| 1 |
<?php |
| 2 |
|
| 3 |
/** |
| 4 |
* @file |
| 5 |
* This file is part of the PdfParser library. |
| 6 |
* |
| 7 |
* @author Sébastien MALOT <sebastien@malot.fr> |
| 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 <sebastien@malot.fr> |
| 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\Element\ElementArray; |
| 36 |
use Smalot\PdfParser\Element\ElementMissing; |
| 37 |
use Smalot\PdfParser\Element\ElementNull; |
| 38 |
use Smalot\PdfParser\Element\ElementXRef; |
| 39 |
|
| 40 |
class Page extends PDFObject |
| 41 |
{ |
| 42 |
/** |
| 43 |
* @var Font[] |
| 44 |
*/ |
| 45 |
protected $fonts; |
| 46 |
|
| 47 |
/** |
| 48 |
* @var PDFObject[] |
| 49 |
*/ |
| 50 |
protected $xobjects; |
| 51 |
|
| 52 |
/** |
| 53 |
* @var array |
| 54 |
*/ |
| 55 |
protected $dataTm; |
| 56 |
|
| 57 |
/** |
| 58 |
* @param array<\Smalot\PdfParser\Font> $fonts |
| 59 |
* |
| 60 |
* @internal |
| 61 |
*/ |
| 62 |
public function setFonts($fonts) |
| 63 |
{ |
| 64 |
if (empty($this->fonts)) { |
| 65 |
$this->fonts = $fonts; |
| 66 |
} |
| 67 |
} |
| 68 |
|
| 69 |
/** |
| 70 |
* @return Font[] |
| 71 |
*/ |
| 72 |
public function getFonts() |
| 73 |
{ |
| 74 |
if (null !== $this->fonts) { |
| 75 |
return $this->fonts; |
| 76 |
} |
| 77 |
|
| 78 |
$resources = $this->get('Resources'); |
| 79 |
|
| 80 |
if (method_exists($resources, 'has') && $resources->has('Font')) { |
| 81 |
if ($resources->get('Font') instanceof ElementMissing) { |
| 82 |
return []; |
| 83 |
} |
| 84 |
|
| 85 |
if ($resources->get('Font') instanceof Header) { |
| 86 |
$fonts = $resources->get('Font')->getElements(); |
| 87 |
} else { |
| 88 |
$fonts = $resources->get('Font')->getHeader()->getElements(); |
| 89 |
} |
| 90 |
|
| 91 |
$table = []; |
| 92 |
|
| 93 |
foreach ($fonts as $id => $font) { |
| 94 |
if ($font instanceof Font) { |
| 95 |
$table[$id] = $font; |
| 96 |
|
| 97 |
// Store too on cleaned id value (only numeric) |
| 98 |
$id = preg_replace('/[^0-9\.\-_]/', '', $id); |
| 99 |
if ('' != $id) { |
| 100 |
$table[$id] = $font; |
| 101 |
} |
| 102 |
} |
| 103 |
} |
| 104 |
|
| 105 |
return $this->fonts = $table; |
| 106 |
} |
| 107 |
|
| 108 |
return []; |
| 109 |
} |
| 110 |
|
| 111 |
public function getFont(string $id): ?Font |
| 112 |
{ |
| 113 |
$fonts = $this->getFonts(); |
| 114 |
|
| 115 |
if (isset($fonts[$id])) { |
| 116 |
return $fonts[$id]; |
| 117 |
} |
| 118 |
|
| 119 |
// According to the PDF specs (https://www.adobe.com/content/dam/acom/en/devnet/pdf/pdfs/PDF32000_2008.pdf, page 238) |
| 120 |
// "The font resource name presented to the Tf operator is arbitrary, as are the names for all kinds of resources" |
| 121 |
// Instead, we search for the unfiltered name first and then do this cleaning as a fallback, so all tests still pass. |
| 122 |
|
| 123 |
if (isset($fonts[$id])) { |
| 124 |
return $fonts[$id]; |
| 125 |
} else { |
| 126 |
$id = preg_replace('/[^0-9\.\-_]/', '', $id); |
| 127 |
if (isset($fonts[$id])) { |
| 128 |
return $fonts[$id]; |
| 129 |
} |
| 130 |
} |
| 131 |
|
| 132 |
return null; |
| 133 |
} |
| 134 |
|
| 135 |
/** |
| 136 |
* Support for XObject |
| 137 |
* |
| 138 |
* @return PDFObject[] |
| 139 |
*/ |
| 140 |
public function getXObjects() |
| 141 |
{ |
| 142 |
if (null !== $this->xobjects) { |
| 143 |
return $this->xobjects; |
| 144 |
} |
| 145 |
|
| 146 |
$resources = $this->get('Resources'); |
| 147 |
|
| 148 |
if (method_exists($resources, 'has') && $resources->has('XObject')) { |
| 149 |
if ($resources->get('XObject') instanceof Header) { |
| 150 |
$xobjects = $resources->get('XObject')->getElements(); |
| 151 |
} else { |
| 152 |
$xobjects = $resources->get('XObject')->getHeader()->getElements(); |
| 153 |
} |
| 154 |
|
| 155 |
$table = []; |
| 156 |
|
| 157 |
foreach ($xobjects as $id => $xobject) { |
| 158 |
$table[$id] = $xobject; |
| 159 |
|
| 160 |
// Store too on cleaned id value (only numeric) |
| 161 |
$id = preg_replace('/[^0-9\.\-_]/', '', $id); |
| 162 |
if ('' != $id) { |
| 163 |
$table[$id] = $xobject; |
| 164 |
} |
| 165 |
} |
| 166 |
|
| 167 |
return $this->xobjects = $table; |
| 168 |
} |
| 169 |
|
| 170 |
return []; |
| 171 |
} |
| 172 |
|
| 173 |
public function getXObject(string $id): ?PDFObject |
| 174 |
{ |
| 175 |
$xobjects = $this->getXObjects(); |
| 176 |
|
| 177 |
if (isset($xobjects[$id])) { |
| 178 |
return $xobjects[$id]; |
| 179 |
} |
| 180 |
|
| 181 |
return null; |
| 182 |
/*$id = preg_replace('/[^0-9\.\-_]/', '', $id); |
| 183 |
|
| 184 |
if (isset($xobjects[$id])) { |
| 185 |
return $xobjects[$id]; |
| 186 |
} else { |
| 187 |
return null; |
| 188 |
}*/ |
| 189 |
} |
| 190 |
|
| 191 |
public function getText(?self $page = null): string |
| 192 |
{ |
| 193 |
if ($contents = $this->get('Contents')) { |
| 194 |
if ($contents instanceof ElementMissing) { |
| 195 |
return ''; |
| 196 |
} elseif ($contents instanceof ElementNull) { |
| 197 |
return ''; |
| 198 |
} elseif ($contents instanceof PDFObject) { |
| 199 |
$elements = $contents->getHeader()->getElements(); |
| 200 |
|
| 201 |
if (is_numeric(key($elements))) { |
| 202 |
$new_content = ''; |
| 203 |
|
| 204 |
foreach ($elements as $element) { |
| 205 |
if ($element instanceof ElementXRef) { |
| 206 |
$new_content .= $element->getObject()->getContent(); |
| 207 |
} else { |
| 208 |
$new_content .= $element->getContent(); |
| 209 |
} |
| 210 |
} |
| 211 |
|
| 212 |
$header = new Header([], $this->document); |
| 213 |
$contents = new PDFObject($this->document, $header, $new_content, $this->config); |
| 214 |
} |
| 215 |
} elseif ($contents instanceof ElementArray) { |
| 216 |
// Create a virtual global content. |
| 217 |
$new_content = ''; |
| 218 |
|
| 219 |
foreach ($contents->getContent() as $content) { |
| 220 |
$new_content .= $content->getContent()."\n"; |
| 221 |
} |
| 222 |
|
| 223 |
$header = new Header([], $this->document); |
| 224 |
$contents = new PDFObject($this->document, $header, $new_content, $this->config); |
| 225 |
} |
| 226 |
|
| 227 |
/* |
| 228 |
* Elements referencing each other on the same page can cause endless loops during text parsing. |
| 229 |
* To combat this we keep a recursionStack containing already parsed elements on the page. |
| 230 |
* The stack is only emptied here after getting text from a page. |
| 231 |
*/ |
| 232 |
$contentsText = $contents->getText($this); |
| 233 |
PDFObject::$recursionStack = []; |
| 234 |
|
| 235 |
return $contentsText; |
| 236 |
} |
| 237 |
|
| 238 |
return ''; |
| 239 |
} |
| 240 |
|
| 241 |
/** |
| 242 |
* Return true if the current page is a (setasign\Fpdi\Fpdi) FPDI/FPDF document |
| 243 |
* |
| 244 |
* The metadata 'Producer' should have the value of "FPDF" . FPDF_VERSION if the |
| 245 |
* pdf file was generated by FPDF/Fpfi. |
| 246 |
* |
| 247 |
* @return bool true is the current page is a FPDI/FPDF document |
| 248 |
*/ |
| 249 |
public function isFpdf(): bool |
| 250 |
{ |
| 251 |
if (\array_key_exists('Producer', $this->document->getDetails()) |
| 252 |
&& \is_string($this->document->getDetails()['Producer']) |
| 253 |
&& 0 === strncmp($this->document->getDetails()['Producer'], 'FPDF', 4)) { |
| 254 |
return true; |
| 255 |
} |
| 256 |
|
| 257 |
return false; |
| 258 |
} |
| 259 |
|
| 260 |
/** |
| 261 |
* Return the page number of the PDF document of the page object |
| 262 |
* |
| 263 |
* @return int the page number |
| 264 |
*/ |
| 265 |
public function getPageNumber(): int |
| 266 |
{ |
| 267 |
$pages = $this->document->getPages(); |
| 268 |
$numOfPages = \count($pages); |
| 269 |
for ($pageNum = 0; $pageNum < $numOfPages; ++$pageNum) { |
| 270 |
if ($pages[$pageNum] === $this) { |
| 271 |
break; |
| 272 |
} |
| 273 |
} |
| 274 |
|
| 275 |
return $pageNum; |
| 276 |
} |
| 277 |
|
| 278 |
/** |
| 279 |
* Return the Object of the page if the document is a FPDF/FPDI document |
| 280 |
* |
| 281 |
* If the document was generated by FPDF/FPDI it returns the |
| 282 |
* PDFObject of the given page |
| 283 |
* |
| 284 |
* @return PDFObject The PDFObject for the page |
| 285 |
*/ |
| 286 |
public function getPDFObjectForFpdf(): PDFObject |
| 287 |
{ |
| 288 |
$pageNum = $this->getPageNumber(); |
| 289 |
$xObjects = $this->getXObjects(); |
| 290 |
|
| 291 |
return $xObjects[$pageNum]; |
| 292 |
} |
| 293 |
|
| 294 |
/** |
| 295 |
* Return a new PDFObject of the document created with FPDF/FPDI |
| 296 |
* |
| 297 |
* For a document generated by FPDF/FPDI, it generates a |
| 298 |
* new PDFObject for that document |
| 299 |
* |
| 300 |
* @return PDFObject The PDFObject |
| 301 |
*/ |
| 302 |
public function createPDFObjectForFpdf(): PDFObject |
| 303 |
{ |
| 304 |
$pdfObject = $this->getPDFObjectForFpdf(); |
| 305 |
$new_content = $pdfObject->getContent(); |
| 306 |
$header = $pdfObject->getHeader(); |
| 307 |
$config = $pdfObject->config; |
| 308 |
|
| 309 |
return new PDFObject($pdfObject->document, $header, $new_content, $config); |
| 310 |
} |
| 311 |
|
| 312 |
/** |
| 313 |
* Return page if document is a FPDF/FPDI document |
| 314 |
* |
| 315 |
* @return Page The page |
| 316 |
*/ |
| 317 |
public function createPageForFpdf(): self |
| 318 |
{ |
| 319 |
$pdfObject = $this->getPDFObjectForFpdf(); |
| 320 |
$new_content = $pdfObject->getContent(); |
| 321 |
$header = $pdfObject->getHeader(); |
| 322 |
$config = $pdfObject->config; |
| 323 |
|
| 324 |
return new self($pdfObject->document, $header, $new_content, $config); |
| 325 |
} |
| 326 |
|
| 327 |
public function getTextArray(?self $page = null): array |
| 328 |
{ |
| 329 |
if ($this->isFpdf()) { |
| 330 |
$pdfObject = $this->getPDFObjectForFpdf(); |
| 331 |
$newPdfObject = $this->createPDFObjectForFpdf(); |
| 332 |
|
| 333 |
return $newPdfObject->getTextArray($pdfObject); |
| 334 |
} else { |
| 335 |
if ($contents = $this->get('Contents')) { |
| 336 |
if ($contents instanceof ElementMissing) { |
| 337 |
return []; |
| 338 |
} elseif ($contents instanceof ElementNull) { |
| 339 |
return []; |
| 340 |
} elseif ($contents instanceof PDFObject) { |
| 341 |
$elements = $contents->getHeader()->getElements(); |
| 342 |
|
| 343 |
if (is_numeric(key($elements))) { |
| 344 |
$new_content = ''; |
| 345 |
|
| 346 |
/** @var PDFObject $element */ |
| 347 |
foreach ($elements as $element) { |
| 348 |
if ($element instanceof ElementXRef) { |
| 349 |
$new_content .= $element->getObject()->getContent(); |
| 350 |
} else { |
| 351 |
$new_content .= $element->getContent(); |
| 352 |
} |
| 353 |
} |
| 354 |
|
| 355 |
$header = new Header([], $this->document); |
| 356 |
$contents = new PDFObject($this->document, $header, $new_content, $this->config); |
| 357 |
} else { |
| 358 |
try { |
| 359 |
$contents->getTextArray($this); |
| 360 |
} catch (\Throwable $e) { |
| 361 |
return $contents->getTextArray(); |
| 362 |
} |
| 363 |
} |
| 364 |
} elseif ($contents instanceof ElementArray) { |
| 365 |
// Create a virtual global content. |
| 366 |
$new_content = ''; |
| 367 |
|
| 368 |
/** @var PDFObject $content */ |
| 369 |
foreach ($contents->getContent() as $content) { |
| 370 |
$new_content .= $content->getContent()."\n"; |
| 371 |
} |
| 372 |
|
| 373 |
$header = new Header([], $this->document); |
| 374 |
$contents = new PDFObject($this->document, $header, $new_content, $this->config); |
| 375 |
} |
| 376 |
|
| 377 |
return $contents->getTextArray($this); |
| 378 |
} |
| 379 |
|
| 380 |
return []; |
| 381 |
} |
| 382 |
} |
| 383 |
|
| 384 |
/** |
| 385 |
* Gets all the text data with its internal representation of the page. |
| 386 |
* |
| 387 |
* Returns an array with the data and the internal representation |
| 388 |
*/ |
| 389 |
public function extractRawData(): array |
| 390 |
{ |
| 391 |
/* |
| 392 |
* Now you can get the complete content of the object with the text on it |
| 393 |
*/ |
| 394 |
$extractedData = []; |
| 395 |
$content = $this->get('Contents'); |
| 396 |
$values = $content->getContent(); |
| 397 |
if (isset($values) && \is_array($values)) { |
| 398 |
$text = ''; |
| 399 |
foreach ($values as $section) { |
| 400 |
$text .= $section->getContent(); |
| 401 |
} |
| 402 |
$sectionsText = $this->getSectionsText($text); |
| 403 |
foreach ($sectionsText as $sectionText) { |
| 404 |
$commandsText = $this->getCommandsText($sectionText); |
| 405 |
foreach ($commandsText as $command) { |
| 406 |
$extractedData[] = $command; |
| 407 |
} |
| 408 |
} |
| 409 |
} else { |
| 410 |
if ($this->isFpdf()) { |
| 411 |
$content = $this->getPDFObjectForFpdf(); |
| 412 |
} |
| 413 |
$sectionsText = $content->getSectionsText($content->getContent()); |
| 414 |
foreach ($sectionsText as $sectionText) { |
| 415 |
$commandsText = $content->getCommandsText($sectionText); |
| 416 |
foreach ($commandsText as $command) { |
| 417 |
$extractedData[] = $command; |
| 418 |
} |
| 419 |
} |
| 420 |
} |
| 421 |
|
| 422 |
return $extractedData; |
| 423 |
} |
| 424 |
|
| 425 |
/** |
| 426 |
* Gets all the decoded text data with it internal representation from a page. |
| 427 |
* |
| 428 |
* @param array $extractedRawData the extracted data return by extractRawData or |
| 429 |
* null if extractRawData should be called |
| 430 |
* |
| 431 |
* @return array An array with the data and the internal representation |
| 432 |
*/ |
| 433 |
public function extractDecodedRawData(?array $extractedRawData = null): array |
| 434 |
{ |
| 435 |
if (!isset($extractedRawData) || !$extractedRawData) { |
| 436 |
$extractedRawData = $this->extractRawData(); |
| 437 |
} |
| 438 |
$currentFont = null; /** @var Font $currentFont */ |
| 439 |
$clippedFont = null; |
| 440 |
$fpdfPage = null; |
| 441 |
if ($this->isFpdf()) { |
| 442 |
$fpdfPage = $this->createPageForFpdf(); |
| 443 |
} |
| 444 |
foreach ($extractedRawData as &$command) { |
| 445 |
if ('Tj' == $command['o'] || 'TJ' == $command['o']) { |
| 446 |
$data = $command['c']; |
| 447 |
if (!\is_array($data)) { |
| 448 |
$tmpText = ''; |
| 449 |
if (isset($currentFont)) { |
| 450 |
$tmpText = $currentFont->decodeOctal($data); |
| 451 |
// $tmpText = $currentFont->decodeHexadecimal($tmpText, false); |
| 452 |
} |
| 453 |
$tmpText = str_replace( |
| 454 |
['\\\\', '\(', '\)', '\n', '\r', '\t', '\ '], |
| 455 |
['\\', '(', ')', "\n", "\r", "\t", ' '], |
| 456 |
$tmpText |
| 457 |
); |
| 458 |
$tmpText = mb_convert_encoding($tmpText, 'UTF-8', 'ISO-8859-1'); |
| 459 |
if (isset($currentFont)) { |
| 460 |
$tmpText = $currentFont->decodeContent($tmpText); |
| 461 |
} |
| 462 |
$command['c'] = $tmpText; |
| 463 |
continue; |
| 464 |
} |
| 465 |
$numText = \count($data); |
| 466 |
for ($i = 0; $i < $numText; ++$i) { |
| 467 |
if (0 != ($i % 2)) { |
| 468 |
continue; |
| 469 |
} |
| 470 |
$tmpText = $data[$i]['c']; |
| 471 |
$decodedText = isset($currentFont) ? $currentFont->decodeOctal($tmpText) : $tmpText; |
| 472 |
$decodedText = str_replace( |
| 473 |
['\\\\', '\(', '\)', '\n', '\r', '\t', '\ '], |
| 474 |
['\\', '(', ')', "\n", "\r", "\t", ' '], |
| 475 |
$decodedText |
| 476 |
); |
| 477 |
|
| 478 |
$decodedText = mb_convert_encoding($decodedText, 'UTF-8', 'ISO-8859-1'); |
| 479 |
|
| 480 |
if (isset($currentFont)) { |
| 481 |
$decodedText = $currentFont->decodeContent($decodedText); |
| 482 |
} |
| 483 |
$command['c'][$i]['c'] = $decodedText; |
| 484 |
continue; |
| 485 |
} |
| 486 |
} elseif ('Tf' == $command['o'] || 'TF' == $command['o']) { |
| 487 |
$fontId = explode(' ', $command['c'])[0]; |
| 488 |
// If document is a FPDI/FPDF the $page has the correct font |
| 489 |
$currentFont = isset($fpdfPage) ? $fpdfPage->getFont($fontId) : $this->getFont($fontId); |
| 490 |
continue; |
| 491 |
} elseif ('Q' == $command['o']) { |
| 492 |
$currentFont = $clippedFont; |
| 493 |
} elseif ('q' == $command['o']) { |
| 494 |
$clippedFont = $currentFont; |
| 495 |
} |
| 496 |
} |
| 497 |
|
| 498 |
return $extractedRawData; |
| 499 |
} |
| 500 |
|
| 501 |
/** |
| 502 |
* Gets just the Text commands that are involved in text positions and |
| 503 |
* Text Matrix (Tm) |
| 504 |
* |
| 505 |
* It extract just the PDF commands that are involved with text positions, and |
| 506 |
* the Text Matrix (Tm). These are: BT, ET, TL, Td, TD, Tm, T*, Tj, ', ", and TJ |
| 507 |
* |
| 508 |
* @param array $extractedDecodedRawData The data extracted by extractDecodeRawData. |
| 509 |
* If it is null, the method extractDecodeRawData is called. |
| 510 |
* |
| 511 |
* @return array An array with the text command of the page |
| 512 |
*/ |
| 513 |
public function getDataCommands(?array $extractedDecodedRawData = null): array |
| 514 |
{ |
| 515 |
if (!isset($extractedDecodedRawData) || !$extractedDecodedRawData) { |
| 516 |
$extractedDecodedRawData = $this->extractDecodedRawData(); |
| 517 |
} |
| 518 |
$extractedData = []; |
| 519 |
foreach ($extractedDecodedRawData as $command) { |
| 520 |
switch ($command['o']) { |
| 521 |
/* |
| 522 |
* BT |
| 523 |
* Begin a text object, inicializind the Tm and Tlm to identity matrix |
| 524 |
*/ |
| 525 |
case 'BT': |
| 526 |
$extractedData[] = $command; |
| 527 |
break; |
| 528 |
/* |
| 529 |
* cm |
| 530 |
* Concatenation Matrix that will transform all following Tm |
| 531 |
*/ |
| 532 |
case 'cm': |
| 533 |
$extractedData[] = $command; |
| 534 |
break; |
| 535 |
/* |
| 536 |
* ET |
| 537 |
* End a text object, discarding the text matrix |
| 538 |
*/ |
| 539 |
case 'ET': |
| 540 |
$extractedData[] = $command; |
| 541 |
break; |
| 542 |
|
| 543 |
/* |
| 544 |
* leading TL |
| 545 |
* Set the text leading, Tl, to leading. Tl is used by the T*, ' and " operators. |
| 546 |
* Initial value: 0 |
| 547 |
*/ |
| 548 |
case 'TL': |
| 549 |
$extractedData[] = $command; |
| 550 |
break; |
| 551 |
|
| 552 |
/* |
| 553 |
* tx ty Td |
| 554 |
* Move to the start of the next line, offset form the start of the |
| 555 |
* current line by tx, ty. |
| 556 |
*/ |
| 557 |
case 'Td': |
| 558 |
$extractedData[] = $command; |
| 559 |
break; |
| 560 |
|
| 561 |
/* |
| 562 |
* tx ty TD |
| 563 |
* Move to the start of the next line, offset form the start of the |
| 564 |
* current line by tx, ty. As a side effect, this operator set the leading |
| 565 |
* parameter in the text state. This operator has the same effect as the |
| 566 |
* code: |
| 567 |
* -ty TL |
| 568 |
* tx ty Td |
| 569 |
*/ |
| 570 |
case 'TD': |
| 571 |
$extractedData[] = $command; |
| 572 |
break; |
| 573 |
|
| 574 |
/* |
| 575 |
* a b c d e f Tm |
| 576 |
* Set the text matrix, Tm, and the text line matrix, Tlm. The operands are |
| 577 |
* all numbers, and the initial value for Tm and Tlm is the identity matrix |
| 578 |
* [1 0 0 1 0 0] |
| 579 |
*/ |
| 580 |
case 'Tm': |
| 581 |
$extractedData[] = $command; |
| 582 |
break; |
| 583 |
|
| 584 |
/* |
| 585 |
* T* |
| 586 |
* Move to the start of the next line. This operator has the same effect |
| 587 |
* as the code: |
| 588 |
* 0 Tl Td |
| 589 |
* Where Tl is the current leading parameter in the text state. |
| 590 |
*/ |
| 591 |
case 'T*': |
| 592 |
$extractedData[] = $command; |
| 593 |
break; |
| 594 |
|
| 595 |
/* |
| 596 |
* string Tj |
| 597 |
* Show a Text String |
| 598 |
*/ |
| 599 |
case 'Tj': |
| 600 |
$extractedData[] = $command; |
| 601 |
break; |
| 602 |
|
| 603 |
/* |
| 604 |
* string ' |
| 605 |
* Move to the next line and show a text string. This operator has the |
| 606 |
* same effect as the code: |
| 607 |
* T* |
| 608 |
* string Tj |
| 609 |
*/ |
| 610 |
case "'": |
| 611 |
$extractedData[] = $command; |
| 612 |
break; |
| 613 |
|
| 614 |
/* |
| 615 |
* aw ac string " |
| 616 |
* Move to the next lkine and show a text string, using aw as the word |
| 617 |
* spacing and ac as the character spacing. This operator has the same |
| 618 |
* effect as the code: |
| 619 |
* aw Tw |
| 620 |
* ac Tc |
| 621 |
* string ' |
| 622 |
* Tw set the word spacing, Tw, to wordSpace. |
| 623 |
* Tc Set the character spacing, Tc, to charsSpace. |
| 624 |
*/ |
| 625 |
case '"': |
| 626 |
$extractedData[] = $command; |
| 627 |
break; |
| 628 |
|
| 629 |
case 'Tf': |
| 630 |
case 'TF': |
| 631 |
$extractedData[] = $command; |
| 632 |
break; |
| 633 |
|
| 634 |
/* |
| 635 |
* array TJ |
| 636 |
* Show one or more text strings allow individual glyph positioning. |
| 637 |
* Each lement of array con be a string or a number. If the element is |
| 638 |
* a string, this operator shows the string. If it is a number, the |
| 639 |
* operator adjust the text position by that amount; that is, it translates |
| 640 |
* the text matrix, Tm. This amount is substracted form the current |
| 641 |
* horizontal or vertical coordinate, depending on the writing mode. |
| 642 |
* in the default coordinate system, a positive adjustment has the effect |
| 643 |
* of moving the next glyph painted either to the left or down by the given |
| 644 |
* amount. |
| 645 |
*/ |
| 646 |
case 'TJ': |
| 647 |
$extractedData[] = $command; |
| 648 |
break; |
| 649 |
/* |
| 650 |
* q |
| 651 |
* Save current graphics state to stack |
| 652 |
*/ |
| 653 |
case 'q': |
| 654 |
/* |
| 655 |
* Q |
| 656 |
* Load last saved graphics state from stack |
| 657 |
*/ |
| 658 |
case 'Q': |
| 659 |
$extractedData[] = $command; |
| 660 |
break; |
| 661 |
default: |
| 662 |
} |
| 663 |
} |
| 664 |
|
| 665 |
return $extractedData; |
| 666 |
} |
| 667 |
|
| 668 |
/** |
| 669 |
* Gets the Text Matrix of the text in the page |
| 670 |
* |
| 671 |
* Return an array where every item is an array where the first item is the |
| 672 |
* Text Matrix (Tm) and the second is a string with the text data. The Text matrix |
| 673 |
* is an array of 6 numbers. The last 2 numbers are the coordinates X and Y of the |
| 674 |
* text. The first 4 numbers has to be with Scalation, Rotation and Skew of the text. |
| 675 |
* |
| 676 |
* @param array $dataCommands the data extracted by getDataCommands |
| 677 |
* if null getDataCommands is called |
| 678 |
* |
| 679 |
* @return array an array with the data of the page including the Tm information |
| 680 |
* of any text in the page |
| 681 |
*/ |
| 682 |
public function getDataTm(?array $dataCommands = null): array |
| 683 |
{ |
| 684 |
if (!isset($dataCommands) || !$dataCommands) { |
| 685 |
$dataCommands = $this->getDataCommands(); |
| 686 |
} |
| 687 |
|
| 688 |
/* |
| 689 |
* At the beginning of a text object Tm is the identity matrix |
| 690 |
*/ |
| 691 |
$defaultTm = ['1', '0', '0', '1', '0', '0']; |
| 692 |
$concatTm = ['1', '0', '0', '1', '0', '0']; |
| 693 |
$graphicsStatesStack = []; |
| 694 |
/* |
| 695 |
* Set the text leading used by T*, ' and " operators |
| 696 |
*/ |
| 697 |
$defaultTl = 0; |
| 698 |
|
| 699 |
/* |
| 700 |
* Set default values for font data |
| 701 |
*/ |
| 702 |
$defaultFontId = -1; |
| 703 |
$defaultFontSize = 1; |
| 704 |
|
| 705 |
/* |
| 706 |
* Indexes of horizontal/vertical scaling and X,Y-coordinates in the matrix (Tm) |
| 707 |
*/ |
| 708 |
$hSc = 0; // horizontal scaling |
| 709 |
/** |
| 710 |
* index of vertical scaling in the array that encodes the text matrix. |
| 711 |
* for more information: https://github.com/smalot/pdfparser/pull/559#discussion_r1053415500 |
| 712 |
*/ |
| 713 |
$vSc = 3; |
| 714 |
$x = 4; |
| 715 |
$y = 5; |
| 716 |
|
| 717 |
/* |
| 718 |
* x,y-coordinates of text space origin in user units |
| 719 |
* |
| 720 |
* These will be assigned the value of the currently printed string |
| 721 |
*/ |
| 722 |
$Tx = 0; |
| 723 |
$Ty = 0; |
| 724 |
|
| 725 |
$Tm = $defaultTm; |
| 726 |
$Tl = $defaultTl; |
| 727 |
$fontId = $defaultFontId; |
| 728 |
$fontSize = $defaultFontSize; // reflects fontSize set by Tf or Tfs |
| 729 |
|
| 730 |
$extractedTexts = $this->getTextArray(); |
| 731 |
$extractedData = []; |
| 732 |
foreach ($dataCommands as $command) { |
| 733 |
// If we've used up all the texts from getTextArray(), exit |
| 734 |
// so we aren't accessing non-existent array indices |
| 735 |
// Fixes 'undefined array key' errors in Issues #575, #576 |
| 736 |
if (\count($extractedTexts) <= \count($extractedData)) { |
| 737 |
break; |
| 738 |
} |
| 739 |
$currentText = $extractedTexts[\count($extractedData)]; |
| 740 |
switch ($command['o']) { |
| 741 |
/* |
| 742 |
* BT |
| 743 |
* Begin a text object, initializing the Tm and Tlm to identity matrix |
| 744 |
*/ |
| 745 |
case 'BT': |
| 746 |
$Tm = $defaultTm; |
| 747 |
$Tl = $defaultTl; |
| 748 |
$Tx = 0; |
| 749 |
$Ty = 0; |
| 750 |
break; |
| 751 |
|
| 752 |
case 'cm': |
| 753 |
$newConcatTm = (array) explode(' ', $command['c']); |
| 754 |
$TempMatrix = []; |
| 755 |
// Multiply with previous concatTm |
| 756 |
$TempMatrix[0] = (float) $concatTm[0] * (float) $newConcatTm[0] + (float) $concatTm[1] * (float) $newConcatTm[2]; |
| 757 |
$TempMatrix[1] = (float) $concatTm[0] * (float) $newConcatTm[1] + (float) $concatTm[1] * (float) $newConcatTm[3]; |
| 758 |
$TempMatrix[2] = (float) $concatTm[2] * (float) $newConcatTm[0] + (float) $concatTm[3] * (float) $newConcatTm[2]; |
| 759 |
$TempMatrix[3] = (float) $concatTm[2] * (float) $newConcatTm[1] + (float) $concatTm[3] * (float) $newConcatTm[3]; |
| 760 |
$TempMatrix[4] = (float) $concatTm[4] * (float) $newConcatTm[0] + (float) $concatTm[5] * (float) $newConcatTm[2] + (float) $newConcatTm[4]; |
| 761 |
$TempMatrix[5] = (float) $concatTm[4] * (float) $newConcatTm[1] + (float) $concatTm[5] * (float) $newConcatTm[3] + (float) $newConcatTm[5]; |
| 762 |
$concatTm = $TempMatrix; |
| 763 |
break; |
| 764 |
/* |
| 765 |
* ET |
| 766 |
* End a text object |
| 767 |
*/ |
| 768 |
case 'ET': |
| 769 |
break; |
| 770 |
|
| 771 |
/* |
| 772 |
* text leading TL |
| 773 |
* Set the text leading, Tl, to leading. Tl is used by the T*, ' and " operators. |
| 774 |
* Initial value: 0 |
| 775 |
*/ |
| 776 |
case 'TL': |
| 777 |
// scaled text leading |
| 778 |
$Tl = (float) $command['c'] * (float) $Tm[$vSc]; |
| 779 |
break; |
| 780 |
|
| 781 |
/* |
| 782 |
* tx ty Td |
| 783 |
* Move to the start of the next line, offset from the start of the |
| 784 |
* current line by tx, ty. |
| 785 |
*/ |
| 786 |
case 'Td': |
| 787 |
$coord = explode(' ', $command['c']); |
| 788 |
$Tx += (float) $coord[0] * (float) $Tm[$hSc]; |
| 789 |
$Ty += (float) $coord[1] * (float) $Tm[$vSc]; |
| 790 |
$Tm[$x] = (string) $Tx; |
| 791 |
$Tm[$y] = (string) $Ty; |
| 792 |
break; |
| 793 |
|
| 794 |
/* |
| 795 |
* tx ty TD |
| 796 |
* Move to the start of the next line, offset form the start of the |
| 797 |
* current line by tx, ty. As a side effect, this operator set the leading |
| 798 |
* parameter in the text state. This operator has the same effect as the |
| 799 |
* code: |
| 800 |
* -ty TL |
| 801 |
* tx ty Td |
| 802 |
*/ |
| 803 |
case 'TD': |
| 804 |
$coord = explode(' ', $command['c']); |
| 805 |
$Tl = -((float) $coord[1] * (float) $Tm[$vSc]); |
| 806 |
$Tx += (float) $coord[0] * (float) $Tm[$hSc]; |
| 807 |
$Ty += (float) $coord[1] * (float) $Tm[$vSc]; |
| 808 |
$Tm[$x] = (string) $Tx; |
| 809 |
$Tm[$y] = (string) $Ty; |
| 810 |
break; |
| 811 |
|
| 812 |
/* |
| 813 |
* a b c d e f Tm |
| 814 |
* Set the text matrix, Tm, and the text line matrix, Tlm. The operands are |
| 815 |
* all numbers, and the initial value for Tm and Tlm is the identity matrix |
| 816 |
* [1 0 0 1 0 0] |
| 817 |
*/ |
| 818 |
case 'Tm': |
| 819 |
$Tm = explode(' ', $command['c']); |
| 820 |
$TempMatrix = []; |
| 821 |
$TempMatrix[0] = (float) $Tm[0] * (float) $concatTm[0] + (float) $Tm[1] * (float) $concatTm[2]; |
| 822 |
$TempMatrix[1] = (float) $Tm[0] * (float) $concatTm[1] + (float) $Tm[1] * (float) $concatTm[3]; |
| 823 |
$TempMatrix[2] = (float) $Tm[2] * (float) $concatTm[0] + (float) $Tm[3] * (float) $concatTm[2]; |
| 824 |
$TempMatrix[3] = (float) $Tm[2] * (float) $concatTm[1] + (float) $Tm[3] * (float) $concatTm[3]; |
| 825 |
$TempMatrix[4] = (float) $Tm[4] * (float) $concatTm[0] + (float) $Tm[5] * (float) $concatTm[2] + (float) $concatTm[4]; |
| 826 |
$TempMatrix[5] = (float) $Tm[4] * (float) $concatTm[1] + (float) $Tm[5] * (float) $concatTm[3] + (float) $concatTm[5]; |
| 827 |
$Tm = $TempMatrix; |
| 828 |
$Tx = (float) $Tm[$x]; |
| 829 |
$Ty = (float) $Tm[$y]; |
| 830 |
break; |
| 831 |
|
| 832 |
/* |
| 833 |
* T* |
| 834 |
* Move to the start of the next line. This operator has the same effect |
| 835 |
* as the code: |
| 836 |
* 0 Tl Td |
| 837 |
* Where Tl is the current leading parameter in the text state. |
| 838 |
*/ |
| 839 |
case 'T*': |
| 840 |
$Ty -= $Tl; |
| 841 |
$Tm[$y] = (string) $Ty; |
| 842 |
break; |
| 843 |
|
| 844 |
/* |
| 845 |
* string Tj |
| 846 |
* Show a Text String |
| 847 |
*/ |
| 848 |
case 'Tj': |
| 849 |
$data = [$Tm, $currentText]; |
| 850 |
if ($this->config->getDataTmFontInfoHasToBeIncluded()) { |
| 851 |
$data[] = $fontId; |
| 852 |
$data[] = $fontSize; |
| 853 |
} |
| 854 |
$extractedData[] = $data; |
| 855 |
break; |
| 856 |
|
| 857 |
/* |
| 858 |
* string ' |
| 859 |
* Move to the next line and show a text string. This operator has the |
| 860 |
* same effect as the code: |
| 861 |
* T* |
| 862 |
* string Tj |
| 863 |
*/ |
| 864 |
case "'": |
| 865 |
$Ty -= $Tl; |
| 866 |
$Tm[$y] = (string) $Ty; |
| 867 |
$extractedData[] = [$Tm, $currentText]; |
| 868 |
break; |
| 869 |
|
| 870 |
/* |
| 871 |
* aw ac string " |
| 872 |
* Move to the next line and show a text string, using aw as the word |
| 873 |
* spacing and ac as the character spacing. This operator has the same |
| 874 |
* effect as the code: |
| 875 |
* aw Tw |
| 876 |
* ac Tc |
| 877 |
* string ' |
| 878 |
* Tw set the word spacing, Tw, to wordSpace. |
| 879 |
* Tc Set the character spacing, Tc, to charsSpace. |
| 880 |
*/ |
| 881 |
case '"': |
| 882 |
$data = explode(' ', $currentText); |
| 883 |
$Ty -= $Tl; |
| 884 |
$Tm[$y] = (string) $Ty; |
| 885 |
$extractedData[] = [$Tm, $data[2]]; // Verify |
| 886 |
break; |
| 887 |
|
| 888 |
case 'Tf': |
| 889 |
/* |
| 890 |
* From PDF 1.0 specification, page 106: |
| 891 |
* fontname size Tf Set font and size |
| 892 |
* Sets the text font and text size in the graphics state. There is no default value for |
| 893 |
* either fontname or size; they must be selected using Tf before drawing any text. |
| 894 |
* fontname is a resource name. size is a number expressed in text space units. |
| 895 |
* |
| 896 |
* Source: https://ia902503.us.archive.org/10/items/pdfy-0vt8s-egqFwDl7L2/PDF%20Reference%201.0.pdf |
| 897 |
* Introduced with https://github.com/smalot/pdfparser/pull/516 |
| 898 |
*/ |
| 899 |
list($fontId, $fontSize) = explode(' ', $command['c'], 2); |
| 900 |
break; |
| 901 |
|
| 902 |
/* |
| 903 |
* array TJ |
| 904 |
* Show one or more text strings allow individual glyph positioning. |
| 905 |
* Each lement of array con be a string or a number. If the element is |
| 906 |
* a string, this operator shows the string. If it is a number, the |
| 907 |
* operator adjust the text position by that amount; that is, it translates |
| 908 |
* the text matrix, Tm. This amount is substracted form the current |
| 909 |
* horizontal or vertical coordinate, depending on the writing mode. |
| 910 |
* in the default coordinate system, a positive adjustment has the effect |
| 911 |
* of moving the next glyph painted either to the left or down by the given |
| 912 |
* amount. |
| 913 |
*/ |
| 914 |
case 'TJ': |
| 915 |
$data = [$Tm, $currentText]; |
| 916 |
if ($this->config->getDataTmFontInfoHasToBeIncluded()) { |
| 917 |
$data[] = $fontId; |
| 918 |
$data[] = $fontSize; |
| 919 |
} |
| 920 |
$extractedData[] = $data; |
| 921 |
break; |
| 922 |
/* |
| 923 |
* q |
| 924 |
* Save current graphics state to stack |
| 925 |
*/ |
| 926 |
case 'q': |
| 927 |
$graphicsStatesStack[] = $concatTm; |
| 928 |
break; |
| 929 |
/* |
| 930 |
* Q |
| 931 |
* Load last saved graphics state from stack |
| 932 |
*/ |
| 933 |
case 'Q': |
| 934 |
$concatTm = array_pop($graphicsStatesStack); |
| 935 |
break; |
| 936 |
default: |
| 937 |
} |
| 938 |
} |
| 939 |
$this->dataTm = $extractedData; |
| 940 |
|
| 941 |
return $extractedData; |
| 942 |
} |
| 943 |
|
| 944 |
/** |
| 945 |
* Gets text data that are around the given coordinates (X,Y) |
| 946 |
* |
| 947 |
* If the text is in near the given coordinates (X,Y) (or the TM info), |
| 948 |
* the text is returned. The extractedData return by getDataTm, could be use to see |
| 949 |
* where is the coordinates of a given text, using the TM info for it. |
| 950 |
* |
| 951 |
* @param float $x The X value of the coordinate to search for. if null |
| 952 |
* just the Y value is considered (same Row) |
| 953 |
* @param float $y The Y value of the coordinate to search for |
| 954 |
* just the X value is considered (same column) |
| 955 |
* @param float $xError The value less or more to consider an X to be "near" |
| 956 |
* @param float $yError The value less or more to consider an Y to be "near" |
| 957 |
* |
| 958 |
* @return array An array of text that are near the given coordinates. If no text |
| 959 |
* "near" the x,y coordinate, an empty array is returned. If Both, x |
| 960 |
* and y coordinates are null, null is returned. |
| 961 |
*/ |
| 962 |
public function getTextXY(?float $x = null, ?float $y = null, float $xError = 0, float $yError = 0): array |
| 963 |
{ |
| 964 |
if (!isset($this->dataTm) || !$this->dataTm) { |
| 965 |
$this->getDataTm(); |
| 966 |
} |
| 967 |
|
| 968 |
if (null !== $x) { |
| 969 |
$x = (float) $x; |
| 970 |
} |
| 971 |
|
| 972 |
if (null !== $y) { |
| 973 |
$y = (float) $y; |
| 974 |
} |
| 975 |
|
| 976 |
if (null === $x && null === $y) { |
| 977 |
return []; |
| 978 |
} |
| 979 |
|
| 980 |
$xError = (float) $xError; |
| 981 |
$yError = (float) $yError; |
| 982 |
|
| 983 |
$extractedData = []; |
| 984 |
foreach ($this->dataTm as $item) { |
| 985 |
$tm = $item[0]; |
| 986 |
$xTm = (float) $tm[4]; |
| 987 |
$yTm = (float) $tm[5]; |
| 988 |
$text = $item[1]; |
| 989 |
if (null === $y) { |
| 990 |
if (($xTm >= ($x - $xError)) |
| 991 |
&& ($xTm <= ($x + $xError))) { |
| 992 |
$extractedData[] = [$tm, $text]; |
| 993 |
continue; |
| 994 |
} |
| 995 |
} |
| 996 |
if (null === $x) { |
| 997 |
if (($yTm >= ($y - $yError)) |
| 998 |
&& ($yTm <= ($y + $yError))) { |
| 999 |
$extractedData[] = [$tm, $text]; |
| 1000 |
continue; |
| 1001 |
} |
| 1002 |
} |
| 1003 |
if (($xTm >= ($x - $xError)) |
| 1004 |
&& ($xTm <= ($x + $xError)) |
| 1005 |
&& ($yTm >= ($y - $yError)) |
| 1006 |
&& ($yTm <= ($y + $yError))) { |
| 1007 |
$extractedData[] = [$tm, $text]; |
| 1008 |
continue; |
| 1009 |
} |
| 1010 |
} |
| 1011 |
|
| 1012 |
return $extractedData; |
| 1013 |
} |
| 1014 |
} |
| 1015 |
|
