|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
/** |
|
4
|
|
|
* This file is based on code of tecnickcom/TCPDF PDF library. |
|
5
|
|
|
* |
|
6
|
|
|
* Original author Nicola Asuni ([email protected]) and |
|
7
|
|
|
* contributors (https://github.com/tecnickcom/TCPDF/graphs/contributors). |
|
8
|
|
|
* |
|
9
|
|
|
* @see https://github.com/tecnickcom/TCPDF |
|
10
|
|
|
* |
|
11
|
|
|
* Original code was licensed on the terms of the LGPL v3. |
|
12
|
|
|
* |
|
13
|
|
|
* ------------------------------------------------------------------------------ |
|
14
|
|
|
* |
|
15
|
|
|
* @file This file is part of the PdfParser library. |
|
16
|
|
|
* |
|
17
|
|
|
* @author Konrad Abicht <[email protected]> |
|
18
|
|
|
* @date 2020-01-06 |
|
19
|
|
|
* |
|
20
|
|
|
* @license LGPLv3 |
|
21
|
|
|
* @url <https://github.com/smalot/pdfparser> |
|
22
|
|
|
* |
|
23
|
|
|
* PdfParser is a pdf library written in PHP, extraction oriented. |
|
24
|
|
|
* Copyright (C) 2017 - Sébastien MALOT <[email protected]> |
|
25
|
|
|
* |
|
26
|
|
|
* This program is free software: you can redistribute it and/or modify |
|
27
|
|
|
* it under the terms of the GNU Lesser General Public License as published by |
|
28
|
|
|
* the Free Software Foundation, either version 3 of the License, or |
|
29
|
|
|
* (at your option) any later version. |
|
30
|
|
|
* |
|
31
|
|
|
* This program is distributed in the hope that it will be useful, |
|
32
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
33
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
34
|
|
|
* GNU Lesser General Public License for more details. |
|
35
|
|
|
* |
|
36
|
|
|
* You should have received a copy of the GNU Lesser General Public License |
|
37
|
|
|
* along with this program. |
|
38
|
|
|
* If not, see <http://www.pdfparser.org/sites/default/LICENSE.txt>. |
|
39
|
|
|
*/ |
|
40
|
|
|
|
|
41
|
|
|
namespace Smalot\PdfParser\RawData; |
|
42
|
|
|
|
|
43
|
|
|
use Exception; |
|
44
|
|
|
|
|
45
|
|
|
class RawDataParser |
|
46
|
|
|
{ |
|
47
|
|
|
/** |
|
48
|
|
|
* Configuration array. |
|
49
|
|
|
*/ |
|
50
|
|
|
protected $cfg = [ |
|
51
|
|
|
// if `true` ignore filter decoding errors |
|
52
|
|
|
'ignore_filter_decoding_errors' => true, |
|
53
|
|
|
// if `true` ignore missing filter decoding errors |
|
54
|
|
|
'ignore_missing_filter_decoders' => true, |
|
55
|
|
|
]; |
|
56
|
|
|
|
|
57
|
|
|
/** |
|
58
|
|
|
* @param array $cfg Configuration array, default is [] |
|
59
|
|
|
*/ |
|
60
|
|
|
public function __construct($cfg = []) |
|
61
|
|
|
{ |
|
62
|
|
|
// merge given array with default values |
|
63
|
|
|
$this->cfg = array_merge($this->cfg, $cfg); |
|
64
|
|
|
|
|
65
|
|
|
$this->filterHelper = new FilterHelper(); |
|
|
|
|
|
|
66
|
|
|
} |
|
67
|
|
|
|
|
68
|
|
|
/** |
|
69
|
|
|
* Decode the specified stream. |
|
70
|
|
|
* |
|
71
|
|
|
* @param string $pdfData PDF data |
|
72
|
|
|
* @param array $sdic Stream's dictionary array |
|
73
|
|
|
* @param string $stream Stream to decode |
|
74
|
|
|
* |
|
75
|
|
|
* @return array containing decoded stream data and remaining filters |
|
76
|
|
|
*/ |
|
77
|
|
|
public function decodeStream($pdfData, $sdic, $stream) |
|
78
|
|
|
{ |
|
79
|
|
|
// get stream length and filters |
|
80
|
|
|
$slength = \strlen($stream); |
|
81
|
|
|
if ($slength <= 0) { |
|
82
|
|
|
return ['', []]; |
|
83
|
|
|
} |
|
84
|
|
|
$filters = []; |
|
85
|
|
|
foreach ($sdic as $k => $v) { |
|
86
|
|
|
if ('/' == $v[0]) { |
|
87
|
|
|
if (('Length' == $v[1]) and (isset($sdic[($k + 1)])) and ('numeric' == $sdic[($k + 1)][0])) { |
|
88
|
|
|
// get declared stream length |
|
89
|
|
|
$declength = (int) ($sdic[($k + 1)][1]); |
|
90
|
|
|
if ($declength < $slength) { |
|
91
|
|
|
$stream = substr($stream, 0, $declength); |
|
92
|
|
|
$slength = $declength; |
|
93
|
|
|
} |
|
94
|
|
|
} elseif (('Filter' == $v[1]) and (isset($sdic[($k + 1)]))) { |
|
95
|
|
|
// resolve indirect object |
|
96
|
|
|
$objval = $this->getObjectVal($pdfData, $sdic[($k + 1)]); |
|
97
|
|
|
if ('/' == $objval[0]) { |
|
98
|
|
|
// single filter |
|
99
|
|
|
$filters[] = $objval[1]; |
|
100
|
|
|
} elseif ('[' == $objval[0]) { |
|
101
|
|
|
// array of filters |
|
102
|
|
|
foreach ($objval[1] as $flt) { |
|
103
|
|
|
if ('/' == $flt[0]) { |
|
104
|
|
|
$filters[] = $flt[1]; |
|
105
|
|
|
} |
|
106
|
|
|
} |
|
107
|
|
|
} |
|
108
|
|
|
} |
|
109
|
|
|
} |
|
110
|
|
|
} |
|
111
|
|
|
|
|
112
|
|
|
// decode the stream |
|
113
|
|
|
$remaining_filters = []; |
|
114
|
|
|
foreach ($filters as $filter) { |
|
115
|
|
|
if (\in_array($filter, $this->filterHelper->getAvailableFilters())) { |
|
116
|
|
|
try { |
|
117
|
|
|
$stream = $this->filterHelper->decodeFilter($filter, $stream); |
|
118
|
|
|
} catch (Exception $e) { |
|
119
|
|
|
$emsg = $e->getMessage(); |
|
120
|
|
|
if ((('~' == $emsg[0]) && !$this->cfg['ignore_missing_filter_decoders']) |
|
121
|
|
|
|| (('~' != $emsg[0]) && !$this->cfg['ignore_filter_decoding_errors']) |
|
122
|
|
|
) { |
|
123
|
|
|
throw new Exception($e->getMessage()); |
|
124
|
|
|
} |
|
125
|
|
|
} |
|
126
|
|
|
} else { |
|
127
|
|
|
// add missing filter to array |
|
128
|
|
|
$remaining_filters[] = $filter; |
|
129
|
|
|
} |
|
130
|
|
|
} |
|
131
|
|
|
|
|
132
|
|
|
return [$stream, $remaining_filters]; |
|
133
|
|
|
} |
|
134
|
|
|
|
|
135
|
|
|
/** |
|
136
|
|
|
* Decode the Cross-Reference section |
|
137
|
|
|
* |
|
138
|
|
|
* @param string $pdfData PDF data |
|
139
|
|
|
* @param int $startxref Offset at which the xref section starts (position of the 'xref' keyword) |
|
140
|
|
|
* @param array $xref Previous xref array (if any) |
|
141
|
|
|
* |
|
142
|
|
|
* @return array containing xref and trailer data |
|
143
|
|
|
*/ |
|
144
|
|
|
public function decodeXref($pdfData, $startxref, $xref = []) |
|
145
|
|
|
{ |
|
146
|
|
|
$startxref += 4; // 4 is the length of the word 'xref' |
|
147
|
|
|
// 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) |
|
148
|
|
|
$offset = $startxref + strspn($pdfData, "\x00\x09\x0a\x0c\x0d\x20", $startxref); |
|
149
|
|
|
// initialize object number |
|
150
|
|
|
$obj_num = 0; |
|
151
|
|
|
// search for cross-reference entries or subsection |
|
152
|
|
|
while (preg_match('/([0-9]+)[\x20]([0-9]+)[\x20]?([nf]?)(\r\n|[\x20]?[\r\n])/', $pdfData, $matches, PREG_OFFSET_CAPTURE, $offset) > 0) { |
|
153
|
|
|
if ($matches[0][1] != $offset) { |
|
154
|
|
|
// we are on another section |
|
155
|
|
|
break; |
|
156
|
|
|
} |
|
157
|
|
|
$offset += \strlen($matches[0][0]); |
|
158
|
|
|
if ('n' == $matches[3][0]) { |
|
159
|
|
|
// create unique object index: [object number]_[generation number] |
|
160
|
|
|
$index = $obj_num.'_'.(int) ($matches[2][0]); |
|
161
|
|
|
// check if object already exist |
|
162
|
|
|
if (!isset($xref['xref'][$index])) { |
|
163
|
|
|
// store object offset position |
|
164
|
|
|
$xref['xref'][$index] = (int) ($matches[1][0]); |
|
165
|
|
|
} |
|
166
|
|
|
++$obj_num; |
|
167
|
|
|
} elseif ('f' == $matches[3][0]) { |
|
168
|
|
|
++$obj_num; |
|
169
|
|
|
} else { |
|
170
|
|
|
// object number (index) |
|
171
|
|
|
$obj_num = (int) ($matches[1][0]); |
|
172
|
|
|
} |
|
173
|
|
|
} |
|
174
|
|
|
// get trailer data |
|
175
|
|
|
if (preg_match('/trailer[\s]*<<(.*)>>/isU', $pdfData, $matches, PREG_OFFSET_CAPTURE, $offset) > 0) { |
|
176
|
|
|
$trailer_data = $matches[1][0]; |
|
177
|
|
|
if (!isset($xref['trailer']) or empty($xref['trailer'])) { |
|
178
|
|
|
// get only the last updated version |
|
179
|
|
|
$xref['trailer'] = []; |
|
180
|
|
|
// parse trailer_data |
|
181
|
|
|
if (preg_match('/Size[\s]+([0-9]+)/i', $trailer_data, $matches) > 0) { |
|
182
|
|
|
$xref['trailer']['size'] = (int) ($matches[1]); |
|
183
|
|
|
} |
|
184
|
|
|
if (preg_match('/Root[\s]+([0-9]+)[\s]+([0-9]+)[\s]+R/i', $trailer_data, $matches) > 0) { |
|
185
|
|
|
$xref['trailer']['root'] = (int) ($matches[1]).'_'.(int) ($matches[2]); |
|
186
|
|
|
} |
|
187
|
|
|
if (preg_match('/Encrypt[\s]+([0-9]+)[\s]+([0-9]+)[\s]+R/i', $trailer_data, $matches) > 0) { |
|
188
|
|
|
$xref['trailer']['encrypt'] = (int) ($matches[1]).'_'.(int) ($matches[2]); |
|
189
|
|
|
} |
|
190
|
|
|
if (preg_match('/Info[\s]+([0-9]+)[\s]+([0-9]+)[\s]+R/i', $trailer_data, $matches) > 0) { |
|
191
|
|
|
$xref['trailer']['info'] = (int) ($matches[1]).'_'.(int) ($matches[2]); |
|
192
|
|
|
} |
|
193
|
|
|
if (preg_match('/ID[\s]*[\[][\s]*[<]([^>]*)[>][\s]*[<]([^>]*)[>]/i', $trailer_data, $matches) > 0) { |
|
194
|
|
|
$xref['trailer']['id'] = []; |
|
195
|
|
|
$xref['trailer']['id'][0] = $matches[1]; |
|
196
|
|
|
$xref['trailer']['id'][1] = $matches[2]; |
|
197
|
|
|
} |
|
198
|
|
|
} |
|
199
|
|
|
if (preg_match('/Prev[\s]+([0-9]+)/i', $trailer_data, $matches) > 0) { |
|
200
|
|
|
// get previous xref |
|
201
|
|
|
$xref = $this->getXrefData($pdfData, (int) ($matches[1]), $xref); |
|
202
|
|
|
} |
|
203
|
|
|
} else { |
|
204
|
|
|
throw new Exception('Unable to find trailer'); |
|
205
|
|
|
} |
|
206
|
|
|
|
|
207
|
|
|
return $xref; |
|
208
|
|
|
} |
|
209
|
|
|
|
|
210
|
|
|
/** |
|
211
|
|
|
* Decode the Cross-Reference Stream section |
|
212
|
|
|
* |
|
213
|
|
|
* @param string $pdfData PDF data |
|
214
|
|
|
* @param int $startxref Offset at which the xref section starts |
|
215
|
|
|
* @param array $xref Previous xref array (if any) |
|
216
|
|
|
* |
|
217
|
|
|
* @return array containing xref and trailer data |
|
218
|
|
|
* |
|
219
|
|
|
* @throws Exception if unknown PNG predictor detected |
|
220
|
|
|
*/ |
|
221
|
|
|
public function decodeXrefStream($pdfData, $startxref, $xref = []) |
|
222
|
|
|
{ |
|
223
|
|
|
// try to read Cross-Reference Stream |
|
224
|
|
|
$xrefobj = $this->getRawObject($pdfData, $startxref); |
|
225
|
|
|
$xrefcrs = $this->getIndirectObject($pdfData, $xrefobj[1], $startxref, true); |
|
226
|
|
|
if (!isset($xref['trailer']) or empty($xref['trailer'])) { |
|
227
|
|
|
// get only the last updated version |
|
228
|
|
|
$xref['trailer'] = []; |
|
229
|
|
|
$filltrailer = true; |
|
230
|
|
|
} else { |
|
231
|
|
|
$filltrailer = false; |
|
232
|
|
|
} |
|
233
|
|
|
if (!isset($xref['xref'])) { |
|
234
|
|
|
$xref['xref'] = []; |
|
235
|
|
|
} |
|
236
|
|
|
$valid_crs = false; |
|
237
|
|
|
$columns = 0; |
|
238
|
|
|
$sarr = $xrefcrs[0][1]; |
|
239
|
|
|
if (!\is_array($sarr)) { |
|
240
|
|
|
$sarr = []; |
|
241
|
|
|
} |
|
242
|
|
|
foreach ($sarr as $k => $v) { |
|
243
|
|
|
if (('/' == $v[0]) and ('Type' == $v[1]) and (isset($sarr[($k + 1)]) and ('/' == $sarr[($k + 1)][0]) and ('XRef' == $sarr[($k + 1)][1]))) { |
|
244
|
|
|
$valid_crs = true; |
|
245
|
|
|
} elseif (('/' == $v[0]) and ('Index' == $v[1]) and (isset($sarr[($k + 1)]))) { |
|
246
|
|
|
// first object number in the subsection |
|
247
|
|
|
$index_first = (int) ($sarr[($k + 1)][1][0][1]); |
|
248
|
|
|
// number of entries in the subsection |
|
249
|
|
|
$index_entries = (int) ($sarr[($k + 1)][1][1][1]); |
|
|
|
|
|
|
250
|
|
|
} elseif (('/' == $v[0]) and ('Prev' == $v[1]) and (isset($sarr[($k + 1)]) and ('numeric' == $sarr[($k + 1)][0]))) { |
|
251
|
|
|
// get previous xref offset |
|
252
|
|
|
$prevxref = (int) ($sarr[($k + 1)][1]); |
|
253
|
|
|
} elseif (('/' == $v[0]) and ('W' == $v[1]) and (isset($sarr[($k + 1)]))) { |
|
254
|
|
|
// number of bytes (in the decoded stream) of the corresponding field |
|
255
|
|
|
$wb = []; |
|
256
|
|
|
$wb[0] = (int) ($sarr[($k + 1)][1][0][1]); |
|
257
|
|
|
$wb[1] = (int) ($sarr[($k + 1)][1][1][1]); |
|
258
|
|
|
$wb[2] = (int) ($sarr[($k + 1)][1][2][1]); |
|
259
|
|
|
} elseif (('/' == $v[0]) and ('DecodeParms' == $v[1]) and (isset($sarr[($k + 1)][1]))) { |
|
260
|
|
|
$decpar = $sarr[($k + 1)][1]; |
|
261
|
|
|
foreach ($decpar as $kdc => $vdc) { |
|
262
|
|
|
if (('/' == $vdc[0]) and ('Columns' == $vdc[1]) and (isset($decpar[($kdc + 1)]) and ('numeric' == $decpar[($kdc + 1)][0]))) { |
|
263
|
|
|
$columns = (int) ($decpar[($kdc + 1)][1]); |
|
264
|
|
|
} elseif (('/' == $vdc[0]) and ('Predictor' == $vdc[1]) and (isset($decpar[($kdc + 1)]) and ('numeric' == $decpar[($kdc + 1)][0]))) { |
|
265
|
|
|
$predictor = (int) ($decpar[($kdc + 1)][1]); |
|
|
|
|
|
|
266
|
|
|
} |
|
267
|
|
|
} |
|
268
|
|
|
} elseif ($filltrailer) { |
|
269
|
|
|
if (('/' == $v[0]) and ('Size' == $v[1]) and (isset($sarr[($k + 1)]) and ('numeric' == $sarr[($k + 1)][0]))) { |
|
270
|
|
|
$xref['trailer']['size'] = $sarr[($k + 1)][1]; |
|
271
|
|
|
} elseif (('/' == $v[0]) and ('Root' == $v[1]) and (isset($sarr[($k + 1)]) and ('objref' == $sarr[($k + 1)][0]))) { |
|
272
|
|
|
$xref['trailer']['root'] = $sarr[($k + 1)][1]; |
|
273
|
|
|
} elseif (('/' == $v[0]) and ('Info' == $v[1]) and (isset($sarr[($k + 1)]) and ('objref' == $sarr[($k + 1)][0]))) { |
|
274
|
|
|
$xref['trailer']['info'] = $sarr[($k + 1)][1]; |
|
275
|
|
|
} elseif (('/' == $v[0]) and ('Encrypt' == $v[1]) and (isset($sarr[($k + 1)]) and ('objref' == $sarr[($k + 1)][0]))) { |
|
276
|
|
|
$xref['trailer']['encrypt'] = $sarr[($k + 1)][1]; |
|
277
|
|
|
} elseif (('/' == $v[0]) and ('ID' == $v[1]) and (isset($sarr[($k + 1)]))) { |
|
278
|
|
|
$xref['trailer']['id'] = []; |
|
279
|
|
|
$xref['trailer']['id'][0] = $sarr[($k + 1)][1][0][1]; |
|
280
|
|
|
$xref['trailer']['id'][1] = $sarr[($k + 1)][1][1][1]; |
|
281
|
|
|
} |
|
282
|
|
|
} |
|
283
|
|
|
} |
|
284
|
|
|
// decode data |
|
285
|
|
|
if ($valid_crs and isset($xrefcrs[1][3][0])) { |
|
286
|
|
|
// number of bytes in a row |
|
287
|
|
|
$rowlen = ($columns + 1); |
|
288
|
|
|
// convert the stream into an array of integers |
|
289
|
|
|
$sdata = unpack('C*', $xrefcrs[1][3][0]); |
|
290
|
|
|
// split the rows |
|
291
|
|
|
$sdata = array_chunk($sdata, $rowlen); |
|
|
|
|
|
|
292
|
|
|
// initialize decoded array |
|
293
|
|
|
$ddata = []; |
|
294
|
|
|
// initialize first row with zeros |
|
295
|
|
|
$prev_row = array_fill(0, $rowlen, 0); |
|
296
|
|
|
// for each row apply PNG unpredictor |
|
297
|
|
|
foreach ($sdata as $k => $row) { |
|
298
|
|
|
// initialize new row |
|
299
|
|
|
$ddata[$k] = []; |
|
300
|
|
|
// get PNG predictor value |
|
301
|
|
|
$predictor = (10 + $row[0]); |
|
302
|
|
|
// for each byte on the row |
|
303
|
|
|
for ($i = 1; $i <= $columns; ++$i) { |
|
304
|
|
|
// new index |
|
305
|
|
|
$j = ($i - 1); |
|
306
|
|
|
$row_up = $prev_row[$j]; |
|
307
|
|
|
if (1 == $i) { |
|
308
|
|
|
$row_left = 0; |
|
309
|
|
|
$row_upleft = 0; |
|
310
|
|
|
} else { |
|
311
|
|
|
$row_left = $row[($i - 1)]; |
|
312
|
|
|
$row_upleft = $prev_row[($j - 1)]; |
|
313
|
|
|
} |
|
314
|
|
|
switch ($predictor) { |
|
315
|
|
|
case 10: // PNG prediction (on encoding, PNG None on all rows) |
|
316
|
|
|
$ddata[$k][$j] = $row[$i]; |
|
317
|
|
|
break; |
|
318
|
|
|
|
|
319
|
|
|
case 11: // PNG prediction (on encoding, PNG Sub on all rows) |
|
320
|
|
|
$ddata[$k][$j] = (($row[$i] + $row_left) & 0xff); |
|
321
|
|
|
break; |
|
322
|
|
|
|
|
323
|
|
|
case 12: // PNG prediction (on encoding, PNG Up on all rows) |
|
324
|
|
|
$ddata[$k][$j] = (($row[$i] + $row_up) & 0xff); |
|
325
|
|
|
break; |
|
326
|
|
|
|
|
327
|
|
|
case 13: // PNG prediction (on encoding, PNG Average on all rows) |
|
328
|
|
|
$ddata[$k][$j] = (($row[$i] + (($row_left + $row_up) / 2)) & 0xff); |
|
329
|
|
|
break; |
|
330
|
|
|
|
|
331
|
|
|
case 14: // PNG prediction (on encoding, PNG Paeth on all rows) |
|
332
|
|
|
// initial estimate |
|
333
|
|
|
$p = ($row_left + $row_up - $row_upleft); |
|
334
|
|
|
// distances |
|
335
|
|
|
$pa = abs($p - $row_left); |
|
336
|
|
|
$pb = abs($p - $row_up); |
|
337
|
|
|
$pc = abs($p - $row_upleft); |
|
338
|
|
|
$pmin = min($pa, $pb, $pc); |
|
339
|
|
|
// return minimum distance |
|
340
|
|
|
switch ($pmin) { |
|
341
|
|
|
case $pa: |
|
342
|
|
|
$ddata[$k][$j] = (($row[$i] + $row_left) & 0xff); |
|
343
|
|
|
break; |
|
344
|
|
|
|
|
345
|
|
|
case $pb: |
|
346
|
|
|
$ddata[$k][$j] = (($row[$i] + $row_up) & 0xff); |
|
347
|
|
|
break; |
|
348
|
|
|
|
|
349
|
|
|
case $pc: |
|
350
|
|
|
$ddata[$k][$j] = (($row[$i] + $row_upleft) & 0xff); |
|
351
|
|
|
break; |
|
352
|
|
|
} |
|
353
|
|
|
break; |
|
354
|
|
|
|
|
355
|
|
|
default: // PNG prediction (on encoding, PNG optimum) |
|
356
|
|
|
throw new Exception('Unknown PNG predictor'); |
|
357
|
|
|
break; |
|
358
|
|
|
} |
|
359
|
|
|
} |
|
360
|
|
|
$prev_row = $ddata[$k]; |
|
361
|
|
|
} // end for each row |
|
362
|
|
|
// complete decoding |
|
363
|
|
|
$sdata = []; |
|
364
|
|
|
// for every row |
|
365
|
|
|
foreach ($ddata as $k => $row) { |
|
366
|
|
|
// initialize new row |
|
367
|
|
|
$sdata[$k] = [0, 0, 0]; |
|
368
|
|
|
if (0 == $wb[0]) { |
|
|
|
|
|
|
369
|
|
|
// default type field |
|
370
|
|
|
$sdata[$k][0] = 1; |
|
371
|
|
|
} |
|
372
|
|
|
$i = 0; // count bytes in the row |
|
373
|
|
|
// for every column |
|
374
|
|
|
for ($c = 0; $c < 3; ++$c) { |
|
375
|
|
|
// for every byte on the column |
|
376
|
|
|
for ($b = 0; $b < $wb[$c]; ++$b) { |
|
377
|
|
|
if (isset($row[$i])) { |
|
378
|
|
|
$sdata[$k][$c] += ($row[$i] << (($wb[$c] - 1 - $b) * 8)); |
|
379
|
|
|
} |
|
380
|
|
|
++$i; |
|
381
|
|
|
} |
|
382
|
|
|
} |
|
383
|
|
|
} |
|
384
|
|
|
$ddata = []; |
|
|
|
|
|
|
385
|
|
|
// fill xref |
|
386
|
|
|
if (isset($index_first)) { |
|
387
|
|
|
$obj_num = $index_first; |
|
388
|
|
|
} else { |
|
389
|
|
|
$obj_num = 0; |
|
390
|
|
|
} |
|
391
|
|
|
foreach ($sdata as $k => $row) { |
|
392
|
|
|
switch ($row[0]) { |
|
393
|
|
|
case 0: // (f) linked list of free objects |
|
394
|
|
|
break; |
|
395
|
|
|
|
|
396
|
|
|
case 1: // (n) objects that are in use but are not compressed |
|
397
|
|
|
// create unique object index: [object number]_[generation number] |
|
398
|
|
|
$index = $obj_num.'_'.$row[2]; |
|
399
|
|
|
// check if object already exist |
|
400
|
|
|
if (!isset($xref['xref'][$index])) { |
|
401
|
|
|
// store object offset position |
|
402
|
|
|
$xref['xref'][$index] = $row[1]; |
|
403
|
|
|
} |
|
404
|
|
|
break; |
|
405
|
|
|
|
|
406
|
|
|
case 2: // compressed objects |
|
407
|
|
|
// $row[1] = object number of the object stream in which this object is stored |
|
408
|
|
|
// $row[2] = index of this object within the object stream |
|
409
|
|
|
$index = $row[1].'_0_'.$row[2]; |
|
410
|
|
|
$xref['xref'][$index] = -1; |
|
411
|
|
|
break; |
|
412
|
|
|
|
|
413
|
|
|
default: // null objects |
|
414
|
|
|
break; |
|
415
|
|
|
} |
|
416
|
|
|
++$obj_num; |
|
417
|
|
|
} |
|
418
|
|
|
} // end decoding data |
|
419
|
|
|
if (isset($prevxref)) { |
|
420
|
|
|
// get previous xref |
|
421
|
|
|
$xref = $this->getXrefData($pdfData, $prevxref, $xref); |
|
422
|
|
|
} |
|
423
|
|
|
|
|
424
|
|
|
return $xref; |
|
425
|
|
|
} |
|
426
|
|
|
|
|
427
|
|
|
/** |
|
428
|
|
|
* Get content of indirect object. |
|
429
|
|
|
* |
|
430
|
|
|
* @param string $pdfData PDF data |
|
431
|
|
|
* @param string $obj_ref Object number and generation number separated by underscore character |
|
432
|
|
|
* @param int $offset Object offset |
|
433
|
|
|
* @param bool $decoding If true decode streams |
|
434
|
|
|
* |
|
435
|
|
|
* @return array containing object data |
|
436
|
|
|
* |
|
437
|
|
|
* @throws Exception if invalid object reference found |
|
438
|
|
|
*/ |
|
439
|
|
|
public function getIndirectObject($pdfData, $obj_ref, $offset = 0, $decoding = true) |
|
440
|
|
|
{ |
|
441
|
|
|
$obj = explode('_', $obj_ref); |
|
442
|
|
|
if ((false === $obj) or (2 != \count($obj))) { |
|
443
|
|
|
throw new Exception('Invalid object reference: '.$obj); |
|
|
|
|
|
|
444
|
|
|
|
|
445
|
|
|
return; |
|
|
|
|
|
|
446
|
|
|
} |
|
447
|
|
|
$objref = $obj[0].' '.$obj[1].' obj'; |
|
448
|
|
|
// ignore leading zeros |
|
449
|
|
|
$offset += strspn($pdfData, '0', $offset); |
|
450
|
|
|
if (strpos($pdfData, $objref, $offset) != $offset) { |
|
451
|
|
|
// an indirect reference to an undefined object shall be considered a reference to the null object |
|
452
|
|
|
return ['null', 'null', $offset]; |
|
453
|
|
|
} |
|
454
|
|
|
// starting position of object content |
|
455
|
|
|
$offset += \strlen($objref); |
|
456
|
|
|
// get array of object content |
|
457
|
|
|
$objdata = []; |
|
458
|
|
|
$i = 0; // object main index |
|
459
|
|
|
do { |
|
460
|
|
|
$oldoffset = $offset; |
|
461
|
|
|
// get element |
|
462
|
|
|
$element = $this->getRawObject($pdfData, $offset); |
|
463
|
|
|
$offset = $element[2]; |
|
464
|
|
|
// decode stream using stream's dictionary information |
|
465
|
|
|
if ($decoding and ('stream' == $element[0]) and (isset($objdata[($i - 1)][0])) and ('<<' == $objdata[($i - 1)][0])) { |
|
466
|
|
|
$element[3] = $this->decodeStream($pdfData, $objdata[($i - 1)][1], $element[1]); |
|
467
|
|
|
} |
|
468
|
|
|
$objdata[$i] = $element; |
|
469
|
|
|
++$i; |
|
470
|
|
|
} while (('endobj' != $element[0]) and ($offset != $oldoffset)); |
|
471
|
|
|
// remove closing delimiter |
|
472
|
|
|
array_pop($objdata); |
|
473
|
|
|
// return raw object content |
|
474
|
|
|
return $objdata; |
|
475
|
|
|
} |
|
476
|
|
|
|
|
477
|
|
|
/** |
|
478
|
|
|
* Get the content of object, resolving indect object reference if necessary. |
|
479
|
|
|
* |
|
480
|
|
|
* @param string $pdfData PDF data |
|
481
|
|
|
* @param string $obj Object value |
|
482
|
|
|
* |
|
483
|
|
|
* @return array containing object data |
|
484
|
|
|
*/ |
|
485
|
|
|
public function getObjectVal($pdfData, $obj) |
|
486
|
|
|
{ |
|
487
|
|
|
if ('objref' == $obj[0]) { |
|
488
|
|
|
// reference to indirect object |
|
489
|
|
|
if (isset($this->objects[$obj[1]])) { |
|
490
|
|
|
// this object has been already parsed |
|
491
|
|
|
return $this->objects[$obj[1]]; |
|
492
|
|
|
} elseif (isset($this->xref[$obj[1]])) { |
|
|
|
|
|
|
493
|
|
|
// parse new object |
|
494
|
|
|
$this->objects[$obj[1]] = $this->getIndirectObject($pdfData, $obj[1], $this->xref[$obj[1]], false); |
|
|
|
|
|
|
495
|
|
|
|
|
496
|
|
|
return $this->objects[$obj[1]]; |
|
497
|
|
|
} |
|
498
|
|
|
} |
|
499
|
|
|
|
|
500
|
|
|
return $obj; |
|
|
|
|
|
|
501
|
|
|
} |
|
502
|
|
|
|
|
503
|
|
|
/** |
|
504
|
|
|
* Parses a PDF and returns extracted data. |
|
505
|
|
|
* |
|
506
|
|
|
* @param string $data PDF data to parse |
|
507
|
|
|
* |
|
508
|
|
|
* @return array array of parsed PDF document objects |
|
509
|
|
|
* |
|
510
|
|
|
* @throws Exception if empty PDF data given |
|
511
|
|
|
* @throws Exception if PDF data missing %PDF header |
|
512
|
|
|
*/ |
|
513
|
|
|
public function getParsedData($data) |
|
514
|
|
|
{ |
|
515
|
|
|
if (empty($data)) { |
|
516
|
|
|
throw new Exception('Empty PDF data given.'); |
|
517
|
|
|
} |
|
518
|
|
|
// find the pdf header starting position |
|
519
|
|
|
if (false === ($trimpos = strpos($data, '%PDF-'))) { |
|
520
|
|
|
throw new Exception('Invalid PDF data: missing %PDF header.'); |
|
521
|
|
|
} |
|
522
|
|
|
|
|
523
|
|
|
// get PDF content string |
|
524
|
|
|
$pdfData = substr($data, $trimpos); |
|
525
|
|
|
|
|
526
|
|
|
// get xref and trailer data |
|
527
|
|
|
$xref = $this->getXrefData($pdfData); |
|
528
|
|
|
|
|
529
|
|
|
// parse all document objects |
|
530
|
|
|
$objects = []; |
|
531
|
|
|
foreach ($xref['xref'] as $obj => $offset) { |
|
532
|
|
|
if (!isset($objects[$obj]) and ($offset > 0)) { |
|
533
|
|
|
// decode objects with positive offset |
|
534
|
|
|
$objects[$obj] = $this->getIndirectObject($pdfData, $obj, $offset, true); |
|
535
|
|
|
} |
|
536
|
|
|
} |
|
537
|
|
|
// release some memory |
|
538
|
|
|
unset($pdfData); |
|
539
|
|
|
$pdfData = ''; |
|
|
|
|
|
|
540
|
|
|
|
|
541
|
|
|
return [$xref, $objects]; |
|
542
|
|
|
} |
|
543
|
|
|
|
|
544
|
|
|
/** |
|
545
|
|
|
* Get object type, raw value and offset to next object |
|
546
|
|
|
* |
|
547
|
|
|
* @param int $offset Object offset |
|
548
|
|
|
* |
|
549
|
|
|
* @return array containing object type, raw value and offset to next object |
|
550
|
|
|
*/ |
|
551
|
|
|
public function getRawObject($pdfData, $offset = 0) |
|
552
|
|
|
{ |
|
553
|
|
|
$objtype = ''; // object type to be returned |
|
554
|
|
|
$objval = ''; // object value to be returned |
|
555
|
|
|
|
|
556
|
|
|
/* |
|
557
|
|
|
* skip initial white space chars: |
|
558
|
|
|
* \x00 null (NUL) |
|
559
|
|
|
* \x09 horizontal tab (HT) |
|
560
|
|
|
* \x0A line feed (LF) |
|
561
|
|
|
* \x0C form feed (FF) |
|
562
|
|
|
* \x0D carriage return (CR) |
|
563
|
|
|
* \x20 space (SP) |
|
564
|
|
|
*/ |
|
565
|
|
|
$offset += strspn($pdfData, "\x00\x09\x0a\x0c\x0d\x20", $offset); |
|
566
|
|
|
|
|
567
|
|
|
// get first char |
|
568
|
|
|
$char = $pdfData[$offset]; |
|
569
|
|
|
// get object type |
|
570
|
|
|
switch ($char) { |
|
571
|
|
|
case '%': // \x25 PERCENT SIGN |
|
572
|
|
|
// skip comment and search for next token |
|
573
|
|
|
$next = strcspn($pdfData, "\r\n", $offset); |
|
574
|
|
|
if ($next > 0) { |
|
575
|
|
|
$offset += $next; |
|
576
|
|
|
|
|
577
|
|
|
return $this->getRawObject($pdfData, $offset); |
|
578
|
|
|
} |
|
579
|
|
|
break; |
|
580
|
|
|
|
|
581
|
|
|
case '/': // \x2F SOLIDUS |
|
582
|
|
|
// name object |
|
583
|
|
|
$objtype = $char; |
|
584
|
|
|
++$offset; |
|
585
|
|
|
$pregResult = preg_match( |
|
586
|
|
|
'/^([^\x00\x09\x0a\x0c\x0d\x20\s\x28\x29\x3c\x3e\x5b\x5d\x7b\x7d\x2f\x25]+)/', |
|
587
|
|
|
substr($pdfData, $offset, 256), |
|
588
|
|
|
$matches |
|
589
|
|
|
); |
|
590
|
|
|
if (1 == $pregResult) { |
|
591
|
|
|
$objval = $matches[1]; // unescaped value |
|
592
|
|
|
$offset += \strlen($objval); |
|
593
|
|
|
} |
|
594
|
|
|
break; |
|
595
|
|
|
|
|
596
|
|
|
case '(': // \x28 LEFT PARENTHESIS |
|
597
|
|
|
case ')': // \x29 RIGHT PARENTHESIS |
|
598
|
|
|
// literal string object |
|
599
|
|
|
$objtype = $char; |
|
600
|
|
|
++$offset; |
|
601
|
|
|
$strpos = $offset; |
|
602
|
|
|
if ('(' == $char) { |
|
603
|
|
|
$open_bracket = 1; |
|
604
|
|
|
while ($open_bracket > 0) { |
|
605
|
|
|
if (!isset($pdfData[$strpos])) { |
|
606
|
|
|
break; |
|
607
|
|
|
} |
|
608
|
|
|
$ch = $pdfData[$strpos]; |
|
609
|
|
|
switch ($ch) { |
|
610
|
|
|
case '\\': // REVERSE SOLIDUS (5Ch) (Backslash) |
|
611
|
|
|
// skip next character |
|
612
|
|
|
++$strpos; |
|
613
|
|
|
break; |
|
614
|
|
|
|
|
615
|
|
|
case '(': // LEFT PARENHESIS (28h) |
|
616
|
|
|
++$open_bracket; |
|
617
|
|
|
break; |
|
618
|
|
|
|
|
619
|
|
|
case ')': // RIGHT PARENTHESIS (29h) |
|
620
|
|
|
--$open_bracket; |
|
621
|
|
|
break; |
|
622
|
|
|
} |
|
623
|
|
|
++$strpos; |
|
624
|
|
|
} |
|
625
|
|
|
$objval = substr($pdfData, $offset, ($strpos - $offset - 1)); |
|
626
|
|
|
$offset = $strpos; |
|
627
|
|
|
} |
|
628
|
|
|
break; |
|
629
|
|
|
|
|
630
|
|
|
case '[': // \x5B LEFT SQUARE BRACKET |
|
631
|
|
|
case ']': // \x5D RIGHT SQUARE BRACKET |
|
632
|
|
|
// array object |
|
633
|
|
|
$objtype = $char; |
|
634
|
|
|
++$offset; |
|
635
|
|
|
if ('[' == $char) { |
|
636
|
|
|
// get array content |
|
637
|
|
|
$objval = []; |
|
638
|
|
|
do { |
|
639
|
|
|
// get element |
|
640
|
|
|
$element = $this->getRawObject($pdfData, $offset); |
|
641
|
|
|
$offset = $element[2]; |
|
642
|
|
|
$objval[] = $element; |
|
643
|
|
|
} while (']' != $element[0]); |
|
644
|
|
|
// remove closing delimiter |
|
645
|
|
|
array_pop($objval); |
|
646
|
|
|
} |
|
647
|
|
|
break; |
|
648
|
|
|
|
|
649
|
|
|
case '<': // \x3C LESS-THAN SIGN |
|
650
|
|
|
case '>': // \x3E GREATER-THAN SIGN |
|
651
|
|
|
if (isset($pdfData[($offset + 1)]) and ($pdfData[($offset + 1)] == $char)) { |
|
652
|
|
|
// dictionary object |
|
653
|
|
|
$objtype = $char.$char; |
|
654
|
|
|
$offset += 2; |
|
655
|
|
|
if ('<' == $char) { |
|
656
|
|
|
// get array content |
|
657
|
|
|
$objval = []; |
|
658
|
|
|
do { |
|
659
|
|
|
// get element |
|
660
|
|
|
$element = $this->getRawObject($pdfData, $offset); |
|
661
|
|
|
$offset = $element[2]; |
|
662
|
|
|
$objval[] = $element; |
|
663
|
|
|
} while ('>>' != $element[0]); |
|
664
|
|
|
// remove closing delimiter |
|
665
|
|
|
array_pop($objval); |
|
666
|
|
|
} |
|
667
|
|
|
} else { |
|
668
|
|
|
// hexadecimal string object |
|
669
|
|
|
$objtype = $char; |
|
670
|
|
|
++$offset; |
|
671
|
|
|
$pregResult = preg_match( |
|
672
|
|
|
'/^([0-9A-Fa-f\x09\x0a\x0c\x0d\x20]+)>/iU', |
|
673
|
|
|
substr($pdfData, $offset), |
|
674
|
|
|
$matches |
|
675
|
|
|
); |
|
676
|
|
|
if (('<' == $char) && 1 == $pregResult) { |
|
677
|
|
|
// remove white space characters |
|
678
|
|
|
$objval = strtr($matches[1], "\x09\x0a\x0c\x0d\x20", ''); |
|
679
|
|
|
$offset += \strlen($matches[0]); |
|
680
|
|
|
} elseif (false !== ($endpos = strpos($pdfData, '>', $offset))) { |
|
681
|
|
|
$offset = $endpos + 1; |
|
682
|
|
|
} |
|
683
|
|
|
} |
|
684
|
|
|
break; |
|
685
|
|
|
|
|
686
|
|
|
default: |
|
687
|
|
|
if ('endobj' == substr($pdfData, $offset, 6)) { |
|
688
|
|
|
// indirect object |
|
689
|
|
|
$objtype = 'endobj'; |
|
690
|
|
|
$offset += 6; |
|
691
|
|
|
} elseif ('null' == substr($pdfData, $offset, 4)) { |
|
692
|
|
|
// null object |
|
693
|
|
|
$objtype = 'null'; |
|
694
|
|
|
$offset += 4; |
|
695
|
|
|
$objval = 'null'; |
|
696
|
|
|
} elseif ('true' == substr($pdfData, $offset, 4)) { |
|
697
|
|
|
// boolean true object |
|
698
|
|
|
$objtype = 'boolean'; |
|
699
|
|
|
$offset += 4; |
|
700
|
|
|
$objval = 'true'; |
|
701
|
|
|
} elseif ('false' == substr($pdfData, $offset, 5)) { |
|
702
|
|
|
// boolean false object |
|
703
|
|
|
$objtype = 'boolean'; |
|
704
|
|
|
$offset += 5; |
|
705
|
|
|
$objval = 'false'; |
|
706
|
|
|
} elseif ('stream' == substr($pdfData, $offset, 6)) { |
|
707
|
|
|
// start stream object |
|
708
|
|
|
$objtype = 'stream'; |
|
709
|
|
|
$offset += 6; |
|
710
|
|
|
if (1 == preg_match('/^([\r]?[\n])/isU', substr($pdfData, $offset), $matches)) { |
|
711
|
|
|
$offset += \strlen($matches[0]); |
|
712
|
|
|
$pregResult = preg_match( |
|
713
|
|
|
'/(endstream)[\x09\x0a\x0c\x0d\x20]/isU', |
|
714
|
|
|
substr($pdfData, $offset), |
|
715
|
|
|
$matches, |
|
716
|
|
|
PREG_OFFSET_CAPTURE |
|
717
|
|
|
); |
|
718
|
|
|
if (1 == $pregResult) { |
|
719
|
|
|
$objval = substr($pdfData, $offset, $matches[0][1]); |
|
720
|
|
|
$offset += $matches[1][1]; |
|
721
|
|
|
} |
|
722
|
|
|
} |
|
723
|
|
|
} elseif ('endstream' == substr($pdfData, $offset, 9)) { |
|
724
|
|
|
// end stream object |
|
725
|
|
|
$objtype = 'endstream'; |
|
726
|
|
|
$offset += 9; |
|
727
|
|
|
} elseif (1 == preg_match('/^([0-9]+)[\s]+([0-9]+)[\s]+R/iU', substr($pdfData, $offset, 33), $matches)) { |
|
728
|
|
|
// indirect object reference |
|
729
|
|
|
$objtype = 'objref'; |
|
730
|
|
|
$offset += \strlen($matches[0]); |
|
731
|
|
|
$objval = (int) ($matches[1]).'_'.(int) ($matches[2]); |
|
732
|
|
|
} elseif (1 == preg_match('/^([0-9]+)[\s]+([0-9]+)[\s]+obj/iU', substr($pdfData, $offset, 33), $matches)) { |
|
733
|
|
|
// object start |
|
734
|
|
|
$objtype = 'obj'; |
|
735
|
|
|
$objval = (int) ($matches[1]).'_'.(int) ($matches[2]); |
|
736
|
|
|
$offset += \strlen($matches[0]); |
|
737
|
|
|
} elseif (($numlen = strspn($pdfData, '+-.0123456789', $offset)) > 0) { |
|
738
|
|
|
// numeric object |
|
739
|
|
|
$objtype = 'numeric'; |
|
740
|
|
|
$objval = substr($pdfData, $offset, $numlen); |
|
741
|
|
|
$offset += $numlen; |
|
742
|
|
|
} |
|
743
|
|
|
break; |
|
744
|
|
|
} |
|
745
|
|
|
|
|
746
|
|
|
return [$objtype, $objval, $offset]; |
|
747
|
|
|
} |
|
748
|
|
|
|
|
749
|
|
|
/** |
|
750
|
|
|
* Get Cross-Reference (xref) table and trailer data from PDF document data. |
|
751
|
|
|
* |
|
752
|
|
|
* @param string $pdfData |
|
753
|
|
|
* @param int $offset xref offset (if know) |
|
754
|
|
|
* @param array $xref previous xref array (if any) |
|
755
|
|
|
* |
|
756
|
|
|
* @return array containing xref and trailer data |
|
757
|
|
|
* |
|
758
|
|
|
* @throws Exception if it was unable to find startxref |
|
759
|
|
|
* @throws Exception if it was unable to find xref |
|
760
|
|
|
*/ |
|
761
|
|
|
public function getXrefData($pdfData, $offset = 0, $xref = []) |
|
762
|
|
|
{ |
|
763
|
|
|
$startxrefPreg = preg_match( |
|
764
|
|
|
'/[\r\n]startxref[\s]*[\r\n]+([0-9]+)[\s]*[\r\n]+%%EOF/i', |
|
765
|
|
|
$pdfData, |
|
766
|
|
|
$matches, |
|
767
|
|
|
PREG_OFFSET_CAPTURE, |
|
768
|
|
|
$offset |
|
769
|
|
|
); |
|
770
|
|
|
|
|
771
|
|
|
if (0 == $offset) { |
|
772
|
|
|
// find last startxref |
|
773
|
|
|
$pregResult = preg_match_all( |
|
774
|
|
|
'/[\r\n]startxref[\s]*[\r\n]+([0-9]+)[\s]*[\r\n]+%%EOF/i', |
|
775
|
|
|
$pdfData, $matches, |
|
776
|
|
|
PREG_SET_ORDER, |
|
777
|
|
|
$offset |
|
778
|
|
|
); |
|
779
|
|
|
if (0 == $pregResult) { |
|
780
|
|
|
throw new Exception('Unable to find startxref'); |
|
781
|
|
|
} |
|
782
|
|
|
$matches = array_pop($matches); |
|
783
|
|
|
$startxref = $matches[1]; |
|
784
|
|
|
} elseif (strpos($pdfData, 'xref', $offset) == $offset) { |
|
785
|
|
|
// Already pointing at the xref table |
|
786
|
|
|
$startxref = $offset; |
|
787
|
|
|
} elseif (preg_match('/([0-9]+[\s][0-9]+[\s]obj)/i', $pdfData, $matches, PREG_OFFSET_CAPTURE, $offset)) { |
|
788
|
|
|
// Cross-Reference Stream object |
|
789
|
|
|
$startxref = $offset; |
|
790
|
|
|
} elseif ($startxrefPreg) { |
|
791
|
|
|
// startxref found |
|
792
|
|
|
$startxref = $matches[1][0]; |
|
793
|
|
|
} else { |
|
794
|
|
|
throw new Exception('Unable to find startxref'); |
|
795
|
|
|
} |
|
796
|
|
|
|
|
797
|
|
|
// check xref position |
|
798
|
|
|
if (strpos($pdfData, 'xref', $startxref) == $startxref) { |
|
799
|
|
|
// Cross-Reference |
|
800
|
|
|
$xref = $this->decodeXref($pdfData, $startxref, $xref); |
|
801
|
|
|
} else { |
|
802
|
|
|
// Cross-Reference Stream |
|
803
|
|
|
$xref = $this->decodeXrefStream($pdfData, $startxref, $xref); |
|
804
|
|
|
} |
|
805
|
|
|
if (empty($xref)) { |
|
806
|
|
|
throw new Exception('Unable to find xref'); |
|
807
|
|
|
} |
|
808
|
|
|
|
|
809
|
|
|
return $xref; |
|
810
|
|
|
} |
|
811
|
|
|
} |
|
812
|
|
|
|
smalot /
pdfparser
Loading history...