|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
/** |
|
4
|
|
|
* @file |
|
5
|
|
|
* This file is part of the PdfParser library. |
|
6
|
|
|
* |
|
7
|
|
|
* @author Sébastien MALOT <[email protected]> |
|
8
|
|
|
* |
|
9
|
|
|
* @date 2017-01-03 |
|
10
|
|
|
* |
|
11
|
|
|
* @license LGPLv3 |
|
12
|
|
|
* |
|
13
|
|
|
* @url <https://github.com/smalot/pdfparser> |
|
14
|
|
|
* |
|
15
|
|
|
* PdfParser is a pdf library written in PHP, extraction oriented. |
|
16
|
|
|
* Copyright (C) 2017 - Sébastien MALOT <[email protected]> |
|
17
|
|
|
* |
|
18
|
|
|
* This program is free software: you can redistribute it and/or modify |
|
19
|
|
|
* it under the terms of the GNU Lesser General Public License as published by |
|
20
|
|
|
* the Free Software Foundation, either version 3 of the License, or |
|
21
|
|
|
* (at your option) any later version. |
|
22
|
|
|
* |
|
23
|
|
|
* This program is distributed in the hope that it will be useful, |
|
24
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
25
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
26
|
|
|
* GNU Lesser General Public License for more details. |
|
27
|
|
|
* |
|
28
|
|
|
* You should have received a copy of the GNU Lesser General Public License |
|
29
|
|
|
* along with this program. |
|
30
|
|
|
* If not, see <http://www.pdfparser.org/sites/default/LICENSE.txt>. |
|
31
|
|
|
*/ |
|
32
|
|
|
|
|
33
|
|
|
namespace Smalot\PdfParser; |
|
34
|
|
|
|
|
35
|
|
|
use Smalot\PdfParser\XObject\Form; |
|
36
|
|
|
use Smalot\PdfParser\XObject\Image; |
|
37
|
|
|
|
|
38
|
|
|
/** |
|
39
|
|
|
* Class PDFObject |
|
40
|
|
|
*/ |
|
41
|
|
|
class PDFObject |
|
42
|
|
|
{ |
|
43
|
|
|
public const TYPE = 't'; |
|
44
|
|
|
|
|
45
|
|
|
public const OPERATOR = 'o'; |
|
46
|
|
|
|
|
47
|
|
|
public const COMMAND = 'c'; |
|
48
|
|
|
|
|
49
|
|
|
/** |
|
50
|
|
|
* The recursion stack. |
|
51
|
|
|
* |
|
52
|
|
|
* @var array |
|
53
|
|
|
*/ |
|
54
|
|
|
public static $recursionStack = []; |
|
55
|
|
|
|
|
56
|
|
|
/** |
|
57
|
|
|
* @var Document|null |
|
58
|
|
|
*/ |
|
59
|
|
|
protected $document; |
|
60
|
|
|
|
|
61
|
|
|
/** |
|
62
|
|
|
* @var Header |
|
63
|
|
|
*/ |
|
64
|
|
|
protected $header; |
|
65
|
|
|
|
|
66
|
|
|
/** |
|
67
|
|
|
* @var string |
|
68
|
|
|
*/ |
|
69
|
|
|
protected $content; |
|
70
|
|
|
|
|
71
|
|
|
/** |
|
72
|
|
|
* @var Config|null |
|
73
|
|
|
*/ |
|
74
|
|
|
protected $config; |
|
75
|
|
|
|
|
76
|
|
|
/** |
|
77
|
|
|
* @var bool |
|
78
|
|
|
*/ |
|
79
|
|
|
protected $addPositionWhitespace = false; |
|
80
|
|
|
|
|
81
|
93 |
|
public function __construct( |
|
82
|
|
|
Document $document, |
|
83
|
|
|
?Header $header = null, |
|
84
|
|
|
?string $content = null, |
|
85
|
|
|
?Config $config = null |
|
86
|
|
|
) { |
|
87
|
93 |
|
$this->document = $document; |
|
88
|
93 |
|
$this->header = $header ?? new Header(); |
|
89
|
93 |
|
$this->content = $content; |
|
90
|
93 |
|
$this->config = $config; |
|
91
|
|
|
} |
|
92
|
|
|
|
|
93
|
72 |
|
public function init() |
|
94
|
|
|
{ |
|
95
|
72 |
|
} |
|
96
|
|
|
|
|
97
|
4 |
|
public function getDocument(): Document |
|
98
|
|
|
{ |
|
99
|
4 |
|
return $this->document; |
|
|
|
|
|
|
100
|
|
|
} |
|
101
|
|
|
|
|
102
|
72 |
|
public function getHeader(): ?Header |
|
103
|
|
|
{ |
|
104
|
72 |
|
return $this->header; |
|
105
|
|
|
} |
|
106
|
|
|
|
|
107
|
4 |
|
public function getConfig(): ?Config |
|
108
|
|
|
{ |
|
109
|
4 |
|
return $this->config; |
|
110
|
|
|
} |
|
111
|
|
|
|
|
112
|
|
|
/** |
|
113
|
|
|
* @return Element|PDFObject|Header |
|
114
|
|
|
*/ |
|
115
|
74 |
|
public function get(string $name) |
|
116
|
|
|
{ |
|
117
|
74 |
|
return $this->header->get($name); |
|
118
|
|
|
} |
|
119
|
|
|
|
|
120
|
73 |
|
public function has(string $name): bool |
|
121
|
|
|
{ |
|
122
|
73 |
|
return $this->header->has($name); |
|
123
|
|
|
} |
|
124
|
|
|
|
|
125
|
4 |
|
public function getDetails(bool $deep = true): array |
|
126
|
|
|
{ |
|
127
|
4 |
|
return $this->header->getDetails($deep); |
|
128
|
|
|
} |
|
129
|
|
|
|
|
130
|
59 |
|
public function getContent(): ?string |
|
131
|
|
|
{ |
|
132
|
59 |
|
return $this->content; |
|
133
|
|
|
} |
|
134
|
|
|
|
|
135
|
|
|
/** |
|
136
|
|
|
* Creates a duplicate of the document stream with |
|
137
|
|
|
* strings and other items replaced by $char. Formerly |
|
138
|
|
|
* getSectionsText() used this output to more easily gather offset |
|
139
|
|
|
* values to extract text from the *actual* document stream. |
|
140
|
|
|
* |
|
141
|
|
|
* @deprecated function is no longer used and will be removed in a future release |
|
142
|
|
|
* |
|
143
|
|
|
* @internal |
|
144
|
|
|
*/ |
|
145
|
1 |
|
public function cleanContent(string $content, string $char = 'X') |
|
146
|
|
|
{ |
|
147
|
1 |
|
$char = $char[0]; |
|
148
|
1 |
|
$content = str_replace(['\\\\', '\\)', '\\('], $char.$char, $content); |
|
149
|
|
|
|
|
150
|
|
|
// Remove image bloc with binary content |
|
151
|
1 |
|
preg_match_all('/\s(BI\s.*?(\sID\s).*?(\sEI))\s/s', $content, $matches, \PREG_OFFSET_CAPTURE); |
|
152
|
1 |
|
foreach ($matches[0] as $part) { |
|
153
|
|
|
$content = substr_replace($content, str_repeat($char, \strlen($part[0])), $part[1], \strlen($part[0])); |
|
154
|
|
|
} |
|
155
|
|
|
|
|
156
|
|
|
// Clean content in square brackets [.....] |
|
157
|
1 |
|
preg_match_all('/\[((\(.*?\)|[0-9\.\-\s]*)*)\]/s', $content, $matches, \PREG_OFFSET_CAPTURE); |
|
|
|
|
|
|
158
|
1 |
|
foreach ($matches[1] as $part) { |
|
159
|
1 |
|
$content = substr_replace($content, str_repeat($char, \strlen($part[0])), $part[1], \strlen($part[0])); |
|
160
|
|
|
} |
|
161
|
|
|
|
|
162
|
|
|
// Clean content in round brackets (.....) |
|
163
|
1 |
|
preg_match_all('/\((.*?)\)/s', $content, $matches, \PREG_OFFSET_CAPTURE); |
|
164
|
1 |
|
foreach ($matches[1] as $part) { |
|
165
|
1 |
|
$content = substr_replace($content, str_repeat($char, \strlen($part[0])), $part[1], \strlen($part[0])); |
|
166
|
|
|
} |
|
167
|
|
|
|
|
168
|
|
|
// Clean structure |
|
169
|
1 |
|
if ($parts = preg_split('/(<|>)/s', $content, -1, \PREG_SPLIT_NO_EMPTY | \PREG_SPLIT_DELIM_CAPTURE)) { |
|
|
|
|
|
|
170
|
1 |
|
$content = ''; |
|
171
|
1 |
|
$level = 0; |
|
172
|
1 |
|
foreach ($parts as $part) { |
|
173
|
1 |
|
if ('<' == $part) { |
|
174
|
1 |
|
++$level; |
|
175
|
|
|
} |
|
176
|
|
|
|
|
177
|
1 |
|
$content .= (0 == $level ? $part : str_repeat($char, \strlen($part))); |
|
178
|
|
|
|
|
179
|
1 |
|
if ('>' == $part) { |
|
180
|
1 |
|
--$level; |
|
181
|
|
|
} |
|
182
|
|
|
} |
|
183
|
|
|
} |
|
184
|
|
|
|
|
185
|
|
|
// Clean BDC and EMC markup |
|
186
|
1 |
|
preg_match_all( |
|
187
|
1 |
|
'/(\/[A-Za-z0-9\_]*\s*'.preg_quote($char).'*BDC)/s', |
|
188
|
1 |
|
$content, |
|
189
|
1 |
|
$matches, |
|
190
|
1 |
|
\PREG_OFFSET_CAPTURE |
|
191
|
1 |
|
); |
|
192
|
1 |
|
foreach ($matches[1] as $part) { |
|
193
|
1 |
|
$content = substr_replace($content, str_repeat($char, \strlen($part[0])), $part[1], \strlen($part[0])); |
|
194
|
|
|
} |
|
195
|
|
|
|
|
196
|
1 |
|
preg_match_all('/\s(EMC)\s/s', $content, $matches, \PREG_OFFSET_CAPTURE); |
|
197
|
1 |
|
foreach ($matches[1] as $part) { |
|
198
|
1 |
|
$content = substr_replace($content, str_repeat($char, \strlen($part[0])), $part[1], \strlen($part[0])); |
|
199
|
|
|
} |
|
200
|
|
|
|
|
201
|
1 |
|
return $content; |
|
202
|
|
|
} |
|
203
|
|
|
|
|
204
|
|
|
/** |
|
205
|
|
|
* Takes a string of PDF document stream text and formats |
|
206
|
|
|
* it into a multi-line string with one PDF command on each line, |
|
207
|
|
|
* separated by \r\n. If the given string is null, or binary data |
|
208
|
|
|
* is detected instead of a document stream then return an empty |
|
209
|
|
|
* string. |
|
210
|
|
|
*/ |
|
211
|
52 |
|
private function formatContent(?string $content): string |
|
212
|
|
|
{ |
|
213
|
52 |
|
if (null === $content) { |
|
214
|
3 |
|
return ''; |
|
215
|
|
|
} |
|
216
|
|
|
|
|
217
|
|
|
// Outside of (String) content in PDF document streams, all |
|
218
|
|
|
// text should conform to UTF-8. Test for binary content by |
|
219
|
|
|
// deleting everything after the first open-parenthesis ( which |
|
220
|
|
|
// indicates the beginning of a string. Then test what remains |
|
221
|
|
|
// for valid UTF-8. If it's not UTF-8, return an empty string |
|
222
|
|
|
// as this $content is most likely binary. |
|
223
|
49 |
|
if (false === mb_check_encoding(preg_replace('/\(.*$/s', '', $content), 'UTF-8')) { |
|
224
|
1 |
|
return ''; |
|
225
|
|
|
} |
|
226
|
|
|
|
|
227
|
|
|
// Find all strings () and replace them so they aren't affected |
|
228
|
|
|
// by the next steps |
|
229
|
49 |
|
$pdfstrings = []; |
|
230
|
49 |
|
$attempt = '('; |
|
231
|
49 |
|
while (preg_match('/'.preg_quote($attempt, '/').'.*?(?<![^\\\\]\\\\)\)/s', $content, $text)) { |
|
232
|
|
|
// PDF strings can contain unescaped parentheses as long as |
|
233
|
|
|
// they're balanced, so check for balanced parentheses |
|
234
|
40 |
|
$left = preg_match_all('/(?<![^\\\\]\\\\)\(/', $text[0]); |
|
235
|
40 |
|
$right = preg_match_all('/(?<![^\\\\]\\\\)\)/', $text[0]); |
|
236
|
|
|
|
|
237
|
40 |
|
if ($left == $right) { |
|
238
|
|
|
// Replace the string with a unique placeholder |
|
239
|
40 |
|
$id = uniqid('STRING_', true); |
|
240
|
40 |
|
$pdfstrings[$id] = $text[0]; |
|
241
|
40 |
|
$content = preg_replace( |
|
242
|
40 |
|
'/'.preg_quote($text[0], '/').'/', |
|
243
|
40 |
|
'@@@'.$id.'@@@', |
|
244
|
40 |
|
$content, |
|
245
|
40 |
|
1 |
|
246
|
40 |
|
); |
|
247
|
|
|
|
|
248
|
|
|
// Reset to search for the next string |
|
249
|
40 |
|
$attempt = '('; |
|
250
|
|
|
} else { |
|
251
|
|
|
// We had unbalanced parentheses, so use the current |
|
252
|
|
|
// match as a base to find a longer string |
|
253
|
|
|
$attempt = $text[0]; |
|
254
|
|
|
} |
|
255
|
|
|
} |
|
256
|
|
|
|
|
257
|
|
|
// Remove all carriage returns and line-feeds from the document stream |
|
258
|
49 |
|
$content = str_replace(["\r", "\n"], ' ', trim($content)); |
|
259
|
|
|
|
|
260
|
|
|
// Find all dictionary << >> commands and replace them so they |
|
261
|
|
|
// aren't affected by the next steps |
|
262
|
49 |
|
$dictstore = []; |
|
263
|
49 |
|
while (preg_match('/(<<.*?>> *)(BDC|BMC|DP|MP)/', $content, $dicttext)) { |
|
264
|
18 |
|
$dictid = uniqid('DICT_', true); |
|
265
|
18 |
|
$dictstore[$dictid] = $dicttext[1]; |
|
266
|
18 |
|
$content = preg_replace( |
|
267
|
18 |
|
'/'.preg_quote($dicttext[0], '/').'/', |
|
268
|
18 |
|
' ###'.$dictid.'###'.$dicttext[2], |
|
269
|
18 |
|
$content, |
|
270
|
18 |
|
1 |
|
271
|
18 |
|
); |
|
272
|
|
|
} |
|
273
|
|
|
|
|
274
|
|
|
// Normalize white-space in the document stream |
|
275
|
49 |
|
$content = preg_replace('/\s{2,}/', ' ', $content); |
|
276
|
|
|
|
|
277
|
|
|
// Find all valid PDF operators and add \r\n after each; this |
|
278
|
|
|
// ensures there is just one command on every line |
|
279
|
|
|
// Source: https://ia801001.us.archive.org/1/items/pdf1.7/pdf_reference_1-7.pdf - Appendix A |
|
280
|
|
|
// Source: https://archive.org/download/pdf320002008/PDF32000_2008.pdf - Annex A |
|
281
|
|
|
// Note: PDF Reference 1.7 lists 'I' and 'rI' as valid commands, while |
|
282
|
|
|
// PDF 32000:2008 lists them as 'i' and 'ri' respectively. Both versions |
|
283
|
|
|
// appear here in the list for completeness. |
|
284
|
49 |
|
$operators = [ |
|
285
|
49 |
|
'b*', 'b', 'BDC', 'BMC', 'B*', 'BI', 'BT', 'BX', 'B', 'cm', 'cs', 'c', 'CS', |
|
286
|
49 |
|
'd0', 'd1', 'd', 'Do', 'DP', 'EMC', 'EI', 'ET', 'EX', 'f*', 'f', 'F', 'gs', |
|
287
|
49 |
|
'g', 'G', 'h', 'i', 'ID', 'I', 'j', 'J', 'k', 'K', 'l', 'm', 'MP', 'M', 'n', |
|
288
|
49 |
|
'q', 'Q', 're', 'rg', 'ri', 'rI', 'RG', 'scn', 'sc', 'sh', 's', 'SCN', 'SC', |
|
289
|
49 |
|
'S', 'T*', 'Tc', 'Td', 'TD', 'Tf', 'TJ', 'Tj', 'TL', 'Tm', 'Tr', 'Ts', 'Tw', |
|
290
|
49 |
|
'Tz', 'v', 'w', 'W*', 'W', 'y', '\'', '"', |
|
291
|
49 |
|
]; |
|
292
|
49 |
|
foreach ($operators as $operator) { |
|
293
|
49 |
|
$content = preg_replace( |
|
294
|
49 |
|
'/(?<!\w|\/)'.preg_quote($operator, '/').'(?![\w10\*])/', |
|
295
|
49 |
|
$operator."\r\n", |
|
296
|
49 |
|
$content |
|
297
|
49 |
|
); |
|
298
|
|
|
} |
|
299
|
|
|
|
|
300
|
|
|
// Restore the original content of the dictionary << >> commands |
|
301
|
49 |
|
$dictstore = array_reverse($dictstore, true); |
|
302
|
49 |
|
foreach ($dictstore as $id => $dict) { |
|
303
|
18 |
|
$content = str_replace('###'.$id.'###', $dict, $content); |
|
304
|
|
|
} |
|
305
|
|
|
|
|
306
|
|
|
// Restore the original string content |
|
307
|
49 |
|
$pdfstrings = array_reverse($pdfstrings, true); |
|
308
|
49 |
|
foreach ($pdfstrings as $id => $text) { |
|
309
|
|
|
// Strings may contain escaped newlines, or literal newlines |
|
310
|
|
|
// and we should clean these up before replacing the string |
|
311
|
|
|
// back into the content stream; this ensures no strings are |
|
312
|
|
|
// split between two lines (every command must be on one line) |
|
313
|
40 |
|
$text = str_replace( |
|
314
|
40 |
|
["\\\r\n", "\\\r", "\\\n", "\r", "\n"], |
|
315
|
40 |
|
['', '', '', '\r', '\n'], |
|
316
|
40 |
|
$text |
|
317
|
40 |
|
); |
|
318
|
|
|
|
|
319
|
40 |
|
$content = str_replace('@@@'.$id.'@@@', $text, $content); |
|
320
|
|
|
} |
|
321
|
|
|
|
|
322
|
49 |
|
$content = trim(preg_replace(['/(\r\n){2,}/', '/\r\n +/'], "\r\n", $content)); |
|
323
|
|
|
|
|
324
|
49 |
|
return $content; |
|
325
|
|
|
} |
|
326
|
|
|
|
|
327
|
|
|
/** |
|
328
|
|
|
* getSectionsText() now takes an entire, unformatted |
|
329
|
|
|
* document stream as a string, cleans it, then filters out |
|
330
|
|
|
* commands that aren't needed for text positioning/extraction. It |
|
331
|
|
|
* returns an array of unprocessed PDF commands, one command per |
|
332
|
|
|
* element. |
|
333
|
|
|
* |
|
334
|
|
|
* @internal |
|
335
|
|
|
*/ |
|
336
|
52 |
|
public function getSectionsText(?string $content): array |
|
337
|
|
|
{ |
|
338
|
52 |
|
$sections = []; |
|
339
|
|
|
|
|
340
|
|
|
// A cleaned stream has one command on every line, so split the |
|
341
|
|
|
// cleaned stream content on \r\n into an array |
|
342
|
52 |
|
$textCleaned = preg_split( |
|
343
|
52 |
|
'/(\r\n|\n|\r)/', |
|
344
|
52 |
|
$this->formatContent($content), |
|
345
|
52 |
|
-1, |
|
346
|
52 |
|
\PREG_SPLIT_NO_EMPTY |
|
347
|
52 |
|
); |
|
348
|
|
|
|
|
349
|
52 |
|
$inTextBlock = false; |
|
350
|
52 |
|
foreach ($textCleaned as $line) { |
|
351
|
49 |
|
$line = trim($line); |
|
352
|
|
|
|
|
353
|
|
|
// Skip empty lines |
|
354
|
49 |
|
if ('' === $line) { |
|
355
|
|
|
continue; |
|
356
|
|
|
} |
|
357
|
|
|
|
|
358
|
|
|
// If a 'BT' is encountered, set the $inTextBlock flag |
|
359
|
49 |
|
if (preg_match('/BT$/', $line)) { |
|
360
|
49 |
|
$inTextBlock = true; |
|
361
|
49 |
|
$sections[] = $line; |
|
362
|
|
|
|
|
363
|
|
|
// If an 'ET' is encountered, unset the $inTextBlock flag |
|
364
|
49 |
|
} elseif ('ET' == $line) { |
|
365
|
49 |
|
$inTextBlock = false; |
|
366
|
49 |
|
$sections[] = $line; |
|
367
|
49 |
|
} elseif ($inTextBlock) { |
|
368
|
|
|
// If we are inside a BT ... ET text block, save all lines |
|
369
|
49 |
|
$sections[] = trim($line); |
|
370
|
|
|
} else { |
|
371
|
|
|
// Otherwise, if we are outside of a text block, only |
|
372
|
|
|
// save specific, necessary lines. Care should be taken |
|
373
|
|
|
// to ensure a command being checked for *only* matches |
|
374
|
|
|
// that command. For instance, a simple search for 'c' |
|
375
|
|
|
// may also match the 'sc' command. See the command |
|
376
|
|
|
// list in the formatContent() method above. |
|
377
|
|
|
// Add more commands to save here as you find them in |
|
378
|
|
|
// weird PDFs! |
|
379
|
48 |
|
if ('q' == $line[-1] || 'Q' == $line[-1]) { |
|
380
|
|
|
// Save and restore graphics state commands |
|
381
|
42 |
|
$sections[] = $line; |
|
382
|
48 |
|
} elseif (preg_match('/(?<!\w)B[DM]C$/', $line)) { |
|
383
|
|
|
// Begin marked content sequence |
|
384
|
16 |
|
$sections[] = $line; |
|
385
|
48 |
|
} elseif (preg_match('/(?<!\w)[DM]P$/', $line)) { |
|
386
|
|
|
// Marked content point |
|
387
|
1 |
|
$sections[] = $line; |
|
388
|
47 |
|
} elseif (preg_match('/(?<!\w)EMC$/', $line)) { |
|
389
|
|
|
// End marked content sequence |
|
390
|
15 |
|
$sections[] = $line; |
|
391
|
45 |
|
} elseif (preg_match('/(?<!\w)cm$/', $line)) { |
|
392
|
|
|
// Graphics position change commands |
|
393
|
33 |
|
$sections[] = $line; |
|
394
|
45 |
|
} elseif (preg_match('/(?<!\w)Tf$/', $line)) { |
|
395
|
|
|
// Font change commands |
|
396
|
3 |
|
$sections[] = $line; |
|
397
|
45 |
|
} elseif (preg_match('/(?<!\w)Do$/', $line)) { |
|
398
|
|
|
// Invoke named XObject command |
|
399
|
15 |
|
$sections[] = $line; |
|
400
|
|
|
} |
|
401
|
|
|
} |
|
402
|
|
|
} |
|
403
|
|
|
|
|
404
|
52 |
|
return $sections; |
|
405
|
|
|
} |
|
406
|
|
|
|
|
407
|
46 |
|
private function getDefaultFont(?Page $page = null): Font |
|
408
|
|
|
{ |
|
409
|
46 |
|
$fonts = []; |
|
410
|
46 |
|
if (null !== $page) { |
|
411
|
44 |
|
$fonts = $page->getFonts(); |
|
412
|
|
|
} |
|
413
|
|
|
|
|
414
|
46 |
|
$firstFont = $this->document->getFirstFont(); |
|
|
|
|
|
|
415
|
46 |
|
if (null !== $firstFont) { |
|
416
|
43 |
|
$fonts[] = $firstFont; |
|
417
|
|
|
} |
|
418
|
|
|
|
|
419
|
46 |
|
if (\count($fonts) > 0) { |
|
420
|
43 |
|
return reset($fonts); |
|
421
|
|
|
} |
|
422
|
|
|
|
|
423
|
3 |
|
return new Font($this->document, null, null, $this->config); |
|
|
|
|
|
|
424
|
|
|
} |
|
425
|
|
|
|
|
426
|
|
|
/** |
|
427
|
|
|
* Decode a '[]TJ' command and attempt to use alternate |
|
428
|
|
|
* fonts if the current font results in output that contains |
|
429
|
|
|
* Unicode control characters. |
|
430
|
|
|
* |
|
431
|
|
|
* @internal |
|
432
|
|
|
* |
|
433
|
|
|
* @param array<int,array<string,string|bool>> $command |
|
434
|
|
|
*/ |
|
435
|
43 |
|
private function getTJUsingFontFallback(Font $font, array $command, ?Page $page = null, float $fontFactor = 4): string |
|
436
|
|
|
{ |
|
437
|
43 |
|
$orig_text = $font->decodeText($command, $fontFactor); |
|
438
|
43 |
|
$text = $orig_text; |
|
439
|
|
|
|
|
440
|
|
|
// If we make this a Config option, we can add a check if it's |
|
441
|
|
|
// enabled here. |
|
442
|
43 |
|
if (null !== $page) { |
|
443
|
43 |
|
$font_ids = array_keys($page->getFonts()); |
|
444
|
|
|
|
|
445
|
|
|
// If the decoded text contains UTF-8 control characters |
|
446
|
|
|
// then the font page being used is probably the wrong one. |
|
447
|
|
|
// Loop through the rest of the fonts to see if we can get |
|
448
|
|
|
// a good decode. Allow x09 to x0d which are whitespace. |
|
449
|
43 |
|
while (preg_match('/[\x00-\x08\x0e-\x1f\x7f]/u', $text) || false !== strpos(bin2hex($text), '00')) { |
|
450
|
|
|
// If we're out of font IDs, then give up and use the |
|
451
|
|
|
// original string |
|
452
|
3 |
|
if (0 == \count($font_ids)) { |
|
453
|
3 |
|
return $orig_text; |
|
454
|
|
|
} |
|
455
|
|
|
|
|
456
|
|
|
// Try the next font ID |
|
457
|
3 |
|
$font = $page->getFont(array_shift($font_ids)); |
|
458
|
3 |
|
$text = $font->decodeText($command, $fontFactor); |
|
459
|
|
|
} |
|
460
|
|
|
} |
|
461
|
|
|
|
|
462
|
43 |
|
return $text; |
|
463
|
|
|
} |
|
464
|
|
|
|
|
465
|
|
|
/** |
|
466
|
|
|
* Expects a string that is a full PDF dictionary object, |
|
467
|
|
|
* including the outer enclosing << >> angle brackets |
|
468
|
|
|
* |
|
469
|
|
|
* @internal |
|
470
|
|
|
* |
|
471
|
|
|
* @throws \Exception |
|
472
|
|
|
*/ |
|
473
|
17 |
|
public function parseDictionary(string $dictionary): array |
|
474
|
|
|
{ |
|
475
|
|
|
// Normalize whitespace |
|
476
|
17 |
|
$dictionary = preg_replace(['/\r/', '/\n/', '/\s{2,}/'], ' ', trim($dictionary)); |
|
477
|
|
|
|
|
478
|
17 |
|
if ('<<' != substr($dictionary, 0, 2)) { |
|
479
|
|
|
throw new \Exception('Not a valid dictionary object.'); |
|
480
|
|
|
} |
|
481
|
|
|
|
|
482
|
17 |
|
$parsed = []; |
|
483
|
17 |
|
$stack = []; |
|
484
|
17 |
|
$currentName = ''; |
|
485
|
17 |
|
$arrayTypeNumeric = false; |
|
486
|
|
|
|
|
487
|
|
|
// Remove outer layer of dictionary, and split on tokens |
|
488
|
17 |
|
$split = preg_split( |
|
489
|
17 |
|
'/(<<|>>|\[|\]|\/[^\s\/\[\]\(\)<>]*)/', |
|
490
|
17 |
|
trim(preg_replace('/^<<|>>$/', '', $dictionary)), |
|
491
|
17 |
|
-1, |
|
492
|
17 |
|
\PREG_SPLIT_NO_EMPTY | \PREG_SPLIT_DELIM_CAPTURE |
|
493
|
17 |
|
); |
|
494
|
|
|
|
|
495
|
17 |
|
foreach ($split as $token) { |
|
496
|
17 |
|
$token = trim($token); |
|
497
|
|
|
switch ($token) { |
|
498
|
17 |
|
case '': |
|
499
|
7 |
|
break; |
|
500
|
|
|
|
|
501
|
|
|
// Open numeric array |
|
502
|
17 |
|
case '[': |
|
503
|
7 |
|
$parsed[$currentName] = []; |
|
504
|
7 |
|
$arrayTypeNumeric = true; |
|
505
|
|
|
|
|
506
|
|
|
// Move up one level in the stack |
|
507
|
7 |
|
$stack[\count($stack)] = &$parsed; |
|
508
|
7 |
|
$parsed = &$parsed[$currentName]; |
|
509
|
7 |
|
$currentName = ''; |
|
510
|
7 |
|
break; |
|
511
|
|
|
|
|
512
|
|
|
// Open hashed array |
|
513
|
17 |
|
case '<<': |
|
514
|
1 |
|
$parsed[$currentName] = []; |
|
515
|
1 |
|
$arrayTypeNumeric = false; |
|
516
|
|
|
|
|
517
|
|
|
// Move up one level in the stack |
|
518
|
1 |
|
$stack[\count($stack)] = &$parsed; |
|
519
|
1 |
|
$parsed = &$parsed[$currentName]; |
|
520
|
1 |
|
$currentName = ''; |
|
521
|
1 |
|
break; |
|
522
|
|
|
|
|
523
|
|
|
// Close numeric array |
|
524
|
17 |
|
case ']': |
|
525
|
|
|
// Revert string type arrays back to a single element |
|
526
|
7 |
|
if (\is_array($parsed) && 1 == \count($parsed) |
|
527
|
7 |
|
&& isset($parsed[0]) && \is_string($parsed[0]) |
|
528
|
7 |
|
&& '' !== $parsed[0] && '/' != $parsed[0][0]) { |
|
529
|
6 |
|
$parsed = '['.$parsed[0].']'; |
|
530
|
|
|
} |
|
531
|
|
|
// Close hashed array |
|
532
|
|
|
// no break |
|
533
|
17 |
|
case '>>': |
|
534
|
7 |
|
$arrayTypeNumeric = false; |
|
535
|
|
|
|
|
536
|
|
|
// Move down one level in the stack |
|
537
|
7 |
|
$parsed = &$stack[\count($stack) - 1]; |
|
538
|
7 |
|
unset($stack[\count($stack) - 1]); |
|
539
|
7 |
|
break; |
|
540
|
|
|
|
|
541
|
|
|
default: |
|
542
|
|
|
// If value begins with a slash, then this is a name |
|
543
|
|
|
// Add it to the appropriate array |
|
544
|
17 |
|
if ('/' == substr($token, 0, 1)) { |
|
545
|
17 |
|
$currentName = substr($token, 1); |
|
546
|
17 |
|
if (true == $arrayTypeNumeric) { |
|
|
|
|
|
|
547
|
6 |
|
$parsed[] = $currentName; |
|
548
|
17 |
|
$currentName = ''; |
|
549
|
|
|
} |
|
550
|
17 |
|
} elseif ('' != $currentName) { |
|
551
|
17 |
|
if (false == $arrayTypeNumeric) { |
|
|
|
|
|
|
552
|
17 |
|
$parsed[$currentName] = $token; |
|
553
|
|
|
} |
|
554
|
17 |
|
$currentName = ''; |
|
555
|
5 |
|
} elseif ('' == $currentName) { |
|
556
|
5 |
|
$parsed[] = $token; |
|
557
|
|
|
} |
|
558
|
|
|
} |
|
559
|
|
|
} |
|
560
|
|
|
|
|
561
|
17 |
|
return $parsed; |
|
562
|
|
|
} |
|
563
|
|
|
|
|
564
|
|
|
/** |
|
565
|
|
|
* Returns the text content of a PDF as a string. Attempts to add |
|
566
|
|
|
* whitespace for spacing and line-breaks where appropriate. |
|
567
|
|
|
* |
|
568
|
|
|
* getText() leverages getTextArray() to get the content |
|
569
|
|
|
* of the document, setting the addPositionWhitespace flag to true |
|
570
|
|
|
* so whitespace is inserted in a logical way for reading by |
|
571
|
|
|
* humans. |
|
572
|
|
|
*/ |
|
573
|
37 |
|
public function getText(?Page $page = null): string |
|
574
|
|
|
{ |
|
575
|
37 |
|
$this->addPositionWhitespace = true; |
|
576
|
37 |
|
$result = $this->getTextArray($page); |
|
577
|
37 |
|
$this->addPositionWhitespace = false; |
|
578
|
|
|
|
|
579
|
37 |
|
return implode('', $result).' '; |
|
580
|
|
|
} |
|
581
|
|
|
|
|
582
|
|
|
/** |
|
583
|
|
|
* Returns the text content of a PDF as an array of strings. No |
|
584
|
|
|
* extra whitespace is inserted besides what is actually encoded in |
|
585
|
|
|
* the PDF text. |
|
586
|
|
|
* |
|
587
|
|
|
* @throws \Exception |
|
588
|
|
|
*/ |
|
589
|
46 |
|
public function getTextArray(?Page $page = null): array |
|
590
|
|
|
{ |
|
591
|
46 |
|
$result = []; |
|
592
|
46 |
|
$text = []; |
|
593
|
|
|
|
|
594
|
46 |
|
$marked_stack = []; |
|
595
|
46 |
|
$last_written_position = false; |
|
596
|
|
|
|
|
597
|
46 |
|
$sections = $this->getSectionsText($this->content); |
|
598
|
46 |
|
$current_font = $this->getDefaultFont($page); |
|
599
|
46 |
|
$current_font_size = 1; |
|
600
|
46 |
|
$current_text_leading = 0; |
|
601
|
|
|
|
|
602
|
46 |
|
$current_position = ['x' => false, 'y' => false]; |
|
603
|
46 |
|
$current_position_tm = [ |
|
604
|
46 |
|
'a' => 1, 'b' => 0, 'c' => 0, |
|
605
|
46 |
|
'i' => 0, 'j' => 1, 'k' => 0, |
|
606
|
46 |
|
'x' => 0, 'y' => 0, 'z' => 1, |
|
607
|
46 |
|
]; |
|
608
|
46 |
|
$current_position_td = ['x' => 0, 'y' => 0]; |
|
609
|
46 |
|
$current_position_cm = [ |
|
610
|
46 |
|
'a' => 1, 'b' => 0, 'c' => 0, |
|
611
|
46 |
|
'i' => 0, 'j' => 1, 'k' => 0, |
|
612
|
46 |
|
'x' => 0, 'y' => 0, 'z' => 1, |
|
613
|
46 |
|
]; |
|
614
|
|
|
|
|
615
|
46 |
|
$clipped_font = []; |
|
616
|
46 |
|
$clipped_position_cm = []; |
|
617
|
|
|
|
|
618
|
46 |
|
self::$recursionStack[] = $this->getUniqueId(); |
|
619
|
|
|
|
|
620
|
46 |
|
foreach ($sections as $section) { |
|
621
|
43 |
|
$commands = $this->getCommandsText($section); |
|
622
|
43 |
|
foreach ($commands as $command) { |
|
623
|
43 |
|
switch ($command[self::OPERATOR]) { |
|
624
|
|
|
// Begin text object |
|
625
|
43 |
|
case 'BT': |
|
626
|
|
|
// Reset text positioning matrices |
|
627
|
43 |
|
$current_position_tm = [ |
|
628
|
43 |
|
'a' => 1, 'b' => 0, 'c' => 0, |
|
629
|
43 |
|
'i' => 0, 'j' => 1, 'k' => 0, |
|
630
|
43 |
|
'x' => 0, 'y' => 0, 'z' => 1, |
|
631
|
43 |
|
]; |
|
632
|
43 |
|
$current_position_td = ['x' => 0, 'y' => 0]; |
|
633
|
43 |
|
$current_text_leading = 0; |
|
634
|
43 |
|
break; |
|
635
|
|
|
|
|
636
|
|
|
// Begin marked content sequence with property list |
|
637
|
43 |
|
case 'BDC': |
|
638
|
16 |
|
if (preg_match('/(<<.*>>)$/', $command[self::COMMAND], $match)) { |
|
639
|
16 |
|
$dict = $this->parseDictionary($match[1]); |
|
640
|
|
|
|
|
641
|
|
|
// Check for ActualText block |
|
642
|
16 |
|
if (isset($dict['ActualText']) && \is_string($dict['ActualText']) && '' !== $dict['ActualText']) { |
|
643
|
4 |
|
if ('[' == $dict['ActualText'][0]) { |
|
644
|
|
|
// Simulate a 'TJ' command on the stack |
|
645
|
|
|
$marked_stack[] = [ |
|
646
|
|
|
'ActualText' => $this->getCommandsText($dict['ActualText'].'TJ')[0], |
|
647
|
|
|
]; |
|
648
|
4 |
|
} elseif ('<' == $dict['ActualText'][0] || '(' == $dict['ActualText'][0]) { |
|
649
|
|
|
// Simulate a 'Tj' command on the stack |
|
650
|
4 |
|
$marked_stack[] = [ |
|
651
|
4 |
|
'ActualText' => $this->getCommandsText($dict['ActualText'].'Tj')[0], |
|
652
|
4 |
|
]; |
|
653
|
|
|
} |
|
654
|
|
|
} |
|
655
|
|
|
} |
|
656
|
16 |
|
break; |
|
657
|
|
|
|
|
658
|
|
|
// Begin marked content sequence |
|
659
|
43 |
|
case 'BMC': |
|
660
|
2 |
|
if ('ReversedChars' == $command[self::COMMAND]) { |
|
661
|
|
|
// Upon encountering a ReversedChars command, |
|
662
|
|
|
// add the characters we've built up so far to |
|
663
|
|
|
// the result array |
|
664
|
1 |
|
$result = array_merge($result, $text); |
|
665
|
|
|
|
|
666
|
|
|
// Start a fresh $text array that will contain |
|
667
|
|
|
// reversed characters |
|
668
|
1 |
|
$text = []; |
|
669
|
|
|
|
|
670
|
|
|
// Add the reversed text flag to the stack |
|
671
|
1 |
|
$marked_stack[] = ['ReversedChars' => true]; |
|
672
|
|
|
} |
|
673
|
2 |
|
break; |
|
674
|
|
|
|
|
675
|
|
|
// set graphics position matrix |
|
676
|
43 |
|
case 'cm': |
|
677
|
29 |
|
$args = preg_split('/\s+/s', $command[self::COMMAND]); |
|
678
|
29 |
|
$current_position_cm = [ |
|
679
|
29 |
|
'a' => (float) $args[0], 'b' => (float) $args[1], 'c' => 0, |
|
680
|
29 |
|
'i' => (float) $args[2], 'j' => (float) $args[3], 'k' => 0, |
|
681
|
29 |
|
'x' => (float) $args[4], 'y' => (float) $args[5], 'z' => 1, |
|
682
|
29 |
|
]; |
|
683
|
29 |
|
break; |
|
684
|
|
|
|
|
685
|
43 |
|
case 'Do': |
|
686
|
15 |
|
if (null !== $page) { |
|
687
|
15 |
|
$args = preg_split('/\s/s', $command[self::COMMAND]); |
|
688
|
15 |
|
$id = trim(array_pop($args), '/ '); |
|
689
|
15 |
|
$xobject = $page->getXObject($id); |
|
690
|
|
|
|
|
691
|
|
|
// @todo $xobject could be a ElementXRef object, which would then throw an error |
|
692
|
15 |
|
if (\is_object($xobject) && $xobject instanceof self && !\in_array($xobject->getUniqueId(), self::$recursionStack, true)) { |
|
693
|
|
|
// Not a circular reference. |
|
694
|
15 |
|
$text[] = $xobject->getText($page); |
|
695
|
|
|
} |
|
696
|
|
|
} |
|
697
|
15 |
|
break; |
|
698
|
|
|
|
|
699
|
|
|
// Marked content point with (DP) & without (MP) property list |
|
700
|
43 |
|
case 'DP': |
|
701
|
43 |
|
case 'MP': |
|
702
|
1 |
|
break; |
|
703
|
|
|
|
|
704
|
|
|
// End text object |
|
705
|
43 |
|
case 'ET': |
|
706
|
43 |
|
break; |
|
707
|
|
|
|
|
708
|
|
|
// Store current selected font and graphics matrix |
|
709
|
43 |
|
case 'q': |
|
710
|
37 |
|
$clipped_font[] = [$current_font, $current_font_size]; |
|
711
|
37 |
|
$clipped_position_cm[] = $current_position_cm; |
|
712
|
37 |
|
break; |
|
713
|
|
|
|
|
714
|
|
|
// Restore previous selected font and graphics matrix |
|
715
|
43 |
|
case 'Q': |
|
716
|
37 |
|
list($current_font, $current_font_size) = array_pop($clipped_font); |
|
717
|
37 |
|
$current_position_cm = array_pop($clipped_position_cm); |
|
718
|
37 |
|
break; |
|
719
|
|
|
|
|
720
|
|
|
// End marked content sequence |
|
721
|
43 |
|
case 'EMC': |
|
722
|
17 |
|
$data = false; |
|
723
|
17 |
|
if (\count($marked_stack)) { |
|
724
|
5 |
|
$marked = array_pop($marked_stack); |
|
725
|
5 |
|
$action = key($marked); |
|
726
|
5 |
|
$data = $marked[$action]; |
|
727
|
|
|
|
|
728
|
|
|
switch ($action) { |
|
729
|
|
|
// If we are in ReversedChars mode... |
|
730
|
5 |
|
case 'ReversedChars': |
|
731
|
|
|
// Reverse the characters we've built up so far |
|
732
|
1 |
|
foreach ($text as $key => $t) { |
|
733
|
1 |
|
$text[$key] = implode('', array_reverse( |
|
734
|
1 |
|
mb_str_split($t, 1, mb_internal_encoding()) |
|
|
|
|
|
|
735
|
1 |
|
)); |
|
736
|
|
|
} |
|
737
|
|
|
|
|
738
|
|
|
// Add these characters to the result array |
|
739
|
1 |
|
$result = array_merge($result, $text); |
|
740
|
|
|
|
|
741
|
|
|
// Start a fresh $text array that will contain |
|
742
|
|
|
// non-reversed characters |
|
743
|
1 |
|
$text = []; |
|
744
|
1 |
|
break; |
|
745
|
|
|
|
|
746
|
4 |
|
case 'ActualText': |
|
747
|
|
|
// Use the content of the ActualText as a command |
|
748
|
4 |
|
$command = $data; |
|
749
|
4 |
|
break; |
|
750
|
|
|
} |
|
751
|
|
|
} |
|
752
|
|
|
|
|
753
|
|
|
// If this EMC command has been transformed into a 'Tj' |
|
754
|
|
|
// or 'TJ' command because of being ActualText, then bypass |
|
755
|
|
|
// the break to proceed to the writing section below. |
|
756
|
17 |
|
if ('Tj' != $command[self::OPERATOR] && 'TJ' != $command[self::OPERATOR]) { |
|
757
|
17 |
|
break; |
|
758
|
|
|
} |
|
759
|
|
|
|
|
760
|
|
|
// no break |
|
761
|
43 |
|
case "'": |
|
762
|
43 |
|
case '"': |
|
763
|
4 |
|
if ("'" == $command[self::OPERATOR] || '"' == $command[self::OPERATOR]) { |
|
764
|
|
|
// Move to next line and write text |
|
765
|
|
|
$current_position['x'] = 0; |
|
766
|
|
|
$current_position_td['x'] = 0; |
|
767
|
|
|
$current_position_td['y'] += $current_text_leading; |
|
768
|
|
|
} |
|
769
|
|
|
// no break |
|
770
|
43 |
|
case 'Tj': |
|
771
|
35 |
|
$command[self::COMMAND] = [$command]; |
|
772
|
|
|
// no break |
|
773
|
43 |
|
case 'TJ': |
|
774
|
|
|
// Check the marked content stack for flags |
|
775
|
43 |
|
$actual_text = false; |
|
776
|
43 |
|
$reverse_text = false; |
|
777
|
43 |
|
foreach ($marked_stack as $marked) { |
|
778
|
5 |
|
if (isset($marked['ActualText'])) { |
|
779
|
4 |
|
$actual_text = true; |
|
780
|
|
|
} |
|
781
|
5 |
|
if (isset($marked['ReversedChars'])) { |
|
782
|
1 |
|
$reverse_text = true; |
|
783
|
|
|
} |
|
784
|
|
|
} |
|
785
|
|
|
|
|
786
|
|
|
// Account for text position ONLY just before we write text |
|
787
|
43 |
|
if (false === $actual_text && \is_array($last_written_position)) { |
|
788
|
|
|
// If $last_written_position is an array, that |
|
789
|
|
|
// means we have stored text position coordinates |
|
790
|
|
|
// for placing an ActualText |
|
791
|
4 |
|
$currentX = $last_written_position[0]; |
|
792
|
4 |
|
$currentY = $last_written_position[1]; |
|
793
|
4 |
|
$last_written_position = false; |
|
794
|
|
|
} else { |
|
795
|
43 |
|
$currentX = $current_position_cm['x'] + $current_position_tm['x'] + $current_position_td['x']; |
|
796
|
43 |
|
$currentY = $current_position_cm['y'] + $current_position_tm['y'] + $current_position_td['y']; |
|
797
|
|
|
} |
|
798
|
43 |
|
$whiteSpace = ''; |
|
799
|
|
|
|
|
800
|
43 |
|
$factorX = -$current_font_size * $current_position_tm['a'] - $current_font_size * $current_position_tm['i']; |
|
801
|
43 |
|
$factorY = $current_font_size * $current_position_tm['b'] + $current_font_size * $current_position_tm['j']; |
|
802
|
|
|
|
|
803
|
43 |
|
if (true === $this->addPositionWhitespace && false !== $current_position['x']) { |
|
804
|
31 |
|
$curY = $currentY - $current_position['y']; |
|
805
|
31 |
|
if (abs($curY) >= abs($factorY) / 4) { |
|
806
|
30 |
|
$whiteSpace = "\n"; |
|
807
|
|
|
} else { |
|
808
|
30 |
|
if (true === $reverse_text) { |
|
809
|
1 |
|
$curX = $current_position['x'] - $currentX; |
|
810
|
|
|
} else { |
|
811
|
30 |
|
$curX = $currentX - $current_position['x']; |
|
812
|
|
|
} |
|
813
|
|
|
|
|
814
|
|
|
// In abs($factorX * 7) below, the 7 is chosen arbitrarily |
|
815
|
|
|
// as the number of apparent "spaces" in a document we |
|
816
|
|
|
// would need before considering them a "tab". In the |
|
817
|
|
|
// future, we might offer this value to users as a config |
|
818
|
|
|
// option. |
|
819
|
30 |
|
if ($curX >= abs($factorX * 7)) { |
|
820
|
20 |
|
$whiteSpace = "\t"; |
|
821
|
29 |
|
} elseif ($curX >= abs($factorX * 2)) { |
|
822
|
17 |
|
$whiteSpace = ' '; |
|
823
|
|
|
} |
|
824
|
|
|
} |
|
825
|
|
|
} |
|
826
|
|
|
|
|
827
|
43 |
|
$newtext = $this->getTJUsingFontFallback( |
|
828
|
43 |
|
$current_font, |
|
829
|
43 |
|
$command[self::COMMAND], |
|
830
|
43 |
|
$page, |
|
831
|
43 |
|
$factorX |
|
832
|
43 |
|
); |
|
833
|
|
|
|
|
834
|
|
|
// If there is no ActualText pending then write |
|
835
|
43 |
|
if (false === $actual_text) { |
|
836
|
43 |
|
$newtext = str_replace(["\r", "\n"], '', $newtext); |
|
837
|
43 |
|
if (false !== $reverse_text) { |
|
838
|
|
|
// If we are in ReversedChars mode, add the whitespace last |
|
839
|
1 |
|
$text[] = preg_replace('/ $/', ' ', $newtext.$whiteSpace); |
|
840
|
|
|
} else { |
|
841
|
|
|
// Otherwise add the whitespace first |
|
842
|
43 |
|
if (' ' === $whiteSpace && isset($text[\count($text) - 1])) { |
|
843
|
16 |
|
$text[\count($text) - 1] = preg_replace('/ $/', '', $text[\count($text) - 1]); |
|
844
|
|
|
} |
|
845
|
43 |
|
$text[] = preg_replace('/^[ \t]{2}/', ' ', $whiteSpace.$newtext); |
|
846
|
|
|
} |
|
847
|
|
|
|
|
848
|
|
|
// Record the position of this inserted text for comparison |
|
849
|
|
|
// with the next text block. |
|
850
|
|
|
// Provide a 'fudge' factor guess on how wide this text block |
|
851
|
|
|
// is based on the number of characters. This helps limit the |
|
852
|
|
|
// number of tabs inserted, but isn't perfect. |
|
853
|
43 |
|
$factor = $factorX / 2; |
|
854
|
43 |
|
$current_position = [ |
|
855
|
43 |
|
'x' => $currentX - mb_strlen($newtext) * $factor, |
|
856
|
43 |
|
'y' => $currentY, |
|
857
|
43 |
|
]; |
|
858
|
4 |
|
} elseif (false === $last_written_position) { |
|
859
|
|
|
// If there is an ActualText in the pipeline |
|
860
|
|
|
// store the position this undisplayed text |
|
861
|
|
|
// *would* have been written to, so the |
|
862
|
|
|
// ActualText is displayed in the right spot |
|
863
|
4 |
|
$last_written_position = [$currentX, $currentY]; |
|
864
|
4 |
|
$current_position['x'] = $currentX; |
|
865
|
|
|
} |
|
866
|
43 |
|
break; |
|
867
|
|
|
|
|
868
|
|
|
// move to start of next line |
|
869
|
43 |
|
case 'T*': |
|
870
|
13 |
|
$current_position['x'] = 0; |
|
871
|
13 |
|
$current_position_td['x'] = 0; |
|
872
|
13 |
|
$current_position_td['y'] += $current_text_leading; |
|
873
|
13 |
|
break; |
|
874
|
|
|
|
|
875
|
|
|
// set character spacing |
|
876
|
43 |
|
case 'Tc': |
|
877
|
13 |
|
break; |
|
878
|
|
|
|
|
879
|
|
|
// move text current point and set leading |
|
880
|
43 |
|
case 'Td': |
|
881
|
43 |
|
case 'TD': |
|
882
|
|
|
// move text current point |
|
883
|
32 |
|
$args = preg_split('/\s+/s', $command[self::COMMAND]); |
|
884
|
32 |
|
$y = (float) array_pop($args); |
|
885
|
32 |
|
$x = (float) array_pop($args); |
|
886
|
|
|
|
|
887
|
32 |
|
if ('TD' == $command[self::OPERATOR]) { |
|
888
|
7 |
|
$current_text_leading = -$y * $current_position_tm['b'] - $y * $current_position_tm['j']; |
|
889
|
|
|
} |
|
890
|
|
|
|
|
891
|
32 |
|
$current_position_td = [ |
|
892
|
32 |
|
'x' => $current_position_td['x'] + $x * $current_position_tm['a'] + $x * $current_position_tm['i'], |
|
893
|
32 |
|
'y' => $current_position_td['y'] + $y * $current_position_tm['b'] + $y * $current_position_tm['j'], |
|
894
|
32 |
|
]; |
|
895
|
32 |
|
break; |
|
896
|
|
|
|
|
897
|
43 |
|
case 'Tf': |
|
898
|
43 |
|
$args = preg_split('/\s/s', $command[self::COMMAND]); |
|
899
|
43 |
|
$size = (float) array_pop($args); |
|
900
|
43 |
|
$id = trim(array_pop($args), '/'); |
|
901
|
43 |
|
if (null !== $page) { |
|
902
|
43 |
|
$new_font = $page->getFont($id); |
|
903
|
|
|
// If an invalid font ID is given, do not update the font. |
|
904
|
|
|
// This should theoretically never happen, as the PDF spec states for the Tf operator: |
|
905
|
|
|
// "The specified font value shall match a resource name in the Font entry of the default resource dictionary" |
|
906
|
|
|
// (https://www.adobe.com/content/dam/acom/en/devnet/pdf/pdfs/PDF32000_2008.pdf, page 435) |
|
907
|
|
|
// But we want to make sure that malformed PDFs do not simply crash. |
|
908
|
43 |
|
if (null !== $new_font) { |
|
909
|
39 |
|
$current_font = $new_font; |
|
910
|
39 |
|
$current_font_size = $size; |
|
911
|
|
|
} |
|
912
|
|
|
} |
|
913
|
43 |
|
break; |
|
914
|
|
|
|
|
915
|
|
|
// set leading |
|
916
|
37 |
|
case 'TL': |
|
917
|
6 |
|
$y = (float) $command[self::COMMAND]; |
|
918
|
6 |
|
$current_text_leading = -$y * $current_position_tm['b'] + -$y * $current_position_tm['j']; |
|
919
|
6 |
|
break; |
|
920
|
|
|
|
|
921
|
|
|
// set text position matrix |
|
922
|
37 |
|
case 'Tm': |
|
923
|
34 |
|
$args = preg_split('/\s+/s', $command[self::COMMAND]); |
|
924
|
34 |
|
$current_position_tm = [ |
|
925
|
34 |
|
'a' => (float) $args[0], 'b' => (float) $args[1], 'c' => 0, |
|
926
|
34 |
|
'i' => (float) $args[2], 'j' => (float) $args[3], 'k' => 0, |
|
927
|
34 |
|
'x' => (float) $args[4], 'y' => (float) $args[5], 'z' => 1, |
|
928
|
34 |
|
]; |
|
929
|
34 |
|
break; |
|
930
|
|
|
|
|
931
|
|
|
// set text rendering mode |
|
932
|
22 |
|
case 'Ts': |
|
933
|
|
|
break; |
|
934
|
|
|
|
|
935
|
|
|
// set super/subscripting text rise |
|
936
|
22 |
|
case 'Ts': |
|
937
|
|
|
break; |
|
938
|
|
|
|
|
939
|
|
|
// set word spacing |
|
940
|
22 |
|
case 'Tw': |
|
941
|
9 |
|
break; |
|
942
|
|
|
|
|
943
|
|
|
// set horizontal scaling |
|
944
|
22 |
|
case 'Tz': |
|
945
|
|
|
break; |
|
946
|
|
|
|
|
947
|
|
|
default: |
|
948
|
|
|
} |
|
949
|
|
|
} |
|
950
|
|
|
} |
|
951
|
|
|
|
|
952
|
46 |
|
$result = array_merge($result, $text); |
|
953
|
|
|
|
|
954
|
46 |
|
return $result; |
|
955
|
|
|
} |
|
956
|
|
|
|
|
957
|
|
|
/** |
|
958
|
|
|
* getCommandsText() expects the content of $text_part to be an |
|
959
|
|
|
* already formatted, single-line command from a document stream. |
|
960
|
|
|
* The companion function getSectionsText() returns a document |
|
961
|
|
|
* stream as an array of single commands for just this purpose. |
|
962
|
|
|
* Because of this, the argument $offset is no longer used, and |
|
963
|
|
|
* may be removed in a future PdfParser release. |
|
964
|
|
|
* |
|
965
|
|
|
* A better name for this function would be getCommandText() |
|
966
|
|
|
* since it now always works on just one command. |
|
967
|
|
|
*/ |
|
968
|
50 |
|
public function getCommandsText(string $text_part, int &$offset = 0): array |
|
|
|
|
|
|
969
|
|
|
{ |
|
970
|
50 |
|
$commands = $matches = []; |
|
971
|
|
|
|
|
972
|
50 |
|
preg_match('/^(([\/\[\(<])?.*)(?<!\w)([a-z01\'\"*]+)$/i', $text_part, $matches); |
|
973
|
|
|
|
|
974
|
|
|
// If no valid command is detected, return an empty array |
|
975
|
50 |
|
if (!isset($matches[1]) || !isset($matches[2]) || !isset($matches[3])) { |
|
976
|
1 |
|
return []; |
|
977
|
|
|
} |
|
978
|
|
|
|
|
979
|
50 |
|
$type = $matches[2]; |
|
980
|
50 |
|
$operator = $matches[3]; |
|
981
|
50 |
|
$command = trim($matches[1]); |
|
982
|
|
|
|
|
983
|
50 |
|
if ('TJ' == $operator) { |
|
984
|
40 |
|
$subcommand = []; |
|
985
|
40 |
|
$command = trim($command, '[]'); |
|
986
|
|
|
do { |
|
987
|
40 |
|
$oldCommand = $command; |
|
988
|
|
|
|
|
989
|
|
|
// Search for parentheses string () format |
|
990
|
40 |
|
if (preg_match('/^ *\((.*?)(?<![^\\\\]\\\\)\) *(-?[\d.]+)?/', $command, $tjmatch)) { |
|
991
|
34 |
|
$subcommand[] = [ |
|
992
|
34 |
|
self::TYPE => '(', |
|
993
|
34 |
|
self::OPERATOR => 'TJ', |
|
994
|
34 |
|
self::COMMAND => $tjmatch[1], |
|
995
|
34 |
|
]; |
|
996
|
34 |
|
if (isset($tjmatch[2]) && trim($tjmatch[2])) { |
|
997
|
28 |
|
$subcommand[] = [ |
|
998
|
28 |
|
self::TYPE => 'n', |
|
999
|
28 |
|
self::OPERATOR => '', |
|
1000
|
28 |
|
self::COMMAND => $tjmatch[2], |
|
1001
|
28 |
|
]; |
|
1002
|
|
|
} |
|
1003
|
34 |
|
$command = substr($command, \strlen($tjmatch[0])); |
|
1004
|
|
|
} |
|
1005
|
|
|
|
|
1006
|
|
|
// Search for hexadecimal <> format |
|
1007
|
40 |
|
if (preg_match('/^ *<([0-9a-f\s]*)> *(-?[\d.]+)?/i', $command, $tjmatch)) { |
|
1008
|
19 |
|
$tjmatch[1] = preg_replace('/\s/', '', $tjmatch[1]); |
|
1009
|
19 |
|
$subcommand[] = [ |
|
1010
|
19 |
|
self::TYPE => '<', |
|
1011
|
19 |
|
self::OPERATOR => 'TJ', |
|
1012
|
19 |
|
self::COMMAND => $tjmatch[1], |
|
1013
|
19 |
|
]; |
|
1014
|
19 |
|
if (isset($tjmatch[2]) && trim($tjmatch[2])) { |
|
1015
|
18 |
|
$subcommand[] = [ |
|
1016
|
18 |
|
self::TYPE => 'n', |
|
1017
|
18 |
|
self::OPERATOR => '', |
|
1018
|
18 |
|
self::COMMAND => $tjmatch[2], |
|
1019
|
18 |
|
]; |
|
1020
|
|
|
} |
|
1021
|
19 |
|
$command = substr($command, \strlen($tjmatch[0])); |
|
1022
|
|
|
} |
|
1023
|
40 |
|
} while ($command != $oldCommand); |
|
1024
|
|
|
|
|
1025
|
40 |
|
$command = $subcommand; |
|
1026
|
50 |
|
} elseif ('Tj' == $operator || "'" == $operator || '"' == $operator) { |
|
1027
|
|
|
// Depending on the string type, trim the data of the |
|
1028
|
|
|
// appropriate delimiters |
|
1029
|
39 |
|
if ('(' == $type) { |
|
1030
|
|
|
// Don't use trim() here since a () string may end with |
|
1031
|
|
|
// a balanced or escaped right parentheses, and trim() |
|
1032
|
|
|
// will delete both. Both strings below are valid: |
|
1033
|
|
|
// eg. (String()) |
|
1034
|
|
|
// eg. (String\)) |
|
1035
|
33 |
|
$command = preg_replace('/^\(|\)$/', '', $command); |
|
1036
|
15 |
|
} elseif ('<' == $type) { |
|
1037
|
39 |
|
$command = trim($command, '<>'); |
|
1038
|
|
|
} |
|
1039
|
50 |
|
} elseif ('/' == $type) { |
|
1040
|
49 |
|
$command = substr($command, 1); |
|
1041
|
|
|
} |
|
1042
|
|
|
|
|
1043
|
50 |
|
$commands[] = [ |
|
1044
|
50 |
|
self::TYPE => $type, |
|
1045
|
50 |
|
self::OPERATOR => $operator, |
|
1046
|
50 |
|
self::COMMAND => $command, |
|
1047
|
50 |
|
]; |
|
1048
|
|
|
|
|
1049
|
50 |
|
return $commands; |
|
1050
|
|
|
} |
|
1051
|
|
|
|
|
1052
|
65 |
|
public static function factory( |
|
1053
|
|
|
Document $document, |
|
1054
|
|
|
Header $header, |
|
1055
|
|
|
?string $content, |
|
1056
|
|
|
?Config $config = null |
|
1057
|
|
|
): self { |
|
1058
|
65 |
|
switch ($header->get('Type')->getContent()) { |
|
1059
|
65 |
|
case 'XObject': |
|
1060
|
19 |
|
switch ($header->get('Subtype')->getContent()) { |
|
1061
|
19 |
|
case 'Image': |
|
1062
|
12 |
|
return new Image($document, $header, $config->getRetainImageContent() ? $content : null, $config); |
|
|
|
|
|
|
1063
|
|
|
|
|
1064
|
8 |
|
case 'Form': |
|
1065
|
8 |
|
return new Form($document, $header, $content, $config); |
|
1066
|
|
|
} |
|
1067
|
|
|
|
|
1068
|
|
|
return new self($document, $header, $content, $config); |
|
1069
|
|
|
|
|
1070
|
65 |
|
case 'Pages': |
|
1071
|
64 |
|
return new Pages($document, $header, $content, $config); |
|
1072
|
|
|
|
|
1073
|
65 |
|
case 'Page': |
|
1074
|
64 |
|
return new Page($document, $header, $content, $config); |
|
1075
|
|
|
|
|
1076
|
65 |
|
case 'Encoding': |
|
1077
|
12 |
|
return new Encoding($document, $header, $content, $config); |
|
1078
|
|
|
|
|
1079
|
65 |
|
case 'Font': |
|
1080
|
64 |
|
$subtype = $header->get('Subtype')->getContent(); |
|
1081
|
64 |
|
$classname = '\Smalot\PdfParser\Font\Font'.$subtype; |
|
1082
|
|
|
|
|
1083
|
64 |
|
if (class_exists($classname)) { |
|
1084
|
64 |
|
return new $classname($document, $header, $content, $config); |
|
1085
|
|
|
} |
|
1086
|
|
|
|
|
1087
|
|
|
return new Font($document, $header, $content, $config); |
|
1088
|
|
|
|
|
1089
|
|
|
default: |
|
1090
|
65 |
|
return new self($document, $header, $content, $config); |
|
1091
|
|
|
} |
|
1092
|
|
|
} |
|
1093
|
|
|
|
|
1094
|
|
|
/** |
|
1095
|
|
|
* Returns unique id identifying the object. |
|
1096
|
|
|
*/ |
|
1097
|
46 |
|
protected function getUniqueId(): string |
|
1098
|
|
|
{ |
|
1099
|
46 |
|
return spl_object_hash($this); |
|
1100
|
|
|
} |
|
1101
|
|
|
} |
|
1102
|
|
|
|
smalot /
pdfparser
Loading history...
