Matthias Andreas Benkard | b382b10 | 2021-01-02 15:32:21 +0100 | [diff] [blame^] | 1 | <?php |
| 2 | |
| 3 | namespace PhpMimeMailParser; |
| 4 | |
| 5 | use PhpMimeMailParser\Contracts\CharsetManager; |
| 6 | |
| 7 | /** |
| 8 | * Parser of php-mime-mail-parser |
| 9 | * |
| 10 | * Fully Tested Mailparse Extension Wrapper for PHP 5.4+ |
| 11 | * |
| 12 | */ |
| 13 | class Parser |
| 14 | { |
| 15 | /** |
| 16 | * Attachment filename argument option for ->saveAttachments(). |
| 17 | */ |
| 18 | const ATTACHMENT_DUPLICATE_THROW = 'DuplicateThrow'; |
| 19 | const ATTACHMENT_DUPLICATE_SUFFIX = 'DuplicateSuffix'; |
| 20 | const ATTACHMENT_RANDOM_FILENAME = 'RandomFilename'; |
| 21 | |
| 22 | /** |
| 23 | * PHP MimeParser Resource ID |
| 24 | * |
| 25 | * @var resource $resource |
| 26 | */ |
| 27 | protected $resource; |
| 28 | |
| 29 | /** |
| 30 | * A file pointer to email |
| 31 | * |
| 32 | * @var resource $stream |
| 33 | */ |
| 34 | protected $stream; |
| 35 | |
| 36 | /** |
| 37 | * A text of an email |
| 38 | * |
| 39 | * @var string $data |
| 40 | */ |
| 41 | protected $data; |
| 42 | |
| 43 | /** |
| 44 | * Parts of an email |
| 45 | * |
| 46 | * @var array $parts |
| 47 | */ |
| 48 | protected $parts; |
| 49 | |
| 50 | /** |
| 51 | * @var CharsetManager object |
| 52 | */ |
| 53 | protected $charset; |
| 54 | |
| 55 | /** |
| 56 | * Valid stream modes for reading |
| 57 | * |
| 58 | * @var array |
| 59 | */ |
| 60 | protected static $readableModes = [ |
| 61 | 'r', 'r+', 'w+', 'a+', 'x+', 'c+', 'rb', 'r+b', 'w+b', 'a+b', |
| 62 | 'x+b', 'c+b', 'rt', 'r+t', 'w+t', 'a+t', 'x+t', 'c+t' |
| 63 | ]; |
| 64 | |
| 65 | /** |
| 66 | * Stack of middleware registered to process data |
| 67 | * |
| 68 | * @var MiddlewareStack |
| 69 | */ |
| 70 | protected $middlewareStack; |
| 71 | |
| 72 | /** |
| 73 | * Parser constructor. |
| 74 | * |
| 75 | * @param CharsetManager|null $charset |
| 76 | */ |
| 77 | public function __construct(CharsetManager $charset = null) |
| 78 | { |
| 79 | if ($charset == null) { |
| 80 | $charset = new Charset(); |
| 81 | } |
| 82 | |
| 83 | $this->charset = $charset; |
| 84 | $this->middlewareStack = new MiddlewareStack(); |
| 85 | } |
| 86 | |
| 87 | /** |
| 88 | * Free the held resources |
| 89 | * |
| 90 | * @return void |
| 91 | */ |
| 92 | public function __destruct() |
| 93 | { |
| 94 | // clear the email file resource |
| 95 | if (is_resource($this->stream)) { |
| 96 | fclose($this->stream); |
| 97 | } |
| 98 | // clear the MailParse resource |
| 99 | if (is_resource($this->resource)) { |
| 100 | mailparse_msg_free($this->resource); |
| 101 | } |
| 102 | } |
| 103 | |
| 104 | /** |
| 105 | * Set the file path we use to get the email text |
| 106 | * |
| 107 | * @param string $path File path to the MIME mail |
| 108 | * |
| 109 | * @return Parser MimeMailParser Instance |
| 110 | */ |
| 111 | public function setPath($path) |
| 112 | { |
| 113 | if (is_writable($path)) { |
| 114 | $file = fopen($path, 'a+'); |
| 115 | fseek($file, -1, SEEK_END); |
| 116 | if (fread($file, 1) != "\n") { |
| 117 | fwrite($file, PHP_EOL); |
| 118 | } |
| 119 | fclose($file); |
| 120 | } |
| 121 | |
| 122 | // should parse message incrementally from file |
| 123 | $this->resource = mailparse_msg_parse_file($path); |
| 124 | $this->stream = fopen($path, 'r'); |
| 125 | $this->parse(); |
| 126 | |
| 127 | return $this; |
| 128 | } |
| 129 | |
| 130 | /** |
| 131 | * Set the Stream resource we use to get the email text |
| 132 | * |
| 133 | * @param resource $stream |
| 134 | * |
| 135 | * @return Parser MimeMailParser Instance |
| 136 | * @throws Exception |
| 137 | */ |
| 138 | public function setStream($stream) |
| 139 | { |
| 140 | // streams have to be cached to file first |
| 141 | $meta = @stream_get_meta_data($stream); |
| 142 | if (!$meta || !$meta['mode'] || !in_array($meta['mode'], self::$readableModes, true) || $meta['eof']) { |
| 143 | throw new Exception( |
| 144 | 'setStream() expects parameter stream to be readable stream resource.' |
| 145 | ); |
| 146 | } |
| 147 | |
| 148 | /** @var resource $tmp_fp */ |
| 149 | $tmp_fp = tmpfile(); |
| 150 | if ($tmp_fp) { |
| 151 | while (!feof($stream)) { |
| 152 | fwrite($tmp_fp, fread($stream, 2028)); |
| 153 | } |
| 154 | |
| 155 | if (fread($tmp_fp, 1) != "\n") { |
| 156 | fwrite($tmp_fp, PHP_EOL); |
| 157 | } |
| 158 | |
| 159 | fseek($tmp_fp, 0); |
| 160 | $this->stream = &$tmp_fp; |
| 161 | } else { |
| 162 | throw new Exception( |
| 163 | 'Could not create temporary files for attachments. Your tmp directory may be unwritable by PHP.' |
| 164 | ); |
| 165 | } |
| 166 | fclose($stream); |
| 167 | |
| 168 | $this->resource = mailparse_msg_create(); |
| 169 | // parses the message incrementally (low memory usage but slower) |
| 170 | while (!feof($this->stream)) { |
| 171 | mailparse_msg_parse($this->resource, fread($this->stream, 2082)); |
| 172 | } |
| 173 | $this->parse(); |
| 174 | |
| 175 | return $this; |
| 176 | } |
| 177 | |
| 178 | /** |
| 179 | * Set the email text |
| 180 | * |
| 181 | * @param string $data |
| 182 | * |
| 183 | * @return Parser MimeMailParser Instance |
| 184 | */ |
| 185 | public function setText($data) |
| 186 | { |
| 187 | if (empty($data)) { |
| 188 | throw new Exception('You must not call MimeMailParser::setText with an empty string parameter'); |
| 189 | } |
| 190 | |
| 191 | if (substr($data, -1) != "\n") { |
| 192 | $data = $data.PHP_EOL; |
| 193 | } |
| 194 | |
| 195 | $this->resource = mailparse_msg_create(); |
| 196 | // does not parse incrementally, fast memory hog might explode |
| 197 | mailparse_msg_parse($this->resource, $data); |
| 198 | $this->data = $data; |
| 199 | $this->parse(); |
| 200 | |
| 201 | return $this; |
| 202 | } |
| 203 | |
| 204 | /** |
| 205 | * Parse the Message into parts |
| 206 | * |
| 207 | * @return void |
| 208 | */ |
| 209 | protected function parse() |
| 210 | { |
| 211 | $structure = mailparse_msg_get_structure($this->resource); |
| 212 | $this->parts = []; |
| 213 | foreach ($structure as $part_id) { |
| 214 | $part = mailparse_msg_get_part($this->resource, $part_id); |
| 215 | $part_data = mailparse_msg_get_part_data($part); |
| 216 | $mimePart = new MimePart($part_id, $part_data); |
| 217 | // let each middleware parse the part before saving |
| 218 | $this->parts[$part_id] = $this->middlewareStack->parse($mimePart)->getPart(); |
| 219 | } |
| 220 | } |
| 221 | |
| 222 | /** |
| 223 | * Retrieve a specific Email Header, without charset conversion. |
| 224 | * |
| 225 | * @param string $name Header name (case-insensitive) |
| 226 | * |
| 227 | * @return string|bool |
| 228 | * @throws Exception |
| 229 | */ |
| 230 | public function getRawHeader($name) |
| 231 | { |
| 232 | $name = strtolower($name); |
| 233 | if (isset($this->parts[1])) { |
| 234 | $headers = $this->getPart('headers', $this->parts[1]); |
| 235 | |
| 236 | return isset($headers[$name]) ? $headers[$name] : false; |
| 237 | } else { |
| 238 | throw new Exception( |
| 239 | 'setPath() or setText() or setStream() must be called before retrieving email headers.' |
| 240 | ); |
| 241 | } |
| 242 | } |
| 243 | |
| 244 | /** |
| 245 | * Retrieve a specific Email Header |
| 246 | * |
| 247 | * @param string $name Header name (case-insensitive) |
| 248 | * |
| 249 | * @return string|bool |
| 250 | */ |
| 251 | public function getHeader($name) |
| 252 | { |
| 253 | $rawHeader = $this->getRawHeader($name); |
| 254 | if ($rawHeader === false) { |
| 255 | return false; |
| 256 | } |
| 257 | |
| 258 | return $this->decodeHeader($rawHeader); |
| 259 | } |
| 260 | |
| 261 | /** |
| 262 | * Retrieve all mail headers |
| 263 | * |
| 264 | * @return array |
| 265 | * @throws Exception |
| 266 | */ |
| 267 | public function getHeaders() |
| 268 | { |
| 269 | if (isset($this->parts[1])) { |
| 270 | $headers = $this->getPart('headers', $this->parts[1]); |
| 271 | foreach ($headers as &$value) { |
| 272 | if (is_array($value)) { |
| 273 | foreach ($value as &$v) { |
| 274 | $v = $this->decodeSingleHeader($v); |
| 275 | } |
| 276 | } else { |
| 277 | $value = $this->decodeSingleHeader($value); |
| 278 | } |
| 279 | } |
| 280 | |
| 281 | return $headers; |
| 282 | } else { |
| 283 | throw new Exception( |
| 284 | 'setPath() or setText() or setStream() must be called before retrieving email headers.' |
| 285 | ); |
| 286 | } |
| 287 | } |
| 288 | |
| 289 | /** |
| 290 | * Retrieve the raw mail headers as a string |
| 291 | * |
| 292 | * @return string |
| 293 | * @throws Exception |
| 294 | */ |
| 295 | public function getHeadersRaw() |
| 296 | { |
| 297 | if (isset($this->parts[1])) { |
| 298 | return $this->getPartHeader($this->parts[1]); |
| 299 | } else { |
| 300 | throw new Exception( |
| 301 | 'setPath() or setText() or setStream() must be called before retrieving email headers.' |
| 302 | ); |
| 303 | } |
| 304 | } |
| 305 | |
| 306 | /** |
| 307 | * Retrieve the raw Header of a MIME part |
| 308 | * |
| 309 | * @return String |
| 310 | * @param $part Object |
| 311 | * @throws Exception |
| 312 | */ |
| 313 | protected function getPartHeader(&$part) |
| 314 | { |
| 315 | $header = ''; |
| 316 | if ($this->stream) { |
| 317 | $header = $this->getPartHeaderFromFile($part); |
| 318 | } elseif ($this->data) { |
| 319 | $header = $this->getPartHeaderFromText($part); |
| 320 | } |
| 321 | return $header; |
| 322 | } |
| 323 | |
| 324 | /** |
| 325 | * Retrieve the Header from a MIME part from file |
| 326 | * |
| 327 | * @return String Mime Header Part |
| 328 | * @param $part Array |
| 329 | */ |
| 330 | protected function getPartHeaderFromFile(&$part) |
| 331 | { |
| 332 | $start = $part['starting-pos']; |
| 333 | $end = $part['starting-pos-body']; |
| 334 | fseek($this->stream, $start, SEEK_SET); |
| 335 | $header = fread($this->stream, $end - $start); |
| 336 | return $header; |
| 337 | } |
| 338 | |
| 339 | /** |
| 340 | * Retrieve the Header from a MIME part from text |
| 341 | * |
| 342 | * @return String Mime Header Part |
| 343 | * @param $part Array |
| 344 | */ |
| 345 | protected function getPartHeaderFromText(&$part) |
| 346 | { |
| 347 | $start = $part['starting-pos']; |
| 348 | $end = $part['starting-pos-body']; |
| 349 | $header = substr($this->data, $start, $end - $start); |
| 350 | return $header; |
| 351 | } |
| 352 | |
| 353 | /** |
| 354 | * Checks whether a given part ID is a child of another part |
| 355 | * eg. an RFC822 attachment may have one or more text parts |
| 356 | * |
| 357 | * @param string $partId |
| 358 | * @param string $parentPartId |
| 359 | * @return bool |
| 360 | */ |
| 361 | protected function partIdIsChildOfPart($partId, $parentPartId) |
| 362 | { |
| 363 | $parentPartId = $parentPartId.'.'; |
| 364 | return substr($partId, 0, strlen($parentPartId)) == $parentPartId; |
| 365 | } |
| 366 | |
| 367 | /** |
| 368 | * Whether the given part ID is a child of any attachment part in the message. |
| 369 | * |
| 370 | * @param string $checkPartId |
| 371 | * @return bool |
| 372 | */ |
| 373 | protected function partIdIsChildOfAnAttachment($checkPartId) |
| 374 | { |
| 375 | foreach ($this->parts as $partId => $part) { |
| 376 | if ($this->getPart('content-disposition', $part) == 'attachment') { |
| 377 | if ($this->partIdIsChildOfPart($checkPartId, $partId)) { |
| 378 | return true; |
| 379 | } |
| 380 | } |
| 381 | } |
| 382 | return false; |
| 383 | } |
| 384 | |
| 385 | /** |
| 386 | * Returns the email message body in the specified format |
| 387 | * |
| 388 | * @param string $type text, html or htmlEmbedded |
| 389 | * |
| 390 | * @return string Body |
| 391 | * @throws Exception |
| 392 | */ |
| 393 | public function getMessageBody($type = 'text') |
| 394 | { |
| 395 | $mime_types = [ |
| 396 | 'text' => 'text/plain', |
| 397 | 'html' => 'text/html', |
| 398 | 'htmlEmbedded' => 'text/html', |
| 399 | ]; |
| 400 | |
| 401 | if (in_array($type, array_keys($mime_types))) { |
| 402 | $part_type = $type === 'htmlEmbedded' ? 'html' : $type; |
| 403 | $inline_parts = $this->getInlineParts($part_type); |
| 404 | $body = empty($inline_parts) ? '' : $inline_parts[0]; |
| 405 | } else { |
| 406 | throw new Exception( |
| 407 | 'Invalid type specified for getMessageBody(). Expected: text, html or htmlEmbeded.' |
| 408 | ); |
| 409 | } |
| 410 | |
| 411 | if ($type == 'htmlEmbedded') { |
| 412 | $attachments = $this->getAttachments(); |
| 413 | foreach ($attachments as $attachment) { |
| 414 | if ($attachment->getContentID() != '') { |
| 415 | $body = str_replace( |
| 416 | '"cid:'.$attachment->getContentID().'"', |
| 417 | '"'.$this->getEmbeddedData($attachment->getContentID()).'"', |
| 418 | $body |
| 419 | ); |
| 420 | } |
| 421 | } |
| 422 | } |
| 423 | |
| 424 | return $body; |
| 425 | } |
| 426 | |
| 427 | /** |
| 428 | * Returns the embedded data structure |
| 429 | * |
| 430 | * @param string $contentId Content-Id |
| 431 | * |
| 432 | * @return string |
| 433 | */ |
| 434 | protected function getEmbeddedData($contentId) |
| 435 | { |
| 436 | foreach ($this->parts as $part) { |
| 437 | if ($this->getPart('content-id', $part) == $contentId) { |
| 438 | $embeddedData = 'data:'; |
| 439 | $embeddedData .= $this->getPart('content-type', $part); |
| 440 | $embeddedData .= ';'.$this->getPart('transfer-encoding', $part); |
| 441 | $embeddedData .= ','.$this->getPartBody($part); |
| 442 | return $embeddedData; |
| 443 | } |
| 444 | } |
| 445 | return ''; |
| 446 | } |
| 447 | |
| 448 | /** |
| 449 | * Return an array with the following keys display, address, is_group |
| 450 | * |
| 451 | * @param string $name Header name (case-insensitive) |
| 452 | * |
| 453 | * @return array |
| 454 | */ |
| 455 | public function getAddresses($name) |
| 456 | { |
| 457 | $value = $this->getRawHeader($name); |
| 458 | $value = (is_array($value)) ? $value[0] : $value; |
| 459 | $addresses = mailparse_rfc822_parse_addresses($value); |
| 460 | foreach ($addresses as $i => $item) { |
| 461 | $addresses[$i]['display'] = $this->decodeHeader($item['display']); |
| 462 | } |
| 463 | return $addresses; |
| 464 | } |
| 465 | |
| 466 | /** |
| 467 | * Returns the attachments contents in order of appearance |
| 468 | * |
| 469 | * @return Attachment[] |
| 470 | */ |
| 471 | public function getInlineParts($type = 'text') |
| 472 | { |
| 473 | $inline_parts = []; |
| 474 | $mime_types = [ |
| 475 | 'text' => 'text/plain', |
| 476 | 'html' => 'text/html', |
| 477 | ]; |
| 478 | |
| 479 | if (!in_array($type, array_keys($mime_types))) { |
| 480 | throw new Exception('Invalid type specified for getInlineParts(). "type" can either be text or html.'); |
| 481 | } |
| 482 | |
| 483 | foreach ($this->parts as $partId => $part) { |
| 484 | if ($this->getPart('content-type', $part) == $mime_types[$type] |
| 485 | && $this->getPart('content-disposition', $part) != 'attachment' |
| 486 | && !$this->partIdIsChildOfAnAttachment($partId) |
| 487 | ) { |
| 488 | $headers = $this->getPart('headers', $part); |
| 489 | $encodingType = array_key_exists('content-transfer-encoding', $headers) ? |
| 490 | $headers['content-transfer-encoding'] : ''; |
| 491 | $undecoded_body = $this->decodeContentTransfer($this->getPartBody($part), $encodingType); |
| 492 | $inline_parts[] = $this->charset->decodeCharset($undecoded_body, $this->getPartCharset($part)); |
| 493 | } |
| 494 | } |
| 495 | |
| 496 | return $inline_parts; |
| 497 | } |
| 498 | |
| 499 | /** |
| 500 | * Returns the attachments contents in order of appearance |
| 501 | * |
| 502 | * @return Attachment[] |
| 503 | */ |
| 504 | public function getAttachments($include_inline = true) |
| 505 | { |
| 506 | $attachments = []; |
| 507 | $dispositions = $include_inline ? ['attachment', 'inline'] : ['attachment']; |
| 508 | $non_attachment_types = ['text/plain', 'text/html']; |
| 509 | $nonameIter = 0; |
| 510 | |
| 511 | foreach ($this->parts as $part) { |
| 512 | $disposition = $this->getPart('content-disposition', $part); |
| 513 | $filename = 'noname'; |
| 514 | |
| 515 | if (isset($part['disposition-filename'])) { |
| 516 | $filename = $this->decodeHeader($part['disposition-filename']); |
| 517 | } elseif (isset($part['content-name'])) { |
| 518 | // if we have no disposition but we have a content-name, it's a valid attachment. |
| 519 | // we simulate the presence of an attachment disposition with a disposition filename |
| 520 | $filename = $this->decodeHeader($part['content-name']); |
| 521 | $disposition = 'attachment'; |
| 522 | } elseif (in_array($part['content-type'], $non_attachment_types, true) |
| 523 | && $disposition !== 'attachment') { |
| 524 | // it is a message body, no attachment |
| 525 | continue; |
| 526 | } elseif (substr($part['content-type'], 0, 10) !== 'multipart/' |
| 527 | && $part['content-type'] !== 'text/plain; (error)') { |
| 528 | // if we cannot get it by getMessageBody(), we assume it is an attachment |
| 529 | $disposition = 'attachment'; |
| 530 | } |
| 531 | if (in_array($disposition, ['attachment', 'inline']) === false && !empty($disposition)) { |
| 532 | $disposition = 'attachment'; |
| 533 | } |
| 534 | |
| 535 | if (in_array($disposition, $dispositions) === true) { |
| 536 | if ($filename == 'noname') { |
| 537 | $nonameIter++; |
| 538 | $filename = 'noname'.$nonameIter; |
| 539 | } else { |
| 540 | // Escape all potentially unsafe characters from the filename |
| 541 | $filename = preg_replace('((^\.)|\/|[\n|\r|\n\r]|(\.$))', '_', $filename); |
| 542 | } |
| 543 | |
| 544 | $headersAttachments = $this->getPart('headers', $part); |
| 545 | $contentidAttachments = $this->getPart('content-id', $part); |
| 546 | |
| 547 | $attachmentStream = $this->getAttachmentStream($part); |
| 548 | $mimePartStr = $this->getPartComplete($part); |
| 549 | |
| 550 | $attachments[] = new Attachment( |
| 551 | $filename, |
| 552 | $this->getPart('content-type', $part), |
| 553 | $attachmentStream, |
| 554 | $disposition, |
| 555 | $contentidAttachments, |
| 556 | $headersAttachments, |
| 557 | $mimePartStr |
| 558 | ); |
| 559 | } |
| 560 | } |
| 561 | |
| 562 | return $attachments; |
| 563 | } |
| 564 | |
| 565 | /** |
| 566 | * Save attachments in a folder |
| 567 | * |
| 568 | * @param string $attach_dir directory |
| 569 | * @param bool $include_inline |
| 570 | * @param string $filenameStrategy How to generate attachment filenames |
| 571 | * |
| 572 | * @return array Saved attachments paths |
| 573 | * @throws Exception |
| 574 | */ |
| 575 | public function saveAttachments( |
| 576 | $attach_dir, |
| 577 | $include_inline = true, |
| 578 | $filenameStrategy = self::ATTACHMENT_DUPLICATE_SUFFIX |
| 579 | ) { |
| 580 | $attachments = $this->getAttachments($include_inline); |
| 581 | |
| 582 | $attachments_paths = []; |
| 583 | foreach ($attachments as $attachment) { |
| 584 | $attachments_paths[] = $attachment->save($attach_dir, $filenameStrategy); |
| 585 | } |
| 586 | |
| 587 | return $attachments_paths; |
| 588 | } |
| 589 | |
| 590 | /** |
| 591 | * Read the attachment Body and save temporary file resource |
| 592 | * |
| 593 | * @param array $part |
| 594 | * |
| 595 | * @return resource Mime Body Part |
| 596 | * @throws Exception |
| 597 | */ |
| 598 | protected function getAttachmentStream(&$part) |
| 599 | { |
| 600 | /** @var resource $temp_fp */ |
| 601 | $temp_fp = tmpfile(); |
| 602 | |
| 603 | $headers = $this->getPart('headers', $part); |
| 604 | $encodingType = array_key_exists('content-transfer-encoding', $headers) ? |
| 605 | $headers['content-transfer-encoding'] : ''; |
| 606 | |
| 607 | if ($temp_fp) { |
| 608 | if ($this->stream) { |
| 609 | $start = $part['starting-pos-body']; |
| 610 | $end = $part['ending-pos-body']; |
| 611 | fseek($this->stream, $start, SEEK_SET); |
| 612 | $len = $end - $start; |
| 613 | $written = 0; |
| 614 | while ($written < $len) { |
| 615 | $write = $len; |
| 616 | $data = fread($this->stream, $write); |
| 617 | fwrite($temp_fp, $this->decodeContentTransfer($data, $encodingType)); |
| 618 | $written += $write; |
| 619 | } |
| 620 | } elseif ($this->data) { |
| 621 | $attachment = $this->decodeContentTransfer($this->getPartBodyFromText($part), $encodingType); |
| 622 | fwrite($temp_fp, $attachment, strlen($attachment)); |
| 623 | } |
| 624 | fseek($temp_fp, 0, SEEK_SET); |
| 625 | } else { |
| 626 | throw new Exception( |
| 627 | 'Could not create temporary files for attachments. Your tmp directory may be unwritable by PHP.' |
| 628 | ); |
| 629 | } |
| 630 | |
| 631 | return $temp_fp; |
| 632 | } |
| 633 | |
| 634 | /** |
| 635 | * Decode the string from Content-Transfer-Encoding |
| 636 | * |
| 637 | * @param string $encodedString The string in its original encoded state |
| 638 | * @param string $encodingType The encoding type from the Content-Transfer-Encoding header of the part. |
| 639 | * |
| 640 | * @return string The decoded string |
| 641 | */ |
| 642 | protected function decodeContentTransfer($encodedString, $encodingType) |
| 643 | { |
| 644 | if (is_array($encodingType)) { |
| 645 | $encodingType = $encodingType[0]; |
| 646 | } |
| 647 | |
| 648 | $encodingType = strtolower($encodingType); |
| 649 | if ($encodingType == 'base64') { |
| 650 | return base64_decode($encodedString); |
| 651 | } elseif ($encodingType == 'quoted-printable') { |
| 652 | return quoted_printable_decode($encodedString); |
| 653 | } else { |
| 654 | return $encodedString; |
| 655 | } |
| 656 | } |
| 657 | |
| 658 | /** |
| 659 | * $input can be a string or array |
| 660 | * |
| 661 | * @param string|array $input |
| 662 | * |
| 663 | * @return string |
| 664 | */ |
| 665 | protected function decodeHeader($input) |
| 666 | { |
| 667 | //Sometimes we have 2 label From so we take only the first |
| 668 | if (is_array($input)) { |
| 669 | return $this->decodeSingleHeader($input[0]); |
| 670 | } |
| 671 | |
| 672 | return $this->decodeSingleHeader($input); |
| 673 | } |
| 674 | |
| 675 | /** |
| 676 | * Decodes a single header (= string) |
| 677 | * |
| 678 | * @param string $input |
| 679 | * |
| 680 | * @return string |
| 681 | */ |
| 682 | protected function decodeSingleHeader($input) |
| 683 | { |
| 684 | // For each encoded-word... |
| 685 | while (preg_match('/(=\?([^?]+)\?(q|b)\?([^?]*)\?=)((\s+)=\?)?/i', $input, $matches)) { |
| 686 | $encoded = $matches[1]; |
| 687 | $charset = $matches[2]; |
| 688 | $encoding = $matches[3]; |
| 689 | $text = $matches[4]; |
| 690 | $space = isset($matches[6]) ? $matches[6] : ''; |
| 691 | |
| 692 | switch (strtolower($encoding)) { |
| 693 | case 'b': |
| 694 | $text = $this->decodeContentTransfer($text, 'base64'); |
| 695 | break; |
| 696 | |
| 697 | case 'q': |
| 698 | $text = str_replace('_', ' ', $text); |
| 699 | preg_match_all('/=([a-f0-9]{2})/i', $text, $matches); |
| 700 | foreach ($matches[1] as $value) { |
| 701 | $text = str_replace('='.$value, chr(hexdec($value)), $text); |
| 702 | } |
| 703 | break; |
| 704 | } |
| 705 | |
| 706 | $text = $this->charset->decodeCharset($text, $this->charset->getCharsetAlias($charset)); |
| 707 | $input = str_replace($encoded.$space, $text, $input); |
| 708 | } |
| 709 | |
| 710 | return $input; |
| 711 | } |
| 712 | |
| 713 | /** |
| 714 | * Return the charset of the MIME part |
| 715 | * |
| 716 | * @param array $part |
| 717 | * |
| 718 | * @return string |
| 719 | */ |
| 720 | protected function getPartCharset($part) |
| 721 | { |
| 722 | if (isset($part['charset'])) { |
| 723 | return $this->charset->getCharsetAlias($part['charset']); |
| 724 | } else { |
| 725 | return 'us-ascii'; |
| 726 | } |
| 727 | } |
| 728 | |
| 729 | /** |
| 730 | * Retrieve a specified MIME part |
| 731 | * |
| 732 | * @param string $type |
| 733 | * @param array $parts |
| 734 | * |
| 735 | * @return string|array |
| 736 | */ |
| 737 | protected function getPart($type, $parts) |
| 738 | { |
| 739 | return (isset($parts[$type])) ? $parts[$type] : false; |
| 740 | } |
| 741 | |
| 742 | /** |
| 743 | * Retrieve the Body of a MIME part |
| 744 | * |
| 745 | * @param array $part |
| 746 | * |
| 747 | * @return string |
| 748 | */ |
| 749 | protected function getPartBody(&$part) |
| 750 | { |
| 751 | $body = ''; |
| 752 | if ($this->stream) { |
| 753 | $body = $this->getPartBodyFromFile($part); |
| 754 | } elseif ($this->data) { |
| 755 | $body = $this->getPartBodyFromText($part); |
| 756 | } |
| 757 | |
| 758 | return $body; |
| 759 | } |
| 760 | |
| 761 | /** |
| 762 | * Retrieve the Body from a MIME part from file |
| 763 | * |
| 764 | * @param array $part |
| 765 | * |
| 766 | * @return string Mime Body Part |
| 767 | */ |
| 768 | protected function getPartBodyFromFile(&$part) |
| 769 | { |
| 770 | $start = $part['starting-pos-body']; |
| 771 | $end = $part['ending-pos-body']; |
| 772 | $body = ''; |
| 773 | if ($end - $start > 0) { |
| 774 | fseek($this->stream, $start, SEEK_SET); |
| 775 | $body = fread($this->stream, $end - $start); |
| 776 | } |
| 777 | |
| 778 | return $body; |
| 779 | } |
| 780 | |
| 781 | /** |
| 782 | * Retrieve the Body from a MIME part from text |
| 783 | * |
| 784 | * @param array $part |
| 785 | * |
| 786 | * @return string Mime Body Part |
| 787 | */ |
| 788 | protected function getPartBodyFromText(&$part) |
| 789 | { |
| 790 | $start = $part['starting-pos-body']; |
| 791 | $end = $part['ending-pos-body']; |
| 792 | |
| 793 | return substr($this->data, $start, $end - $start); |
| 794 | } |
| 795 | |
| 796 | /** |
| 797 | * Retrieve the content of a MIME part |
| 798 | * |
| 799 | * @param array $part |
| 800 | * |
| 801 | * @return string |
| 802 | */ |
| 803 | protected function getPartComplete(&$part) |
| 804 | { |
| 805 | $body = ''; |
| 806 | if ($this->stream) { |
| 807 | $body = $this->getPartFromFile($part); |
| 808 | } elseif ($this->data) { |
| 809 | $body = $this->getPartFromText($part); |
| 810 | } |
| 811 | |
| 812 | return $body; |
| 813 | } |
| 814 | |
| 815 | /** |
| 816 | * Retrieve the content from a MIME part from file |
| 817 | * |
| 818 | * @param array $part |
| 819 | * |
| 820 | * @return string Mime Content |
| 821 | */ |
| 822 | protected function getPartFromFile(&$part) |
| 823 | { |
| 824 | $start = $part['starting-pos']; |
| 825 | $end = $part['ending-pos']; |
| 826 | $body = ''; |
| 827 | if ($end - $start > 0) { |
| 828 | fseek($this->stream, $start, SEEK_SET); |
| 829 | $body = fread($this->stream, $end - $start); |
| 830 | } |
| 831 | |
| 832 | return $body; |
| 833 | } |
| 834 | |
| 835 | /** |
| 836 | * Retrieve the content from a MIME part from text |
| 837 | * |
| 838 | * @param array $part |
| 839 | * |
| 840 | * @return string Mime Content |
| 841 | */ |
| 842 | protected function getPartFromText(&$part) |
| 843 | { |
| 844 | $start = $part['starting-pos']; |
| 845 | $end = $part['ending-pos']; |
| 846 | |
| 847 | return substr($this->data, $start, $end - $start); |
| 848 | } |
| 849 | |
| 850 | /** |
| 851 | * Retrieve the resource |
| 852 | * |
| 853 | * @return resource resource |
| 854 | */ |
| 855 | public function getResource() |
| 856 | { |
| 857 | return $this->resource; |
| 858 | } |
| 859 | |
| 860 | /** |
| 861 | * Retrieve the file pointer to email |
| 862 | * |
| 863 | * @return resource stream |
| 864 | */ |
| 865 | public function getStream() |
| 866 | { |
| 867 | return $this->stream; |
| 868 | } |
| 869 | |
| 870 | /** |
| 871 | * Retrieve the text of an email |
| 872 | * |
| 873 | * @return string data |
| 874 | */ |
| 875 | public function getData() |
| 876 | { |
| 877 | return $this->data; |
| 878 | } |
| 879 | |
| 880 | /** |
| 881 | * Retrieve the parts of an email |
| 882 | * |
| 883 | * @return array parts |
| 884 | */ |
| 885 | public function getParts() |
| 886 | { |
| 887 | return $this->parts; |
| 888 | } |
| 889 | |
| 890 | /** |
| 891 | * Retrieve the charset manager object |
| 892 | * |
| 893 | * @return CharsetManager charset |
| 894 | */ |
| 895 | public function getCharset() |
| 896 | { |
| 897 | return $this->charset; |
| 898 | } |
| 899 | |
| 900 | /** |
| 901 | * Add a middleware to the parser MiddlewareStack |
| 902 | * Each middleware is invoked when: |
| 903 | * a MimePart is retrieved by mailparse_msg_get_part_data() during $this->parse() |
| 904 | * The middleware will receive MimePart $part and the next MiddlewareStack $next |
| 905 | * |
| 906 | * Eg: |
| 907 | * |
| 908 | * $Parser->addMiddleware(function(MimePart $part, MiddlewareStack $next) { |
| 909 | * // do something with the $part |
| 910 | * return $next($part); |
| 911 | * }); |
| 912 | * |
| 913 | * @param callable $middleware Plain Function or Middleware Instance to execute |
| 914 | * @return void |
| 915 | */ |
| 916 | public function addMiddleware(callable $middleware) |
| 917 | { |
| 918 | if (!$middleware instanceof Middleware) { |
| 919 | $middleware = new Middleware($middleware); |
| 920 | } |
| 921 | $this->middlewareStack = $this->middlewareStack->add($middleware); |
| 922 | } |
| 923 | } |