Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
100.00% |
146 / 146 |
|
100.00% |
14 / 14 |
CRAP | |
100.00% |
1 / 1 |
| Tokenizer | |
100.00% |
146 / 146 |
|
100.00% |
14 / 14 |
106 | |
100.00% |
1 / 1 |
| __construct | |
100.00% |
3 / 3 |
|
100.00% |
1 / 1 |
1 | |||
| getPosition | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| setPosition | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getData | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| isWhitespace | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
6 | |||
| isDelimiter | |
100.00% |
3 / 3 |
|
100.00% |
1 / 1 |
10 | |||
| skipWhitespaceAndComments | |
100.00% |
8 / 8 |
|
100.00% |
1 / 1 |
7 | |||
| next | |
100.00% |
25 / 25 |
|
100.00% |
1 / 1 |
15 | |||
| readName | |
100.00% |
12 / 12 |
|
100.00% |
1 / 1 |
8 | |||
| readLiteralString | |
100.00% |
51 / 51 |
|
100.00% |
1 / 1 |
26 | |||
| readAngleOpen | |
100.00% |
15 / 15 |
|
100.00% |
1 / 1 |
10 | |||
| readAngleClose | |
100.00% |
5 / 5 |
|
100.00% |
1 / 1 |
3 | |||
| readNumber | |
100.00% |
11 / 11 |
|
100.00% |
1 / 1 |
12 | |||
| readKeyword | |
100.00% |
9 / 9 |
|
100.00% |
1 / 1 |
5 | |||
| 1 | <?php |
| 2 | declare(strict_types=1); |
| 3 | /** |
| 4 | * Pop PHP Framework (https://www.popphp.org/) |
| 5 | * |
| 6 | * @link https://github.com/popphp/popphp-framework |
| 7 | * @author Nick Sagona, III <nick@popphp.org> |
| 8 | * @copyright Copyright (c) 2009-2026 Nick Sagona, III |
| 9 | * @license https://www.popphp.org/license New BSD License |
| 10 | */ |
| 11 | |
| 12 | /** |
| 13 | * @namespace |
| 14 | */ |
| 15 | namespace Pop\Pdf\Extract; |
| 16 | |
| 17 | /** |
| 18 | * Pdf extract tokenizer class |
| 19 | * |
| 20 | * @category Pop |
| 21 | * @package Pop\Pdf |
| 22 | * @author Nick Sagona, III <nick@popphp.org> |
| 23 | * @copyright Copyright (c) 2009-2026 Nick Sagona, III |
| 24 | * @license https://www.popphp.org/license New BSD License |
| 25 | * @version 6.0.0 |
| 26 | */ |
| 27 | class Tokenizer |
| 28 | { |
| 29 | |
| 30 | /** |
| 31 | * Raw data being tokenized |
| 32 | * @var string |
| 33 | */ |
| 34 | protected string $data; |
| 35 | |
| 36 | /** |
| 37 | * Length of the raw data |
| 38 | * @var int |
| 39 | */ |
| 40 | protected int $length; |
| 41 | |
| 42 | /** |
| 43 | * Current byte offset into the data |
| 44 | * @var int |
| 45 | */ |
| 46 | protected int $pos; |
| 47 | |
| 48 | /** |
| 49 | * Constructor |
| 50 | * |
| 51 | * Instantiate a tokenizer. |
| 52 | * |
| 53 | * @param string $data |
| 54 | * @param int $pos |
| 55 | */ |
| 56 | public function __construct(string $data, int $pos = 0) |
| 57 | { |
| 58 | $this->data = $data; |
| 59 | $this->length = strlen($data); |
| 60 | $this->pos = $pos; |
| 61 | } |
| 62 | |
| 63 | /** |
| 64 | * Get the current byte offset |
| 65 | * |
| 66 | * @return int |
| 67 | */ |
| 68 | public function getPosition(): int |
| 69 | { |
| 70 | return $this->pos; |
| 71 | } |
| 72 | |
| 73 | /** |
| 74 | * Set the current byte offset |
| 75 | * |
| 76 | * @param int $pos |
| 77 | * @return void |
| 78 | */ |
| 79 | public function setPosition(int $pos): void |
| 80 | { |
| 81 | $this->pos = $pos; |
| 82 | } |
| 83 | |
| 84 | /** |
| 85 | * Get the raw data being tokenized |
| 86 | * |
| 87 | * @return string |
| 88 | */ |
| 89 | public function getData(): string |
| 90 | { |
| 91 | return $this->data; |
| 92 | } |
| 93 | |
| 94 | /** |
| 95 | * Determine if a byte is PDF whitespace |
| 96 | * |
| 97 | * @param string $c |
| 98 | * @return bool |
| 99 | */ |
| 100 | public static function isWhitespace(string $c): bool |
| 101 | { |
| 102 | return ($c === ' ') || ($c === "\t") || ($c === "\r") || ($c === "\n") || ($c === "\f") || ($c === "\0"); |
| 103 | } |
| 104 | |
| 105 | /** |
| 106 | * Determine if a byte is a PDF delimiter |
| 107 | * |
| 108 | * @param string $c |
| 109 | * @return bool |
| 110 | */ |
| 111 | public static function isDelimiter(string $c): bool |
| 112 | { |
| 113 | return ($c === '(') || ($c === ')') || ($c === '<') || ($c === '>') || |
| 114 | ($c === '[') || ($c === ']') || ($c === '{') || ($c === '}') || |
| 115 | ($c === '/') || ($c === '%'); |
| 116 | } |
| 117 | |
| 118 | /** |
| 119 | * Skip whitespace and comments |
| 120 | * |
| 121 | * @return void |
| 122 | */ |
| 123 | protected function skipWhitespaceAndComments(): void |
| 124 | { |
| 125 | while ($this->pos < $this->length) { |
| 126 | $c = $this->data[$this->pos]; |
| 127 | if (self::isWhitespace($c)) { |
| 128 | $this->pos++; |
| 129 | } elseif ($c === '%') { |
| 130 | while (($this->pos < $this->length) && ($this->data[$this->pos] !== "\n") && ($this->data[$this->pos] !== "\r")) { |
| 131 | $this->pos++; |
| 132 | } |
| 133 | } else { |
| 134 | break; |
| 135 | } |
| 136 | } |
| 137 | } |
| 138 | |
| 139 | /** |
| 140 | * Read the next token |
| 141 | * |
| 142 | * @return array |
| 143 | */ |
| 144 | public function next(): array |
| 145 | { |
| 146 | // Loop rather than recurse for tokens that may need to skip over |
| 147 | // themselves and continue (e.g. a stray '>' byte) - a malicious |
| 148 | // input with many such bytes in a row must not grow the call stack. |
| 149 | while (true) { |
| 150 | $this->skipWhitespaceAndComments(); |
| 151 | |
| 152 | if ($this->pos >= $this->length) { |
| 153 | return ['type' => 'eof', 'value' => null]; |
| 154 | } |
| 155 | |
| 156 | $c = $this->data[$this->pos]; |
| 157 | |
| 158 | if ($c === '/') { |
| 159 | return $this->readName(); |
| 160 | } elseif ($c === '(') { |
| 161 | return $this->readLiteralString(); |
| 162 | } elseif ($c === '<') { |
| 163 | return $this->readAngleOpen(); |
| 164 | } elseif ($c === '>') { |
| 165 | $token = $this->readAngleClose(); |
| 166 | if ($token === null) { |
| 167 | // Stray '>' - already skipped by readAngleClose(); loop |
| 168 | // back around to tokenize whatever follows. |
| 169 | continue; |
| 170 | } |
| 171 | return $token; |
| 172 | } elseif ($c === '[') { |
| 173 | $this->pos++; |
| 174 | return ['type' => 'array_start', 'value' => '[']; |
| 175 | } elseif ($c === ']') { |
| 176 | $this->pos++; |
| 177 | return ['type' => 'array_end', 'value' => ']']; |
| 178 | } elseif (($c === '+') || ($c === '-') || ($c === '.') || (($c >= '0') && ($c <= '9'))) { |
| 179 | return $this->readNumber(); |
| 180 | } else { |
| 181 | return $this->readKeyword(); |
| 182 | } |
| 183 | } |
| 184 | } |
| 185 | |
| 186 | /** |
| 187 | * Read a name token |
| 188 | * |
| 189 | * @return array |
| 190 | */ |
| 191 | protected function readName(): array |
| 192 | { |
| 193 | $this->pos++; // skip '/' |
| 194 | $name = ''; |
| 195 | |
| 196 | while (($this->pos < $this->length) && |
| 197 | (!self::isWhitespace($this->data[$this->pos])) && |
| 198 | (!self::isDelimiter($this->data[$this->pos]))) { |
| 199 | if (($this->data[$this->pos] === '#') && (($this->pos + 2) < $this->length) && |
| 200 | ctype_xdigit($this->data[$this->pos + 1]) && ctype_xdigit($this->data[$this->pos + 2])) { |
| 201 | $name .= chr((int) hexdec(substr($this->data, $this->pos + 1, 2))); |
| 202 | $this->pos += 3; |
| 203 | } else { |
| 204 | $name .= $this->data[$this->pos]; |
| 205 | $this->pos++; |
| 206 | } |
| 207 | } |
| 208 | |
| 209 | return ['type' => 'name', 'value' => $name]; |
| 210 | } |
| 211 | |
| 212 | /** |
| 213 | * Read a literal string token |
| 214 | * |
| 215 | * @return array |
| 216 | */ |
| 217 | protected function readLiteralString(): array |
| 218 | { |
| 219 | $this->pos++; // skip '(' |
| 220 | $depth = 1; |
| 221 | $value = ''; |
| 222 | |
| 223 | while (($this->pos < $this->length) && ($depth > 0)) { |
| 224 | $c = $this->data[$this->pos]; |
| 225 | |
| 226 | if ($c === '\\') { |
| 227 | $this->pos++; |
| 228 | if ($this->pos >= $this->length) { |
| 229 | break; |
| 230 | } |
| 231 | $esc = $this->data[$this->pos]; |
| 232 | |
| 233 | if ($esc === 'n') { |
| 234 | $value .= "\n"; $this->pos++; |
| 235 | } elseif ($esc === 'r') { |
| 236 | $value .= "\r"; $this->pos++; |
| 237 | } elseif ($esc === 't') { |
| 238 | $value .= "\t"; $this->pos++; |
| 239 | } elseif ($esc === 'b') { |
| 240 | $value .= "\x08"; $this->pos++; |
| 241 | } elseif ($esc === 'f') { |
| 242 | $value .= "\x0C"; $this->pos++; |
| 243 | } elseif (($esc === '(') || ($esc === ')') || ($esc === '\\')) { |
| 244 | $value .= $esc; $this->pos++; |
| 245 | } elseif ($esc === "\n") { |
| 246 | $this->pos++; |
| 247 | } elseif ($esc === "\r") { |
| 248 | $this->pos++; |
| 249 | if (($this->pos < $this->length) && ($this->data[$this->pos] === "\n")) { |
| 250 | $this->pos++; |
| 251 | } |
| 252 | } elseif (($esc >= '0') && ($esc <= '7')) { |
| 253 | $octal = ''; |
| 254 | $count = 0; |
| 255 | while (($count < 3) && ($this->pos < $this->length) && |
| 256 | ($this->data[$this->pos] >= '0') && ($this->data[$this->pos] <= '7')) { |
| 257 | $octal .= $this->data[$this->pos]; |
| 258 | $this->pos++; |
| 259 | $count++; |
| 260 | } |
| 261 | $value .= chr(((int) octdec($octal)) & 0xFF); |
| 262 | } else { |
| 263 | $value .= $esc; |
| 264 | $this->pos++; |
| 265 | } |
| 266 | } elseif ($c === '(') { |
| 267 | $depth++; |
| 268 | $value .= $c; |
| 269 | $this->pos++; |
| 270 | } elseif ($c === ')') { |
| 271 | $depth--; |
| 272 | $this->pos++; |
| 273 | if ($depth > 0) { |
| 274 | $value .= $c; |
| 275 | } |
| 276 | } else { |
| 277 | $value .= $c; |
| 278 | $this->pos++; |
| 279 | } |
| 280 | } |
| 281 | |
| 282 | return ['type' => 'string', 'value' => $value]; |
| 283 | } |
| 284 | |
| 285 | /** |
| 286 | * Read a token starting with '<' (hex string or dict-open) |
| 287 | * |
| 288 | * @return array |
| 289 | */ |
| 290 | protected function readAngleOpen(): array |
| 291 | { |
| 292 | if ((($this->pos + 1) < $this->length) && ($this->data[$this->pos + 1] === '<')) { |
| 293 | $this->pos += 2; |
| 294 | return ['type' => 'dict_start', 'value' => '<<']; |
| 295 | } |
| 296 | |
| 297 | $this->pos++; // skip '<' |
| 298 | $hex = ''; |
| 299 | |
| 300 | while (($this->pos < $this->length) && ($this->data[$this->pos] !== '>')) { |
| 301 | $c = $this->data[$this->pos]; |
| 302 | if (ctype_xdigit($c)) { |
| 303 | $hex .= $c; |
| 304 | } |
| 305 | $this->pos++; |
| 306 | } |
| 307 | |
| 308 | if (($this->pos < $this->length) && ($this->data[$this->pos] === '>')) { |
| 309 | $this->pos++; |
| 310 | } |
| 311 | |
| 312 | if ((strlen($hex) % 2) !== 0) { |
| 313 | $hex .= '0'; |
| 314 | } |
| 315 | |
| 316 | return ['type' => 'string', 'value' => ($hex === '') ? '' : (string) hex2bin($hex)]; |
| 317 | } |
| 318 | |
| 319 | /** |
| 320 | * Read a token starting with '>' (dict-close, or a stray byte to skip) |
| 321 | * |
| 322 | * @return ?array |
| 323 | */ |
| 324 | protected function readAngleClose(): ?array |
| 325 | { |
| 326 | if ((($this->pos + 1) < $this->length) && ($this->data[$this->pos + 1] === '>')) { |
| 327 | $this->pos += 2; |
| 328 | return ['type' => 'dict_end', 'value' => '>>']; |
| 329 | } |
| 330 | |
| 331 | // Stray '>' outside of a hex string / dict-close - skip it. Return |
| 332 | // null so next() loops back around instead of recursing here. |
| 333 | $this->pos++; |
| 334 | |
| 335 | return null; |
| 336 | } |
| 337 | |
| 338 | /** |
| 339 | * Read a number token |
| 340 | * |
| 341 | * @return array |
| 342 | */ |
| 343 | protected function readNumber(): array |
| 344 | { |
| 345 | $start = $this->pos; |
| 346 | |
| 347 | if (($this->data[$this->pos] === '+') || ($this->data[$this->pos] === '-')) { |
| 348 | $this->pos++; |
| 349 | } |
| 350 | |
| 351 | while (($this->pos < $this->length) && |
| 352 | ((($this->data[$this->pos] >= '0') && ($this->data[$this->pos] <= '9')) || ($this->data[$this->pos] === '.'))) { |
| 353 | $this->pos++; |
| 354 | } |
| 355 | |
| 356 | $raw = substr($this->data, $start, $this->pos - $start); |
| 357 | |
| 358 | if (($raw === '') || ($raw === '+') || ($raw === '-') || ($raw === '.')) { |
| 359 | return ['type' => 'keyword', 'value' => $raw]; |
| 360 | } |
| 361 | |
| 362 | $value = (str_contains($raw, '.')) ? (float) $raw : (int) $raw; |
| 363 | |
| 364 | return ['type' => 'number', 'value' => $value]; |
| 365 | } |
| 366 | |
| 367 | /** |
| 368 | * Read a keyword token |
| 369 | * |
| 370 | * @return array |
| 371 | */ |
| 372 | protected function readKeyword(): array |
| 373 | { |
| 374 | $start = $this->pos; |
| 375 | |
| 376 | while (($this->pos < $this->length) && |
| 377 | (!self::isWhitespace($this->data[$this->pos])) && |
| 378 | (!self::isDelimiter($this->data[$this->pos]))) { |
| 379 | $this->pos++; |
| 380 | } |
| 381 | |
| 382 | if ($this->pos === $start) { |
| 383 | $this->pos++; |
| 384 | return ['type' => 'keyword', 'value' => $this->data[$start]]; |
| 385 | } |
| 386 | |
| 387 | return ['type' => 'keyword', 'value' => substr($this->data, $start, $this->pos - $start)]; |
| 388 | } |
| 389 | |
| 390 | } |