
||
| 448 | * @throws Exception If the Csv control character is not one character only. |
||
| 449 | * |
||
| 450 | * @return static |
||
| 451 | 3 | */ |
|
| 452 | public function setEscape(string $escape): self |
||
| 453 | 3 | { |
|
| 454 | if ($escape === $this->escape) { |
||
| 455 | 3 | return $this; |
|
| 456 | } |
||
| 457 | |||
| 458 | if ('' === $escape || 1 === strlen($escape)) { |
||
| 459 | $this->escape = $escape; |
||
| 460 | $this->resetProperties(); |
||
| 461 | |||
| 462 | return $this; |
||
| 463 | 6 | } |
|
| 464 | |||
| 465 | 6 | throw new InvalidArgument(sprintf('%s() expects escape to be a single character or the empty string %s given', __METHOD__, $escape)); |
|
| 466 | } |
||
| 467 | 6 | ||
| 468 | /** |
||
| 469 | * Enables BOM Stripping. |
||
| 470 | * |
||
| 471 | * @return static |
||
| 472 | */ |
||
| 473 | public function skipInputBOM(): self |
||
| 474 | { |
||
| 475 | 9 | $this->is_input_bom_included = false; |
|
| 476 | |||
| 477 | 9 | return $this; |
|
| 478 | } |
||
| 479 | 9 | ||
| 480 | /** |
||
| 481 | * Disables skipping Input BOM. |
||
| 482 | * |
||
| 483 | * @return static |
||
| 484 | */ |
||
| 485 | public function includeInputBOM(): self |
||
| 486 | { |
||
| 487 | $this->is_input_bom_included = true; |
||
| 488 | |||
| 489 | return $this; |
||
| 490 | } |
||
| 491 | 15 | ||
| 492 | /** |
||
| 493 | 15 | * Sets the BOM sequence to prepend the CSV on output. |
|
| 494 | 3 | * |
|
| 495 | * @return static |
||
| 496 | */ |
||
| 497 | 12 | public function setOutputBOM(string $str): self |
|
| 498 | 9 | { |
|
| 499 | 9 | $this->output_bom = $str; |
|
| 500 | 9 | ||
| 501 | return $this; |
||
| 502 | 9 | } |
|
| 503 | |||
| 504 | /** |
||
| 505 | * append a stream filter. |
||
| 527 |