| 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\Encoding\WinAnsiEncoding; |
| 36 |
use Smalot\PdfParser\Exception\EncodingNotFoundException; |
| 37 |
|
| 38 |
/** |
| 39 |
* Class Font |
| 40 |
*/ |
| 41 |
class Font extends PDFObject |
| 42 |
{ |
| 43 |
public const MISSING = '?'; |
| 44 |
|
| 45 |
/** |
| 46 |
* @var array |
| 47 |
*/ |
| 48 |
protected $table; |
| 49 |
|
| 50 |
/** |
| 51 |
* @var array |
| 52 |
*/ |
| 53 |
protected $tableSizes; |
| 54 |
|
| 55 |
/** |
| 56 |
* Caches results from uchr. |
| 57 |
* |
| 58 |
* @var array |
| 59 |
*/ |
| 60 |
private static $uchrCache = []; |
| 61 |
|
| 62 |
/** |
| 63 |
* In some PDF-files encoding could be referenced by object id but object itself does not contain |
| 64 |
* `/Type /Encoding` in its dictionary. These objects wouldn't be initialized as Encoding in |
| 65 |
* \Smalot\PdfParser\PDFObject::factory() during file parsing (they would be just PDFObject). |
| 66 |
* |
| 67 |
* Therefore, we create an instance of Encoding from them during decoding and cache this value in this property. |
| 68 |
* |
| 69 |
* @var Encoding |
| 70 |
* |
| 71 |
* @see https://github.com/smalot/pdfparser/pull/500 |
| 72 |
*/ |
| 73 |
private $initializedEncodingByPdfObject; |
| 74 |
|
| 75 |
public function init() |
| 76 |
{ |
| 77 |
// Load translate table. |
| 78 |
$this->loadTranslateTable(); |
| 79 |
} |
| 80 |
|
| 81 |
public function getName(): string |
| 82 |
{ |
| 83 |
return $this->has('BaseFont') ? (string) $this->get('BaseFont') : '[Unknown]'; |
| 84 |
} |
| 85 |
|
| 86 |
public function getType(): string |
| 87 |
{ |
| 88 |
return (string) $this->header->get('Subtype'); |
| 89 |
} |
| 90 |
|
| 91 |
public function getDetails(bool $deep = true): array |
| 92 |
{ |
| 93 |
$details = []; |
| 94 |
|
| 95 |
$details['Name'] = $this->getName(); |
| 96 |
$details['Type'] = $this->getType(); |
| 97 |
$details['Encoding'] = ($this->has('Encoding') ? (string) $this->get('Encoding') : 'Ansi'); |
| 98 |
|
| 99 |
$details += parent::getDetails($deep); |
| 100 |
|
| 101 |
return $details; |
| 102 |
} |
| 103 |
|
| 104 |
/** |
| 105 |
* @return string|bool |
| 106 |
*/ |
| 107 |
public function translateChar(string $char, bool $use_default = true) |
| 108 |
{ |
| 109 |
$dec = hexdec(bin2hex($char)); |
| 110 |
|
| 111 |
if (\array_key_exists($dec, $this->table)) { |
| 112 |
return $this->table[$dec]; |
| 113 |
} |
| 114 |
|
| 115 |
// fallback for decoding single-byte ANSI characters that are not in the lookup table |
| 116 |
$fallbackDecoded = $char; |
| 117 |
if ( |
| 118 |
\strlen($char) < 2 |
| 119 |
&& $this->has('Encoding') |
| 120 |
&& $this->get('Encoding') instanceof Encoding |
| 121 |
) { |
| 122 |
try { |
| 123 |
if (WinAnsiEncoding::class === $this->get('Encoding')->__toString()) { |
| 124 |
$fallbackDecoded = self::uchr($dec); |
| 125 |
} |
| 126 |
} catch (EncodingNotFoundException $e) { |
| 127 |
// Encoding->getEncodingClass() throws EncodingNotFoundException when BaseEncoding doesn't exists |
| 128 |
// See table 5.11 on PDF 1.5 specs for more info |
| 129 |
} |
| 130 |
} |
| 131 |
|
| 132 |
return $use_default ? self::MISSING : $fallbackDecoded; |
| 133 |
} |
| 134 |
|
| 135 |
/** |
| 136 |
* Convert unicode character code to "utf-8" encoded string. |
| 137 |
* |
| 138 |
* @param int|float $code Unicode character code. Will be casted to int internally! |
| 139 |
*/ |
| 140 |
public static function uchr($code): string |
| 141 |
{ |
| 142 |
// note: |
| 143 |
// $code was typed as int before, but changed in https://github.com/smalot/pdfparser/pull/623 |
| 144 |
// because in some cases uchr was called with a float instead of an integer. |
| 145 |
$code = (int) $code; |
| 146 |
|
| 147 |
if (!isset(self::$uchrCache[$code])) { |
| 148 |
// html_entity_decode() will not work with UTF-16 or UTF-32 char entities, |
| 149 |
// therefore, we use mb_convert_encoding() instead |
| 150 |
self::$uchrCache[$code] = mb_convert_encoding("&#{$code};", 'UTF-8', 'HTML-ENTITIES'); |
| 151 |
} |
| 152 |
|
| 153 |
return self::$uchrCache[$code]; |
| 154 |
} |
| 155 |
|
| 156 |
/** |
| 157 |
* Init internal chars translation table by ToUnicode CMap. |
| 158 |
*/ |
| 159 |
public function loadTranslateTable(): array |
| 160 |
{ |
| 161 |
if (null !== $this->table) { |
| 162 |
return $this->table; |
| 163 |
} |
| 164 |
|
| 165 |
$this->table = []; |
| 166 |
$this->tableSizes = [ |
| 167 |
'from' => 1, |
| 168 |
'to' => 1, |
| 169 |
]; |
| 170 |
|
| 171 |
if ($this->has('ToUnicode')) { |
| 172 |
$content = $this->get('ToUnicode')->getContent(); |
| 173 |
$matches = []; |
| 174 |
|
| 175 |
// Support for multiple spacerange sections |
| 176 |
if (preg_match_all('/begincodespacerange(?P<sections>.*?)endcodespacerange/s', $content, $matches)) { |
| 177 |
foreach ($matches['sections'] as $section) { |
| 178 |
$regexp = '/<(?P<from>[0-9A-F]+)> *<(?P<to>[0-9A-F]+)>[ \r\n]+/is'; |
| 179 |
|
| 180 |
preg_match_all($regexp, $section, $matches); |
| 181 |
|
| 182 |
$this->tableSizes = [ |
| 183 |
'from' => max(1, \strlen(current($matches['from'])) / 2), |
| 184 |
'to' => max(1, \strlen(current($matches['to'])) / 2), |
| 185 |
]; |
| 186 |
|
| 187 |
break; |
| 188 |
} |
| 189 |
} |
| 190 |
|
| 191 |
// Support for multiple bfchar sections |
| 192 |
if (preg_match_all('/beginbfchar(?P<sections>.*?)endbfchar/s', $content, $matches)) { |
| 193 |
foreach ($matches['sections'] as $section) { |
| 194 |
$regexp = '/<(?P<from>[0-9A-F]+)> *<(?P<to>[0-9A-F]+)>[ \r\n]+/is'; |
| 195 |
|
| 196 |
preg_match_all($regexp, $section, $matches); |
| 197 |
|
| 198 |
$this->tableSizes['from'] = max(1, \strlen(current($matches['from'])) / 2); |
| 199 |
|
| 200 |
foreach ($matches['from'] as $key => $from) { |
| 201 |
$parts = preg_split( |
| 202 |
'/([0-9A-F]{4})/i', |
| 203 |
$matches['to'][$key], |
| 204 |
0, |
| 205 |
\PREG_SPLIT_NO_EMPTY | \PREG_SPLIT_DELIM_CAPTURE |
| 206 |
); |
| 207 |
$text = ''; |
| 208 |
foreach ($parts as $part) { |
| 209 |
$text .= self::uchr(hexdec($part)); |
| 210 |
} |
| 211 |
$this->table[hexdec($from)] = $text; |
| 212 |
} |
| 213 |
} |
| 214 |
} |
| 215 |
|
| 216 |
// Support for multiple bfrange sections |
| 217 |
if (preg_match_all('/beginbfrange(?P<sections>.*?)endbfrange/s', $content, $matches)) { |
| 218 |
foreach ($matches['sections'] as $section) { |
| 219 |
// Support for : <srcCode1> <srcCode2> <dstString> |
| 220 |
$regexp = '/<(?P<from>[0-9A-F]+)> *<(?P<to>[0-9A-F]+)> *<(?P<offset>[0-9A-F]+)>[ \r\n]+/is'; |
| 221 |
|
| 222 |
preg_match_all($regexp, $section, $matches); |
| 223 |
|
| 224 |
foreach ($matches['from'] as $key => $from) { |
| 225 |
$char_from = hexdec($from); |
| 226 |
$char_to = hexdec($matches['to'][$key]); |
| 227 |
$offset = hexdec($matches['offset'][$key]); |
| 228 |
|
| 229 |
for ($char = $char_from; $char <= $char_to; ++$char) { |
| 230 |
$this->table[$char] = self::uchr($char - $char_from + $offset); |
| 231 |
} |
| 232 |
} |
| 233 |
|
| 234 |
// Support for : <srcCode1> <srcCodeN> [<dstString1> <dstString2> ... <dstStringN>] |
| 235 |
// Some PDF file has 2-byte Unicode values on new lines > added \r\n |
| 236 |
$regexp = '/<(?P<from>[0-9A-F]+)> *<(?P<to>[0-9A-F]+)> *\[(?P<strings>[\r\n<>0-9A-F ]+)\][ \r\n]+/is'; |
| 237 |
|
| 238 |
preg_match_all($regexp, $section, $matches); |
| 239 |
|
| 240 |
foreach ($matches['from'] as $key => $from) { |
| 241 |
$char_from = hexdec($from); |
| 242 |
$strings = []; |
| 243 |
|
| 244 |
preg_match_all('/<(?P<string>[0-9A-F]+)> */is', $matches['strings'][$key], $strings); |
| 245 |
|
| 246 |
foreach ($strings['string'] as $position => $string) { |
| 247 |
$parts = preg_split( |
| 248 |
'/([0-9A-F]{4})/i', |
| 249 |
$string, |
| 250 |
0, |
| 251 |
\PREG_SPLIT_NO_EMPTY | \PREG_SPLIT_DELIM_CAPTURE |
| 252 |
); |
| 253 |
$text = ''; |
| 254 |
foreach ($parts as $part) { |
| 255 |
$text .= self::uchr(hexdec($part)); |
| 256 |
} |
| 257 |
$this->table[$char_from + $position] = $text; |
| 258 |
} |
| 259 |
} |
| 260 |
} |
| 261 |
} |
| 262 |
} |
| 263 |
|
| 264 |
return $this->table; |
| 265 |
} |
| 266 |
|
| 267 |
/** |
| 268 |
* Set custom char translation table where: |
| 269 |
* - key - integer character code; |
| 270 |
* - value - "utf-8" encoded value; |
| 271 |
* |
| 272 |
* @return void |
| 273 |
*/ |
| 274 |
public function setTable(array $table) |
| 275 |
{ |
| 276 |
$this->table = $table; |
| 277 |
} |
| 278 |
|
| 279 |
/** |
| 280 |
* Calculate text width with data from header 'Widths'. If width of character is not found then character is added to missing array. |
| 281 |
*/ |
| 282 |
public function calculateTextWidth(string $text, ?array &$missing = null): ?float |
| 283 |
{ |
| 284 |
$index_map = array_flip($this->table); |
| 285 |
$details = $this->getDetails(); |
| 286 |
|
| 287 |
// Usually, Widths key is set in $details array, but if it isn't use an empty array instead. |
| 288 |
$widths = $details['Widths'] ?? []; |
| 289 |
|
| 290 |
/* |
| 291 |
* Widths array is zero indexed but table is not. We must map them based on FirstChar and LastChar |
| 292 |
* |
| 293 |
* Note: Without the change you would see warnings in PHP 8.4 because the values of FirstChar or LastChar |
| 294 |
* can be null sometimes. |
| 295 |
*/ |
| 296 |
$width_map = array_flip(range((int) $details['FirstChar'], (int) $details['LastChar'])); |
| 297 |
|
| 298 |
$width = null; |
| 299 |
$missing = []; |
| 300 |
$textLength = mb_strlen($text); |
| 301 |
for ($i = 0; $i < $textLength; ++$i) { |
| 302 |
$char = mb_substr($text, $i, 1); |
| 303 |
if ( |
| 304 |
!\array_key_exists($char, $index_map) |
| 305 |
|| !\array_key_exists($index_map[$char], $width_map) |
| 306 |
|| !\array_key_exists($width_map[$index_map[$char]], $widths) |
| 307 |
) { |
| 308 |
$missing[] = $char; |
| 309 |
continue; |
| 310 |
} |
| 311 |
$width_index = $width_map[$index_map[$char]]; |
| 312 |
$width += $widths[$width_index]; |
| 313 |
} |
| 314 |
|
| 315 |
return $width; |
| 316 |
} |
| 317 |
|
| 318 |
/** |
| 319 |
* Decode hexadecimal encoded string. If $add_braces is true result value would be wrapped by parentheses. |
| 320 |
*/ |
| 321 |
public static function decodeHexadecimal(string $hexa, bool $add_braces = false): string |
| 322 |
{ |
| 323 |
// Special shortcut for XML content. |
| 324 |
if (false !== stripos($hexa, '<?xml')) { |
| 325 |
return $hexa; |
| 326 |
} |
| 327 |
|
| 328 |
$text = ''; |
| 329 |
$parts = preg_split('/(<[a-f0-9\s]+>)/si', $hexa, -1, \PREG_SPLIT_NO_EMPTY | \PREG_SPLIT_DELIM_CAPTURE); |
| 330 |
|
| 331 |
foreach ($parts as $part) { |
| 332 |
if (preg_match('/^<[a-f0-9\s]+>$/si', $part)) { |
| 333 |
// strip whitespace |
| 334 |
$part = preg_replace("/\s/", '', $part); |
| 335 |
$part = trim($part, '<>'); |
| 336 |
if ($add_braces) { |
| 337 |
$text .= '('; |
| 338 |
} |
| 339 |
|
| 340 |
$part = pack('H*', $part); |
| 341 |
$text .= ($add_braces ? preg_replace('/\\\/s', '\\\\\\', $part) : $part); |
| 342 |
|
| 343 |
if ($add_braces) { |
| 344 |
$text .= ')'; |
| 345 |
} |
| 346 |
} else { |
| 347 |
$text .= $part; |
| 348 |
} |
| 349 |
} |
| 350 |
|
| 351 |
return $text; |
| 352 |
} |
| 353 |
|
| 354 |
/** |
| 355 |
* Decode string with octal-decoded chunks. |
| 356 |
*/ |
| 357 |
public static function decodeOctal(string $text): string |
| 358 |
{ |
| 359 |
// Replace all double backslashes \\ with a special string |
| 360 |
$text = strtr($text, ['\\\\' => '[**pdfparserdblslsh**]']); |
| 361 |
|
| 362 |
// Now we can replace all octal codes without worrying about |
| 363 |
// escaped backslashes |
| 364 |
$text = preg_replace_callback('/\\\\([0-7]{1,3})/', function ($m) { |
| 365 |
return \chr(octdec($m[1])); |
| 366 |
}, $text); |
| 367 |
|
| 368 |
// Unescape any parentheses |
| 369 |
$text = str_replace(['\\(', '\\)'], ['(', ')'], $text); |
| 370 |
|
| 371 |
// Replace instances of the special string with a single backslash |
| 372 |
return str_replace('[**pdfparserdblslsh**]', '\\', $text); |
| 373 |
} |
| 374 |
|
| 375 |
/** |
| 376 |
* Decode string with html entity encoded chars. |
| 377 |
*/ |
| 378 |
public static function decodeEntities(string $text): string |
| 379 |
{ |
| 380 |
return preg_replace_callback('/#([0-9a-f]{2})/i', function ($m) { |
| 381 |
return \chr(hexdec($m[1])); |
| 382 |
}, $text); |
| 383 |
} |
| 384 |
|
| 385 |
/** |
| 386 |
* Check if given string is Unicode text (by BOM); |
| 387 |
* If true - decode to "utf-8" encoded string. |
| 388 |
* Otherwise - return text as is. |
| 389 |
* |
| 390 |
* @todo Rename in next major release to make the name correspond to reality (for ex. decodeIfUnicode()) |
| 391 |
*/ |
| 392 |
public static function decodeUnicode(string $text): string |
| 393 |
{ |
| 394 |
if ("\xFE\xFF" === substr($text, 0, 2)) { |
| 395 |
// Strip U+FEFF byte order marker. |
| 396 |
$decode = substr($text, 2); |
| 397 |
$text = ''; |
| 398 |
$length = \strlen($decode); |
| 399 |
|
| 400 |
for ($i = 0; $i < $length; $i += 2) { |
| 401 |
$text .= self::uchr(hexdec(bin2hex(substr($decode, $i, 2)))); |
| 402 |
} |
| 403 |
} |
| 404 |
|
| 405 |
return $text; |
| 406 |
} |
| 407 |
|
| 408 |
/** |
| 409 |
* @todo Deprecated, use $this->config->getFontSpaceLimit() instead. |
| 410 |
*/ |
| 411 |
protected function getFontSpaceLimit(): int |
| 412 |
{ |
| 413 |
return $this->config->getFontSpaceLimit(); |
| 414 |
} |
| 415 |
|
| 416 |
/** |
| 417 |
* Decode text by commands array. |
| 418 |
*/ |
| 419 |
public function decodeText(array $commands, float $fontFactor = 4): string |
| 420 |
{ |
| 421 |
$word_position = 0; |
| 422 |
$words = []; |
| 423 |
$font_space = $this->getFontSpaceLimit() * abs($fontFactor) / 4; |
| 424 |
|
| 425 |
foreach ($commands as $command) { |
| 426 |
switch ($command[PDFObject::TYPE]) { |
| 427 |
case 'n': |
| 428 |
$offset = (float) trim($command[PDFObject::COMMAND]); |
| 429 |
if ($offset - (float) $font_space < 0) { |
| 430 |
$word_position = \count($words); |
| 431 |
} |
| 432 |
continue 2; |
| 433 |
case '<': |
| 434 |
// Decode hexadecimal. |
| 435 |
$text = self::decodeHexadecimal('<'.$command[PDFObject::COMMAND].'>'); |
| 436 |
break; |
| 437 |
|
| 438 |
default: |
| 439 |
// Decode octal (if necessary). |
| 440 |
$text = self::decodeOctal($command[PDFObject::COMMAND]); |
| 441 |
} |
| 442 |
|
| 443 |
// replace escaped chars |
| 444 |
$text = str_replace( |
| 445 |
['\\\\', '\(', '\)', '\n', '\r', '\t', '\f', '\ ', '\b'], |
| 446 |
[\chr(92), \chr(40), \chr(41), \chr(10), \chr(13), \chr(9), \chr(12), \chr(32), \chr(8)], |
| 447 |
$text |
| 448 |
); |
| 449 |
|
| 450 |
// add content to result string |
| 451 |
if (isset($words[$word_position])) { |
| 452 |
$words[$word_position] .= $text; |
| 453 |
} else { |
| 454 |
$words[$word_position] = $text; |
| 455 |
} |
| 456 |
} |
| 457 |
|
| 458 |
foreach ($words as &$word) { |
| 459 |
$word = $this->decodeContent($word); |
| 460 |
$word = str_replace("\t", ' ', $word); |
| 461 |
} |
| 462 |
|
| 463 |
// Remove internal "words" that are just spaces, but leave them |
| 464 |
// if they are at either end of the array of words. This fixes, |
| 465 |
// for example, lines that are justified to fill |
| 466 |
// a whole row. |
| 467 |
for ($x = \count($words) - 2; $x >= 1; --$x) { |
| 468 |
if ('' === trim($words[$x], ' ')) { |
| 469 |
unset($words[$x]); |
| 470 |
} |
| 471 |
} |
| 472 |
$words = array_values($words); |
| 473 |
|
| 474 |
// Cut down on the number of unnecessary internal spaces by |
| 475 |
// imploding the string on the null byte, and checking if the |
| 476 |
// text includes extra spaces on either side. If so, merge |
| 477 |
// where appropriate. |
| 478 |
$words = implode("\x00\x00", $words); |
| 479 |
$words = str_replace( |
| 480 |
[" \x00\x00 ", "\x00\x00 ", " \x00\x00", "\x00\x00"], |
| 481 |
[' ', ' ', ' ', ' '], |
| 482 |
$words |
| 483 |
); |
| 484 |
|
| 485 |
return $words; |
| 486 |
} |
| 487 |
|
| 488 |
/** |
| 489 |
* Decode given $text to "utf-8" encoded string. |
| 490 |
* |
| 491 |
* @param bool $unicode This parameter is deprecated and might be removed in a future release |
| 492 |
*/ |
| 493 |
public function decodeContent(string $text, ?bool &$unicode = null): string |
| 494 |
{ |
| 495 |
// If this string begins with a UTF-16BE BOM, then decode it |
| 496 |
// directly as Unicode |
| 497 |
if ("\xFE\xFF" === substr($text, 0, 2)) { |
| 498 |
return $this->decodeUnicode($text); |
| 499 |
} |
| 500 |
|
| 501 |
if ($this->has('ToUnicode')) { |
| 502 |
return $this->decodeContentByToUnicodeCMapOrDescendantFonts($text); |
| 503 |
} |
| 504 |
|
| 505 |
if ($this->has('Encoding')) { |
| 506 |
$result = $this->decodeContentByEncoding($text); |
| 507 |
|
| 508 |
if (null !== $result) { |
| 509 |
return $result; |
| 510 |
} |
| 511 |
} |
| 512 |
|
| 513 |
return $this->decodeContentByAutodetectIfNecessary($text); |
| 514 |
} |
| 515 |
|
| 516 |
/** |
| 517 |
* First try to decode $text by ToUnicode CMap. |
| 518 |
* If char translation not found in ToUnicode CMap tries: |
| 519 |
* - If DescendantFonts exists tries to decode char by one of that fonts. |
| 520 |
* - If have no success to decode by DescendantFonts interpret $text as a string with "Windows-1252" encoding. |
| 521 |
* - If DescendantFonts does not exist just return "?" as decoded char. |
| 522 |
* |
| 523 |
* @todo Seems this is invalid algorithm that do not follow pdf-format specification. Must be rewritten. |
| 524 |
*/ |
| 525 |
private function decodeContentByToUnicodeCMapOrDescendantFonts(string $text): string |
| 526 |
{ |
| 527 |
$bytes = $this->tableSizes['from']; |
| 528 |
|
| 529 |
if ($bytes) { |
| 530 |
$result = ''; |
| 531 |
$length = \strlen($text); |
| 532 |
|
| 533 |
for ($i = 0; $i < $length; $i += $bytes) { |
| 534 |
$char = substr($text, $i, $bytes); |
| 535 |
|
| 536 |
if (false !== ($decoded = $this->translateChar($char, false))) { |
| 537 |
$char = $decoded; |
| 538 |
} elseif ($this->has('DescendantFonts')) { |
| 539 |
if ($this->get('DescendantFonts') instanceof PDFObject) { |
| 540 |
$fonts = $this->get('DescendantFonts')->getHeader()->getElements(); |
| 541 |
} else { |
| 542 |
$fonts = $this->get('DescendantFonts')->getContent(); |
| 543 |
} |
| 544 |
$decoded = false; |
| 545 |
|
| 546 |
foreach ($fonts as $font) { |
| 547 |
if ($font instanceof self) { |
| 548 |
if (false !== ($decoded = $font->translateChar($char, false))) { |
| 549 |
$decoded = mb_convert_encoding($decoded, 'UTF-8', 'Windows-1252'); |
| 550 |
break; |
| 551 |
} |
| 552 |
} |
| 553 |
} |
| 554 |
|
| 555 |
if (false !== $decoded) { |
| 556 |
$char = $decoded; |
| 557 |
} else { |
| 558 |
$char = mb_convert_encoding($char, 'UTF-8', 'Windows-1252'); |
| 559 |
} |
| 560 |
} else { |
| 561 |
$char = self::MISSING; |
| 562 |
} |
| 563 |
|
| 564 |
$result .= $char; |
| 565 |
} |
| 566 |
|
| 567 |
$text = $result; |
| 568 |
} |
| 569 |
|
| 570 |
return $text; |
| 571 |
} |
| 572 |
|
| 573 |
/** |
| 574 |
* Decode content by any type of Encoding (dictionary's item) instance. |
| 575 |
*/ |
| 576 |
private function decodeContentByEncoding(string $text): ?string |
| 577 |
{ |
| 578 |
$encoding = $this->get('Encoding'); |
| 579 |
|
| 580 |
// When Encoding referenced by object id (/Encoding 520 0 R) but object itself does not contain `/Type /Encoding` in it's dictionary. |
| 581 |
if ($encoding instanceof PDFObject) { |
| 582 |
$encoding = $this->getInitializedEncodingByPdfObject($encoding); |
| 583 |
} |
| 584 |
|
| 585 |
// When Encoding referenced by object id (/Encoding 520 0 R) but object itself contains `/Type /Encoding` in it's dictionary. |
| 586 |
if ($encoding instanceof Encoding) { |
| 587 |
return $this->decodeContentByEncodingEncoding($text, $encoding); |
| 588 |
} |
| 589 |
|
| 590 |
// When Encoding is just string (/Encoding /WinAnsiEncoding) |
| 591 |
if ($encoding instanceof Element) { // todo: ElementString class must by used? |
| 592 |
return $this->decodeContentByEncodingElement($text, $encoding); |
| 593 |
} |
| 594 |
|
| 595 |
// don't double-encode strings already in UTF-8 |
| 596 |
if (!mb_check_encoding($text, 'UTF-8')) { |
| 597 |
return mb_convert_encoding($text, 'UTF-8', 'Windows-1252'); |
| 598 |
} |
| 599 |
|
| 600 |
return $text; |
| 601 |
} |
| 602 |
|
| 603 |
/** |
| 604 |
* Returns already created or create a new one if not created before Encoding instance by PDFObject instance. |
| 605 |
*/ |
| 606 |
private function getInitializedEncodingByPdfObject(PDFObject $PDFObject): Encoding |
| 607 |
{ |
| 608 |
if (!$this->initializedEncodingByPdfObject) { |
| 609 |
$this->initializedEncodingByPdfObject = $this->createInitializedEncodingByPdfObject($PDFObject); |
| 610 |
} |
| 611 |
|
| 612 |
return $this->initializedEncodingByPdfObject; |
| 613 |
} |
| 614 |
|
| 615 |
/** |
| 616 |
* Decode content when $encoding (given by $this->get('Encoding')) is instance of Encoding. |
| 617 |
*/ |
| 618 |
private function decodeContentByEncodingEncoding(string $text, Encoding $encoding): string |
| 619 |
{ |
| 620 |
$result = ''; |
| 621 |
$length = \strlen($text); |
| 622 |
|
| 623 |
for ($i = 0; $i < $length; ++$i) { |
| 624 |
$dec_av = hexdec(bin2hex($text[$i])); |
| 625 |
$dec_ap = $encoding->translateChar($dec_av); |
| 626 |
$result .= self::uchr($dec_ap ?? $dec_av); |
| 627 |
} |
| 628 |
|
| 629 |
return $result; |
| 630 |
} |
| 631 |
|
| 632 |
/** |
| 633 |
* Decode content when $encoding (given by $this->get('Encoding')) is instance of Element. |
| 634 |
*/ |
| 635 |
private function decodeContentByEncodingElement(string $text, Element $encoding): ?string |
| 636 |
{ |
| 637 |
$pdfEncodingName = $encoding->getContent(); |
| 638 |
|
| 639 |
// mb_convert_encoding does not support MacRoman/macintosh, |
| 640 |
// so we use iconv() here |
| 641 |
$iconvEncodingName = $this->getIconvEncodingNameOrNullByPdfEncodingName($pdfEncodingName); |
| 642 |
|
| 643 |
return $iconvEncodingName ? iconv($iconvEncodingName, 'UTF-8//TRANSLIT//IGNORE', $text) : null; |
| 644 |
} |
| 645 |
|
| 646 |
/** |
| 647 |
* Convert PDF encoding name to iconv-known encoding name. |
| 648 |
*/ |
| 649 |
private function getIconvEncodingNameOrNullByPdfEncodingName(string $pdfEncodingName): ?string |
| 650 |
{ |
| 651 |
$pdfToIconvEncodingNameMap = [ |
| 652 |
'StandardEncoding' => 'ISO-8859-1', |
| 653 |
'MacRomanEncoding' => 'MACINTOSH', |
| 654 |
'WinAnsiEncoding' => 'CP1252', |
| 655 |
]; |
| 656 |
|
| 657 |
return \array_key_exists($pdfEncodingName, $pdfToIconvEncodingNameMap) |
| 658 |
? $pdfToIconvEncodingNameMap[$pdfEncodingName] |
| 659 |
: null; |
| 660 |
} |
| 661 |
|
| 662 |
/** |
| 663 |
* If string seems like "utf-8" encoded string do nothing and just return given string as is. |
| 664 |
* Otherwise, interpret string as "Window-1252" encoded string. |
| 665 |
* |
| 666 |
* @return string|false |
| 667 |
*/ |
| 668 |
private function decodeContentByAutodetectIfNecessary(string $text) |
| 669 |
{ |
| 670 |
if (mb_check_encoding($text, 'UTF-8')) { |
| 671 |
return $text; |
| 672 |
} |
| 673 |
|
| 674 |
return mb_convert_encoding($text, 'UTF-8', 'Windows-1252'); |
| 675 |
// todo: Why exactly `Windows-1252` used? |
| 676 |
} |
| 677 |
|
| 678 |
/** |
| 679 |
* Create Encoding instance by PDFObject instance and init it. |
| 680 |
*/ |
| 681 |
private function createInitializedEncodingByPdfObject(PDFObject $PDFObject): Encoding |
| 682 |
{ |
| 683 |
$encoding = $this->createEncodingByPdfObject($PDFObject); |
| 684 |
$encoding->init(); |
| 685 |
|
| 686 |
return $encoding; |
| 687 |
} |
| 688 |
|
| 689 |
/** |
| 690 |
* Create Encoding instance by PDFObject instance (without init). |
| 691 |
*/ |
| 692 |
private function createEncodingByPdfObject(PDFObject $PDFObject): Encoding |
| 693 |
{ |
| 694 |
$document = $PDFObject->getDocument(); |
| 695 |
$header = $PDFObject->getHeader(); |
| 696 |
$content = $PDFObject->getContent(); |
| 697 |
$config = $PDFObject->getConfig(); |
| 698 |
|
| 699 |
return new Encoding($document, $header, $content, $config); |
| 700 |
} |
| 701 |
} |
| 702 |
|
