root / drupal7 / sites / all / libraries / tcpdf-version / tcpdf_parser.php @ a84ddea4
| 1 |
<?php
|
|---|---|
| 2 |
//============================================================+
|
| 3 |
// File name : tcpdf_parser.php
|
| 4 |
// Version : 1.0.010
|
| 5 |
// Begin : 2011-05-23
|
| 6 |
// Last Update : 2013-09-25
|
| 7 |
// Author : Nicola Asuni - Tecnick.com LTD - www.tecnick.com - info@tecnick.com
|
| 8 |
// License : http://www.tecnick.com/pagefiles/tcpdf/LICENSE.TXT GNU-LGPLv3
|
| 9 |
// -------------------------------------------------------------------
|
| 10 |
// Copyright (C) 2011-2013 Nicola Asuni - Tecnick.com LTD
|
| 11 |
//
|
| 12 |
// This file is part of TCPDF software library.
|
| 13 |
//
|
| 14 |
// TCPDF is free software: you can redistribute it and/or modify it
|
| 15 |
// under the terms of the GNU Lesser General Public License as
|
| 16 |
// published by the Free Software Foundation, either version 3 of the
|
| 17 |
// License, or (at your option) any later version.
|
| 18 |
//
|
| 19 |
// TCPDF is distributed in the hope that it will be useful, but
|
| 20 |
// WITHOUT ANY WARRANTY; without even the implied warranty of
|
| 21 |
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
|
| 22 |
// See the GNU Lesser General Public License for more details.
|
| 23 |
//
|
| 24 |
// You should have received a copy of the License
|
| 25 |
// along with TCPDF. If not, see
|
| 26 |
// <http://www.tecnick.com/pagefiles/tcpdf/LICENSE.TXT>.
|
| 27 |
//
|
| 28 |
// See LICENSE.TXT file for more information.
|
| 29 |
// -------------------------------------------------------------------
|
| 30 |
//
|
| 31 |
// Description : This is a PHP class for parsing PDF documents.
|
| 32 |
//
|
| 33 |
//============================================================+
|
| 34 |
|
| 35 |
/**
|
| 36 |
* @file
|
| 37 |
* This is a PHP class for parsing PDF documents.<br>
|
| 38 |
* @package com.tecnick.tcpdf
|
| 39 |
* @author Nicola Asuni
|
| 40 |
* @version 1.0.010
|
| 41 |
*/
|
| 42 |
|
| 43 |
// include class for decoding filters
|
| 44 |
require_once(dirname(__FILE__).'/include/tcpdf_filters.php'); |
| 45 |
|
| 46 |
/**
|
| 47 |
* @class TCPDF_PARSER
|
| 48 |
* This is a PHP class for parsing PDF documents.<br>
|
| 49 |
* @package com.tecnick.tcpdf
|
| 50 |
* @brief This is a PHP class for parsing PDF documents..
|
| 51 |
* @version 1.0.010
|
| 52 |
* @author Nicola Asuni - info@tecnick.com
|
| 53 |
*/
|
| 54 |
class TCPDF_PARSER { |
| 55 |
|
| 56 |
/**
|
| 57 |
* Raw content of the PDF document.
|
| 58 |
* @private
|
| 59 |
*/
|
| 60 |
private $pdfdata = ''; |
| 61 |
|
| 62 |
/**
|
| 63 |
* XREF data.
|
| 64 |
* @protected
|
| 65 |
*/
|
| 66 |
protected $xref = array(); |
| 67 |
|
| 68 |
/**
|
| 69 |
* Array of PDF objects.
|
| 70 |
* @protected
|
| 71 |
*/
|
| 72 |
protected $objects = array(); |
| 73 |
|
| 74 |
/**
|
| 75 |
* Class object for decoding filters.
|
| 76 |
* @private
|
| 77 |
*/
|
| 78 |
private $FilterDecoders; |
| 79 |
|
| 80 |
/**
|
| 81 |
* Array of configuration parameters.
|
| 82 |
* @private
|
| 83 |
*/
|
| 84 |
private $cfg = array( |
| 85 |
'die_for_errors' => false, |
| 86 |
'ignore_filter_decoding_errors' => true, |
| 87 |
'ignore_missing_filter_decoders' => true, |
| 88 |
); |
| 89 |
|
| 90 |
// -----------------------------------------------------------------------------
|
| 91 |
|
| 92 |
/**
|
| 93 |
* Parse a PDF document an return an array of objects.
|
| 94 |
* @param $data (string) PDF data to parse.
|
| 95 |
* @param $cfg (array) Array of configuration parameters:
|
| 96 |
* 'die_for_errors' : if true termitate the program execution in case of error, otherwise thows an exception;
|
| 97 |
* 'ignore_filter_decoding_errors' : if true ignore filter decoding errors;
|
| 98 |
* 'ignore_missing_filter_decoders' : if true ignore missing filter decoding errors.
|
| 99 |
* @public
|
| 100 |
* @since 1.0.000 (2011-05-24)
|
| 101 |
*/
|
| 102 |
public function __construct($data, $cfg=array()) { |
| 103 |
if (empty($data)) { |
| 104 |
$this->Error('Empty PDF data.'); |
| 105 |
} |
| 106 |
// set configuration parameters
|
| 107 |
$this->setConfig($cfg); |
| 108 |
// get PDF content string
|
| 109 |
$this->pdfdata = $data; |
| 110 |
// get length
|
| 111 |
$pdflen = strlen($this->pdfdata); |
| 112 |
// get xref and trailer data
|
| 113 |
$this->xref = $this->getXrefData(); |
| 114 |
// parse all document objects
|
| 115 |
$this->objects = array(); |
| 116 |
foreach ($this->xref['xref'] as $obj => $offset) { |
| 117 |
if (!isset($this->objects[$obj]) AND ($offset > 0)) { |
| 118 |
// decode objects with positive offset
|
| 119 |
$this->objects[$obj] = $this->getIndirectObject($obj, $offset, true); |
| 120 |
} |
| 121 |
} |
| 122 |
// release some memory
|
| 123 |
unset($this->pdfdata); |
| 124 |
$this->pdfdata = ''; |
| 125 |
} |
| 126 |
|
| 127 |
/**
|
| 128 |
* Set the configuration parameters.
|
| 129 |
* @param $cfg (array) Array of configuration parameters:
|
| 130 |
* 'die_for_errors' : if true termitate the program execution in case of error, otherwise thows an exception;
|
| 131 |
* 'ignore_filter_decoding_errors' : if true ignore filter decoding errors;
|
| 132 |
* 'ignore_missing_filter_decoders' : if true ignore missing filter decoding errors.
|
| 133 |
* @public
|
| 134 |
*/
|
| 135 |
protected function setConfig($cfg) { |
| 136 |
if (isset($cfg['die_for_errors'])) { |
| 137 |
$this->cfg['die_for_errors'] = !!$cfg['die_for_errors']; |
| 138 |
} |
| 139 |
if (isset($cfg['ignore_filter_decoding_errors'])) { |
| 140 |
$this->cfg['ignore_filter_decoding_errors'] = !!$cfg['ignore_filter_decoding_errors']; |
| 141 |
} |
| 142 |
if (isset($cfg['ignore_missing_filter_decoders'])) { |
| 143 |
$this->cfg['ignore_missing_filter_decoders'] = !!$cfg['ignore_missing_filter_decoders']; |
| 144 |
} |
| 145 |
} |
| 146 |
|
| 147 |
/**
|
| 148 |
* Return an array of parsed PDF document objects.
|
| 149 |
* @return (array) Array of parsed PDF document objects.
|
| 150 |
* @public
|
| 151 |
* @since 1.0.000 (2011-06-26)
|
| 152 |
*/
|
| 153 |
public function getParsedData() { |
| 154 |
return array($this->xref, $this->objects); |
| 155 |
} |
| 156 |
|
| 157 |
/**
|
| 158 |
* Get Cross-Reference (xref) table and trailer data from PDF document data.
|
| 159 |
* @param $offset (int) xref offset (if know).
|
| 160 |
* @param $xref (array) previous xref array (if any).
|
| 161 |
* @return Array containing xref and trailer data.
|
| 162 |
* @protected
|
| 163 |
* @since 1.0.000 (2011-05-24)
|
| 164 |
*/
|
| 165 |
protected function getXrefData($offset=0, $xref=array()) { |
| 166 |
if ($offset == 0) { |
| 167 |
// find last startxref
|
| 168 |
if (preg_match_all('/[\r\n]startxref[\s]*[\r\n]+([0-9]+)[\s]*[\r\n]+%%EOF/i', $this->pdfdata, $matches, PREG_SET_ORDER, $offset) == 0) { |
| 169 |
$this->Error('Unable to find startxref'); |
| 170 |
} |
| 171 |
$matches = array_pop($matches); |
| 172 |
$startxref = $matches[1]; |
| 173 |
} elseif (strpos($this->pdfdata, 'xref', $offset) == $offset) { |
| 174 |
// Already pointing at the xref table
|
| 175 |
$startxref = $offset; |
| 176 |
} elseif (preg_match('/([0-9]+[\s][0-9]+[\s]obj)/i', $this->pdfdata, $matches, PREG_OFFSET_CAPTURE, $offset)) { |
| 177 |
// Cross-Reference Stream object
|
| 178 |
$startxref = $offset; |
| 179 |
} elseif (preg_match('/[\r\n]startxref[\s]*[\r\n]+([0-9]+)[\s]*[\r\n]+%%EOF/i', $this->pdfdata, $matches, PREG_OFFSET_CAPTURE, $offset)) { |
| 180 |
// startxref found
|
| 181 |
$startxref = $matches[1][0]; |
| 182 |
} else {
|
| 183 |
$this->Error('Unable to find startxref'); |
| 184 |
} |
| 185 |
// check xref position
|
| 186 |
if (strpos($this->pdfdata, 'xref', $startxref) == $startxref) { |
| 187 |
// Cross-Reference
|
| 188 |
$xref = $this->decodeXref($startxref, $xref); |
| 189 |
} else {
|
| 190 |
// Cross-Reference Stream
|
| 191 |
$xref = $this->decodeXrefStream($startxref, $xref); |
| 192 |
} |
| 193 |
if (empty($xref)) { |
| 194 |
$this->Error('Unable to find xref'); |
| 195 |
} |
| 196 |
return $xref; |
| 197 |
} |
| 198 |
|
| 199 |
/**
|
| 200 |
* Decode the Cross-Reference section
|
| 201 |
* @param $startxref (int) Offset at which the xref section starts (position of the 'xref' keyword).
|
| 202 |
* @param $xref (array) Previous xref array (if any).
|
| 203 |
* @return Array containing xref and trailer data.
|
| 204 |
* @protected
|
| 205 |
* @since 1.0.000 (2011-06-20)
|
| 206 |
*/
|
| 207 |
protected function decodeXref($startxref, $xref=array()) { |
| 208 |
$startxref += 4; // 4 is the lenght of the word 'xref' |
| 209 |
// skip initial white space chars: \x00 null (NUL), \x09 horizontal tab (HT), \x0A line feed (LF), \x0C form feed (FF), \x0D carriage return (CR), \x20 space (SP)
|
| 210 |
$offset = $startxref + strspn($this->pdfdata, "\x00\x09\x0a\x0c\x0d\x20", $startxref); |
| 211 |
// initialize object number
|
| 212 |
$obj_num = 0; |
| 213 |
// search for cross-reference entries or subsection
|
| 214 |
while (preg_match('/([0-9]+)[\x20]([0-9]+)[\x20]?([nf]?)(\r\n|[\x20]?[\r\n])/', $this->pdfdata, $matches, PREG_OFFSET_CAPTURE, $offset) > 0) { |
| 215 |
if ($matches[0][1] != $offset) { |
| 216 |
// we are on another section
|
| 217 |
break;
|
| 218 |
} |
| 219 |
$offset += strlen($matches[0][0]); |
| 220 |
if ($matches[3][0] == 'n') { |
| 221 |
// create unique object index: [object number]_[generation number]
|
| 222 |
$index = $obj_num.'_'.intval($matches[2][0]); |
| 223 |
// check if object already exist
|
| 224 |
if (!isset($xref['xref'][$index])) { |
| 225 |
// store object offset position
|
| 226 |
$xref['xref'][$index] = intval($matches[1][0]); |
| 227 |
} |
| 228 |
++$obj_num;
|
| 229 |
} elseif ($matches[3][0] == 'f') { |
| 230 |
++$obj_num;
|
| 231 |
} else {
|
| 232 |
// object number (index)
|
| 233 |
$obj_num = intval($matches[1][0]); |
| 234 |
} |
| 235 |
} |
| 236 |
// get trailer data
|
| 237 |
if (preg_match('/trailer[\s]*<<(.*)>>[\s]*[\r\n]+startxref[\s]*[\r\n]+/isU', $this->pdfdata, $matches, PREG_OFFSET_CAPTURE, $offset) > 0) { |
| 238 |
$trailer_data = $matches[1][0]; |
| 239 |
if (!isset($xref['trailer']) OR empty($xref['trailer'])) { |
| 240 |
// get only the last updated version
|
| 241 |
$xref['trailer'] = array(); |
| 242 |
// parse trailer_data
|
| 243 |
if (preg_match('/Size[\s]+([0-9]+)/i', $trailer_data, $matches) > 0) { |
| 244 |
$xref['trailer']['size'] = intval($matches[1]); |
| 245 |
} |
| 246 |
if (preg_match('/Root[\s]+([0-9]+)[\s]+([0-9]+)[\s]+R/i', $trailer_data, $matches) > 0) { |
| 247 |
$xref['trailer']['root'] = intval($matches[1]).'_'.intval($matches[2]); |
| 248 |
} |
| 249 |
if (preg_match('/Encrypt[\s]+([0-9]+)[\s]+([0-9]+)[\s]+R/i', $trailer_data, $matches) > 0) { |
| 250 |
$xref['trailer']['encrypt'] = intval($matches[1]).'_'.intval($matches[2]); |
| 251 |
} |
| 252 |
if (preg_match('/Info[\s]+([0-9]+)[\s]+([0-9]+)[\s]+R/i', $trailer_data, $matches) > 0) { |
| 253 |
$xref['trailer']['info'] = intval($matches[1]).'_'.intval($matches[2]); |
| 254 |
} |
| 255 |
if (preg_match('/ID[\s]*[\[][\s]*[<]([^>]*)[>][\s]*[<]([^>]*)[>]/i', $trailer_data, $matches) > 0) { |
| 256 |
$xref['trailer']['id'] = array(); |
| 257 |
$xref['trailer']['id'][0] = $matches[1]; |
| 258 |
$xref['trailer']['id'][1] = $matches[2]; |
| 259 |
} |
| 260 |
} |
| 261 |
if (preg_match('/Prev[\s]+([0-9]+)/i', $trailer_data, $matches) > 0) { |
| 262 |
// get previous xref
|
| 263 |
$xref = $this->getXrefData(intval($matches[1]), $xref); |
| 264 |
} |
| 265 |
} else {
|
| 266 |
$this->Error('Unable to find trailer'); |
| 267 |
} |
| 268 |
return $xref; |
| 269 |
} |
| 270 |
|
| 271 |
/**
|
| 272 |
* Decode the Cross-Reference Stream section
|
| 273 |
* @param $startxref (int) Offset at which the xref section starts.
|
| 274 |
* @param $xref (array) Previous xref array (if any).
|
| 275 |
* @return Array containing xref and trailer data.
|
| 276 |
* @protected
|
| 277 |
* @since 1.0.003 (2013-03-16)
|
| 278 |
*/
|
| 279 |
protected function decodeXrefStream($startxref, $xref=array()) { |
| 280 |
// try to read Cross-Reference Stream
|
| 281 |
$xrefobj = $this->getRawObject($startxref); |
| 282 |
$xrefcrs = $this->getIndirectObject($xrefobj[1], $startxref, true); |
| 283 |
if (!isset($xref['trailer']) OR empty($xref['trailer'])) { |
| 284 |
// get only the last updated version
|
| 285 |
$xref['trailer'] = array(); |
| 286 |
$filltrailer = true; |
| 287 |
} else {
|
| 288 |
$filltrailer = false; |
| 289 |
} |
| 290 |
$valid_crs = false; |
| 291 |
$sarr = $xrefcrs[0][1]; |
| 292 |
foreach ($sarr as $k => $v) { |
| 293 |
if (($v[0] == '/') AND ($v[1] == 'Type') AND (isset($sarr[($k +1)]) AND ($sarr[($k +1)][0] == '/') AND ($sarr[($k +1)][1] == 'XRef'))) { |
| 294 |
$valid_crs = true; |
| 295 |
} elseif (($v[0] == '/') AND ($v[1] == 'Index') AND (isset($sarr[($k +1)]))) { |
| 296 |
// first object number in the subsection
|
| 297 |
$index_first = intval($sarr[($k +1)][1][0][1]); |
| 298 |
// number of entries in the subsection
|
| 299 |
$index_entries = intval($sarr[($k +1)][1][1][1]); |
| 300 |
} elseif (($v[0] == '/') AND ($v[1] == 'Prev') AND (isset($sarr[($k +1)]) AND ($sarr[($k +1)][0] == 'numeric'))) { |
| 301 |
// get previous xref offset
|
| 302 |
$prevxref = intval($sarr[($k +1)][1]); |
| 303 |
} elseif (($v[0] == '/') AND ($v[1] == 'W') AND (isset($sarr[($k +1)]))) { |
| 304 |
// number of bytes (in the decoded stream) of the corresponding field
|
| 305 |
$wb = array(); |
| 306 |
$wb[0] = intval($sarr[($k +1)][1][0][1]); |
| 307 |
$wb[1] = intval($sarr[($k +1)][1][1][1]); |
| 308 |
$wb[2] = intval($sarr[($k +1)][1][2][1]); |
| 309 |
} elseif (($v[0] == '/') AND ($v[1] == 'DecodeParms') AND (isset($sarr[($k +1)][1]))) { |
| 310 |
$decpar = $sarr[($k +1)][1]; |
| 311 |
foreach ($decpar as $kdc => $vdc) { |
| 312 |
if (($vdc[0] == '/') AND ($vdc[1] == 'Columns') AND (isset($decpar[($kdc +1)]) AND ($decpar[($kdc +1)][0] == 'numeric'))) { |
| 313 |
$columns = intval($decpar[($kdc +1)][1]); |
| 314 |
} elseif (($vdc[0] == '/') AND ($vdc[1] == 'Predictor') AND (isset($decpar[($kdc +1)]) AND ($decpar[($kdc +1)][0] == 'numeric'))) { |
| 315 |
$predictor = intval($decpar[($kdc +1)][1]); |
| 316 |
} |
| 317 |
} |
| 318 |
} elseif ($filltrailer) { |
| 319 |
if (($v[0] == '/') AND ($v[1] == 'Size') AND (isset($sarr[($k +1)]) AND ($sarr[($k +1)][0] == 'numeric'))) { |
| 320 |
$xref['trailer']['size'] = $sarr[($k +1)][1]; |
| 321 |
} elseif (($v[0] == '/') AND ($v[1] == 'Root') AND (isset($sarr[($k +1)]) AND ($sarr[($k +1)][0] == 'ojbref'))) { |
| 322 |
$xref['trailer']['root'] = $sarr[($k +1)][1]; |
| 323 |
} elseif (($v[0] == '/') AND ($v[1] == 'Info') AND (isset($sarr[($k +1)]) AND ($sarr[($k +1)][0] == 'ojbref'))) { |
| 324 |
$xref['trailer']['info'] = $sarr[($k +1)][1]; |
| 325 |
} elseif (($v[0] == '/') AND ($v[1] == 'ID') AND (isset($sarr[($k +1)]))) { |
| 326 |
$xref['trailer']['id'] = array(); |
| 327 |
$xref['trailer']['id'][0] = $sarr[($k +1)][1][0][1]; |
| 328 |
$xref['trailer']['id'][1] = $sarr[($k +1)][1][1][1]; |
| 329 |
} |
| 330 |
} |
| 331 |
} |
| 332 |
// decode data
|
| 333 |
if ($valid_crs AND isset($xrefcrs[1][3][0])) { |
| 334 |
// number of bytes in a row
|
| 335 |
$rowlen = ($columns + 1); |
| 336 |
// convert the stream into an array of integers
|
| 337 |
$sdata = unpack('C*', $xrefcrs[1][3][0]); |
| 338 |
// split the rows
|
| 339 |
$sdata = array_chunk($sdata, $rowlen); |
| 340 |
// initialize decoded array
|
| 341 |
$ddata = array(); |
| 342 |
// initialize first row with zeros
|
| 343 |
$prev_row = array_fill (0, $rowlen, 0); |
| 344 |
// for each row apply PNG unpredictor
|
| 345 |
foreach ($sdata as $k => $row) { |
| 346 |
// initialize new row
|
| 347 |
$ddata[$k] = array(); |
| 348 |
// get PNG predictor value
|
| 349 |
$predictor = (10 + $row[0]); |
| 350 |
// for each byte on the row
|
| 351 |
for ($i=1; $i<=$columns; ++$i) { |
| 352 |
// new index
|
| 353 |
$j = ($i - 1); |
| 354 |
$row_up = $prev_row[$j]; |
| 355 |
if ($i == 1) { |
| 356 |
$row_left = 0; |
| 357 |
$row_upleft = 0; |
| 358 |
} else {
|
| 359 |
$row_left = $row[($i - 1)]; |
| 360 |
$row_upleft = $prev_row[($j - 1)]; |
| 361 |
} |
| 362 |
switch ($predictor) { |
| 363 |
case 10: { // PNG prediction (on encoding, PNG None on all rows) |
| 364 |
$ddata[$k][$j] = $row[$i]; |
| 365 |
break;
|
| 366 |
} |
| 367 |
case 11: { // PNG prediction (on encoding, PNG Sub on all rows) |
| 368 |
$ddata[$k][$j] = (($row[$i] + $row_left) & 0xff); |
| 369 |
break;
|
| 370 |
} |
| 371 |
case 12: { // PNG prediction (on encoding, PNG Up on all rows) |
| 372 |
$ddata[$k][$j] = (($row[$i] + $row_up) & 0xff); |
| 373 |
break;
|
| 374 |
} |
| 375 |
case 13: { // PNG prediction (on encoding, PNG Average on all rows) |
| 376 |
$ddata[$k][$j] = (($row[$i] + (($row_left + $row_up) / 2)) & 0xff); |
| 377 |
break;
|
| 378 |
} |
| 379 |
case 14: { // PNG prediction (on encoding, PNG Paeth on all rows) |
| 380 |
// initial estimate
|
| 381 |
$p = ($row_left + $row_up - $row_upleft); |
| 382 |
// distances
|
| 383 |
$pa = abs($p - $row_left); |
| 384 |
$pb = abs($p - $row_up); |
| 385 |
$pc = abs($p - $row_upleft); |
| 386 |
$pmin = min($pa, $pb, $pc); |
| 387 |
// return minumum distance
|
| 388 |
switch ($pmin) { |
| 389 |
case $pa: { |
| 390 |
$ddata[$k][$j] = (($row[$i] + $row_left) & 0xff); |
| 391 |
break;
|
| 392 |
} |
| 393 |
case $pb: { |
| 394 |
$ddata[$k][$j] = (($row[$i] + $row_up) & 0xff); |
| 395 |
break;
|
| 396 |
} |
| 397 |
case $pc: { |
| 398 |
$ddata[$k][$j] = (($row[$i] + $row_upleft) & 0xff); |
| 399 |
break;
|
| 400 |
} |
| 401 |
} |
| 402 |
break;
|
| 403 |
} |
| 404 |
default: { // PNG prediction (on encoding, PNG optimum) |
| 405 |
$this->Error('Unknown PNG predictor'); |
| 406 |
break;
|
| 407 |
} |
| 408 |
} |
| 409 |
} |
| 410 |
$prev_row = $ddata[$k]; |
| 411 |
} // end for each row
|
| 412 |
// complete decoding
|
| 413 |
$sdata = array(); |
| 414 |
// for every row
|
| 415 |
foreach ($ddata as $k => $row) { |
| 416 |
// initialize new row
|
| 417 |
$sdata[$k] = array(0, 0, 0); |
| 418 |
if ($wb[0] == 0) { |
| 419 |
// default type field
|
| 420 |
$sdata[$k][0] = 1; |
| 421 |
} |
| 422 |
$i = 0; // count bytes on the row |
| 423 |
// for every column
|
| 424 |
for ($c = 0; $c < 3; ++$c) { |
| 425 |
// for every byte on the column
|
| 426 |
for ($b = 0; $b < $wb[$c]; ++$b) { |
| 427 |
$sdata[$k][$c] += ($row[$i] << (($wb[$c] - 1 - $b) * 8)); |
| 428 |
++$i;
|
| 429 |
} |
| 430 |
} |
| 431 |
} |
| 432 |
$ddata = array(); |
| 433 |
// fill xref
|
| 434 |
if (isset($index_first)) { |
| 435 |
$obj_num = $index_first; |
| 436 |
} else {
|
| 437 |
$obj_num = 0; |
| 438 |
} |
| 439 |
foreach ($sdata as $k => $row) { |
| 440 |
switch ($row[0]) { |
| 441 |
case 0: { // (f) linked list of free objects |
| 442 |
++$obj_num;
|
| 443 |
break;
|
| 444 |
} |
| 445 |
case 1: { // (n) objects that are in use but are not compressed |
| 446 |
// create unique object index: [object number]_[generation number]
|
| 447 |
$index = $obj_num.'_'.$row[2]; |
| 448 |
// check if object already exist
|
| 449 |
if (!isset($xref['xref'][$index])) { |
| 450 |
// store object offset position
|
| 451 |
$xref['xref'][$index] = $row[1]; |
| 452 |
} |
| 453 |
++$obj_num;
|
| 454 |
break;
|
| 455 |
} |
| 456 |
case 2: { // compressed objects |
| 457 |
// $row[1] = object number of the object stream in which this object is stored
|
| 458 |
// $row[2] = index of this object within the object stream
|
| 459 |
$index = $row[1].'_0_'.$row[2]; |
| 460 |
$xref['xref'][$index] = -1; |
| 461 |
break;
|
| 462 |
} |
| 463 |
default: { // null objects |
| 464 |
break;
|
| 465 |
} |
| 466 |
} |
| 467 |
} |
| 468 |
} // end decoding data
|
| 469 |
if (isset($prevxref)) { |
| 470 |
// get previous xref
|
| 471 |
$xref = $this->getXrefData($prevxref, $xref); |
| 472 |
} |
| 473 |
return $xref; |
| 474 |
} |
| 475 |
|
| 476 |
/**
|
| 477 |
* Get object type, raw value and offset to next object
|
| 478 |
* @param $offset (int) Object offset.
|
| 479 |
* @return array containing object type, raw value and offset to next object
|
| 480 |
* @protected
|
| 481 |
* @since 1.0.000 (2011-06-20)
|
| 482 |
*/
|
| 483 |
protected function getRawObject($offset=0) { |
| 484 |
$objtype = ''; // object type to be returned |
| 485 |
$objval = ''; // object value to be returned |
| 486 |
// skip initial white space chars: \x00 null (NUL), \x09 horizontal tab (HT), \x0A line feed (LF), \x0C form feed (FF), \x0D carriage return (CR), \x20 space (SP)
|
| 487 |
$offset += strspn($this->pdfdata, "\x00\x09\x0a\x0c\x0d\x20", $offset); |
| 488 |
// get first char
|
| 489 |
$char = $this->pdfdata[$offset]; |
| 490 |
// get object type
|
| 491 |
switch ($char) { |
| 492 |
case '%': { // \x25 PERCENT SIGN |
| 493 |
// skip comment and search for next token
|
| 494 |
$next = strcspn($this->pdfdata, "\r\n", $offset); |
| 495 |
if ($next > 0) { |
| 496 |
$offset += $next; |
| 497 |
return $this->getRawObject($this->pdfdata, $offset); |
| 498 |
} |
| 499 |
break;
|
| 500 |
} |
| 501 |
case '/': { // \x2F SOLIDUS |
| 502 |
// name object
|
| 503 |
$objtype = $char; |
| 504 |
++$offset;
|
| 505 |
if (preg_match('/^([^\x00\x09\x0a\x0c\x0d\x20\s\x28\x29\x3c\x3e\x5b\x5d\x7b\x7d\x2f\x25]+)/', substr($this->pdfdata, $offset, 256), $matches) == 1) { |
| 506 |
$objval = $matches[1]; // unescaped value |
| 507 |
$offset += strlen($objval); |
| 508 |
} |
| 509 |
break;
|
| 510 |
} |
| 511 |
case '(': // \x28 LEFT PARENTHESIS |
| 512 |
case ')': { // \x29 RIGHT PARENTHESIS |
| 513 |
// literal string object
|
| 514 |
$objtype = $char; |
| 515 |
++$offset;
|
| 516 |
$strpos = $offset; |
| 517 |
if ($char == '(') { |
| 518 |
$open_bracket = 1; |
| 519 |
while ($open_bracket > 0) { |
| 520 |
if (!isset($this->pdfdata{$strpos})) { |
| 521 |
break;
|
| 522 |
} |
| 523 |
$ch = $this->pdfdata{$strpos}; |
| 524 |
switch ($ch) { |
| 525 |
case '\\': { // REVERSE SOLIDUS (5Ch) (Backslash) |
| 526 |
// skip next character
|
| 527 |
++$strpos;
|
| 528 |
break;
|
| 529 |
} |
| 530 |
case '(': { // LEFT PARENHESIS (28h) |
| 531 |
++$open_bracket;
|
| 532 |
break;
|
| 533 |
} |
| 534 |
case ')': { // RIGHT PARENTHESIS (29h) |
| 535 |
--$open_bracket;
|
| 536 |
break;
|
| 537 |
} |
| 538 |
} |
| 539 |
++$strpos;
|
| 540 |
} |
| 541 |
$objval = substr($this->pdfdata, $offset, ($strpos - $offset - 1)); |
| 542 |
$offset = $strpos; |
| 543 |
} |
| 544 |
break;
|
| 545 |
} |
| 546 |
case '[': // \x5B LEFT SQUARE BRACKET |
| 547 |
case ']': { // \x5D RIGHT SQUARE BRACKET |
| 548 |
// array object
|
| 549 |
$objtype = $char; |
| 550 |
++$offset;
|
| 551 |
if ($char == '[') { |
| 552 |
// get array content
|
| 553 |
$objval = array(); |
| 554 |
do {
|
| 555 |
// get element
|
| 556 |
$element = $this->getRawObject($offset); |
| 557 |
$offset = $element[2]; |
| 558 |
$objval[] = $element; |
| 559 |
} while ($element[0] != ']'); |
| 560 |
// remove closing delimiter
|
| 561 |
array_pop($objval); |
| 562 |
} |
| 563 |
break;
|
| 564 |
} |
| 565 |
case '<': // \x3C LESS-THAN SIGN |
| 566 |
case '>': { // \x3E GREATER-THAN SIGN |
| 567 |
if (isset($this->pdfdata{($offset + 1)}) AND ($this->pdfdata{($offset + 1)} == $char)) { |
| 568 |
// dictionary object
|
| 569 |
$objtype = $char.$char; |
| 570 |
$offset += 2; |
| 571 |
if ($char == '<') { |
| 572 |
// get array content
|
| 573 |
$objval = array(); |
| 574 |
do {
|
| 575 |
// get element
|
| 576 |
$element = $this->getRawObject($offset); |
| 577 |
$offset = $element[2]; |
| 578 |
$objval[] = $element; |
| 579 |
} while ($element[0] != '>>'); |
| 580 |
// remove closing delimiter
|
| 581 |
array_pop($objval); |
| 582 |
} |
| 583 |
} else {
|
| 584 |
// hexadecimal string object
|
| 585 |
$objtype = $char; |
| 586 |
++$offset;
|
| 587 |
if (($char == '<') AND (preg_match('/^([0-9A-Fa-f\x09\x0a\x0c\x0d\x20]+)>/iU', substr($this->pdfdata, $offset), $matches) == 1)) { |
| 588 |
// remove white space characters
|
| 589 |
$objval = strtr($matches[1], "\x09\x0a\x0c\x0d\x20", ''); |
| 590 |
$offset += strlen($matches[0]); |
| 591 |
} |
| 592 |
} |
| 593 |
break;
|
| 594 |
} |
| 595 |
default: {
|
| 596 |
if (substr($this->pdfdata, $offset, 6) == 'endobj') { |
| 597 |
// indirect object
|
| 598 |
$objtype = 'endobj'; |
| 599 |
$offset += 6; |
| 600 |
} elseif (substr($this->pdfdata, $offset, 4) == 'null') { |
| 601 |
// null object
|
| 602 |
$objtype = 'null'; |
| 603 |
$offset += 4; |
| 604 |
$objval = 'null'; |
| 605 |
} elseif (substr($this->pdfdata, $offset, 4) == 'true') { |
| 606 |
// boolean true object
|
| 607 |
$objtype = 'boolean'; |
| 608 |
$offset += 4; |
| 609 |
$objval = 'true'; |
| 610 |
} elseif (substr($this->pdfdata, $offset, 5) == 'false') { |
| 611 |
// boolean false object
|
| 612 |
$objtype = 'boolean'; |
| 613 |
$offset += 5; |
| 614 |
$objval = 'false'; |
| 615 |
} elseif (substr($this->pdfdata, $offset, 6) == 'stream') { |
| 616 |
// start stream object
|
| 617 |
$objtype = 'stream'; |
| 618 |
$offset += 6; |
| 619 |
if (preg_match('/^([\r]?[\n])/isU', substr($this->pdfdata, $offset), $matches) == 1) { |
| 620 |
$offset += strlen($matches[0]); |
| 621 |
if (preg_match('/([\r]?[\n])?(endstream)[\x09\x0a\x0c\x0d\x20]/isU', substr($this->pdfdata, $offset), $matches, PREG_OFFSET_CAPTURE) == 1) { |
| 622 |
$objval = substr($this->pdfdata, $offset, $matches[0][1]); |
| 623 |
$offset += $matches[2][1]; |
| 624 |
} |
| 625 |
} |
| 626 |
} elseif (substr($this->pdfdata, $offset, 9) == 'endstream') { |
| 627 |
// end stream object
|
| 628 |
$objtype = 'endstream'; |
| 629 |
$offset += 9; |
| 630 |
} elseif (preg_match('/^([0-9]+)[\s]+([0-9]+)[\s]+R/iU', substr($this->pdfdata, $offset, 33), $matches) == 1) { |
| 631 |
// indirect object reference
|
| 632 |
$objtype = 'ojbref'; |
| 633 |
$offset += strlen($matches[0]); |
| 634 |
$objval = intval($matches[1]).'_'.intval($matches[2]); |
| 635 |
} elseif (preg_match('/^([0-9]+)[\s]+([0-9]+)[\s]+obj/iU', substr($this->pdfdata, $offset, 33), $matches) == 1) { |
| 636 |
// object start
|
| 637 |
$objtype = 'ojb'; |
| 638 |
$objval = intval($matches[1]).'_'.intval($matches[2]); |
| 639 |
$offset += strlen ($matches[0]); |
| 640 |
} elseif (($numlen = strspn($this->pdfdata, '+-.0123456789', $offset)) > 0) { |
| 641 |
// numeric object
|
| 642 |
$objtype = 'numeric'; |
| 643 |
$objval = substr($this->pdfdata, $offset, $numlen); |
| 644 |
$offset += $numlen; |
| 645 |
} |
| 646 |
break;
|
| 647 |
} |
| 648 |
} |
| 649 |
return array($objtype, $objval, $offset); |
| 650 |
} |
| 651 |
|
| 652 |
/**
|
| 653 |
* Get content of indirect object.
|
| 654 |
* @param $obj_ref (string) Object number and generation number separated by underscore character.
|
| 655 |
* @param $offset (int) Object offset.
|
| 656 |
* @param $decoding (boolean) If true decode streams.
|
| 657 |
* @return array containing object data.
|
| 658 |
* @protected
|
| 659 |
* @since 1.0.000 (2011-05-24)
|
| 660 |
*/
|
| 661 |
protected function getIndirectObject($obj_ref, $offset=0, $decoding=true) { |
| 662 |
$obj = explode('_', $obj_ref); |
| 663 |
if (($obj === false) OR (count($obj) != 2)) { |
| 664 |
$this->Error('Invalid object reference: '.$obj); |
| 665 |
return;
|
| 666 |
} |
| 667 |
$objref = $obj[0].' '.$obj[1].' obj'; |
| 668 |
if (strpos($this->pdfdata, $objref, $offset) != $offset) { |
| 669 |
// an indirect reference to an undefined object shall be considered a reference to the null object
|
| 670 |
return array('null', 'null', $offset); |
| 671 |
} |
| 672 |
// starting position of object content
|
| 673 |
$offset += strlen($objref); |
| 674 |
// get array of object content
|
| 675 |
$objdata = array(); |
| 676 |
$i = 0; // object main index |
| 677 |
do {
|
| 678 |
// get element
|
| 679 |
$element = $this->getRawObject($offset); |
| 680 |
$offset = $element[2]; |
| 681 |
// decode stream using stream's dictionary information
|
| 682 |
if ($decoding AND ($element[0] == 'stream') AND (isset($objdata[($i - 1)][0])) AND ($objdata[($i - 1)][0] == '<<')) { |
| 683 |
$element[3] = $this->decodeStream($objdata[($i - 1)][1], $element[1]); |
| 684 |
} |
| 685 |
$objdata[$i] = $element; |
| 686 |
++$i;
|
| 687 |
} while ($element[0] != 'endobj'); |
| 688 |
// remove closing delimiter
|
| 689 |
array_pop($objdata); |
| 690 |
// return raw object content
|
| 691 |
return $objdata; |
| 692 |
} |
| 693 |
|
| 694 |
/**
|
| 695 |
* Get the content of object, resolving indect object reference if necessary.
|
| 696 |
* @param $obj (string) Object value.
|
| 697 |
* @return array containing object data.
|
| 698 |
* @protected
|
| 699 |
* @since 1.0.000 (2011-06-26)
|
| 700 |
*/
|
| 701 |
protected function getObjectVal($obj) { |
| 702 |
if ($obj[0] == 'objref') { |
| 703 |
// reference to indirect object
|
| 704 |
if (isset($this->objects[$obj[1]])) { |
| 705 |
// this object has been already parsed
|
| 706 |
return $this->objects[$obj[1]]; |
| 707 |
} elseif (isset($this->xref[$obj[1]])) { |
| 708 |
// parse new object
|
| 709 |
$this->objects[$obj[1]] = $this->getIndirectObject($obj[1], $this->xref[$obj[1]], false); |
| 710 |
return $this->objects[$obj[1]]; |
| 711 |
} |
| 712 |
} |
| 713 |
return $obj; |
| 714 |
} |
| 715 |
|
| 716 |
/**
|
| 717 |
* Decode the specified stream.
|
| 718 |
* @param $sdic (array) Stream's dictionary array.
|
| 719 |
* @param $stream (string) Stream to decode.
|
| 720 |
* @return array containing decoded stream data and remaining filters.
|
| 721 |
* @protected
|
| 722 |
* @since 1.0.000 (2011-06-22)
|
| 723 |
*/
|
| 724 |
protected function decodeStream($sdic, $stream) { |
| 725 |
// get stream lenght and filters
|
| 726 |
$slength = strlen($stream); |
| 727 |
if ($slength <= 0) { |
| 728 |
return array('', array()); |
| 729 |
} |
| 730 |
$filters = array(); |
| 731 |
foreach ($sdic as $k => $v) { |
| 732 |
if ($v[0] == '/') { |
| 733 |
if (($v[1] == 'Length') AND (isset($sdic[($k + 1)])) AND ($sdic[($k + 1)][0] == 'numeric')) { |
| 734 |
// get declared stream lenght
|
| 735 |
$declength = intval($sdic[($k + 1)][1]); |
| 736 |
if ($declength < $slength) { |
| 737 |
$stream = substr($stream, 0, $declength); |
| 738 |
$slength = $declength; |
| 739 |
} |
| 740 |
} elseif (($v[1] == 'Filter') AND (isset($sdic[($k + 1)]))) { |
| 741 |
// resolve indirect object
|
| 742 |
$objval = $this->getObjectVal($sdic[($k + 1)]); |
| 743 |
if ($objval[0] == '/') { |
| 744 |
// single filter
|
| 745 |
$filters[] = $objval[1]; |
| 746 |
} elseif ($objval[0] == '[') { |
| 747 |
// array of filters
|
| 748 |
foreach ($objval[1] as $flt) { |
| 749 |
if ($flt[0] == '/') { |
| 750 |
$filters[] = $flt[1]; |
| 751 |
} |
| 752 |
} |
| 753 |
} |
| 754 |
} |
| 755 |
} |
| 756 |
} |
| 757 |
// decode the stream
|
| 758 |
$remaining_filters = array(); |
| 759 |
foreach ($filters as $filter) { |
| 760 |
if (in_array($filter, TCPDF_FILTERS::getAvailableFilters())) { |
| 761 |
try {
|
| 762 |
$stream = TCPDF_FILTERS::decodeFilter($filter, $stream); |
| 763 |
} catch (Exception $e) { |
| 764 |
$emsg = $e->getMessage(); |
| 765 |
if ((($emsg[0] == '~') AND !$this->cfg['ignore_missing_filter_decoders']) |
| 766 |
OR (($emsg[0] != '~') AND !$this->cfg['ignore_filter_decoding_errors'])) { |
| 767 |
$this->Error($e->getMessage()); |
| 768 |
} |
| 769 |
} |
| 770 |
} else {
|
| 771 |
// add missing filter to array
|
| 772 |
$remaining_filters[] = $filter; |
| 773 |
} |
| 774 |
} |
| 775 |
return array($stream, $remaining_filters); |
| 776 |
} |
| 777 |
|
| 778 |
/**
|
| 779 |
* Throw an exception or print an error message and die if the K_TCPDF_PARSER_THROW_EXCEPTION_ERROR constant is set to true.
|
| 780 |
* @param $msg (string) The error message
|
| 781 |
* @public
|
| 782 |
* @since 1.0.000 (2011-05-23)
|
| 783 |
*/
|
| 784 |
public function Error($msg) { |
| 785 |
if ($this->cfg['die_for_errors']) { |
| 786 |
die('<strong>TCPDF_PARSER ERROR: </strong>'.$msg); |
| 787 |
} else {
|
| 788 |
throw new Exception('TCPDF_PARSER ERROR: '.$msg); |
| 789 |
} |
| 790 |
} |
| 791 |
|
| 792 |
} // END OF TCPDF_PARSER CLASS
|
| 793 |
|
| 794 |
//============================================================+
|
| 795 |
// END OF FILE
|
| 796 |
//============================================================+
|