root / drupal7 / sites / all / libraries / fpdi-1.5.4 / pdf_parser.php @ 4b706e38
| 1 |
<?php
|
|---|---|
| 2 |
//
|
| 3 |
// FPDI - Version 1.5.4
|
| 4 |
//
|
| 5 |
// Copyright 2004-2015 Setasign - Jan Slabon
|
| 6 |
//
|
| 7 |
// Licensed under the Apache License, Version 2.0 (the "License");
|
| 8 |
// you may not use this file except in compliance with the License.
|
| 9 |
// You may obtain a copy of the License at
|
| 10 |
//
|
| 11 |
// http://www.apache.org/licenses/LICENSE-2.0
|
| 12 |
//
|
| 13 |
// Unless required by applicable law or agreed to in writing, software
|
| 14 |
// distributed under the License is distributed on an "AS IS" BASIS,
|
| 15 |
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
| 16 |
// See the License for the specific language governing permissions and
|
| 17 |
// limitations under the License.
|
| 18 |
//
|
| 19 |
|
| 20 |
/**
|
| 21 |
* Class pdf_parser
|
| 22 |
*/
|
| 23 |
class pdf_parser |
| 24 |
{
|
| 25 |
/**
|
| 26 |
* Type constant
|
| 27 |
*
|
| 28 |
* @var integer
|
| 29 |
*/
|
| 30 |
const TYPE_NULL = 0; |
| 31 |
|
| 32 |
/**
|
| 33 |
* Type constant
|
| 34 |
*
|
| 35 |
* @var integer
|
| 36 |
*/
|
| 37 |
const TYPE_NUMERIC = 1; |
| 38 |
|
| 39 |
/**
|
| 40 |
* Type constant
|
| 41 |
*
|
| 42 |
* @var integer
|
| 43 |
*/
|
| 44 |
const TYPE_TOKEN = 2; |
| 45 |
|
| 46 |
/**
|
| 47 |
* Type constant
|
| 48 |
*
|
| 49 |
* @var integer
|
| 50 |
*/
|
| 51 |
const TYPE_HEX = 3; |
| 52 |
|
| 53 |
/**
|
| 54 |
* Type constant
|
| 55 |
*
|
| 56 |
* @var integer
|
| 57 |
*/
|
| 58 |
const TYPE_STRING = 4; |
| 59 |
|
| 60 |
/**
|
| 61 |
* Type constant
|
| 62 |
*
|
| 63 |
* @var integer
|
| 64 |
*/
|
| 65 |
const TYPE_DICTIONARY = 5; |
| 66 |
|
| 67 |
/**
|
| 68 |
* Type constant
|
| 69 |
*
|
| 70 |
* @var integer
|
| 71 |
*/
|
| 72 |
const TYPE_ARRAY = 6; |
| 73 |
|
| 74 |
/**
|
| 75 |
* Type constant
|
| 76 |
*
|
| 77 |
* @var integer
|
| 78 |
*/
|
| 79 |
const TYPE_OBJDEC = 7; |
| 80 |
|
| 81 |
/**
|
| 82 |
* Type constant
|
| 83 |
*
|
| 84 |
* @var integer
|
| 85 |
*/
|
| 86 |
const TYPE_OBJREF = 8; |
| 87 |
|
| 88 |
/**
|
| 89 |
* Type constant
|
| 90 |
*
|
| 91 |
* @var integer
|
| 92 |
*/
|
| 93 |
const TYPE_OBJECT = 9; |
| 94 |
|
| 95 |
/**
|
| 96 |
* Type constant
|
| 97 |
*
|
| 98 |
* @var integer
|
| 99 |
*/
|
| 100 |
const TYPE_STREAM = 10; |
| 101 |
|
| 102 |
/**
|
| 103 |
* Type constant
|
| 104 |
*
|
| 105 |
* @var integer
|
| 106 |
*/
|
| 107 |
const TYPE_BOOLEAN = 11; |
| 108 |
|
| 109 |
/**
|
| 110 |
* Type constant
|
| 111 |
*
|
| 112 |
* @var integer
|
| 113 |
*/
|
| 114 |
const TYPE_REAL = 12; |
| 115 |
|
| 116 |
/**
|
| 117 |
* Define the amount of byte in which the initial keyword of a PDF document should be searched.
|
| 118 |
*
|
| 119 |
* @var int
|
| 120 |
*/
|
| 121 |
static public $searchForStartxrefLength = 5500; |
| 122 |
|
| 123 |
/**
|
| 124 |
* Filename
|
| 125 |
*
|
| 126 |
* @var string
|
| 127 |
*/
|
| 128 |
public $filename; |
| 129 |
|
| 130 |
/**
|
| 131 |
* File resource
|
| 132 |
*
|
| 133 |
* @var resource
|
| 134 |
*/
|
| 135 |
protected $_f; |
| 136 |
|
| 137 |
/**
|
| 138 |
* PDF Context
|
| 139 |
*
|
| 140 |
* @var pdf_context
|
| 141 |
*/
|
| 142 |
protected $_c; |
| 143 |
|
| 144 |
/**
|
| 145 |
* xref-Data
|
| 146 |
*
|
| 147 |
* @var array
|
| 148 |
*/
|
| 149 |
protected $_xref; |
| 150 |
|
| 151 |
/**
|
| 152 |
* Data of the Root object
|
| 153 |
*
|
| 154 |
* @var array
|
| 155 |
*/
|
| 156 |
protected $_root; |
| 157 |
|
| 158 |
/**
|
| 159 |
* PDF version of the loaded document
|
| 160 |
*
|
| 161 |
* @var string
|
| 162 |
*/
|
| 163 |
protected $_pdfVersion; |
| 164 |
|
| 165 |
/**
|
| 166 |
* For reading encrypted documents and xref/object streams are in use
|
| 167 |
*
|
| 168 |
* @var boolean
|
| 169 |
*/
|
| 170 |
protected $_readPlain = true; |
| 171 |
|
| 172 |
/**
|
| 173 |
* The current read object
|
| 174 |
*
|
| 175 |
* @var array
|
| 176 |
*/
|
| 177 |
protected $_currentObj; |
| 178 |
|
| 179 |
/**
|
| 180 |
* Constructor
|
| 181 |
*
|
| 182 |
* @param string $filename Source filename
|
| 183 |
* @throws InvalidArgumentException
|
| 184 |
*/
|
| 185 |
public function __construct($filename) |
| 186 |
{
|
| 187 |
$this->filename = $filename; |
| 188 |
|
| 189 |
$this->_f = @fopen($this->filename, 'rb'); |
| 190 |
|
| 191 |
if (!$this->_f) { |
| 192 |
throw new InvalidArgumentException(sprintf('Cannot open %s !', $filename)); |
| 193 |
} |
| 194 |
|
| 195 |
$this->getPdfVersion();
|
| 196 |
|
| 197 |
if (!class_exists('pdf_context')) { |
| 198 |
require_once('pdf_context.php'); |
| 199 |
} |
| 200 |
$this->_c = new pdf_context($this->_f); |
| 201 |
|
| 202 |
// Read xref-Data
|
| 203 |
$this->_xref = array(); |
| 204 |
$this->_readXref($this->_xref, $this->_findXref()); |
| 205 |
|
| 206 |
// Check for Encryption
|
| 207 |
$this->getEncryption();
|
| 208 |
|
| 209 |
// Read root
|
| 210 |
$this->_readRoot();
|
| 211 |
} |
| 212 |
|
| 213 |
/**
|
| 214 |
* Destructor
|
| 215 |
*/
|
| 216 |
public function __destruct() |
| 217 |
{
|
| 218 |
$this->closeFile();
|
| 219 |
} |
| 220 |
|
| 221 |
/**
|
| 222 |
* Close the opened file
|
| 223 |
*/
|
| 224 |
public function closeFile() |
| 225 |
{
|
| 226 |
if (isset($this->_f) && is_resource($this->_f)) { |
| 227 |
fclose($this->_f); |
| 228 |
unset($this->_f); |
| 229 |
} |
| 230 |
} |
| 231 |
|
| 232 |
/**
|
| 233 |
* Check Trailer for Encryption
|
| 234 |
*
|
| 235 |
* @throws Exception
|
| 236 |
*/
|
| 237 |
public function getEncryption() |
| 238 |
{
|
| 239 |
if (isset($this->_xref['trailer'][1]['/Encrypt'])) { |
| 240 |
throw new Exception('File is encrypted!'); |
| 241 |
} |
| 242 |
} |
| 243 |
|
| 244 |
/**
|
| 245 |
* Get PDF-Version
|
| 246 |
*
|
| 247 |
* @return string
|
| 248 |
*/
|
| 249 |
public function getPdfVersion() |
| 250 |
{
|
| 251 |
if ($this->_pdfVersion === null) { |
| 252 |
fseek($this->_f, 0); |
| 253 |
preg_match('/\d\.\d/', fread($this->_f, 16), $m); |
| 254 |
if (isset($m[0])) |
| 255 |
$this->_pdfVersion = $m[0]; |
| 256 |
} |
| 257 |
|
| 258 |
return $this->_pdfVersion; |
| 259 |
} |
| 260 |
|
| 261 |
/**
|
| 262 |
* Read the /Root dictionary
|
| 263 |
*/
|
| 264 |
protected function _readRoot() |
| 265 |
{
|
| 266 |
if ($this->_xref['trailer'][1]['/Root'][0] != self::TYPE_OBJREF) { |
| 267 |
throw new Exception('Wrong Type of Root-Element! Must be an indirect reference'); |
| 268 |
} |
| 269 |
|
| 270 |
$this->_root = $this->resolveObject($this->_xref['trailer'][1]['/Root']); |
| 271 |
} |
| 272 |
|
| 273 |
/**
|
| 274 |
* Find the xref table
|
| 275 |
*
|
| 276 |
* @return integer
|
| 277 |
* @throws Exception
|
| 278 |
*/
|
| 279 |
protected function _findXref() |
| 280 |
{
|
| 281 |
$toRead = self::$searchForStartxrefLength; |
| 282 |
|
| 283 |
$stat = fseek($this->_f, -$toRead, SEEK_END); |
| 284 |
if ($stat === -1) { |
| 285 |
fseek($this->_f, 0); |
| 286 |
} |
| 287 |
|
| 288 |
$data = fread($this->_f, $toRead); |
| 289 |
|
| 290 |
$keywordPos = strpos(strrev($data), strrev('startxref')); |
| 291 |
if (false === $keywordPos) { |
| 292 |
$keywordPos = strpos(strrev($data), strrev('startref')); |
| 293 |
} |
| 294 |
|
| 295 |
if (false === $keywordPos) { |
| 296 |
throw new Exception('Unable to find "startxref" keyword.'); |
| 297 |
} |
| 298 |
|
| 299 |
$pos = strlen($data) - $keywordPos; |
| 300 |
$data = substr($data, $pos); |
| 301 |
|
| 302 |
if (!preg_match('/\s*(\d+).*$/s', $data, $matches)) { |
| 303 |
throw new Exception('Unable to find pointer to xref table.'); |
| 304 |
} |
| 305 |
|
| 306 |
return (int) $matches[1]; |
| 307 |
} |
| 308 |
|
| 309 |
/**
|
| 310 |
* Read the xref table
|
| 311 |
*
|
| 312 |
* @param array $result Array of xref table entries
|
| 313 |
* @param integer $offset of xref table
|
| 314 |
* @return boolean
|
| 315 |
* @throws Exception
|
| 316 |
*/
|
| 317 |
protected function _readXref(&$result, $offset) |
| 318 |
{
|
| 319 |
$tempPos = $offset - min(20, $offset); |
| 320 |
fseek($this->_f, $tempPos); // set some bytes backwards to fetch corrupted docs |
| 321 |
|
| 322 |
$data = fread($this->_f, 100); |
| 323 |
|
| 324 |
$xrefPos = strrpos($data, 'xref'); |
| 325 |
|
| 326 |
if ($xrefPos === false) { |
| 327 |
$this->_c->reset($offset); |
| 328 |
$xrefStreamObjDec = $this->_readValue($this->_c); |
| 329 |
|
| 330 |
if (is_array($xrefStreamObjDec) && isset($xrefStreamObjDec[0]) && $xrefStreamObjDec[0] == self::TYPE_OBJDEC) { |
| 331 |
throw new Exception( |
| 332 |
sprintf(
|
| 333 |
'This document (%s) probably uses a compression technique which is not supported by the ' .
|
| 334 |
'free parser shipped with FPDI. (See https://www.setasign.com/fpdi-pdf-parser for more details)',
|
| 335 |
$this->filename
|
| 336 |
) |
| 337 |
); |
| 338 |
} else {
|
| 339 |
throw new Exception('Unable to find xref table.'); |
| 340 |
} |
| 341 |
} |
| 342 |
|
| 343 |
if (!isset($result['xrefLocation'])) { |
| 344 |
$result['xrefLocation'] = $tempPos + $xrefPos; |
| 345 |
$result['maxObject'] = 0; |
| 346 |
} |
| 347 |
|
| 348 |
$cycles = -1; |
| 349 |
$bytesPerCycle = 100; |
| 350 |
|
| 351 |
fseek($this->_f, $tempPos = $tempPos + $xrefPos + 4); // set the handle directly after the "xref"-keyword |
| 352 |
$data = fread($this->_f, $bytesPerCycle); |
| 353 |
|
| 354 |
while (($trailerPos = strpos($data, 'trailer', max($bytesPerCycle * $cycles++, 0))) === false && !feof($this->_f)) { |
| 355 |
$data .= fread($this->_f, $bytesPerCycle); |
| 356 |
} |
| 357 |
|
| 358 |
if ($trailerPos === false) { |
| 359 |
throw new Exception('Trailer keyword not found after xref table'); |
| 360 |
} |
| 361 |
|
| 362 |
$data = ltrim(substr($data, 0, $trailerPos)); |
| 363 |
|
| 364 |
// get Line-Ending
|
| 365 |
$found = preg_match_all("/(\r\n|\n|\r)/", substr($data, 0, 100), $m); // check the first 100 bytes for line breaks |
| 366 |
if ($found === 0) { |
| 367 |
throw new Exception('Xref table seems to be corrupted.'); |
| 368 |
} |
| 369 |
$differentLineEndings = count(array_unique($m[0])); |
| 370 |
if ($differentLineEndings > 1) { |
| 371 |
$lines = preg_split("/(\r\n|\n|\r)/", $data, -1, PREG_SPLIT_NO_EMPTY); |
| 372 |
} else {
|
| 373 |
$lines = explode($m[0][0], $data); |
| 374 |
} |
| 375 |
|
| 376 |
$data = $differentLineEndings = $m = null; |
| 377 |
unset($data, $differentLineEndings, $m); |
| 378 |
|
| 379 |
$linesCount = count($lines); |
| 380 |
|
| 381 |
$start = 1; |
| 382 |
|
| 383 |
for ($i = 0; $i < $linesCount; $i++) { |
| 384 |
$line = trim($lines[$i]); |
| 385 |
if ($line) { |
| 386 |
$pieces = explode(' ', $line); |
| 387 |
$c = count($pieces); |
| 388 |
switch($c) { |
| 389 |
case 2: |
| 390 |
$start = (int)$pieces[0]; |
| 391 |
$end = $start + (int)$pieces[1]; |
| 392 |
if ($end > $result['maxObject']) |
| 393 |
$result['maxObject'] = $end; |
| 394 |
break;
|
| 395 |
case 3: |
| 396 |
if (!isset($result['xref'][$start])) |
| 397 |
$result['xref'][$start] = array(); |
| 398 |
|
| 399 |
if (!array_key_exists($gen = (int) $pieces[1], $result['xref'][$start])) { |
| 400 |
$result['xref'][$start][$gen] = $pieces[2] == 'n' ? (int) $pieces[0] : null; |
| 401 |
} |
| 402 |
$start++;
|
| 403 |
break;
|
| 404 |
default:
|
| 405 |
throw new Exception('Unexpected data in xref table'); |
| 406 |
} |
| 407 |
} |
| 408 |
} |
| 409 |
|
| 410 |
$lines = $pieces = $line = $start = $end = $gen = null; |
| 411 |
unset($lines, $pieces, $line, $start, $end, $gen); |
| 412 |
|
| 413 |
$this->_c->reset($tempPos + $trailerPos + 7); |
| 414 |
$trailer = $this->_readValue($this->_c); |
| 415 |
|
| 416 |
if (!isset($result['trailer'])) { |
| 417 |
$result['trailer'] = $trailer; |
| 418 |
} |
| 419 |
|
| 420 |
if (isset($trailer[1]['/Prev'])) { |
| 421 |
$this->_readXref($result, $trailer[1]['/Prev'][1]); |
| 422 |
} |
| 423 |
|
| 424 |
$trailer = null; |
| 425 |
unset($trailer); |
| 426 |
|
| 427 |
return true; |
| 428 |
} |
| 429 |
|
| 430 |
/**
|
| 431 |
* Reads a PDF value
|
| 432 |
*
|
| 433 |
* @param pdf_context $c
|
| 434 |
* @param string $token A token
|
| 435 |
* @return mixed
|
| 436 |
* @throws Exception
|
| 437 |
*/
|
| 438 |
protected function _readValue(&$c, $token = null) |
| 439 |
{
|
| 440 |
if (is_null($token)) { |
| 441 |
$token = $this->_readToken($c); |
| 442 |
} |
| 443 |
|
| 444 |
if ($token === false) { |
| 445 |
return false; |
| 446 |
} |
| 447 |
|
| 448 |
switch ($token) { |
| 449 |
case '<': |
| 450 |
// This is a hex string.
|
| 451 |
// Read the value, then the terminator
|
| 452 |
|
| 453 |
$pos = $c->offset; |
| 454 |
|
| 455 |
while(1) { |
| 456 |
|
| 457 |
$match = strpos($c->buffer, '>', $pos); |
| 458 |
|
| 459 |
// If you can't find it, try
|
| 460 |
// reading more data from the stream
|
| 461 |
|
| 462 |
if ($match === false) { |
| 463 |
if (!$c->increaseLength()) { |
| 464 |
return false; |
| 465 |
} else {
|
| 466 |
continue;
|
| 467 |
} |
| 468 |
} |
| 469 |
|
| 470 |
$result = substr($c->buffer, $c->offset, $match - $c->offset); |
| 471 |
$c->offset = $match + 1; |
| 472 |
|
| 473 |
return array (self::TYPE_HEX, $result); |
| 474 |
} |
| 475 |
break;
|
| 476 |
|
| 477 |
case '<<': |
| 478 |
// This is a dictionary.
|
| 479 |
|
| 480 |
$result = array(); |
| 481 |
|
| 482 |
// Recurse into this function until we reach
|
| 483 |
// the end of the dictionary.
|
| 484 |
while (($key = $this->_readToken($c)) !== '>>') { |
| 485 |
if ($key === false) { |
| 486 |
return false; |
| 487 |
} |
| 488 |
|
| 489 |
if (($value = $this->_readValue($c)) === false) { |
| 490 |
return false; |
| 491 |
} |
| 492 |
|
| 493 |
// Catch missing value
|
| 494 |
if ($value[0] == self::TYPE_TOKEN && $value[1] == '>>') { |
| 495 |
$result[$key] = array(self::TYPE_NULL); |
| 496 |
break;
|
| 497 |
} |
| 498 |
|
| 499 |
$result[$key] = $value; |
| 500 |
} |
| 501 |
|
| 502 |
return array (self::TYPE_DICTIONARY, $result); |
| 503 |
|
| 504 |
case '[': |
| 505 |
// This is an array.
|
| 506 |
|
| 507 |
$result = array(); |
| 508 |
|
| 509 |
// Recurse into this function until we reach
|
| 510 |
// the end of the array.
|
| 511 |
while (($token = $this->_readToken($c)) !== ']') { |
| 512 |
if ($token === false) { |
| 513 |
return false; |
| 514 |
} |
| 515 |
|
| 516 |
if (($value = $this->_readValue($c, $token)) === false) { |
| 517 |
return false; |
| 518 |
} |
| 519 |
|
| 520 |
$result[] = $value; |
| 521 |
} |
| 522 |
|
| 523 |
return array (self::TYPE_ARRAY, $result); |
| 524 |
|
| 525 |
case '(': |
| 526 |
// This is a string
|
| 527 |
$pos = $c->offset; |
| 528 |
|
| 529 |
$openBrackets = 1; |
| 530 |
do {
|
| 531 |
for (; $openBrackets != 0 && $pos < $c->length; $pos++) { |
| 532 |
switch (ord($c->buffer[$pos])) { |
| 533 |
case 0x28: // '(' |
| 534 |
$openBrackets++;
|
| 535 |
break;
|
| 536 |
case 0x29: // ')' |
| 537 |
$openBrackets--;
|
| 538 |
break;
|
| 539 |
case 0x5C: // backslash |
| 540 |
$pos++;
|
| 541 |
} |
| 542 |
} |
| 543 |
} while($openBrackets != 0 && $c->increaseLength()); |
| 544 |
|
| 545 |
$result = substr($c->buffer, $c->offset, $pos - $c->offset - 1); |
| 546 |
$c->offset = $pos; |
| 547 |
|
| 548 |
return array (self::TYPE_STRING, $result); |
| 549 |
|
| 550 |
case 'stream': |
| 551 |
$tempPos = $c->getPos() - strlen($c->buffer); |
| 552 |
$tempOffset = $c->offset; |
| 553 |
|
| 554 |
$c->reset($startPos = $tempPos + $tempOffset); |
| 555 |
|
| 556 |
// Find the first "newline"
|
| 557 |
while ($c->buffer[0] !== chr(10) && $c->buffer[0] !== chr(13)) { |
| 558 |
$c->reset(++$startPos); |
| 559 |
if ($c->ensureContent() === false) { |
| 560 |
throw new Exception( |
| 561 |
'Unable to parse stream data. No newline followed the stream keyword.'
|
| 562 |
); |
| 563 |
} |
| 564 |
} |
| 565 |
|
| 566 |
$e = 0; // ensure line breaks in front of the stream |
| 567 |
if ($c->buffer[0] == chr(10) || $c->buffer[0] == chr(13)) |
| 568 |
$e++;
|
| 569 |
if ($c->buffer[1] == chr(10) && $c->buffer[0] != chr(10)) |
| 570 |
$e++;
|
| 571 |
|
| 572 |
if ($this->_currentObj[1][1]['/Length'][0] == self::TYPE_OBJREF) { |
| 573 |
$tmpLength = $this->resolveObject($this->_currentObj[1][1]['/Length']); |
| 574 |
$length = $tmpLength[1][1]; |
| 575 |
} else {
|
| 576 |
$length = $this->_currentObj[1][1]['/Length'][1]; |
| 577 |
} |
| 578 |
|
| 579 |
if ($length > 0) { |
| 580 |
$c->reset($startPos + $e, $length); |
| 581 |
$v = $c->buffer; |
| 582 |
} else {
|
| 583 |
$v = ''; |
| 584 |
} |
| 585 |
|
| 586 |
$c->reset($startPos + $e + $length); |
| 587 |
$endstream = $this->_readToken($c); |
| 588 |
|
| 589 |
if ($endstream != 'endstream') { |
| 590 |
$c->reset($startPos + $e + $length + 9); // 9 = strlen("endstream") |
| 591 |
// We don't throw an error here because the next
|
| 592 |
// round trip will start at a new offset
|
| 593 |
} |
| 594 |
|
| 595 |
return array(self::TYPE_STREAM, $v); |
| 596 |
|
| 597 |
default :
|
| 598 |
if (is_numeric($token)) { |
| 599 |
// A numeric token. Make sure that
|
| 600 |
// it is not part of something else.
|
| 601 |
if (($tok2 = $this->_readToken($c)) !== false) { |
| 602 |
if (is_numeric($tok2)) { |
| 603 |
|
| 604 |
// Two numeric tokens in a row.
|
| 605 |
// In this case, we're probably in
|
| 606 |
// front of either an object reference
|
| 607 |
// or an object specification.
|
| 608 |
// Determine the case and return the data
|
| 609 |
if (($tok3 = $this->_readToken($c)) !== false) { |
| 610 |
switch ($tok3) { |
| 611 |
case 'obj': |
| 612 |
return array(self::TYPE_OBJDEC, (int)$token, (int)$tok2); |
| 613 |
case 'R': |
| 614 |
return array(self::TYPE_OBJREF, (int)$token, (int)$tok2); |
| 615 |
} |
| 616 |
// If we get to this point, that numeric value up
|
| 617 |
// there was just a numeric value. Push the extra
|
| 618 |
// tokens back into the stack and return the value.
|
| 619 |
array_push($c->stack, $tok3); |
| 620 |
} |
| 621 |
} |
| 622 |
|
| 623 |
array_push($c->stack, $tok2); |
| 624 |
} |
| 625 |
|
| 626 |
if ($token === (string)((int)$token)) |
| 627 |
return array(self::TYPE_NUMERIC, (int)$token); |
| 628 |
else
|
| 629 |
return array(self::TYPE_REAL, (float)$token); |
| 630 |
} else if ($token == 'true' || $token == 'false') { |
| 631 |
return array(self::TYPE_BOOLEAN, $token == 'true'); |
| 632 |
} else if ($token == 'null') { |
| 633 |
return array(self::TYPE_NULL); |
| 634 |
} else {
|
| 635 |
// Just a token. Return it.
|
| 636 |
return array(self::TYPE_TOKEN, $token); |
| 637 |
} |
| 638 |
} |
| 639 |
} |
| 640 |
|
| 641 |
/**
|
| 642 |
* Resolve an object
|
| 643 |
*
|
| 644 |
* @param array $objSpec The object-data
|
| 645 |
* @return array|boolean
|
| 646 |
* @throws Exception
|
| 647 |
*/
|
| 648 |
public function resolveObject($objSpec) |
| 649 |
{
|
| 650 |
$c = $this->_c; |
| 651 |
|
| 652 |
// Exit if we get invalid data
|
| 653 |
if (!is_array($objSpec)) { |
| 654 |
return false; |
| 655 |
} |
| 656 |
|
| 657 |
if ($objSpec[0] == self::TYPE_OBJREF) { |
| 658 |
|
| 659 |
// This is a reference, resolve it
|
| 660 |
if (isset($this->_xref['xref'][$objSpec[1]][$objSpec[2]])) { |
| 661 |
|
| 662 |
// Save current file position
|
| 663 |
// This is needed if you want to resolve
|
| 664 |
// references while you're reading another object
|
| 665 |
// (e.g.: if you need to determine the length
|
| 666 |
// of a stream)
|
| 667 |
|
| 668 |
$oldPos = $c->getPos(); |
| 669 |
|
| 670 |
// Reposition the file pointer and
|
| 671 |
// load the object header.
|
| 672 |
|
| 673 |
$c->reset($this->_xref['xref'][$objSpec[1]][$objSpec[2]]); |
| 674 |
|
| 675 |
$header = $this->_readValue($c); |
| 676 |
|
| 677 |
if ($header[0] != self::TYPE_OBJDEC || $header[1] != $objSpec[1] || $header[2] != $objSpec[2]) { |
| 678 |
$toSearchFor = $objSpec[1] . ' ' . $objSpec[2] . ' obj'; |
| 679 |
if (preg_match('/' . $toSearchFor . '/', $c->buffer)) { |
| 680 |
$c->offset = strpos($c->buffer, $toSearchFor) + strlen($toSearchFor); |
| 681 |
// reset stack
|
| 682 |
$c->stack = array(); |
| 683 |
} else {
|
| 684 |
throw new Exception( |
| 685 |
sprintf("Unable to find object (%s, %s) at expected location.", $objSpec[1], $objSpec[2]) |
| 686 |
); |
| 687 |
} |
| 688 |
} |
| 689 |
|
| 690 |
// If we're being asked to store all the information
|
| 691 |
// about the object, we add the object ID and generation
|
| 692 |
// number for later use
|
| 693 |
$result = array ( |
| 694 |
self::TYPE_OBJECT, |
| 695 |
'obj' => $objSpec[1], |
| 696 |
'gen' => $objSpec[2] |
| 697 |
); |
| 698 |
|
| 699 |
$this->_currentObj =& $result; |
| 700 |
|
| 701 |
// Now simply read the object data until
|
| 702 |
// we encounter an end-of-object marker
|
| 703 |
while (true) { |
| 704 |
$value = $this->_readValue($c); |
| 705 |
if ($value === false || count($result) > 4) { |
| 706 |
// in this case the parser couldn't find an "endobj" so we break here
|
| 707 |
break;
|
| 708 |
} |
| 709 |
|
| 710 |
if ($value[0] == self::TYPE_TOKEN && $value[1] === 'endobj') { |
| 711 |
break;
|
| 712 |
} |
| 713 |
|
| 714 |
$result[] = $value; |
| 715 |
} |
| 716 |
|
| 717 |
$c->reset($oldPos); |
| 718 |
|
| 719 |
if (isset($result[2][0]) && $result[2][0] == self::TYPE_STREAM) { |
| 720 |
$result[0] = self::TYPE_STREAM; |
| 721 |
} |
| 722 |
|
| 723 |
} else {
|
| 724 |
throw new Exception( |
| 725 |
sprintf("Unable to find object (%s, %s) at expected location.", $objSpec[1], $objSpec[2]) |
| 726 |
); |
| 727 |
} |
| 728 |
|
| 729 |
return $result; |
| 730 |
} else {
|
| 731 |
return $objSpec; |
| 732 |
} |
| 733 |
} |
| 734 |
|
| 735 |
/**
|
| 736 |
* Reads a token from the context
|
| 737 |
*
|
| 738 |
* @param pdf_context $c
|
| 739 |
* @return mixed
|
| 740 |
*/
|
| 741 |
protected function _readToken($c) |
| 742 |
{
|
| 743 |
// If there is a token available
|
| 744 |
// on the stack, pop it out and
|
| 745 |
// return it.
|
| 746 |
|
| 747 |
if (count($c->stack)) { |
| 748 |
return array_pop($c->stack); |
| 749 |
} |
| 750 |
|
| 751 |
// Strip away any whitespace
|
| 752 |
|
| 753 |
do {
|
| 754 |
if (!$c->ensureContent()) { |
| 755 |
return false; |
| 756 |
} |
| 757 |
$c->offset += strspn($c->buffer, "\x20\x0A\x0C\x0D\x09\x00", $c->offset); |
| 758 |
} while ($c->offset >= $c->length - 1); |
| 759 |
|
| 760 |
// Get the first character in the stream
|
| 761 |
|
| 762 |
$char = $c->buffer[$c->offset++]; |
| 763 |
|
| 764 |
switch ($char) { |
| 765 |
|
| 766 |
case '[': |
| 767 |
case ']': |
| 768 |
case '(': |
| 769 |
case ')': |
| 770 |
|
| 771 |
// This is either an array or literal string
|
| 772 |
// delimiter, Return it
|
| 773 |
|
| 774 |
return $char; |
| 775 |
|
| 776 |
case '<': |
| 777 |
case '>': |
| 778 |
|
| 779 |
// This could either be a hex string or
|
| 780 |
// dictionary delimiter. Determine the
|
| 781 |
// appropriate case and return the token
|
| 782 |
|
| 783 |
if ($c->buffer[$c->offset] == $char) { |
| 784 |
if (!$c->ensureContent()) { |
| 785 |
return false; |
| 786 |
} |
| 787 |
$c->offset++;
|
| 788 |
return $char . $char; |
| 789 |
} else {
|
| 790 |
return $char; |
| 791 |
} |
| 792 |
|
| 793 |
case '%': |
| 794 |
|
| 795 |
// This is a comment - jump over it!
|
| 796 |
|
| 797 |
$pos = $c->offset; |
| 798 |
while(1) { |
| 799 |
$match = preg_match("/(\r\n|\r|\n)/", $c->buffer, $m, PREG_OFFSET_CAPTURE, $pos); |
| 800 |
if ($match === 0) { |
| 801 |
if (!$c->increaseLength()) { |
| 802 |
return false; |
| 803 |
} else {
|
| 804 |
continue;
|
| 805 |
} |
| 806 |
} |
| 807 |
|
| 808 |
$c->offset = $m[0][1] + strlen($m[0][0]); |
| 809 |
|
| 810 |
return $this->_readToken($c); |
| 811 |
} |
| 812 |
|
| 813 |
default:
|
| 814 |
|
| 815 |
// This is "another" type of token (probably
|
| 816 |
// a dictionary entry or a numeric value)
|
| 817 |
// Find the end and return it.
|
| 818 |
|
| 819 |
if (!$c->ensureContent()) { |
| 820 |
return false; |
| 821 |
} |
| 822 |
|
| 823 |
while(1) { |
| 824 |
|
| 825 |
// Determine the length of the token
|
| 826 |
|
| 827 |
$pos = strcspn($c->buffer, "\x20%[]<>()/\x0A\x0C\x0D\x09\x00", $c->offset); |
| 828 |
|
| 829 |
if ($c->offset + $pos <= $c->length - 1) { |
| 830 |
break;
|
| 831 |
} else {
|
| 832 |
// If the script reaches this point,
|
| 833 |
// the token may span beyond the end
|
| 834 |
// of the current buffer. Therefore,
|
| 835 |
// we increase the size of the buffer
|
| 836 |
// and try again--just to be safe.
|
| 837 |
|
| 838 |
$c->increaseLength();
|
| 839 |
} |
| 840 |
} |
| 841 |
|
| 842 |
$result = substr($c->buffer, $c->offset - 1, $pos + 1); |
| 843 |
|
| 844 |
$c->offset += $pos; |
| 845 |
|
| 846 |
return $result; |
| 847 |
} |
| 848 |
} |
| 849 |
|
| 850 |
/**
|
| 851 |
* Un-filter a stream object
|
| 852 |
*
|
| 853 |
* @param array $obj
|
| 854 |
* @return string
|
| 855 |
* @throws Exception
|
| 856 |
*/
|
| 857 |
protected function _unFilterStream($obj) |
| 858 |
{
|
| 859 |
$filters = array(); |
| 860 |
|
| 861 |
if (isset($obj[1][1]['/Filter'])) { |
| 862 |
$filter = $obj[1][1]['/Filter']; |
| 863 |
|
| 864 |
if ($filter[0] == pdf_parser::TYPE_OBJREF) { |
| 865 |
$tmpFilter = $this->resolveObject($filter); |
| 866 |
$filter = $tmpFilter[1]; |
| 867 |
} |
| 868 |
|
| 869 |
if ($filter[0] == pdf_parser::TYPE_TOKEN) { |
| 870 |
$filters[] = $filter; |
| 871 |
} else if ($filter[0] == pdf_parser::TYPE_ARRAY) { |
| 872 |
$filters = $filter[1]; |
| 873 |
} |
| 874 |
} |
| 875 |
|
| 876 |
$stream = $obj[2][1]; |
| 877 |
|
| 878 |
foreach ($filters AS $filter) { |
| 879 |
switch ($filter[1]) { |
| 880 |
case '/FlateDecode': |
| 881 |
case '/Fl': |
| 882 |
if (function_exists('gzuncompress')) { |
| 883 |
$oStream = $stream; |
| 884 |
$stream = (strlen($stream) > 0) ? @gzuncompress($stream) : ''; |
| 885 |
} else {
|
| 886 |
throw new Exception( |
| 887 |
sprintf('To handle %s filter, please compile php with zlib support.', $filter[1]) |
| 888 |
); |
| 889 |
} |
| 890 |
|
| 891 |
if ($stream === false) { |
| 892 |
$tries = 0; |
| 893 |
while ($tries < 8 && ($stream === false || strlen($stream) < strlen($oStream))) { |
| 894 |
$oStream = substr($oStream, 1); |
| 895 |
$stream = @gzinflate($oStream); |
| 896 |
$tries++;
|
| 897 |
} |
| 898 |
|
| 899 |
if ($stream === false) { |
| 900 |
throw new Exception('Error while decompressing stream.'); |
| 901 |
} |
| 902 |
} |
| 903 |
break;
|
| 904 |
case '/LZWDecode': |
| 905 |
if (!class_exists('FilterLZW')) { |
| 906 |
require_once('filters/FilterLZW.php'); |
| 907 |
} |
| 908 |
$decoder = new FilterLZW(); |
| 909 |
$stream = $decoder->decode($stream); |
| 910 |
break;
|
| 911 |
case '/ASCII85Decode': |
| 912 |
if (!class_exists('FilterASCII85')) { |
| 913 |
require_once('filters/FilterASCII85.php'); |
| 914 |
} |
| 915 |
$decoder = new FilterASCII85(); |
| 916 |
$stream = $decoder->decode($stream); |
| 917 |
break;
|
| 918 |
case '/ASCIIHexDecode': |
| 919 |
if (!class_exists('FilterASCIIHexDecode')) { |
| 920 |
require_once('filters/FilterASCIIHexDecode.php'); |
| 921 |
} |
| 922 |
$decoder = new FilterASCIIHexDecode(); |
| 923 |
$stream = $decoder->decode($stream); |
| 924 |
break;
|
| 925 |
case null: |
| 926 |
break;
|
| 927 |
default:
|
| 928 |
throw new Exception(sprintf('Unsupported Filter: %s', $filter[1])); |
| 929 |
} |
| 930 |
} |
| 931 |
|
| 932 |
return $stream; |
| 933 |
} |
| 934 |
} |