|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
/** |
|
4
|
|
|
* @file |
|
5
|
|
|
* This file is part of the PdfParser library. |
|
6
|
|
|
* |
|
7
|
|
|
* @author Sébastien MALOT <[email protected]> |
|
8
|
|
|
* @date 2017-01-03 |
|
9
|
|
|
* |
|
10
|
|
|
* @license LGPLv3 |
|
11
|
|
|
* @url <https://github.com/smalot/pdfparser> |
|
12
|
|
|
* |
|
13
|
|
|
* PdfParser is a pdf library written in PHP, extraction oriented. |
|
14
|
|
|
* Copyright (C) 2017 - Sébastien MALOT <[email protected]> |
|
15
|
|
|
* |
|
16
|
|
|
* This program is free software: you can redistribute it and/or modify |
|
17
|
|
|
* it under the terms of the GNU Lesser General Public License as published by |
|
18
|
|
|
* the Free Software Foundation, either version 3 of the License, or |
|
19
|
|
|
* (at your option) any later version. |
|
20
|
|
|
* |
|
21
|
|
|
* This program is distributed in the hope that it will be useful, |
|
22
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
23
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
24
|
|
|
* GNU Lesser General Public License for more details. |
|
25
|
|
|
* |
|
26
|
|
|
* You should have received a copy of the GNU Lesser General Public License |
|
27
|
|
|
* along with this program. |
|
28
|
|
|
* If not, see <http://www.pdfparser.org/sites/default/LICENSE.txt>. |
|
29
|
|
|
*/ |
|
30
|
|
|
|
|
31
|
|
|
namespace Smalot\PdfParser; |
|
32
|
|
|
|
|
33
|
|
|
use Smalot\PdfParser\XObject\Form; |
|
34
|
|
|
use Smalot\PdfParser\XObject\Image; |
|
35
|
|
|
|
|
36
|
|
|
/** |
|
37
|
|
|
* Class PDFObject |
|
38
|
|
|
*/ |
|
39
|
|
|
class PDFObject |
|
40
|
|
|
{ |
|
41
|
|
|
const TYPE = 't'; |
|
42
|
|
|
|
|
43
|
|
|
const OPERATOR = 'o'; |
|
44
|
|
|
|
|
45
|
|
|
const COMMAND = 'c'; |
|
46
|
|
|
|
|
47
|
|
|
/** |
|
48
|
|
|
* The recursion stack. |
|
49
|
|
|
* |
|
50
|
|
|
* @var array |
|
51
|
|
|
*/ |
|
52
|
|
|
public static $recursionStack = []; |
|
53
|
|
|
|
|
54
|
|
|
/** |
|
55
|
|
|
* @var Document |
|
56
|
|
|
*/ |
|
57
|
|
|
protected $document = null; |
|
58
|
|
|
|
|
59
|
|
|
/** |
|
60
|
|
|
* @var Header |
|
61
|
|
|
*/ |
|
62
|
|
|
protected $header = null; |
|
63
|
|
|
|
|
64
|
|
|
/** |
|
65
|
|
|
* @var string |
|
66
|
|
|
*/ |
|
67
|
|
|
protected $content = null; |
|
68
|
|
|
|
|
69
|
|
|
/** |
|
70
|
|
|
* @param Header $header |
|
71
|
|
|
* @param string $content |
|
72
|
|
|
*/ |
|
73
|
33 |
|
public function __construct(Document $document, Header $header = null, $content = null) |
|
74
|
|
|
{ |
|
75
|
33 |
|
$this->document = $document; |
|
76
|
33 |
|
$this->header = null !== $header ? $header : new Header(); |
|
77
|
33 |
|
$this->content = $content; |
|
78
|
33 |
|
} |
|
79
|
|
|
|
|
80
|
29 |
|
public function init() |
|
81
|
|
|
{ |
|
82
|
29 |
|
} |
|
83
|
|
|
|
|
84
|
|
|
/** |
|
85
|
|
|
* @return Header|null |
|
86
|
|
|
*/ |
|
87
|
29 |
|
public function getHeader() |
|
88
|
|
|
{ |
|
89
|
29 |
|
return $this->header; |
|
90
|
|
|
} |
|
91
|
|
|
|
|
92
|
|
|
/** |
|
93
|
|
|
* @param string $name |
|
94
|
|
|
* |
|
95
|
|
|
* @return Element|PDFObject |
|
96
|
|
|
*/ |
|
97
|
24 |
|
public function get($name) |
|
98
|
|
|
{ |
|
99
|
24 |
|
return $this->header->get($name); |
|
100
|
|
|
} |
|
101
|
|
|
|
|
102
|
|
|
/** |
|
103
|
|
|
* @param string $name |
|
104
|
|
|
* |
|
105
|
|
|
* @return bool |
|
106
|
|
|
*/ |
|
107
|
23 |
|
public function has($name) |
|
108
|
|
|
{ |
|
109
|
23 |
|
return $this->header->has($name); |
|
110
|
|
|
} |
|
111
|
|
|
|
|
112
|
|
|
/** |
|
113
|
|
|
* @param bool $deep |
|
114
|
|
|
* |
|
115
|
|
|
* @return array |
|
116
|
|
|
*/ |
|
117
|
7 |
|
public function getDetails($deep = true) |
|
118
|
|
|
{ |
|
119
|
7 |
|
return $this->header->getDetails($deep); |
|
120
|
|
|
} |
|
121
|
|
|
|
|
122
|
|
|
/** |
|
123
|
|
|
* @return string|null |
|
124
|
|
|
*/ |
|
125
|
20 |
|
public function getContent() |
|
126
|
|
|
{ |
|
127
|
20 |
|
return $this->content; |
|
128
|
|
|
} |
|
129
|
|
|
|
|
130
|
|
|
/** |
|
131
|
|
|
* @param string $content |
|
132
|
|
|
*/ |
|
133
|
15 |
|
public function cleanContent($content, $char = 'X') |
|
134
|
|
|
{ |
|
135
|
15 |
|
$char = $char[0]; |
|
136
|
15 |
|
$content = str_replace(['\\\\', '\\)', '\\('], $char.$char, $content); |
|
137
|
|
|
|
|
138
|
|
|
// Remove image bloc with binary content |
|
139
|
15 |
|
preg_match_all('/\s(BI\s.*?(\sID\s).*?(\sEI))\s/s', $content, $matches, PREG_OFFSET_CAPTURE); |
|
140
|
15 |
|
foreach ($matches[0] as $part) { |
|
141
|
2 |
|
$content = substr_replace($content, str_repeat($char, \strlen($part[0])), $part[1], \strlen($part[0])); |
|
142
|
|
|
} |
|
143
|
|
|
|
|
144
|
|
|
// Clean content in square brackets [.....] |
|
145
|
15 |
|
preg_match_all('/\[((\(.*?\)|[0-9\.\-\s]*)*)\]/s', $content, $matches, PREG_OFFSET_CAPTURE); |
|
146
|
15 |
|
foreach ($matches[1] as $part) { |
|
147
|
10 |
|
$content = substr_replace($content, str_repeat($char, \strlen($part[0])), $part[1], \strlen($part[0])); |
|
148
|
|
|
} |
|
149
|
|
|
|
|
150
|
|
|
// Clean content in round brackets (.....) |
|
151
|
15 |
|
preg_match_all('/\((.*?)\)/s', $content, $matches, PREG_OFFSET_CAPTURE); |
|
152
|
15 |
|
foreach ($matches[1] as $part) { |
|
153
|
11 |
|
$content = substr_replace($content, str_repeat($char, \strlen($part[0])), $part[1], \strlen($part[0])); |
|
154
|
|
|
} |
|
155
|
|
|
|
|
156
|
|
|
// Clean structure |
|
157
|
15 |
|
if ($parts = preg_split('/(<|>)/s', $content, -1, PREG_SPLIT_NO_EMPTY | PREG_SPLIT_DELIM_CAPTURE)) { |
|
158
|
15 |
|
$content = ''; |
|
159
|
15 |
|
$level = 0; |
|
160
|
15 |
|
foreach ($parts as $part) { |
|
161
|
15 |
|
if ('<' == $part) { |
|
162
|
10 |
|
++$level; |
|
163
|
|
|
} |
|
164
|
|
|
|
|
165
|
15 |
|
$content .= (0 == $level ? $part : str_repeat($char, \strlen($part))); |
|
166
|
|
|
|
|
167
|
15 |
|
if ('>' == $part) { |
|
168
|
10 |
|
--$level; |
|
169
|
|
|
} |
|
170
|
|
|
} |
|
171
|
|
|
} |
|
172
|
|
|
|
|
173
|
|
|
// Clean BDC and EMC markup |
|
174
|
15 |
|
preg_match_all( |
|
175
|
15 |
|
'/(\/[A-Za-z0-9\_]*\s*'.preg_quote($char).'*BDC)/s', |
|
176
|
|
|
$content, |
|
177
|
|
|
$matches, |
|
178
|
15 |
|
PREG_OFFSET_CAPTURE |
|
179
|
|
|
); |
|
180
|
15 |
|
foreach ($matches[1] as $part) { |
|
181
|
2 |
|
$content = substr_replace($content, str_repeat($char, \strlen($part[0])), $part[1], \strlen($part[0])); |
|
182
|
|
|
} |
|
183
|
|
|
|
|
184
|
15 |
|
preg_match_all('/\s(EMC)\s/s', $content, $matches, PREG_OFFSET_CAPTURE); |
|
185
|
15 |
|
foreach ($matches[1] as $part) { |
|
186
|
6 |
|
$content = substr_replace($content, str_repeat($char, \strlen($part[0])), $part[1], \strlen($part[0])); |
|
187
|
|
|
} |
|
188
|
|
|
|
|
189
|
15 |
|
return $content; |
|
190
|
|
|
} |
|
191
|
|
|
|
|
192
|
|
|
/** |
|
193
|
|
|
* @param string $content |
|
194
|
|
|
* |
|
195
|
|
|
* @return array |
|
196
|
|
|
*/ |
|
197
|
14 |
|
public function getSectionsText($content) |
|
198
|
|
|
{ |
|
199
|
14 |
|
$sections = []; |
|
200
|
14 |
|
$content = ' '.$content.' '; |
|
201
|
14 |
|
$textCleaned = $this->cleanContent($content, '_'); |
|
202
|
|
|
|
|
203
|
|
|
// Extract text blocks. |
|
204
|
14 |
|
if (preg_match_all('/\s+BT[\s|\(|\[]+(.*?)\s*ET/s', $textCleaned, $matches, PREG_OFFSET_CAPTURE)) { |
|
205
|
14 |
|
foreach ($matches[1] as $part) { |
|
206
|
14 |
|
$text = $part[0]; |
|
207
|
14 |
|
if ('' === $text) { |
|
208
|
|
|
continue; |
|
209
|
|
|
} |
|
210
|
14 |
|
$offset = $part[1]; |
|
211
|
14 |
|
$section = substr($content, $offset, \strlen($text)); |
|
212
|
|
|
|
|
213
|
|
|
// Removes BDC and EMC markup. |
|
214
|
14 |
|
$section = preg_replace('/(\/[A-Za-z0-9]+\s*<<.*?)(>>\s*BDC)(.*?)(EMC\s+)/s', '${3}', $section.' '); |
|
215
|
|
|
|
|
216
|
14 |
|
$sections[] = $section; |
|
217
|
|
|
} |
|
218
|
|
|
} |
|
219
|
|
|
|
|
220
|
|
|
// Extract 'do' commands. |
|
221
|
14 |
|
if (preg_match_all('/(\/[A-Za-z0-9\.\-_]+\s+Do)\s/s', $textCleaned, $matches, PREG_OFFSET_CAPTURE)) { |
|
222
|
5 |
|
foreach ($matches[1] as $part) { |
|
223
|
5 |
|
$text = $part[0]; |
|
224
|
5 |
|
$offset = $part[1]; |
|
225
|
5 |
|
$section = substr($content, $offset, \strlen($text)); |
|
226
|
|
|
|
|
227
|
5 |
|
$sections[] = $section; |
|
228
|
|
|
} |
|
229
|
|
|
} |
|
230
|
|
|
|
|
231
|
14 |
|
return $sections; |
|
232
|
|
|
} |
|
233
|
|
|
|
|
234
|
|
|
/** |
|
235
|
|
|
* @param Page $page |
|
236
|
|
|
* |
|
237
|
|
|
* @return string |
|
238
|
|
|
* |
|
239
|
|
|
* @throws \Exception |
|
240
|
|
|
*/ |
|
241
|
8 |
|
public function getText(Page $page = null) |
|
242
|
|
|
{ |
|
243
|
8 |
|
$text = ''; |
|
244
|
8 |
|
$sections = $this->getSectionsText($this->content); |
|
245
|
8 |
|
$current_font = null; |
|
246
|
8 |
|
$current_font_size = 1; |
|
247
|
|
|
|
|
248
|
8 |
|
foreach ($this->document->getObjects() as $obj) { |
|
249
|
8 |
|
if ($obj instanceof Font) { |
|
250
|
8 |
|
$current_font = $obj; |
|
251
|
8 |
|
break; |
|
252
|
|
|
} |
|
253
|
|
|
} |
|
254
|
|
|
|
|
255
|
8 |
|
if (null === $current_font) { |
|
256
|
|
|
$current_font = new Font($this->document); |
|
257
|
|
|
} |
|
258
|
|
|
|
|
259
|
8 |
|
$current_position_td = ['x' => false, 'y' => false]; |
|
260
|
8 |
|
$current_position_tm = ['x' => false, 'y' => false]; |
|
261
|
|
|
|
|
262
|
8 |
|
array_push(self::$recursionStack, $this->getUniqueId()); |
|
263
|
|
|
|
|
264
|
8 |
|
foreach ($sections as $section) { |
|
265
|
8 |
|
$commands = $this->getCommandsText($section); |
|
266
|
|
|
|
|
267
|
8 |
|
foreach ($commands as $command) { |
|
268
|
8 |
|
switch ($command[self::OPERATOR]) { |
|
269
|
|
|
// set character spacing |
|
270
|
8 |
|
case 'Tc': |
|
271
|
1 |
|
break; |
|
272
|
|
|
|
|
273
|
|
|
// move text current point |
|
274
|
8 |
|
case 'Td': |
|
275
|
6 |
|
$args = preg_split('/\s/s', $command[self::COMMAND]); |
|
276
|
6 |
|
$y = array_pop($args); |
|
|
|
|
|
|
277
|
6 |
|
$x = array_pop($args); |
|
278
|
|
|
|
|
279
|
|
|
/* |
|
280
|
|
|
* When converting pdf into text some offsets between text should at least outputted as spaces. |
|
281
|
|
|
* But small move-text operations may be normal, see https://github.com/smalot/pdfparser/issues/201 |
|
282
|
|
|
* |
|
283
|
|
|
* To find out what we are handling with, get the max char width for current font and check if offset is bigger than that |
|
284
|
|
|
*/ |
|
285
|
|
|
|
|
286
|
6 |
|
$fontMaxWidth = static::getMaxCharWidthFromFont($current_font, $current_font_size); |
|
287
|
|
|
|
|
288
|
6 |
|
$looksLikeCarriageReturn = (float) ($x) <= 0; |
|
289
|
|
|
|
|
290
|
6 |
|
$movingVertically = false !== $current_position_td['y'] |
|
|
|
|
|
|
291
|
6 |
|
&& (float) ($y) < (float) ($current_position_td['y']); |
|
292
|
|
|
|
|
293
|
6 |
|
$lookLikeSpaceChar = ((float) ($x) - (float) ($current_position_td['x'])) > $fontMaxWidth; |
|
294
|
|
|
|
|
295
|
6 |
|
if ( $looksLikeCarriageReturn || $movingVertically) { |
|
296
|
|
|
// vertical offset |
|
297
|
6 |
|
$text .= "\n"; |
|
298
|
6 |
|
} elseif (false !== $current_position_td['x'] && $lookLikeSpaceChar) { |
|
299
|
|
|
// horizontal offset |
|
300
|
4 |
|
$text .= ' '; |
|
301
|
|
|
} |
|
302
|
|
|
|
|
303
|
6 |
|
$current_position_td = ['x' => $x, 'y' => $y]; |
|
304
|
6 |
|
break; |
|
305
|
|
|
|
|
306
|
|
|
// move text current point and set leading |
|
307
|
8 |
|
case 'TD': |
|
308
|
1 |
|
$args = preg_split('/\s/s', $command[self::COMMAND]); |
|
309
|
1 |
|
$y = array_pop($args); |
|
310
|
1 |
|
$x = array_pop($args); |
|
311
|
1 |
|
if ((float) $y < 0) { |
|
312
|
1 |
|
$text .= "\n"; |
|
313
|
|
|
} elseif ((float) $x <= 0) { |
|
314
|
|
|
$text .= ' '; |
|
315
|
|
|
} |
|
316
|
1 |
|
break; |
|
317
|
|
|
|
|
318
|
8 |
|
case 'Tf': |
|
319
|
8 |
|
list($id, $size) = preg_split('/\s/s', $command[self::COMMAND]); |
|
320
|
8 |
|
$current_font_size = (float) $size; |
|
321
|
8 |
|
$id = trim($id, '/'); |
|
322
|
8 |
|
if (null !== $page) { |
|
323
|
8 |
|
$current_font = $page->getFont($id); |
|
324
|
|
|
} |
|
325
|
8 |
|
break; |
|
326
|
|
|
|
|
327
|
8 |
|
case "'": |
|
328
|
8 |
|
case 'Tj': |
|
329
|
6 |
|
$command[self::COMMAND] = [$command]; |
|
330
|
|
|
// no break |
|
331
|
8 |
|
case 'TJ': |
|
332
|
|
|
// Skip if not previously defined, should never happened. |
|
333
|
8 |
|
if (null === $current_font) { |
|
334
|
|
|
// Fallback |
|
335
|
|
|
// TODO : Improve |
|
336
|
1 |
|
$text .= $command[self::COMMAND][0][self::COMMAND]; |
|
337
|
1 |
|
break; |
|
338
|
|
|
} |
|
339
|
|
|
|
|
340
|
8 |
|
$sub_text = $current_font->decodeText($command[self::COMMAND]); |
|
341
|
8 |
|
$text .= $sub_text; |
|
342
|
8 |
|
break; |
|
343
|
|
|
|
|
344
|
|
|
// set leading |
|
345
|
8 |
|
case 'TL': |
|
346
|
3 |
|
$text .= ' '; |
|
347
|
3 |
|
break; |
|
348
|
|
|
|
|
349
|
8 |
|
case 'Tm': |
|
350
|
8 |
|
$args = preg_split('/\s/s', $command[self::COMMAND]); |
|
351
|
8 |
|
$y = array_pop($args); |
|
352
|
8 |
|
$x = array_pop($args); |
|
353
|
8 |
|
if (false !== $current_position_tm['x']) { |
|
354
|
8 |
|
$delta = abs((float) $x - (float) ($current_position_tm['x'])); |
|
355
|
8 |
|
if ($delta > 10) { |
|
356
|
4 |
|
$text .= "\t"; |
|
357
|
|
|
} |
|
358
|
|
|
} |
|
359
|
8 |
|
if (false !== $current_position_tm['y']) { |
|
360
|
8 |
|
$delta = abs((float) $y - (float) ($current_position_tm['y'])); |
|
361
|
8 |
|
if ($delta > 10) { |
|
362
|
5 |
|
$text .= "\n"; |
|
363
|
|
|
} |
|
364
|
|
|
} |
|
365
|
8 |
|
$current_position_tm = ['x' => $x, 'y' => $y]; |
|
366
|
8 |
|
break; |
|
367
|
|
|
|
|
368
|
|
|
// set super/subscripting text rise |
|
369
|
7 |
|
case 'Ts': |
|
370
|
|
|
break; |
|
371
|
|
|
|
|
372
|
|
|
// set word spacing |
|
373
|
7 |
|
case 'Tw': |
|
374
|
|
|
break; |
|
375
|
|
|
|
|
376
|
|
|
// set horizontal scaling |
|
377
|
7 |
|
case 'Tz': |
|
378
|
|
|
$text .= "\n"; |
|
379
|
|
|
break; |
|
380
|
|
|
|
|
381
|
|
|
// move to start of next line |
|
382
|
7 |
|
case 'T*': |
|
383
|
2 |
|
$text .= "\n"; |
|
384
|
2 |
|
break; |
|
385
|
|
|
|
|
386
|
6 |
|
case 'Da': |
|
387
|
|
|
break; |
|
388
|
|
|
|
|
389
|
6 |
|
case 'Do': |
|
390
|
5 |
|
if (null !== $page) { |
|
391
|
5 |
|
$args = preg_split('/\s/s', $command[self::COMMAND]); |
|
392
|
5 |
|
$id = trim(array_pop($args), '/ '); |
|
393
|
5 |
|
$xobject = $page->getXObject($id); |
|
394
|
|
|
|
|
395
|
|
|
// @todo $xobject could be a ElementXRef object, which would then throw an error |
|
396
|
5 |
|
if (\is_object($xobject) && $xobject instanceof self && !\in_array($xobject->getUniqueId(), self::$recursionStack)) { |
|
397
|
|
|
// Not a circular reference. |
|
398
|
5 |
|
$text .= $xobject->getText($page); |
|
399
|
|
|
} |
|
400
|
|
|
} |
|
401
|
5 |
|
break; |
|
402
|
|
|
|
|
403
|
4 |
|
case 'rg': |
|
404
|
4 |
|
case 'RG': |
|
405
|
2 |
|
break; |
|
406
|
|
|
|
|
407
|
4 |
|
case 're': |
|
408
|
|
|
break; |
|
409
|
|
|
|
|
410
|
4 |
|
case 'co': |
|
411
|
|
|
break; |
|
412
|
|
|
|
|
413
|
4 |
|
case 'cs': |
|
414
|
1 |
|
break; |
|
415
|
|
|
|
|
416
|
4 |
|
case 'gs': |
|
417
|
3 |
|
break; |
|
418
|
|
|
|
|
419
|
4 |
|
case 'en': |
|
420
|
|
|
break; |
|
421
|
|
|
|
|
422
|
4 |
|
case 'sc': |
|
423
|
4 |
|
case 'SC': |
|
424
|
|
|
break; |
|
425
|
|
|
|
|
426
|
4 |
|
case 'g': |
|
427
|
3 |
|
case 'G': |
|
428
|
3 |
|
break; |
|
429
|
|
|
|
|
430
|
2 |
|
case 'V': |
|
431
|
|
|
break; |
|
432
|
|
|
|
|
433
|
2 |
|
case 'vo': |
|
434
|
2 |
|
case 'Vo': |
|
435
|
|
|
break; |
|
436
|
|
|
|
|
437
|
|
|
default: |
|
438
|
|
|
} |
|
439
|
|
|
} |
|
440
|
|
|
} |
|
441
|
|
|
|
|
442
|
8 |
|
array_pop(self::$recursionStack); |
|
443
|
|
|
|
|
444
|
8 |
|
return $text.' '; |
|
445
|
|
|
} |
|
446
|
|
|
|
|
447
|
|
|
/** |
|
448
|
|
|
* @param Page $page |
|
449
|
|
|
* |
|
450
|
|
|
* @return array |
|
451
|
|
|
* |
|
452
|
|
|
* @throws \Exception |
|
453
|
|
|
*/ |
|
454
|
|
|
public function getTextArray(Page $page = null) |
|
455
|
|
|
{ |
|
456
|
|
|
$text = []; |
|
457
|
|
|
$sections = $this->getSectionsText($this->content); |
|
458
|
|
|
$current_font = new Font($this->document); |
|
459
|
|
|
|
|
460
|
|
|
foreach ($sections as $section) { |
|
461
|
|
|
$commands = $this->getCommandsText($section); |
|
462
|
|
|
|
|
463
|
|
|
foreach ($commands as $command) { |
|
464
|
|
|
switch ($command[self::OPERATOR]) { |
|
465
|
|
|
// set character spacing |
|
466
|
|
|
case 'Tc': |
|
467
|
|
|
break; |
|
468
|
|
|
|
|
469
|
|
|
// move text current point |
|
470
|
|
|
case 'Td': |
|
471
|
|
|
break; |
|
472
|
|
|
|
|
473
|
|
|
// move text current point and set leading |
|
474
|
|
|
case 'TD': |
|
475
|
|
|
break; |
|
476
|
|
|
|
|
477
|
|
|
case 'Tf': |
|
478
|
|
|
list($id) = preg_split('/\s/s', $command[self::COMMAND]); |
|
479
|
|
|
$id = trim($id, '/'); |
|
480
|
|
|
$current_font = $page->getFont($id); |
|
|
|
|
|
|
481
|
|
|
break; |
|
482
|
|
|
|
|
483
|
|
|
case "'": |
|
484
|
|
|
case 'Tj': |
|
485
|
|
|
$command[self::COMMAND] = [$command]; |
|
486
|
|
|
// no break |
|
487
|
|
|
case 'TJ': |
|
488
|
|
|
// Skip if not previously defined, should never happened. |
|
489
|
|
|
if (null === $current_font) { |
|
490
|
|
|
// Fallback |
|
491
|
|
|
// TODO : Improve |
|
492
|
|
|
$text[] = $command[self::COMMAND][0][self::COMMAND]; |
|
493
|
|
|
break; |
|
494
|
|
|
} |
|
495
|
|
|
|
|
496
|
|
|
$sub_text = $current_font->decodeText($command[self::COMMAND]); |
|
497
|
|
|
$text[] = $sub_text; |
|
498
|
|
|
break; |
|
499
|
|
|
|
|
500
|
|
|
// set leading |
|
501
|
|
|
case 'TL': |
|
502
|
|
|
break; |
|
503
|
|
|
|
|
504
|
|
|
case 'Tm': |
|
505
|
|
|
break; |
|
506
|
|
|
|
|
507
|
|
|
// set super/subscripting text rise |
|
508
|
|
|
case 'Ts': |
|
509
|
|
|
break; |
|
510
|
|
|
|
|
511
|
|
|
// set word spacing |
|
512
|
|
|
case 'Tw': |
|
513
|
|
|
break; |
|
514
|
|
|
|
|
515
|
|
|
// set horizontal scaling |
|
516
|
|
|
case 'Tz': |
|
517
|
|
|
//$text .= "\n"; |
|
518
|
|
|
break; |
|
519
|
|
|
|
|
520
|
|
|
// move to start of next line |
|
521
|
|
|
case 'T*': |
|
522
|
|
|
//$text .= "\n"; |
|
523
|
|
|
break; |
|
524
|
|
|
|
|
525
|
|
|
case 'Da': |
|
526
|
|
|
break; |
|
527
|
|
|
|
|
528
|
|
|
case 'Do': |
|
529
|
|
|
if (null !== $page) { |
|
530
|
|
|
$args = preg_split('/\s/s', $command[self::COMMAND]); |
|
531
|
|
|
$id = trim(array_pop($args), '/ '); |
|
|
|
|
|
|
532
|
|
|
if ($xobject = $page->getXObject($id)) { |
|
533
|
|
|
$text[] = $xobject->getText($page); |
|
534
|
|
|
} |
|
535
|
|
|
} |
|
536
|
|
|
break; |
|
537
|
|
|
|
|
538
|
|
|
case 'rg': |
|
539
|
|
|
case 'RG': |
|
540
|
|
|
break; |
|
541
|
|
|
|
|
542
|
|
|
case 're': |
|
543
|
|
|
break; |
|
544
|
|
|
|
|
545
|
|
|
case 'co': |
|
546
|
|
|
break; |
|
547
|
|
|
|
|
548
|
|
|
case 'cs': |
|
549
|
|
|
break; |
|
550
|
|
|
|
|
551
|
|
|
case 'gs': |
|
552
|
|
|
break; |
|
553
|
|
|
|
|
554
|
|
|
case 'en': |
|
555
|
|
|
break; |
|
556
|
|
|
|
|
557
|
|
|
case 'sc': |
|
558
|
|
|
case 'SC': |
|
559
|
|
|
break; |
|
560
|
|
|
|
|
561
|
|
|
case 'g': |
|
562
|
|
|
case 'G': |
|
563
|
|
|
break; |
|
564
|
|
|
|
|
565
|
|
|
case 'V': |
|
566
|
|
|
break; |
|
567
|
|
|
|
|
568
|
|
|
case 'vo': |
|
569
|
|
|
case 'Vo': |
|
570
|
|
|
break; |
|
571
|
|
|
|
|
572
|
|
|
default: |
|
573
|
|
|
} |
|
574
|
|
|
} |
|
575
|
|
|
} |
|
576
|
|
|
|
|
577
|
|
|
return $text; |
|
578
|
|
|
} |
|
579
|
|
|
|
|
580
|
|
|
/** |
|
581
|
|
|
* @param string $text_part |
|
582
|
|
|
* @param int $offset |
|
583
|
|
|
* |
|
584
|
|
|
* @return array |
|
585
|
|
|
*/ |
|
586
|
14 |
|
public function getCommandsText($text_part, &$offset = 0) |
|
587
|
|
|
{ |
|
588
|
14 |
|
$commands = $matches = []; |
|
589
|
|
|
|
|
590
|
14 |
|
while ($offset < \strlen($text_part)) { |
|
591
|
14 |
|
$offset += strspn($text_part, "\x00\x09\x0a\x0c\x0d\x20", $offset); |
|
592
|
14 |
|
$char = $text_part[$offset]; |
|
593
|
|
|
|
|
594
|
14 |
|
$operator = ''; |
|
595
|
14 |
|
$type = ''; |
|
596
|
14 |
|
$command = false; |
|
597
|
|
|
|
|
598
|
14 |
|
switch ($char) { |
|
599
|
14 |
|
case '/': |
|
600
|
14 |
|
$type = $char; |
|
601
|
14 |
|
if (preg_match( |
|
602
|
14 |
|
'/^\/([A-Z0-9\._,\+]+\s+[0-9.\-]+)\s+([A-Z]+)\s*/si', |
|
603
|
14 |
|
substr($text_part, $offset), |
|
604
|
|
|
$matches |
|
605
|
|
|
) |
|
606
|
|
|
) { |
|
607
|
14 |
|
$operator = $matches[2]; |
|
608
|
14 |
|
$command = $matches[1]; |
|
609
|
14 |
|
$offset += \strlen($matches[0]); |
|
610
|
|
|
} elseif (preg_match( |
|
611
|
6 |
|
'/^\/([A-Z0-9\._,\+]+)\s+([A-Z]+)\s*/si', |
|
612
|
6 |
|
substr($text_part, $offset), |
|
613
|
|
|
$matches |
|
614
|
|
|
) |
|
615
|
|
|
) { |
|
616
|
6 |
|
$operator = $matches[2]; |
|
617
|
6 |
|
$command = $matches[1]; |
|
618
|
6 |
|
$offset += \strlen($matches[0]); |
|
619
|
|
|
} |
|
620
|
14 |
|
break; |
|
621
|
|
|
|
|
622
|
14 |
|
case '[': |
|
623
|
14 |
|
case ']': |
|
624
|
|
|
// array object |
|
625
|
11 |
|
$type = $char; |
|
626
|
11 |
|
if ('[' == $char) { |
|
627
|
11 |
|
++$offset; |
|
628
|
|
|
// get elements |
|
629
|
11 |
|
$command = $this->getCommandsText($text_part, $offset); |
|
630
|
|
|
|
|
631
|
11 |
|
if (preg_match('/^\s*[A-Z]{1,2}\s*/si', substr($text_part, $offset), $matches)) { |
|
632
|
11 |
|
$operator = trim($matches[0]); |
|
633
|
11 |
|
$offset += \strlen($matches[0]); |
|
634
|
|
|
} |
|
635
|
|
|
} else { |
|
636
|
11 |
|
++$offset; |
|
637
|
11 |
|
break; |
|
638
|
|
|
} |
|
639
|
11 |
|
break; |
|
640
|
|
|
|
|
641
|
14 |
|
case '<': |
|
642
|
14 |
|
case '>': |
|
643
|
|
|
// array object |
|
644
|
6 |
|
$type = $char; |
|
645
|
6 |
|
++$offset; |
|
646
|
6 |
|
if ('<' == $char) { |
|
647
|
6 |
|
$strpos = strpos($text_part, '>', $offset); |
|
648
|
6 |
|
$command = substr($text_part, $offset, ($strpos - $offset)); |
|
649
|
6 |
|
$offset = $strpos + 1; |
|
650
|
|
|
} |
|
651
|
|
|
|
|
652
|
6 |
|
if (preg_match('/^\s*[A-Z]{1,2}\s*/si', substr($text_part, $offset), $matches)) { |
|
653
|
5 |
|
$operator = trim($matches[0]); |
|
654
|
5 |
|
$offset += \strlen($matches[0]); |
|
655
|
|
|
} |
|
656
|
6 |
|
break; |
|
657
|
|
|
|
|
658
|
14 |
|
case '(': |
|
659
|
14 |
|
case ')': |
|
660
|
10 |
|
++$offset; |
|
661
|
10 |
|
$type = $char; |
|
662
|
10 |
|
$strpos = $offset; |
|
663
|
10 |
|
if ('(' == $char) { |
|
664
|
10 |
|
$open_bracket = 1; |
|
665
|
10 |
|
while ($open_bracket > 0) { |
|
666
|
10 |
|
if (!isset($text_part[$strpos])) { |
|
667
|
|
|
break; |
|
668
|
|
|
} |
|
669
|
10 |
|
$ch = $text_part[$strpos]; |
|
670
|
10 |
|
switch ($ch) { |
|
671
|
10 |
|
case '\\': |
|
672
|
|
|
// REVERSE SOLIDUS (5Ch) (Backslash) |
|
673
|
|
|
// skip next character |
|
674
|
9 |
|
++$strpos; |
|
675
|
9 |
|
break; |
|
676
|
|
|
|
|
677
|
10 |
|
case '(': |
|
678
|
|
|
// LEFT PARENHESIS (28h) |
|
679
|
|
|
++$open_bracket; |
|
680
|
|
|
break; |
|
681
|
|
|
|
|
682
|
10 |
|
case ')': |
|
683
|
|
|
// RIGHT PARENTHESIS (29h) |
|
684
|
10 |
|
--$open_bracket; |
|
685
|
10 |
|
break; |
|
686
|
|
|
} |
|
687
|
10 |
|
++$strpos; |
|
688
|
|
|
} |
|
689
|
10 |
|
$command = substr($text_part, $offset, ($strpos - $offset - 1)); |
|
690
|
10 |
|
$offset = $strpos; |
|
691
|
|
|
|
|
692
|
10 |
|
if (preg_match('/^\s*([A-Z\']{1,2})\s*/si', substr($text_part, $offset), $matches)) { |
|
693
|
9 |
|
$operator = $matches[1]; |
|
694
|
9 |
|
$offset += \strlen($matches[0]); |
|
695
|
|
|
} |
|
696
|
|
|
} |
|
697
|
10 |
|
break; |
|
698
|
|
|
|
|
699
|
|
|
default: |
|
700
|
|
|
|
|
701
|
14 |
|
if ('ET' == substr($text_part, $offset, 2)) { |
|
702
|
1 |
|
break; |
|
703
|
|
|
} elseif (preg_match( |
|
704
|
14 |
|
'/^\s*(?P<data>([0-9\.\-]+\s*?)+)\s+(?P<id>[A-Z]{1,3})\s*/si', |
|
705
|
14 |
|
substr($text_part, $offset), |
|
706
|
|
|
$matches |
|
707
|
|
|
) |
|
708
|
|
|
) { |
|
709
|
14 |
|
$operator = trim($matches['id']); |
|
710
|
14 |
|
$command = trim($matches['data']); |
|
711
|
14 |
|
$offset += \strlen($matches[0]); |
|
712
|
11 |
|
} elseif (preg_match('/^\s*([0-9\.\-]+\s*?)+\s*/si', substr($text_part, $offset), $matches)) { |
|
713
|
11 |
|
$type = 'n'; |
|
714
|
11 |
|
$command = trim($matches[0]); |
|
715
|
11 |
|
$offset += \strlen($matches[0]); |
|
716
|
8 |
|
} elseif (preg_match('/^\s*([A-Z\*]+)\s*/si', substr($text_part, $offset), $matches)) { |
|
717
|
8 |
|
$type = ''; |
|
718
|
8 |
|
$operator = $matches[1]; |
|
719
|
8 |
|
$command = ''; |
|
720
|
8 |
|
$offset += \strlen($matches[0]); |
|
721
|
|
|
} |
|
722
|
|
|
} |
|
723
|
|
|
|
|
724
|
14 |
|
if (false !== $command) { |
|
725
|
14 |
|
$commands[] = [ |
|
726
|
14 |
|
self::TYPE => $type, |
|
727
|
14 |
|
self::OPERATOR => $operator, |
|
728
|
14 |
|
self::COMMAND => $command, |
|
729
|
|
|
]; |
|
730
|
|
|
} else { |
|
731
|
11 |
|
break; |
|
732
|
|
|
} |
|
733
|
|
|
} |
|
734
|
|
|
|
|
735
|
14 |
|
return $commands; |
|
736
|
|
|
} |
|
737
|
|
|
|
|
738
|
|
|
/** |
|
739
|
|
|
* @param string $content |
|
740
|
|
|
* |
|
741
|
|
|
* @return PDFObject |
|
742
|
|
|
*/ |
|
743
|
21 |
|
public static function factory(Document $document, Header $header, $content) |
|
744
|
|
|
{ |
|
745
|
21 |
|
switch ($header->get('Type')->getContent()) { |
|
746
|
21 |
|
case 'XObject': |
|
747
|
6 |
|
switch ($header->get('Subtype')->getContent()) { |
|
748
|
6 |
|
case 'Image': |
|
749
|
5 |
|
return new Image($document, $header, $content); |
|
750
|
|
|
|
|
751
|
2 |
|
case 'Form': |
|
752
|
2 |
|
return new Form($document, $header, $content); |
|
753
|
|
|
} |
|
754
|
|
|
|
|
755
|
|
|
return new self($document, $header, $content); |
|
756
|
|
|
|
|
757
|
21 |
|
case 'Pages': |
|
758
|
21 |
|
return new Pages($document, $header, $content); |
|
759
|
|
|
|
|
760
|
21 |
|
case 'Page': |
|
761
|
21 |
|
return new Page($document, $header, $content); |
|
762
|
|
|
|
|
763
|
21 |
|
case 'Encoding': |
|
764
|
3 |
|
return new Encoding($document, $header, $content); |
|
765
|
|
|
|
|
766
|
21 |
|
case 'Font': |
|
767
|
21 |
|
$subtype = $header->get('Subtype')->getContent(); |
|
768
|
21 |
|
$classname = '\Smalot\PdfParser\Font\Font'.$subtype; |
|
769
|
|
|
|
|
770
|
21 |
|
if (class_exists($classname)) { |
|
771
|
21 |
|
return new $classname($document, $header, $content); |
|
772
|
|
|
} |
|
773
|
|
|
|
|
774
|
|
|
return new Font($document, $header, $content); |
|
775
|
|
|
|
|
776
|
|
|
default: |
|
777
|
21 |
|
return new self($document, $header, $content); |
|
778
|
|
|
} |
|
779
|
|
|
} |
|
780
|
|
|
|
|
781
|
|
|
/** |
|
782
|
|
|
* Returns unique id identifying the object. |
|
783
|
|
|
* |
|
784
|
|
|
* @return string |
|
785
|
|
|
*/ |
|
786
|
8 |
|
protected function getUniqueId() |
|
787
|
|
|
{ |
|
788
|
8 |
|
return spl_object_hash($this); |
|
789
|
|
|
} |
|
790
|
|
|
|
|
791
|
|
|
/** |
|
792
|
|
|
* This function returns an unscaled text space units for the widest char in current font |
|
793
|
|
|
* |
|
794
|
|
|
* @param Font $currentFont |
|
795
|
|
|
* @param float $currentFontSize |
|
796
|
|
|
* @return int |
|
797
|
|
|
*/ |
|
798
|
6 |
|
protected static function getMaxCharWidthFromFont($currentFont, $currentFontSize) |
|
799
|
|
|
{ |
|
800
|
6 |
|
$fontMaxWidth = 0; |
|
801
|
6 |
|
if (null !== $currentFont) { |
|
802
|
6 |
|
$fontDictonary = $currentFont->getDetails(); |
|
803
|
|
|
// type 0 |
|
804
|
6 |
|
if (isset($fontDictonary['DescendantFonts'])) { |
|
805
|
3 |
|
$fontDictonary = $fontDictonary['DescendantFonts'][0]; |
|
806
|
|
|
} |
|
807
|
|
|
// type 1 |
|
808
|
6 |
|
if (isset($fontDictonary['Widths'])) { |
|
809
|
3 |
|
foreach ($fontDictonary['Widths'] as $width) { |
|
810
|
3 |
|
if (($floatWidth = (float) $width) > $fontMaxWidth) { |
|
811
|
3 |
|
$fontMaxWidth = $floatWidth; |
|
812
|
|
|
} |
|
813
|
|
|
} |
|
814
|
|
|
|
|
815
|
3 |
|
$fontMaxWidth = ($fontMaxWidth / 1000) * $currentFontSize; |
|
816
|
|
|
} |
|
817
|
|
|
// CIDFontType2 |
|
818
|
6 |
|
if ('cidfonttype2' === strtolower($fontDictonary['Type'])) { |
|
819
|
3 |
|
if (isset($fontDictonary['DW']) && $fontDictonary['DW']) { |
|
820
|
|
|
// defaultWidth / maxWidth for CIDFonts |
|
821
|
|
|
$fontMaxWidth = ((float) $fontDictonary['DW'] / 1000) * $currentFontSize; |
|
822
|
|
|
} else { |
|
823
|
|
|
// default |
|
824
|
3 |
|
$fontMaxWidth = $currentFontSize; |
|
825
|
|
|
} |
|
826
|
|
|
} |
|
827
|
|
|
} |
|
828
|
6 |
|
return $fontMaxWidth; |
|
829
|
|
|
} |
|
830
|
|
|
} |
|
831
|
|
|
|
smalot /
pdfparser
Loading history...
