|
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\Element\ElementArray; |
|
34
|
|
|
use Smalot\PdfParser\Element\ElementMissing; |
|
35
|
|
|
use Smalot\PdfParser\Element\ElementNull; |
|
36
|
|
|
use Smalot\PdfParser\Element\ElementXRef; |
|
37
|
|
|
|
|
38
|
|
|
class Page extends PDFObject |
|
39
|
|
|
{ |
|
40
|
|
|
/** |
|
41
|
|
|
* @var Font[] |
|
42
|
|
|
*/ |
|
43
|
|
|
protected $fonts = null; |
|
44
|
|
|
|
|
45
|
|
|
/** |
|
46
|
|
|
* @var PDFObject[] |
|
47
|
|
|
*/ |
|
48
|
|
|
protected $xobjects = null; |
|
49
|
|
|
|
|
50
|
|
|
/** |
|
51
|
|
|
* @var array |
|
52
|
|
|
*/ |
|
53
|
|
|
protected $dataTm = null; |
|
54
|
|
|
|
|
55
|
|
|
/** |
|
56
|
|
|
* @return Font[] |
|
57
|
|
|
*/ |
|
58
|
|
|
public function getFonts() |
|
59
|
|
|
{ |
|
60
|
|
|
if (null !== $this->fonts) { |
|
61
|
|
|
return $this->fonts; |
|
62
|
|
|
} |
|
63
|
|
|
|
|
64
|
|
|
$resources = $this->get('Resources'); |
|
65
|
|
|
|
|
66
|
|
|
if (method_exists($resources, 'has') && $resources->has('Font')) { |
|
67
|
|
|
if ($resources->get('Font') instanceof ElementMissing) { |
|
68
|
|
|
return []; |
|
69
|
|
|
} |
|
70
|
|
|
|
|
71
|
|
|
if ($resources->get('Font') instanceof Header) { |
|
72
|
|
|
$fonts = $resources->get('Font')->getElements(); |
|
73
|
|
|
} else { |
|
74
|
|
|
$fonts = $resources->get('Font')->getHeader()->getElements(); |
|
75
|
|
|
} |
|
76
|
|
|
|
|
77
|
|
|
$table = []; |
|
78
|
|
|
|
|
79
|
|
|
foreach ($fonts as $id => $font) { |
|
80
|
|
|
if ($font instanceof Font) { |
|
81
|
|
|
$table[$id] = $font; |
|
82
|
|
|
|
|
83
|
|
|
// Store too on cleaned id value (only numeric) |
|
84
|
|
|
$id = preg_replace('/[^0-9\.\-_]/', '', $id); |
|
85
|
|
|
if ('' != $id) { |
|
86
|
|
|
$table[$id] = $font; |
|
87
|
|
|
} |
|
88
|
|
|
} |
|
89
|
|
|
} |
|
90
|
|
|
|
|
91
|
|
|
return $this->fonts = $table; |
|
92
|
|
|
} |
|
93
|
|
|
|
|
94
|
|
|
return []; |
|
95
|
|
|
} |
|
96
|
|
|
|
|
97
|
|
|
public function getFont(string $id): ?Font |
|
98
|
|
|
{ |
|
99
|
|
|
$fonts = $this->getFonts(); |
|
100
|
|
|
|
|
101
|
|
|
if (isset($fonts[$id])) { |
|
102
|
|
|
return $fonts[$id]; |
|
103
|
|
|
} |
|
104
|
|
|
|
|
105
|
|
|
// According to the PDF specs (https://www.adobe.com/content/dam/acom/en/devnet/pdf/pdfs/PDF32000_2008.pdf, page 238) |
|
106
|
|
|
// "The font resource name presented to the Tf operator is arbitrary, as are the names for all kinds of resources" |
|
107
|
|
|
// Instead, we search for the unfiltered name first and then do this cleaning as a fallback, so all tests still pass. |
|
108
|
|
|
|
|
109
|
|
|
if (isset($fonts[$id])) { |
|
110
|
|
|
return $fonts[$id]; |
|
111
|
|
|
} else { |
|
112
|
|
|
$id = preg_replace('/[^0-9\.\-_]/', '', $id); |
|
113
|
|
|
if (isset($fonts[$id])) { |
|
114
|
|
|
return $fonts[$id]; |
|
115
|
|
|
} |
|
116
|
|
|
} |
|
117
|
|
|
|
|
118
|
|
|
return null; |
|
119
|
|
|
} |
|
120
|
|
|
|
|
121
|
|
|
/** |
|
122
|
|
|
* Support for XObject |
|
123
|
|
|
* |
|
124
|
|
|
* @return PDFObject[] |
|
125
|
|
|
*/ |
|
126
|
|
|
public function getXObjects() |
|
127
|
|
|
{ |
|
128
|
|
|
if (null !== $this->xobjects) { |
|
129
|
|
|
return $this->xobjects; |
|
130
|
|
|
} |
|
131
|
|
|
|
|
132
|
|
|
$resources = $this->get('Resources'); |
|
133
|
|
|
|
|
134
|
|
|
if (method_exists($resources, 'has') && $resources->has('XObject')) { |
|
135
|
|
|
if ($resources->get('XObject') instanceof Header) { |
|
136
|
|
|
$xobjects = $resources->get('XObject')->getElements(); |
|
137
|
|
|
} else { |
|
138
|
|
|
$xobjects = $resources->get('XObject')->getHeader()->getElements(); |
|
139
|
|
|
} |
|
140
|
|
|
|
|
141
|
|
|
$table = []; |
|
142
|
|
|
|
|
143
|
|
|
foreach ($xobjects as $id => $xobject) { |
|
144
|
|
|
$table[$id] = $xobject; |
|
145
|
|
|
|
|
146
|
|
|
// Store too on cleaned id value (only numeric) |
|
147
|
|
|
$id = preg_replace('/[^0-9\.\-_]/', '', $id); |
|
148
|
|
|
if ('' != $id) { |
|
149
|
|
|
$table[$id] = $xobject; |
|
150
|
|
|
} |
|
151
|
|
|
} |
|
152
|
|
|
|
|
153
|
|
|
return $this->xobjects = $table; |
|
154
|
|
|
} |
|
155
|
|
|
|
|
156
|
|
|
return []; |
|
157
|
|
|
} |
|
158
|
|
|
|
|
159
|
|
|
public function getXObject(string $id): ?PDFObject |
|
160
|
|
|
{ |
|
161
|
|
|
$xobjects = $this->getXObjects(); |
|
162
|
|
|
|
|
163
|
|
|
if (isset($xobjects[$id])) { |
|
164
|
|
|
return $xobjects[$id]; |
|
165
|
|
|
} |
|
166
|
|
|
|
|
167
|
|
|
return null; |
|
168
|
|
|
/*$id = preg_replace('/[^0-9\.\-_]/', '', $id); |
|
169
|
|
|
|
|
170
|
|
|
if (isset($xobjects[$id])) { |
|
171
|
|
|
return $xobjects[$id]; |
|
172
|
|
|
} else { |
|
173
|
|
|
return null; |
|
174
|
|
|
}*/ |
|
175
|
|
|
} |
|
176
|
|
|
|
|
177
|
|
|
public function getText(self $page = null): string |
|
178
|
|
|
{ |
|
179
|
|
|
if ($contents = $this->get('Contents')) { |
|
180
|
|
|
if ($contents instanceof ElementMissing) { |
|
181
|
|
|
return ''; |
|
182
|
|
|
} elseif ($contents instanceof ElementNull) { |
|
183
|
|
|
return ''; |
|
184
|
|
|
} elseif ($contents instanceof PDFObject) { |
|
185
|
|
|
$elements = $contents->getHeader()->getElements(); |
|
186
|
|
|
|
|
187
|
|
|
if (is_numeric(key($elements))) { |
|
188
|
|
|
$new_content = ''; |
|
189
|
|
|
|
|
190
|
|
|
foreach ($elements as $element) { |
|
191
|
|
|
if ($element instanceof ElementXRef) { |
|
192
|
|
|
$new_content .= $element->getObject()->getContent(); |
|
193
|
|
|
} else { |
|
194
|
|
|
$new_content .= $element->getContent(); |
|
195
|
|
|
} |
|
196
|
|
|
} |
|
197
|
|
|
|
|
198
|
|
|
$header = new Header([], $this->document); |
|
199
|
|
|
$contents = new PDFObject($this->document, $header, $new_content, $this->config); |
|
200
|
|
|
} |
|
201
|
|
|
} elseif ($contents instanceof ElementArray) { |
|
202
|
|
|
// Create a virtual global content. |
|
203
|
|
|
$new_content = ''; |
|
204
|
|
|
|
|
205
|
|
|
foreach ($contents->getContent() as $content) { |
|
206
|
|
|
$new_content .= $content->getContent()."\n"; |
|
207
|
|
|
} |
|
208
|
|
|
|
|
209
|
|
|
$header = new Header([], $this->document); |
|
210
|
|
|
$contents = new PDFObject($this->document, $header, $new_content, $this->config); |
|
211
|
|
|
} |
|
212
|
|
|
|
|
213
|
|
|
/* |
|
214
|
|
|
* Elements referencing each other on the same page can cause endless loops during text parsing. |
|
215
|
|
|
* To combat this we keep a recursionStack containing already parsed elements on the page. |
|
216
|
|
|
* The stack is only emptied here after getting text from a page. |
|
217
|
|
|
*/ |
|
218
|
|
|
$contentsText = $contents->getText($this); |
|
219
|
|
|
PDFObject::$recursionStack = []; |
|
220
|
|
|
|
|
221
|
|
|
return $contentsText; |
|
222
|
|
|
} |
|
223
|
|
|
|
|
224
|
|
|
return ''; |
|
225
|
|
|
} |
|
226
|
|
|
|
|
227
|
|
|
/** |
|
228
|
|
|
* Return true if the current page is a (setasign\Fpdi\Fpdi) FPDI/FPDF document |
|
229
|
|
|
* |
|
230
|
|
|
* The metadata 'Producer' should have the value of "FPDF" . FPDF_VERSION if the |
|
231
|
|
|
* pdf file was generated by FPDF/Fpfi. |
|
232
|
|
|
* |
|
233
|
|
|
* @return bool true is the current page is a FPDI/FPDF document |
|
234
|
|
|
*/ |
|
235
|
|
|
public function isFpdf(): bool |
|
236
|
|
|
{ |
|
237
|
|
|
if (\array_key_exists('Producer', $this->document->getDetails()) && |
|
238
|
|
|
\is_string($this->document->getDetails()['Producer']) && |
|
239
|
|
|
0 === strncmp($this->document->getDetails()['Producer'], 'FPDF', \strlen('FPDF')) { |
|
240
|
|
|
return true; |
|
|
|
|
|
|
241
|
|
|
} |
|
242
|
|
|
|
|
243
|
|
|
return false; |
|
244
|
|
|
} |
|
245
|
|
|
|
|
246
|
|
|
/** |
|
247
|
|
|
* Return the page number of the PDF document of the page object |
|
248
|
|
|
* |
|
249
|
|
|
* @return int the page number |
|
250
|
|
|
*/ |
|
251
|
|
|
public function getPageNumber(): int |
|
252
|
|
|
{ |
|
253
|
|
|
$pages = $this->document->getPages(); |
|
254
|
|
|
$numOfPages = \count($pages); |
|
255
|
|
|
for ($pageNum = 0; $pageNum < $numOfPages; ++$pageNum) { |
|
256
|
|
|
if ($pages[$pageNum] === $this) { |
|
257
|
|
|
break; |
|
258
|
|
|
} |
|
259
|
|
|
} |
|
260
|
|
|
|
|
261
|
|
|
return $pageNum; |
|
262
|
|
|
} |
|
263
|
|
|
|
|
264
|
|
|
/** |
|
265
|
|
|
* Return the Object of the page if the document is a FPDF/FPDI document |
|
266
|
|
|
* |
|
267
|
|
|
* If the document was generated by FPDF/FPDI it returns the |
|
268
|
|
|
* PDFObject of the given page |
|
269
|
|
|
* |
|
270
|
|
|
* @return PDFObject The PDFObject for the page |
|
271
|
|
|
*/ |
|
272
|
|
|
public function getPDFObjectForFpdf(): PDFObject |
|
273
|
|
|
{ |
|
274
|
|
|
$pageNum = $this->getPageNumber(); |
|
275
|
|
|
$xObjects = $this->getXObjects(); |
|
276
|
|
|
|
|
277
|
|
|
return $xObjects[$pageNum]; |
|
278
|
|
|
} |
|
279
|
|
|
|
|
280
|
|
|
/** |
|
281
|
|
|
* Return a new PDFObject of the document created with FPDF/FPDI |
|
282
|
|
|
* |
|
283
|
|
|
* For a document generated by FPDF/FPDI, it generates a |
|
284
|
|
|
* new PDFObject for that document |
|
285
|
|
|
* |
|
286
|
|
|
* @return PDFObject The PDFObject |
|
287
|
|
|
*/ |
|
288
|
|
|
public function createPDFObjectForFpdf(): PDFObject |
|
289
|
|
|
{ |
|
290
|
|
|
$pdfObject = $this->getPDFObjectForFpdf(); |
|
291
|
|
|
$new_content = $pdfObject->getContent(); |
|
292
|
|
|
$header = $pdfObject->getHeader(); |
|
293
|
|
|
$config = $pdfObject->config; |
|
294
|
|
|
|
|
295
|
|
|
return new PDFObject($pdfObject->document, $header, $new_content, $config); |
|
296
|
|
|
} |
|
297
|
|
|
|
|
298
|
|
|
/** |
|
299
|
|
|
* Return page if document is a FPDF/FPDI document |
|
300
|
|
|
* |
|
301
|
|
|
* @return Page The page |
|
302
|
|
|
*/ |
|
303
|
|
|
public function createPageForFpdf(): self |
|
304
|
|
|
{ |
|
305
|
|
|
$pdfObject = $this->getPDFObjectForFpdf(); |
|
306
|
|
|
$new_content = $pdfObject->getContent(); |
|
307
|
|
|
$header = $pdfObject->getHeader(); |
|
308
|
|
|
$config = $pdfObject->config; |
|
309
|
|
|
|
|
310
|
|
|
return new self($pdfObject->document, $header, $new_content, $config); |
|
311
|
|
|
} |
|
312
|
|
|
|
|
313
|
|
|
public function getTextArray(self $page = null): array |
|
314
|
|
|
{ |
|
315
|
|
|
if ($this->isFpdf()) { |
|
316
|
|
|
$pdfObject = $this->getPDFObjectForFpdf(); |
|
317
|
|
|
$newPdfObject = $this->createPDFObjectForFpdf(); |
|
318
|
|
|
|
|
319
|
|
|
return $newPdfObject->getTextArray($pdfObject); |
|
320
|
|
|
} else { |
|
321
|
|
|
if ($contents = $this->get('Contents')) { |
|
322
|
|
|
if ($contents instanceof ElementMissing) { |
|
323
|
|
|
return []; |
|
324
|
|
|
} elseif ($contents instanceof ElementNull) { |
|
325
|
|
|
return []; |
|
326
|
|
|
} elseif ($contents instanceof PDFObject) { |
|
327
|
|
|
$elements = $contents->getHeader()->getElements(); |
|
328
|
|
|
|
|
329
|
|
|
if (is_numeric(key($elements))) { |
|
330
|
|
|
$new_content = ''; |
|
331
|
|
|
|
|
332
|
|
|
/** @var PDFObject $element */ |
|
333
|
|
|
foreach ($elements as $element) { |
|
334
|
|
|
if ($element instanceof ElementXRef) { |
|
335
|
|
|
$new_content .= $element->getObject()->getContent(); |
|
336
|
|
|
} else { |
|
337
|
|
|
$new_content .= $element->getContent(); |
|
338
|
|
|
} |
|
339
|
|
|
} |
|
340
|
|
|
|
|
341
|
|
|
$header = new Header([], $this->document); |
|
342
|
|
|
$contents = new PDFObject($this->document, $header, $new_content, $this->config); |
|
343
|
|
|
} else { |
|
344
|
|
|
try { |
|
345
|
|
|
$contents->getTextArray($this); |
|
346
|
|
|
} catch (\Throwable $e) { |
|
347
|
|
|
return $contents->getTextArray(); |
|
348
|
|
|
} |
|
349
|
|
|
} |
|
350
|
|
|
} elseif ($contents instanceof ElementArray) { |
|
351
|
|
|
// Create a virtual global content. |
|
352
|
|
|
$new_content = ''; |
|
353
|
|
|
|
|
354
|
|
|
/** @var PDFObject $content */ |
|
355
|
|
|
foreach ($contents->getContent() as $content) { |
|
356
|
|
|
$new_content .= $content->getContent()."\n"; |
|
357
|
|
|
} |
|
358
|
|
|
|
|
359
|
|
|
$header = new Header([], $this->document); |
|
360
|
|
|
$contents = new PDFObject($this->document, $header, $new_content, $this->config); |
|
361
|
|
|
} |
|
362
|
|
|
|
|
363
|
|
|
return $contents->getTextArray($this); |
|
364
|
|
|
} |
|
365
|
|
|
|
|
366
|
|
|
return []; |
|
367
|
|
|
} |
|
368
|
|
|
} |
|
369
|
|
|
|
|
370
|
|
|
/** |
|
371
|
|
|
* Gets all the text data with its internal representation of the page. |
|
372
|
|
|
* |
|
373
|
|
|
* Returns an array with the data and the internal representation |
|
374
|
|
|
*/ |
|
375
|
|
|
public function extractRawData(): array |
|
376
|
|
|
{ |
|
377
|
|
|
/* |
|
378
|
|
|
* Now you can get the complete content of the object with the text on it |
|
379
|
|
|
*/ |
|
380
|
|
|
$extractedData = []; |
|
381
|
|
|
$content = $this->get('Contents'); |
|
382
|
|
|
$values = $content->getContent(); |
|
383
|
|
|
if (isset($values) && \is_array($values)) { |
|
384
|
|
|
$text = ''; |
|
385
|
|
|
foreach ($values as $section) { |
|
386
|
|
|
$text .= $section->getContent(); |
|
387
|
|
|
} |
|
388
|
|
|
$sectionsText = $this->getSectionsText($text); |
|
389
|
|
|
foreach ($sectionsText as $sectionText) { |
|
390
|
|
|
$commandsText = $this->getCommandsText($sectionText); |
|
391
|
|
|
foreach ($commandsText as $command) { |
|
392
|
|
|
$extractedData[] = $command; |
|
393
|
|
|
} |
|
394
|
|
|
} |
|
395
|
|
|
} else { |
|
396
|
|
|
if ($this->isFpdf()) { |
|
397
|
|
|
$content = $this->getPDFObjectForFpdf(); |
|
398
|
|
|
} |
|
399
|
|
|
$sectionsText = $content->getSectionsText($content->getContent()); |
|
400
|
|
|
foreach ($sectionsText as $sectionText) { |
|
401
|
|
|
$extractedData[] = ['t' => '', 'o' => 'BT', 'c' => '']; |
|
402
|
|
|
|
|
403
|
|
|
$commandsText = $content->getCommandsText($sectionText); |
|
404
|
|
|
foreach ($commandsText as $command) { |
|
405
|
|
|
$extractedData[] = $command; |
|
406
|
|
|
} |
|
407
|
|
|
} |
|
408
|
|
|
} |
|
409
|
|
|
|
|
410
|
|
|
return $extractedData; |
|
411
|
|
|
} |
|
412
|
|
|
|
|
413
|
|
|
/** |
|
414
|
|
|
* Gets all the decoded text data with it internal representation from a page. |
|
415
|
|
|
* |
|
416
|
|
|
* @param array $extractedRawData the extracted data return by extractRawData or |
|
417
|
|
|
* null if extractRawData should be called |
|
418
|
|
|
* |
|
419
|
|
|
* @return array An array with the data and the internal representation |
|
420
|
|
|
*/ |
|
421
|
|
|
public function extractDecodedRawData(array $extractedRawData = null): array |
|
422
|
|
|
{ |
|
423
|
|
|
if (!isset($extractedRawData) || !$extractedRawData) { |
|
424
|
|
|
$extractedRawData = $this->extractRawData(); |
|
425
|
|
|
} |
|
426
|
|
|
$currentFont = null; /** @var Font $currentFont */ |
|
427
|
|
|
$clippedFont = null; |
|
428
|
|
|
$fpdfPage = null; |
|
429
|
|
|
if ($this->isFpdf()) { |
|
430
|
|
|
$fpdfPage = $this->createPageForFpdf(); |
|
431
|
|
|
} |
|
432
|
|
|
foreach ($extractedRawData as &$command) { |
|
433
|
|
|
if ('Tj' == $command['o'] || 'TJ' == $command['o']) { |
|
434
|
|
|
$data = $command['c']; |
|
435
|
|
|
if (!\is_array($data)) { |
|
436
|
|
|
$tmpText = ''; |
|
437
|
|
|
if (isset($currentFont)) { |
|
438
|
|
|
$tmpText = $currentFont->decodeOctal($data); |
|
439
|
|
|
//$tmpText = $currentFont->decodeHexadecimal($tmpText, false); |
|
440
|
|
|
} |
|
441
|
|
|
$tmpText = str_replace( |
|
442
|
|
|
['\\\\', '\(', '\)', '\n', '\r', '\t', '\ '], |
|
443
|
|
|
['\\', '(', ')', "\n", "\r", "\t", ' '], |
|
444
|
|
|
$tmpText |
|
445
|
|
|
); |
|
446
|
|
|
$tmpText = utf8_encode($tmpText); |
|
447
|
|
|
if (isset($currentFont)) { |
|
448
|
|
|
$tmpText = $currentFont->decodeContent($tmpText); |
|
449
|
|
|
} |
|
450
|
|
|
$command['c'] = $tmpText; |
|
451
|
|
|
continue; |
|
452
|
|
|
} |
|
453
|
|
|
$numText = \count($data); |
|
454
|
|
|
for ($i = 0; $i < $numText; ++$i) { |
|
455
|
|
|
if (0 != ($i % 2)) { |
|
456
|
|
|
continue; |
|
457
|
|
|
} |
|
458
|
|
|
$tmpText = $data[$i]['c']; |
|
459
|
|
|
$decodedText = isset($currentFont) ? $currentFont->decodeOctal($tmpText) : $tmpText; |
|
460
|
|
|
$decodedText = str_replace( |
|
461
|
|
|
['\\\\', '\(', '\)', '\n', '\r', '\t', '\ '], |
|
462
|
|
|
['\\', '(', ')', "\n", "\r", "\t", ' '], |
|
463
|
|
|
$decodedText |
|
464
|
|
|
); |
|
465
|
|
|
$decodedText = utf8_encode($decodedText); |
|
466
|
|
|
if (isset($currentFont)) { |
|
467
|
|
|
$decodedText = $currentFont->decodeContent($decodedText); |
|
468
|
|
|
} |
|
469
|
|
|
$command['c'][$i]['c'] = $decodedText; |
|
470
|
|
|
continue; |
|
471
|
|
|
} |
|
472
|
|
|
} elseif ('Tf' == $command['o'] || 'TF' == $command['o']) { |
|
473
|
|
|
$fontId = explode(' ', $command['c'])[0]; |
|
474
|
|
|
// If document is a FPDI/FPDF the $page has the correct font |
|
475
|
|
|
$currentFont = isset($fpdfPage) ? $fpdfPage->getFont($fontId) : $this->getFont($fontId); |
|
476
|
|
|
continue; |
|
477
|
|
|
} elseif ('Q' == $command['o']) { |
|
478
|
|
|
$currentFont = $clippedFont; |
|
479
|
|
|
} elseif ('q' == $command['o']) { |
|
480
|
|
|
$clippedFont = $currentFont; |
|
481
|
|
|
} |
|
482
|
|
|
} |
|
483
|
|
|
|
|
484
|
|
|
return $extractedRawData; |
|
485
|
|
|
} |
|
486
|
|
|
|
|
487
|
|
|
/** |
|
488
|
|
|
* Gets just the Text commands that are involved in text positions and |
|
489
|
|
|
* Text Matrix (Tm) |
|
490
|
|
|
* |
|
491
|
|
|
* It extract just the PDF commands that are involved with text positions, and |
|
492
|
|
|
* the Text Matrix (Tm). These are: BT, ET, TL, Td, TD, Tm, T*, Tj, ', ", and TJ |
|
493
|
|
|
* |
|
494
|
|
|
* @param array $extractedDecodedRawData The data extracted by extractDecodeRawData. |
|
495
|
|
|
* If it is null, the method extractDecodeRawData is called. |
|
496
|
|
|
* |
|
497
|
|
|
* @return array An array with the text command of the page |
|
498
|
|
|
*/ |
|
499
|
|
|
public function getDataCommands(array $extractedDecodedRawData = null): array |
|
500
|
|
|
{ |
|
501
|
|
|
if (!isset($extractedDecodedRawData) || !$extractedDecodedRawData) { |
|
502
|
|
|
$extractedDecodedRawData = $this->extractDecodedRawData(); |
|
503
|
|
|
} |
|
504
|
|
|
$extractedData = []; |
|
505
|
|
|
foreach ($extractedDecodedRawData as $command) { |
|
506
|
|
|
switch ($command['o']) { |
|
507
|
|
|
/* |
|
508
|
|
|
* BT |
|
509
|
|
|
* Begin a text object, inicializind the Tm and Tlm to identity matrix |
|
510
|
|
|
*/ |
|
511
|
|
|
case 'BT': |
|
512
|
|
|
$extractedData[] = $command; |
|
513
|
|
|
break; |
|
514
|
|
|
|
|
515
|
|
|
/* |
|
516
|
|
|
* ET |
|
517
|
|
|
* End a text object, discarding the text matrix |
|
518
|
|
|
*/ |
|
519
|
|
|
case 'ET': |
|
520
|
|
|
$extractedData[] = $command; |
|
521
|
|
|
break; |
|
522
|
|
|
|
|
523
|
|
|
/* |
|
524
|
|
|
* leading TL |
|
525
|
|
|
* Set the text leading, Tl, to leading. Tl is used by the T*, ' and " operators. |
|
526
|
|
|
* Initial value: 0 |
|
527
|
|
|
*/ |
|
528
|
|
|
case 'TL': |
|
529
|
|
|
$extractedData[] = $command; |
|
530
|
|
|
break; |
|
531
|
|
|
|
|
532
|
|
|
/* |
|
533
|
|
|
* tx ty Td |
|
534
|
|
|
* Move to the start of the next line, offset form the start of the |
|
535
|
|
|
* current line by tx, ty. |
|
536
|
|
|
*/ |
|
537
|
|
|
case 'Td': |
|
538
|
|
|
$extractedData[] = $command; |
|
539
|
|
|
break; |
|
540
|
|
|
|
|
541
|
|
|
/* |
|
542
|
|
|
* tx ty TD |
|
543
|
|
|
* Move to the start of the next line, offset form the start of the |
|
544
|
|
|
* current line by tx, ty. As a side effect, this operator set the leading |
|
545
|
|
|
* parameter in the text state. This operator has the same effect as the |
|
546
|
|
|
* code: |
|
547
|
|
|
* -ty TL |
|
548
|
|
|
* tx ty Td |
|
549
|
|
|
*/ |
|
550
|
|
|
case 'TD': |
|
551
|
|
|
$extractedData[] = $command; |
|
552
|
|
|
break; |
|
553
|
|
|
|
|
554
|
|
|
/* |
|
555
|
|
|
* a b c d e f Tm |
|
556
|
|
|
* Set the text matrix, Tm, and the text line matrix, Tlm. The operands are |
|
557
|
|
|
* all numbers, and the initial value for Tm and Tlm is the identity matrix |
|
558
|
|
|
* [1 0 0 1 0 0] |
|
559
|
|
|
*/ |
|
560
|
|
|
case 'Tm': |
|
561
|
|
|
$extractedData[] = $command; |
|
562
|
|
|
break; |
|
563
|
|
|
|
|
564
|
|
|
/* |
|
565
|
|
|
* T* |
|
566
|
|
|
* Move to the start of the next line. This operator has the same effect |
|
567
|
|
|
* as the code: |
|
568
|
|
|
* 0 Tl Td |
|
569
|
|
|
* Where Tl is the current leading parameter in the text state. |
|
570
|
|
|
*/ |
|
571
|
|
|
case 'T*': |
|
572
|
|
|
$extractedData[] = $command; |
|
573
|
|
|
break; |
|
574
|
|
|
|
|
575
|
|
|
/* |
|
576
|
|
|
* string Tj |
|
577
|
|
|
* Show a Text String |
|
578
|
|
|
*/ |
|
579
|
|
|
case 'Tj': |
|
580
|
|
|
$extractedData[] = $command; |
|
581
|
|
|
break; |
|
582
|
|
|
|
|
583
|
|
|
/* |
|
584
|
|
|
* string ' |
|
585
|
|
|
* Move to the next line and show a text string. This operator has the |
|
586
|
|
|
* same effect as the code: |
|
587
|
|
|
* T* |
|
588
|
|
|
* string Tj |
|
589
|
|
|
*/ |
|
590
|
|
|
case "'": |
|
591
|
|
|
$extractedData[] = $command; |
|
592
|
|
|
break; |
|
593
|
|
|
|
|
594
|
|
|
/* |
|
595
|
|
|
* aw ac string " |
|
596
|
|
|
* Move to the next lkine and show a text string, using aw as the word |
|
597
|
|
|
* spacing and ac as the character spacing. This operator has the same |
|
598
|
|
|
* effect as the code: |
|
599
|
|
|
* aw Tw |
|
600
|
|
|
* ac Tc |
|
601
|
|
|
* string ' |
|
602
|
|
|
* Tw set the word spacing, Tw, to wordSpace. |
|
603
|
|
|
* Tc Set the character spacing, Tc, to charsSpace. |
|
604
|
|
|
*/ |
|
605
|
|
|
case '"': |
|
606
|
|
|
$extractedData[] = $command; |
|
607
|
|
|
break; |
|
608
|
|
|
|
|
609
|
|
|
/* |
|
610
|
|
|
* array TJ |
|
611
|
|
|
* Show one or more text strings allow individual glyph positioning. |
|
612
|
|
|
* Each lement of array con be a string or a number. If the element is |
|
613
|
|
|
* a string, this operator shows the string. If it is a number, the |
|
614
|
|
|
* operator adjust the text position by that amount; that is, it translates |
|
615
|
|
|
* the text matrix, Tm. This amount is substracted form the current |
|
616
|
|
|
* horizontal or vertical coordinate, depending on the writing mode. |
|
617
|
|
|
* in the default coordinate system, a positive adjustment has the effect |
|
618
|
|
|
* of moving the next glyph painted either to the left or down by the given |
|
619
|
|
|
* amount. |
|
620
|
|
|
*/ |
|
621
|
|
|
case 'TJ': |
|
622
|
|
|
$extractedData[] = $command; |
|
623
|
|
|
break; |
|
624
|
|
|
default: |
|
625
|
|
|
} |
|
626
|
|
|
} |
|
627
|
|
|
|
|
628
|
|
|
return $extractedData; |
|
629
|
|
|
} |
|
630
|
|
|
|
|
631
|
|
|
/** |
|
632
|
|
|
* Gets the Text Matrix of the text in the page |
|
633
|
|
|
* |
|
634
|
|
|
* Return an array where every item is an array where the first item is the |
|
635
|
|
|
* Text Matrix (Tm) and the second is a string with the text data. The Text matrix |
|
636
|
|
|
* is an array of 6 numbers. The last 2 numbers are the coordinates X and Y of the |
|
637
|
|
|
* text. The first 4 numbers has to be with Scalation, Rotation and Skew of the text. |
|
638
|
|
|
* |
|
639
|
|
|
* @param array $dataCommands the data extracted by getDataCommands |
|
640
|
|
|
* if null getDataCommands is called |
|
641
|
|
|
* |
|
642
|
|
|
* @return array an array with the data of the page including the Tm information |
|
643
|
|
|
* of any text in the page |
|
644
|
|
|
*/ |
|
645
|
|
|
public function getDataTm(array $dataCommands = null): array |
|
646
|
|
|
{ |
|
647
|
|
|
if (!isset($dataCommands) || !$dataCommands) { |
|
648
|
|
|
$dataCommands = $this->getDataCommands(); |
|
649
|
|
|
} |
|
650
|
|
|
|
|
651
|
|
|
/* |
|
652
|
|
|
* At the beginning of a text object Tm is the identity matrix |
|
653
|
|
|
*/ |
|
654
|
|
|
$defaultTm = ['1', '0', '0', '1', '0', '0']; |
|
655
|
|
|
|
|
656
|
|
|
/* |
|
657
|
|
|
* Set the text leading used by T*, ' and " operators |
|
658
|
|
|
*/ |
|
659
|
|
|
$defaultTl = 0; |
|
660
|
|
|
|
|
661
|
|
|
/* |
|
662
|
|
|
* Setting where are the X and Y coordinates in the matrix (Tm) |
|
663
|
|
|
*/ |
|
664
|
|
|
$x = 4; |
|
665
|
|
|
$y = 5; |
|
666
|
|
|
$Tx = 0; |
|
667
|
|
|
$Ty = 0; |
|
668
|
|
|
|
|
669
|
|
|
$Tm = $defaultTm; |
|
670
|
|
|
$Tl = $defaultTl; |
|
671
|
|
|
|
|
672
|
|
|
$extractedTexts = $this->getTextArray(); |
|
673
|
|
|
$extractedData = []; |
|
674
|
|
|
foreach ($dataCommands as $command) { |
|
675
|
|
|
$currentText = $extractedTexts[\count($extractedData)]; |
|
676
|
|
|
switch ($command['o']) { |
|
677
|
|
|
/* |
|
678
|
|
|
* BT |
|
679
|
|
|
* Begin a text object, inicializind the Tm and Tlm to identity matrix |
|
680
|
|
|
*/ |
|
681
|
|
|
case 'BT': |
|
682
|
|
|
$Tm = $defaultTm; |
|
683
|
|
|
$Tl = $defaultTl; //review this. |
|
684
|
|
|
$Tx = 0; |
|
685
|
|
|
$Ty = 0; |
|
686
|
|
|
break; |
|
687
|
|
|
|
|
688
|
|
|
/* |
|
689
|
|
|
* ET |
|
690
|
|
|
* End a text object, discarding the text matrix |
|
691
|
|
|
*/ |
|
692
|
|
|
case 'ET': |
|
693
|
|
|
$Tm = $defaultTm; |
|
694
|
|
|
$Tl = $defaultTl; //review this |
|
695
|
|
|
$Tx = 0; |
|
696
|
|
|
$Ty = 0; |
|
697
|
|
|
break; |
|
698
|
|
|
|
|
699
|
|
|
/* |
|
700
|
|
|
* leading TL |
|
701
|
|
|
* Set the text leading, Tl, to leading. Tl is used by the T*, ' and " operators. |
|
702
|
|
|
* Initial value: 0 |
|
703
|
|
|
*/ |
|
704
|
|
|
case 'TL': |
|
705
|
|
|
$Tl = (float) $command['c']; |
|
706
|
|
|
break; |
|
707
|
|
|
|
|
708
|
|
|
/* |
|
709
|
|
|
* tx ty Td |
|
710
|
|
|
* Move to the start of the next line, offset form the start of the |
|
711
|
|
|
* current line by tx, ty. |
|
712
|
|
|
*/ |
|
713
|
|
|
case 'Td': |
|
714
|
|
|
$coord = explode(' ', $command['c']); |
|
715
|
|
|
$Tx += (float) $coord[0]; |
|
716
|
|
|
$Ty += (float) $coord[1]; |
|
717
|
|
|
$Tm[$x] = (string) $Tx; |
|
718
|
|
|
$Tm[$y] = (string) $Ty; |
|
719
|
|
|
break; |
|
720
|
|
|
|
|
721
|
|
|
/* |
|
722
|
|
|
* tx ty TD |
|
723
|
|
|
* Move to the start of the next line, offset form the start of the |
|
724
|
|
|
* current line by tx, ty. As a side effect, this operator set the leading |
|
725
|
|
|
* parameter in the text state. This operator has the same effect as the |
|
726
|
|
|
* code: |
|
727
|
|
|
* -ty TL |
|
728
|
|
|
* tx ty Td |
|
729
|
|
|
*/ |
|
730
|
|
|
case 'TD': |
|
731
|
|
|
$coord = explode(' ', $command['c']); |
|
732
|
|
|
$Tl = (float) $coord[1]; |
|
733
|
|
|
$Tx += (float) $coord[0]; |
|
734
|
|
|
$Ty -= (float) $coord[1]; |
|
735
|
|
|
$Tm[$x] = (string) $Tx; |
|
736
|
|
|
$Tm[$y] = (string) $Ty; |
|
737
|
|
|
break; |
|
738
|
|
|
|
|
739
|
|
|
/* |
|
740
|
|
|
* a b c d e f Tm |
|
741
|
|
|
* Set the text matrix, Tm, and the text line matrix, Tlm. The operands are |
|
742
|
|
|
* all numbers, and the initial value for Tm and Tlm is the identity matrix |
|
743
|
|
|
* [1 0 0 1 0 0] |
|
744
|
|
|
*/ |
|
745
|
|
|
case 'Tm': |
|
746
|
|
|
$Tm = explode(' ', $command['c']); |
|
747
|
|
|
$Tx = (float) $Tm[$x]; |
|
748
|
|
|
$Ty = (float) $Tm[$y]; |
|
749
|
|
|
break; |
|
750
|
|
|
|
|
751
|
|
|
/* |
|
752
|
|
|
* T* |
|
753
|
|
|
* Move to the start of the next line. This operator has the same effect |
|
754
|
|
|
* as the code: |
|
755
|
|
|
* 0 Tl Td |
|
756
|
|
|
* Where Tl is the current leading parameter in the text state. |
|
757
|
|
|
*/ |
|
758
|
|
|
case 'T*': |
|
759
|
|
|
$Ty -= $Tl; |
|
760
|
|
|
$Tm[$y] = (string) $Ty; |
|
761
|
|
|
break; |
|
762
|
|
|
|
|
763
|
|
|
/* |
|
764
|
|
|
* string Tj |
|
765
|
|
|
* Show a Text String |
|
766
|
|
|
*/ |
|
767
|
|
|
case 'Tj': |
|
768
|
|
|
$extractedData[] = [$Tm, $currentText]; |
|
769
|
|
|
break; |
|
770
|
|
|
|
|
771
|
|
|
/* |
|
772
|
|
|
* string ' |
|
773
|
|
|
* Move to the next line and show a text string. This operator has the |
|
774
|
|
|
* same effect as the code: |
|
775
|
|
|
* T* |
|
776
|
|
|
* string Tj |
|
777
|
|
|
*/ |
|
778
|
|
|
case "'": |
|
779
|
|
|
$Ty -= $Tl; |
|
780
|
|
|
$Tm[$y] = (string) $Ty; |
|
781
|
|
|
$extractedData[] = [$Tm, $currentText]; |
|
782
|
|
|
break; |
|
783
|
|
|
|
|
784
|
|
|
/* |
|
785
|
|
|
* aw ac string " |
|
786
|
|
|
* Move to the next line and show a text string, using aw as the word |
|
787
|
|
|
* spacing and ac as the character spacing. This operator has the same |
|
788
|
|
|
* effect as the code: |
|
789
|
|
|
* aw Tw |
|
790
|
|
|
* ac Tc |
|
791
|
|
|
* string ' |
|
792
|
|
|
* Tw set the word spacing, Tw, to wordSpace. |
|
793
|
|
|
* Tc Set the character spacing, Tc, to charsSpace. |
|
794
|
|
|
*/ |
|
795
|
|
|
case '"': |
|
796
|
|
|
$data = explode(' ', $currentText); |
|
797
|
|
|
$Ty -= $Tl; |
|
798
|
|
|
$Tm[$y] = (string) $Ty; |
|
799
|
|
|
$extractedData[] = [$Tm, $data[2]]; //Verify |
|
800
|
|
|
break; |
|
801
|
|
|
|
|
802
|
|
|
/* |
|
803
|
|
|
* array TJ |
|
804
|
|
|
* Show one or more text strings allow individual glyph positioning. |
|
805
|
|
|
* Each lement of array con be a string or a number. If the element is |
|
806
|
|
|
* a string, this operator shows the string. If it is a number, the |
|
807
|
|
|
* operator adjust the text position by that amount; that is, it translates |
|
808
|
|
|
* the text matrix, Tm. This amount is substracted form the current |
|
809
|
|
|
* horizontal or vertical coordinate, depending on the writing mode. |
|
810
|
|
|
* in the default coordinate system, a positive adjustment has the effect |
|
811
|
|
|
* of moving the next glyph painted either to the left or down by the given |
|
812
|
|
|
* amount. |
|
813
|
|
|
*/ |
|
814
|
|
|
case 'TJ': |
|
815
|
|
|
$extractedData[] = [$Tm, $currentText]; |
|
816
|
|
|
break; |
|
817
|
|
|
default: |
|
818
|
|
|
} |
|
819
|
|
|
} |
|
820
|
|
|
$this->dataTm = $extractedData; |
|
821
|
|
|
|
|
822
|
|
|
return $extractedData; |
|
823
|
|
|
} |
|
824
|
|
|
|
|
825
|
|
|
/** |
|
826
|
|
|
* Gets text data that are around the given coordinates (X,Y) |
|
827
|
|
|
* |
|
828
|
|
|
* If the text is in near the given coordinates (X,Y) (or the TM info), |
|
829
|
|
|
* the text is returned. The extractedData return by getDataTm, could be use to see |
|
830
|
|
|
* where is the coordinates of a given text, using the TM info for it. |
|
831
|
|
|
* |
|
832
|
|
|
* @param float $x The X value of the coordinate to search for. if null |
|
833
|
|
|
* just the Y value is considered (same Row) |
|
834
|
|
|
* @param float $y The Y value of the coordinate to search for |
|
835
|
|
|
* just the X value is considered (same column) |
|
836
|
|
|
* @param float $xError The value less or more to consider an X to be "near" |
|
837
|
|
|
* @param float $yError The value less or more to consider an Y to be "near" |
|
838
|
|
|
* |
|
839
|
|
|
* @return array An array of text that are near the given coordinates. If no text |
|
840
|
|
|
* "near" the x,y coordinate, an empty array is returned. If Both, x |
|
841
|
|
|
* and y coordinates are null, null is returned. |
|
842
|
|
|
*/ |
|
843
|
|
|
public function getTextXY(float $x = null, float $y = null, float $xError = 0, float $yError = 0): array |
|
844
|
|
|
{ |
|
845
|
|
|
if (!isset($this->dataTm) || !$this->dataTm) { |
|
846
|
|
|
$this->getDataTm(); |
|
847
|
|
|
} |
|
848
|
|
|
|
|
849
|
|
|
if (null !== $x) { |
|
850
|
|
|
$x = (float) $x; |
|
851
|
|
|
} |
|
852
|
|
|
|
|
853
|
|
|
if (null !== $y) { |
|
854
|
|
|
$y = (float) $y; |
|
855
|
|
|
} |
|
856
|
|
|
|
|
857
|
|
|
if (null === $x && null === $y) { |
|
858
|
|
|
return []; |
|
859
|
|
|
} |
|
860
|
|
|
|
|
861
|
|
|
$xError = (float) $xError; |
|
862
|
|
|
$yError = (float) $yError; |
|
863
|
|
|
|
|
864
|
|
|
$extractedData = []; |
|
865
|
|
|
foreach ($this->dataTm as $item) { |
|
866
|
|
|
$tm = $item[0]; |
|
867
|
|
|
$xTm = (float) $tm[4]; |
|
868
|
|
|
$yTm = (float) $tm[5]; |
|
869
|
|
|
$text = $item[1]; |
|
870
|
|
|
if (null === $y) { |
|
871
|
|
|
if (($xTm >= ($x - $xError)) && |
|
872
|
|
|
($xTm <= ($x + $xError))) { |
|
873
|
|
|
$extractedData[] = [$tm, $text]; |
|
874
|
|
|
continue; |
|
875
|
|
|
} |
|
876
|
|
|
} |
|
877
|
|
|
if (null === $x) { |
|
878
|
|
|
if (($yTm >= ($y - $yError)) && |
|
879
|
|
|
($yTm <= ($y + $yError))) { |
|
880
|
|
|
$extractedData[] = [$tm, $text]; |
|
881
|
|
|
continue; |
|
882
|
|
|
} |
|
883
|
|
|
} |
|
884
|
|
|
if (($xTm >= ($x - $xError)) && |
|
885
|
|
|
($xTm <= ($x + $xError)) && |
|
886
|
|
|
($yTm >= ($y - $yError)) && |
|
887
|
|
|
($yTm <= ($y + $yError))) { |
|
888
|
|
|
$extractedData[] = [$tm, $text]; |
|
889
|
|
|
continue; |
|
890
|
|
|
} |
|
891
|
|
|
} |
|
892
|
|
|
|
|
893
|
|
|
return $extractedData; |
|
894
|
|
|
} |
|
895
|
|
|
} |
|
896
|
|
|
|
smalot /
pdfparser
Loading history...