[6a3a178] | 1 | const util = require('./util')
|
---|
| 2 |
|
---|
| 3 | let source
|
---|
| 4 | let parseState
|
---|
| 5 | let stack
|
---|
| 6 | let pos
|
---|
| 7 | let line
|
---|
| 8 | let column
|
---|
| 9 | let token
|
---|
| 10 | let key
|
---|
| 11 | let root
|
---|
| 12 |
|
---|
| 13 | module.exports = function parse (text, reviver) {
|
---|
| 14 | source = String(text)
|
---|
| 15 | parseState = 'start'
|
---|
| 16 | stack = []
|
---|
| 17 | pos = 0
|
---|
| 18 | line = 1
|
---|
| 19 | column = 0
|
---|
| 20 | token = undefined
|
---|
| 21 | key = undefined
|
---|
| 22 | root = undefined
|
---|
| 23 |
|
---|
| 24 | do {
|
---|
| 25 | token = lex()
|
---|
| 26 |
|
---|
| 27 | // This code is unreachable.
|
---|
| 28 | // if (!parseStates[parseState]) {
|
---|
| 29 | // throw invalidParseState()
|
---|
| 30 | // }
|
---|
| 31 |
|
---|
| 32 | parseStates[parseState]()
|
---|
| 33 | } while (token.type !== 'eof')
|
---|
| 34 |
|
---|
| 35 | if (typeof reviver === 'function') {
|
---|
| 36 | return internalize({'': root}, '', reviver)
|
---|
| 37 | }
|
---|
| 38 |
|
---|
| 39 | return root
|
---|
| 40 | }
|
---|
| 41 |
|
---|
| 42 | function internalize (holder, name, reviver) {
|
---|
| 43 | const value = holder[name]
|
---|
| 44 | if (value != null && typeof value === 'object') {
|
---|
| 45 | for (const key in value) {
|
---|
| 46 | const replacement = internalize(value, key, reviver)
|
---|
| 47 | if (replacement === undefined) {
|
---|
| 48 | delete value[key]
|
---|
| 49 | } else {
|
---|
| 50 | value[key] = replacement
|
---|
| 51 | }
|
---|
| 52 | }
|
---|
| 53 | }
|
---|
| 54 |
|
---|
| 55 | return reviver.call(holder, name, value)
|
---|
| 56 | }
|
---|
| 57 |
|
---|
| 58 | let lexState
|
---|
| 59 | let buffer
|
---|
| 60 | let doubleQuote
|
---|
| 61 | let sign
|
---|
| 62 | let c
|
---|
| 63 |
|
---|
| 64 | function lex () {
|
---|
| 65 | lexState = 'default'
|
---|
| 66 | buffer = ''
|
---|
| 67 | doubleQuote = false
|
---|
| 68 | sign = 1
|
---|
| 69 |
|
---|
| 70 | for (;;) {
|
---|
| 71 | c = peek()
|
---|
| 72 |
|
---|
| 73 | // This code is unreachable.
|
---|
| 74 | // if (!lexStates[lexState]) {
|
---|
| 75 | // throw invalidLexState(lexState)
|
---|
| 76 | // }
|
---|
| 77 |
|
---|
| 78 | const token = lexStates[lexState]()
|
---|
| 79 | if (token) {
|
---|
| 80 | return token
|
---|
| 81 | }
|
---|
| 82 | }
|
---|
| 83 | }
|
---|
| 84 |
|
---|
| 85 | function peek () {
|
---|
| 86 | if (source[pos]) {
|
---|
| 87 | return String.fromCodePoint(source.codePointAt(pos))
|
---|
| 88 | }
|
---|
| 89 | }
|
---|
| 90 |
|
---|
| 91 | function read () {
|
---|
| 92 | const c = peek()
|
---|
| 93 |
|
---|
| 94 | if (c === '\n') {
|
---|
| 95 | line++
|
---|
| 96 | column = 0
|
---|
| 97 | } else if (c) {
|
---|
| 98 | column += c.length
|
---|
| 99 | } else {
|
---|
| 100 | column++
|
---|
| 101 | }
|
---|
| 102 |
|
---|
| 103 | if (c) {
|
---|
| 104 | pos += c.length
|
---|
| 105 | }
|
---|
| 106 |
|
---|
| 107 | return c
|
---|
| 108 | }
|
---|
| 109 |
|
---|
| 110 | const lexStates = {
|
---|
| 111 | default () {
|
---|
| 112 | switch (c) {
|
---|
| 113 | case '\t':
|
---|
| 114 | case '\v':
|
---|
| 115 | case '\f':
|
---|
| 116 | case ' ':
|
---|
| 117 | case '\u00A0':
|
---|
| 118 | case '\uFEFF':
|
---|
| 119 | case '\n':
|
---|
| 120 | case '\r':
|
---|
| 121 | case '\u2028':
|
---|
| 122 | case '\u2029':
|
---|
| 123 | read()
|
---|
| 124 | return
|
---|
| 125 |
|
---|
| 126 | case '/':
|
---|
| 127 | read()
|
---|
| 128 | lexState = 'comment'
|
---|
| 129 | return
|
---|
| 130 |
|
---|
| 131 | case undefined:
|
---|
| 132 | read()
|
---|
| 133 | return newToken('eof')
|
---|
| 134 | }
|
---|
| 135 |
|
---|
| 136 | if (util.isSpaceSeparator(c)) {
|
---|
| 137 | read()
|
---|
| 138 | return
|
---|
| 139 | }
|
---|
| 140 |
|
---|
| 141 | // This code is unreachable.
|
---|
| 142 | // if (!lexStates[parseState]) {
|
---|
| 143 | // throw invalidLexState(parseState)
|
---|
| 144 | // }
|
---|
| 145 |
|
---|
| 146 | return lexStates[parseState]()
|
---|
| 147 | },
|
---|
| 148 |
|
---|
| 149 | comment () {
|
---|
| 150 | switch (c) {
|
---|
| 151 | case '*':
|
---|
| 152 | read()
|
---|
| 153 | lexState = 'multiLineComment'
|
---|
| 154 | return
|
---|
| 155 |
|
---|
| 156 | case '/':
|
---|
| 157 | read()
|
---|
| 158 | lexState = 'singleLineComment'
|
---|
| 159 | return
|
---|
| 160 | }
|
---|
| 161 |
|
---|
| 162 | throw invalidChar(read())
|
---|
| 163 | },
|
---|
| 164 |
|
---|
| 165 | multiLineComment () {
|
---|
| 166 | switch (c) {
|
---|
| 167 | case '*':
|
---|
| 168 | read()
|
---|
| 169 | lexState = 'multiLineCommentAsterisk'
|
---|
| 170 | return
|
---|
| 171 |
|
---|
| 172 | case undefined:
|
---|
| 173 | throw invalidChar(read())
|
---|
| 174 | }
|
---|
| 175 |
|
---|
| 176 | read()
|
---|
| 177 | },
|
---|
| 178 |
|
---|
| 179 | multiLineCommentAsterisk () {
|
---|
| 180 | switch (c) {
|
---|
| 181 | case '*':
|
---|
| 182 | read()
|
---|
| 183 | return
|
---|
| 184 |
|
---|
| 185 | case '/':
|
---|
| 186 | read()
|
---|
| 187 | lexState = 'default'
|
---|
| 188 | return
|
---|
| 189 |
|
---|
| 190 | case undefined:
|
---|
| 191 | throw invalidChar(read())
|
---|
| 192 | }
|
---|
| 193 |
|
---|
| 194 | read()
|
---|
| 195 | lexState = 'multiLineComment'
|
---|
| 196 | },
|
---|
| 197 |
|
---|
| 198 | singleLineComment () {
|
---|
| 199 | switch (c) {
|
---|
| 200 | case '\n':
|
---|
| 201 | case '\r':
|
---|
| 202 | case '\u2028':
|
---|
| 203 | case '\u2029':
|
---|
| 204 | read()
|
---|
| 205 | lexState = 'default'
|
---|
| 206 | return
|
---|
| 207 |
|
---|
| 208 | case undefined:
|
---|
| 209 | read()
|
---|
| 210 | return newToken('eof')
|
---|
| 211 | }
|
---|
| 212 |
|
---|
| 213 | read()
|
---|
| 214 | },
|
---|
| 215 |
|
---|
| 216 | value () {
|
---|
| 217 | switch (c) {
|
---|
| 218 | case '{':
|
---|
| 219 | case '[':
|
---|
| 220 | return newToken('punctuator', read())
|
---|
| 221 |
|
---|
| 222 | case 'n':
|
---|
| 223 | read()
|
---|
| 224 | literal('ull')
|
---|
| 225 | return newToken('null', null)
|
---|
| 226 |
|
---|
| 227 | case 't':
|
---|
| 228 | read()
|
---|
| 229 | literal('rue')
|
---|
| 230 | return newToken('boolean', true)
|
---|
| 231 |
|
---|
| 232 | case 'f':
|
---|
| 233 | read()
|
---|
| 234 | literal('alse')
|
---|
| 235 | return newToken('boolean', false)
|
---|
| 236 |
|
---|
| 237 | case '-':
|
---|
| 238 | case '+':
|
---|
| 239 | if (read() === '-') {
|
---|
| 240 | sign = -1
|
---|
| 241 | }
|
---|
| 242 |
|
---|
| 243 | lexState = 'sign'
|
---|
| 244 | return
|
---|
| 245 |
|
---|
| 246 | case '.':
|
---|
| 247 | buffer = read()
|
---|
| 248 | lexState = 'decimalPointLeading'
|
---|
| 249 | return
|
---|
| 250 |
|
---|
| 251 | case '0':
|
---|
| 252 | buffer = read()
|
---|
| 253 | lexState = 'zero'
|
---|
| 254 | return
|
---|
| 255 |
|
---|
| 256 | case '1':
|
---|
| 257 | case '2':
|
---|
| 258 | case '3':
|
---|
| 259 | case '4':
|
---|
| 260 | case '5':
|
---|
| 261 | case '6':
|
---|
| 262 | case '7':
|
---|
| 263 | case '8':
|
---|
| 264 | case '9':
|
---|
| 265 | buffer = read()
|
---|
| 266 | lexState = 'decimalInteger'
|
---|
| 267 | return
|
---|
| 268 |
|
---|
| 269 | case 'I':
|
---|
| 270 | read()
|
---|
| 271 | literal('nfinity')
|
---|
| 272 | return newToken('numeric', Infinity)
|
---|
| 273 |
|
---|
| 274 | case 'N':
|
---|
| 275 | read()
|
---|
| 276 | literal('aN')
|
---|
| 277 | return newToken('numeric', NaN)
|
---|
| 278 |
|
---|
| 279 | case '"':
|
---|
| 280 | case "'":
|
---|
| 281 | doubleQuote = (read() === '"')
|
---|
| 282 | buffer = ''
|
---|
| 283 | lexState = 'string'
|
---|
| 284 | return
|
---|
| 285 | }
|
---|
| 286 |
|
---|
| 287 | throw invalidChar(read())
|
---|
| 288 | },
|
---|
| 289 |
|
---|
| 290 | identifierNameStartEscape () {
|
---|
| 291 | if (c !== 'u') {
|
---|
| 292 | throw invalidChar(read())
|
---|
| 293 | }
|
---|
| 294 |
|
---|
| 295 | read()
|
---|
| 296 | const u = unicodeEscape()
|
---|
| 297 | switch (u) {
|
---|
| 298 | case '$':
|
---|
| 299 | case '_':
|
---|
| 300 | break
|
---|
| 301 |
|
---|
| 302 | default:
|
---|
| 303 | if (!util.isIdStartChar(u)) {
|
---|
| 304 | throw invalidIdentifier()
|
---|
| 305 | }
|
---|
| 306 |
|
---|
| 307 | break
|
---|
| 308 | }
|
---|
| 309 |
|
---|
| 310 | buffer += u
|
---|
| 311 | lexState = 'identifierName'
|
---|
| 312 | },
|
---|
| 313 |
|
---|
| 314 | identifierName () {
|
---|
| 315 | switch (c) {
|
---|
| 316 | case '$':
|
---|
| 317 | case '_':
|
---|
| 318 | case '\u200C':
|
---|
| 319 | case '\u200D':
|
---|
| 320 | buffer += read()
|
---|
| 321 | return
|
---|
| 322 |
|
---|
| 323 | case '\\':
|
---|
| 324 | read()
|
---|
| 325 | lexState = 'identifierNameEscape'
|
---|
| 326 | return
|
---|
| 327 | }
|
---|
| 328 |
|
---|
| 329 | if (util.isIdContinueChar(c)) {
|
---|
| 330 | buffer += read()
|
---|
| 331 | return
|
---|
| 332 | }
|
---|
| 333 |
|
---|
| 334 | return newToken('identifier', buffer)
|
---|
| 335 | },
|
---|
| 336 |
|
---|
| 337 | identifierNameEscape () {
|
---|
| 338 | if (c !== 'u') {
|
---|
| 339 | throw invalidChar(read())
|
---|
| 340 | }
|
---|
| 341 |
|
---|
| 342 | read()
|
---|
| 343 | const u = unicodeEscape()
|
---|
| 344 | switch (u) {
|
---|
| 345 | case '$':
|
---|
| 346 | case '_':
|
---|
| 347 | case '\u200C':
|
---|
| 348 | case '\u200D':
|
---|
| 349 | break
|
---|
| 350 |
|
---|
| 351 | default:
|
---|
| 352 | if (!util.isIdContinueChar(u)) {
|
---|
| 353 | throw invalidIdentifier()
|
---|
| 354 | }
|
---|
| 355 |
|
---|
| 356 | break
|
---|
| 357 | }
|
---|
| 358 |
|
---|
| 359 | buffer += u
|
---|
| 360 | lexState = 'identifierName'
|
---|
| 361 | },
|
---|
| 362 |
|
---|
| 363 | sign () {
|
---|
| 364 | switch (c) {
|
---|
| 365 | case '.':
|
---|
| 366 | buffer = read()
|
---|
| 367 | lexState = 'decimalPointLeading'
|
---|
| 368 | return
|
---|
| 369 |
|
---|
| 370 | case '0':
|
---|
| 371 | buffer = read()
|
---|
| 372 | lexState = 'zero'
|
---|
| 373 | return
|
---|
| 374 |
|
---|
| 375 | case '1':
|
---|
| 376 | case '2':
|
---|
| 377 | case '3':
|
---|
| 378 | case '4':
|
---|
| 379 | case '5':
|
---|
| 380 | case '6':
|
---|
| 381 | case '7':
|
---|
| 382 | case '8':
|
---|
| 383 | case '9':
|
---|
| 384 | buffer = read()
|
---|
| 385 | lexState = 'decimalInteger'
|
---|
| 386 | return
|
---|
| 387 |
|
---|
| 388 | case 'I':
|
---|
| 389 | read()
|
---|
| 390 | literal('nfinity')
|
---|
| 391 | return newToken('numeric', sign * Infinity)
|
---|
| 392 |
|
---|
| 393 | case 'N':
|
---|
| 394 | read()
|
---|
| 395 | literal('aN')
|
---|
| 396 | return newToken('numeric', NaN)
|
---|
| 397 | }
|
---|
| 398 |
|
---|
| 399 | throw invalidChar(read())
|
---|
| 400 | },
|
---|
| 401 |
|
---|
| 402 | zero () {
|
---|
| 403 | switch (c) {
|
---|
| 404 | case '.':
|
---|
| 405 | buffer += read()
|
---|
| 406 | lexState = 'decimalPoint'
|
---|
| 407 | return
|
---|
| 408 |
|
---|
| 409 | case 'e':
|
---|
| 410 | case 'E':
|
---|
| 411 | buffer += read()
|
---|
| 412 | lexState = 'decimalExponent'
|
---|
| 413 | return
|
---|
| 414 |
|
---|
| 415 | case 'x':
|
---|
| 416 | case 'X':
|
---|
| 417 | buffer += read()
|
---|
| 418 | lexState = 'hexadecimal'
|
---|
| 419 | return
|
---|
| 420 | }
|
---|
| 421 |
|
---|
| 422 | return newToken('numeric', sign * 0)
|
---|
| 423 | },
|
---|
| 424 |
|
---|
| 425 | decimalInteger () {
|
---|
| 426 | switch (c) {
|
---|
| 427 | case '.':
|
---|
| 428 | buffer += read()
|
---|
| 429 | lexState = 'decimalPoint'
|
---|
| 430 | return
|
---|
| 431 |
|
---|
| 432 | case 'e':
|
---|
| 433 | case 'E':
|
---|
| 434 | buffer += read()
|
---|
| 435 | lexState = 'decimalExponent'
|
---|
| 436 | return
|
---|
| 437 | }
|
---|
| 438 |
|
---|
| 439 | if (util.isDigit(c)) {
|
---|
| 440 | buffer += read()
|
---|
| 441 | return
|
---|
| 442 | }
|
---|
| 443 |
|
---|
| 444 | return newToken('numeric', sign * Number(buffer))
|
---|
| 445 | },
|
---|
| 446 |
|
---|
| 447 | decimalPointLeading () {
|
---|
| 448 | if (util.isDigit(c)) {
|
---|
| 449 | buffer += read()
|
---|
| 450 | lexState = 'decimalFraction'
|
---|
| 451 | return
|
---|
| 452 | }
|
---|
| 453 |
|
---|
| 454 | throw invalidChar(read())
|
---|
| 455 | },
|
---|
| 456 |
|
---|
| 457 | decimalPoint () {
|
---|
| 458 | switch (c) {
|
---|
| 459 | case 'e':
|
---|
| 460 | case 'E':
|
---|
| 461 | buffer += read()
|
---|
| 462 | lexState = 'decimalExponent'
|
---|
| 463 | return
|
---|
| 464 | }
|
---|
| 465 |
|
---|
| 466 | if (util.isDigit(c)) {
|
---|
| 467 | buffer += read()
|
---|
| 468 | lexState = 'decimalFraction'
|
---|
| 469 | return
|
---|
| 470 | }
|
---|
| 471 |
|
---|
| 472 | return newToken('numeric', sign * Number(buffer))
|
---|
| 473 | },
|
---|
| 474 |
|
---|
| 475 | decimalFraction () {
|
---|
| 476 | switch (c) {
|
---|
| 477 | case 'e':
|
---|
| 478 | case 'E':
|
---|
| 479 | buffer += read()
|
---|
| 480 | lexState = 'decimalExponent'
|
---|
| 481 | return
|
---|
| 482 | }
|
---|
| 483 |
|
---|
| 484 | if (util.isDigit(c)) {
|
---|
| 485 | buffer += read()
|
---|
| 486 | return
|
---|
| 487 | }
|
---|
| 488 |
|
---|
| 489 | return newToken('numeric', sign * Number(buffer))
|
---|
| 490 | },
|
---|
| 491 |
|
---|
| 492 | decimalExponent () {
|
---|
| 493 | switch (c) {
|
---|
| 494 | case '+':
|
---|
| 495 | case '-':
|
---|
| 496 | buffer += read()
|
---|
| 497 | lexState = 'decimalExponentSign'
|
---|
| 498 | return
|
---|
| 499 | }
|
---|
| 500 |
|
---|
| 501 | if (util.isDigit(c)) {
|
---|
| 502 | buffer += read()
|
---|
| 503 | lexState = 'decimalExponentInteger'
|
---|
| 504 | return
|
---|
| 505 | }
|
---|
| 506 |
|
---|
| 507 | throw invalidChar(read())
|
---|
| 508 | },
|
---|
| 509 |
|
---|
| 510 | decimalExponentSign () {
|
---|
| 511 | if (util.isDigit(c)) {
|
---|
| 512 | buffer += read()
|
---|
| 513 | lexState = 'decimalExponentInteger'
|
---|
| 514 | return
|
---|
| 515 | }
|
---|
| 516 |
|
---|
| 517 | throw invalidChar(read())
|
---|
| 518 | },
|
---|
| 519 |
|
---|
| 520 | decimalExponentInteger () {
|
---|
| 521 | if (util.isDigit(c)) {
|
---|
| 522 | buffer += read()
|
---|
| 523 | return
|
---|
| 524 | }
|
---|
| 525 |
|
---|
| 526 | return newToken('numeric', sign * Number(buffer))
|
---|
| 527 | },
|
---|
| 528 |
|
---|
| 529 | hexadecimal () {
|
---|
| 530 | if (util.isHexDigit(c)) {
|
---|
| 531 | buffer += read()
|
---|
| 532 | lexState = 'hexadecimalInteger'
|
---|
| 533 | return
|
---|
| 534 | }
|
---|
| 535 |
|
---|
| 536 | throw invalidChar(read())
|
---|
| 537 | },
|
---|
| 538 |
|
---|
| 539 | hexadecimalInteger () {
|
---|
| 540 | if (util.isHexDigit(c)) {
|
---|
| 541 | buffer += read()
|
---|
| 542 | return
|
---|
| 543 | }
|
---|
| 544 |
|
---|
| 545 | return newToken('numeric', sign * Number(buffer))
|
---|
| 546 | },
|
---|
| 547 |
|
---|
| 548 | string () {
|
---|
| 549 | switch (c) {
|
---|
| 550 | case '\\':
|
---|
| 551 | read()
|
---|
| 552 | buffer += escape()
|
---|
| 553 | return
|
---|
| 554 |
|
---|
| 555 | case '"':
|
---|
| 556 | if (doubleQuote) {
|
---|
| 557 | read()
|
---|
| 558 | return newToken('string', buffer)
|
---|
| 559 | }
|
---|
| 560 |
|
---|
| 561 | buffer += read()
|
---|
| 562 | return
|
---|
| 563 |
|
---|
| 564 | case "'":
|
---|
| 565 | if (!doubleQuote) {
|
---|
| 566 | read()
|
---|
| 567 | return newToken('string', buffer)
|
---|
| 568 | }
|
---|
| 569 |
|
---|
| 570 | buffer += read()
|
---|
| 571 | return
|
---|
| 572 |
|
---|
| 573 | case '\n':
|
---|
| 574 | case '\r':
|
---|
| 575 | throw invalidChar(read())
|
---|
| 576 |
|
---|
| 577 | case '\u2028':
|
---|
| 578 | case '\u2029':
|
---|
| 579 | separatorChar(c)
|
---|
| 580 | break
|
---|
| 581 |
|
---|
| 582 | case undefined:
|
---|
| 583 | throw invalidChar(read())
|
---|
| 584 | }
|
---|
| 585 |
|
---|
| 586 | buffer += read()
|
---|
| 587 | },
|
---|
| 588 |
|
---|
| 589 | start () {
|
---|
| 590 | switch (c) {
|
---|
| 591 | case '{':
|
---|
| 592 | case '[':
|
---|
| 593 | return newToken('punctuator', read())
|
---|
| 594 |
|
---|
| 595 | // This code is unreachable since the default lexState handles eof.
|
---|
| 596 | // case undefined:
|
---|
| 597 | // return newToken('eof')
|
---|
| 598 | }
|
---|
| 599 |
|
---|
| 600 | lexState = 'value'
|
---|
| 601 | },
|
---|
| 602 |
|
---|
| 603 | beforePropertyName () {
|
---|
| 604 | switch (c) {
|
---|
| 605 | case '$':
|
---|
| 606 | case '_':
|
---|
| 607 | buffer = read()
|
---|
| 608 | lexState = 'identifierName'
|
---|
| 609 | return
|
---|
| 610 |
|
---|
| 611 | case '\\':
|
---|
| 612 | read()
|
---|
| 613 | lexState = 'identifierNameStartEscape'
|
---|
| 614 | return
|
---|
| 615 |
|
---|
| 616 | case '}':
|
---|
| 617 | return newToken('punctuator', read())
|
---|
| 618 |
|
---|
| 619 | case '"':
|
---|
| 620 | case "'":
|
---|
| 621 | doubleQuote = (read() === '"')
|
---|
| 622 | lexState = 'string'
|
---|
| 623 | return
|
---|
| 624 | }
|
---|
| 625 |
|
---|
| 626 | if (util.isIdStartChar(c)) {
|
---|
| 627 | buffer += read()
|
---|
| 628 | lexState = 'identifierName'
|
---|
| 629 | return
|
---|
| 630 | }
|
---|
| 631 |
|
---|
| 632 | throw invalidChar(read())
|
---|
| 633 | },
|
---|
| 634 |
|
---|
| 635 | afterPropertyName () {
|
---|
| 636 | if (c === ':') {
|
---|
| 637 | return newToken('punctuator', read())
|
---|
| 638 | }
|
---|
| 639 |
|
---|
| 640 | throw invalidChar(read())
|
---|
| 641 | },
|
---|
| 642 |
|
---|
| 643 | beforePropertyValue () {
|
---|
| 644 | lexState = 'value'
|
---|
| 645 | },
|
---|
| 646 |
|
---|
| 647 | afterPropertyValue () {
|
---|
| 648 | switch (c) {
|
---|
| 649 | case ',':
|
---|
| 650 | case '}':
|
---|
| 651 | return newToken('punctuator', read())
|
---|
| 652 | }
|
---|
| 653 |
|
---|
| 654 | throw invalidChar(read())
|
---|
| 655 | },
|
---|
| 656 |
|
---|
| 657 | beforeArrayValue () {
|
---|
| 658 | if (c === ']') {
|
---|
| 659 | return newToken('punctuator', read())
|
---|
| 660 | }
|
---|
| 661 |
|
---|
| 662 | lexState = 'value'
|
---|
| 663 | },
|
---|
| 664 |
|
---|
| 665 | afterArrayValue () {
|
---|
| 666 | switch (c) {
|
---|
| 667 | case ',':
|
---|
| 668 | case ']':
|
---|
| 669 | return newToken('punctuator', read())
|
---|
| 670 | }
|
---|
| 671 |
|
---|
| 672 | throw invalidChar(read())
|
---|
| 673 | },
|
---|
| 674 |
|
---|
| 675 | end () {
|
---|
| 676 | // This code is unreachable since it's handled by the default lexState.
|
---|
| 677 | // if (c === undefined) {
|
---|
| 678 | // read()
|
---|
| 679 | // return newToken('eof')
|
---|
| 680 | // }
|
---|
| 681 |
|
---|
| 682 | throw invalidChar(read())
|
---|
| 683 | },
|
---|
| 684 | }
|
---|
| 685 |
|
---|
| 686 | function newToken (type, value) {
|
---|
| 687 | return {
|
---|
| 688 | type,
|
---|
| 689 | value,
|
---|
| 690 | line,
|
---|
| 691 | column,
|
---|
| 692 | }
|
---|
| 693 | }
|
---|
| 694 |
|
---|
| 695 | function literal (s) {
|
---|
| 696 | for (const c of s) {
|
---|
| 697 | const p = peek()
|
---|
| 698 |
|
---|
| 699 | if (p !== c) {
|
---|
| 700 | throw invalidChar(read())
|
---|
| 701 | }
|
---|
| 702 |
|
---|
| 703 | read()
|
---|
| 704 | }
|
---|
| 705 | }
|
---|
| 706 |
|
---|
| 707 | function escape () {
|
---|
| 708 | const c = peek()
|
---|
| 709 | switch (c) {
|
---|
| 710 | case 'b':
|
---|
| 711 | read()
|
---|
| 712 | return '\b'
|
---|
| 713 |
|
---|
| 714 | case 'f':
|
---|
| 715 | read()
|
---|
| 716 | return '\f'
|
---|
| 717 |
|
---|
| 718 | case 'n':
|
---|
| 719 | read()
|
---|
| 720 | return '\n'
|
---|
| 721 |
|
---|
| 722 | case 'r':
|
---|
| 723 | read()
|
---|
| 724 | return '\r'
|
---|
| 725 |
|
---|
| 726 | case 't':
|
---|
| 727 | read()
|
---|
| 728 | return '\t'
|
---|
| 729 |
|
---|
| 730 | case 'v':
|
---|
| 731 | read()
|
---|
| 732 | return '\v'
|
---|
| 733 |
|
---|
| 734 | case '0':
|
---|
| 735 | read()
|
---|
| 736 | if (util.isDigit(peek())) {
|
---|
| 737 | throw invalidChar(read())
|
---|
| 738 | }
|
---|
| 739 |
|
---|
| 740 | return '\0'
|
---|
| 741 |
|
---|
| 742 | case 'x':
|
---|
| 743 | read()
|
---|
| 744 | return hexEscape()
|
---|
| 745 |
|
---|
| 746 | case 'u':
|
---|
| 747 | read()
|
---|
| 748 | return unicodeEscape()
|
---|
| 749 |
|
---|
| 750 | case '\n':
|
---|
| 751 | case '\u2028':
|
---|
| 752 | case '\u2029':
|
---|
| 753 | read()
|
---|
| 754 | return ''
|
---|
| 755 |
|
---|
| 756 | case '\r':
|
---|
| 757 | read()
|
---|
| 758 | if (peek() === '\n') {
|
---|
| 759 | read()
|
---|
| 760 | }
|
---|
| 761 |
|
---|
| 762 | return ''
|
---|
| 763 |
|
---|
| 764 | case '1':
|
---|
| 765 | case '2':
|
---|
| 766 | case '3':
|
---|
| 767 | case '4':
|
---|
| 768 | case '5':
|
---|
| 769 | case '6':
|
---|
| 770 | case '7':
|
---|
| 771 | case '8':
|
---|
| 772 | case '9':
|
---|
| 773 | throw invalidChar(read())
|
---|
| 774 |
|
---|
| 775 | case undefined:
|
---|
| 776 | throw invalidChar(read())
|
---|
| 777 | }
|
---|
| 778 |
|
---|
| 779 | return read()
|
---|
| 780 | }
|
---|
| 781 |
|
---|
| 782 | function hexEscape () {
|
---|
| 783 | let buffer = ''
|
---|
| 784 | let c = peek()
|
---|
| 785 |
|
---|
| 786 | if (!util.isHexDigit(c)) {
|
---|
| 787 | throw invalidChar(read())
|
---|
| 788 | }
|
---|
| 789 |
|
---|
| 790 | buffer += read()
|
---|
| 791 |
|
---|
| 792 | c = peek()
|
---|
| 793 | if (!util.isHexDigit(c)) {
|
---|
| 794 | throw invalidChar(read())
|
---|
| 795 | }
|
---|
| 796 |
|
---|
| 797 | buffer += read()
|
---|
| 798 |
|
---|
| 799 | return String.fromCodePoint(parseInt(buffer, 16))
|
---|
| 800 | }
|
---|
| 801 |
|
---|
| 802 | function unicodeEscape () {
|
---|
| 803 | let buffer = ''
|
---|
| 804 | let count = 4
|
---|
| 805 |
|
---|
| 806 | while (count-- > 0) {
|
---|
| 807 | const c = peek()
|
---|
| 808 | if (!util.isHexDigit(c)) {
|
---|
| 809 | throw invalidChar(read())
|
---|
| 810 | }
|
---|
| 811 |
|
---|
| 812 | buffer += read()
|
---|
| 813 | }
|
---|
| 814 |
|
---|
| 815 | return String.fromCodePoint(parseInt(buffer, 16))
|
---|
| 816 | }
|
---|
| 817 |
|
---|
| 818 | const parseStates = {
|
---|
| 819 | start () {
|
---|
| 820 | if (token.type === 'eof') {
|
---|
| 821 | throw invalidEOF()
|
---|
| 822 | }
|
---|
| 823 |
|
---|
| 824 | push()
|
---|
| 825 | },
|
---|
| 826 |
|
---|
| 827 | beforePropertyName () {
|
---|
| 828 | switch (token.type) {
|
---|
| 829 | case 'identifier':
|
---|
| 830 | case 'string':
|
---|
| 831 | key = token.value
|
---|
| 832 | parseState = 'afterPropertyName'
|
---|
| 833 | return
|
---|
| 834 |
|
---|
| 835 | case 'punctuator':
|
---|
| 836 | // This code is unreachable since it's handled by the lexState.
|
---|
| 837 | // if (token.value !== '}') {
|
---|
| 838 | // throw invalidToken()
|
---|
| 839 | // }
|
---|
| 840 |
|
---|
| 841 | pop()
|
---|
| 842 | return
|
---|
| 843 |
|
---|
| 844 | case 'eof':
|
---|
| 845 | throw invalidEOF()
|
---|
| 846 | }
|
---|
| 847 |
|
---|
| 848 | // This code is unreachable since it's handled by the lexState.
|
---|
| 849 | // throw invalidToken()
|
---|
| 850 | },
|
---|
| 851 |
|
---|
| 852 | afterPropertyName () {
|
---|
| 853 | // This code is unreachable since it's handled by the lexState.
|
---|
| 854 | // if (token.type !== 'punctuator' || token.value !== ':') {
|
---|
| 855 | // throw invalidToken()
|
---|
| 856 | // }
|
---|
| 857 |
|
---|
| 858 | if (token.type === 'eof') {
|
---|
| 859 | throw invalidEOF()
|
---|
| 860 | }
|
---|
| 861 |
|
---|
| 862 | parseState = 'beforePropertyValue'
|
---|
| 863 | },
|
---|
| 864 |
|
---|
| 865 | beforePropertyValue () {
|
---|
| 866 | if (token.type === 'eof') {
|
---|
| 867 | throw invalidEOF()
|
---|
| 868 | }
|
---|
| 869 |
|
---|
| 870 | push()
|
---|
| 871 | },
|
---|
| 872 |
|
---|
| 873 | beforeArrayValue () {
|
---|
| 874 | if (token.type === 'eof') {
|
---|
| 875 | throw invalidEOF()
|
---|
| 876 | }
|
---|
| 877 |
|
---|
| 878 | if (token.type === 'punctuator' && token.value === ']') {
|
---|
| 879 | pop()
|
---|
| 880 | return
|
---|
| 881 | }
|
---|
| 882 |
|
---|
| 883 | push()
|
---|
| 884 | },
|
---|
| 885 |
|
---|
| 886 | afterPropertyValue () {
|
---|
| 887 | // This code is unreachable since it's handled by the lexState.
|
---|
| 888 | // if (token.type !== 'punctuator') {
|
---|
| 889 | // throw invalidToken()
|
---|
| 890 | // }
|
---|
| 891 |
|
---|
| 892 | if (token.type === 'eof') {
|
---|
| 893 | throw invalidEOF()
|
---|
| 894 | }
|
---|
| 895 |
|
---|
| 896 | switch (token.value) {
|
---|
| 897 | case ',':
|
---|
| 898 | parseState = 'beforePropertyName'
|
---|
| 899 | return
|
---|
| 900 |
|
---|
| 901 | case '}':
|
---|
| 902 | pop()
|
---|
| 903 | }
|
---|
| 904 |
|
---|
| 905 | // This code is unreachable since it's handled by the lexState.
|
---|
| 906 | // throw invalidToken()
|
---|
| 907 | },
|
---|
| 908 |
|
---|
| 909 | afterArrayValue () {
|
---|
| 910 | // This code is unreachable since it's handled by the lexState.
|
---|
| 911 | // if (token.type !== 'punctuator') {
|
---|
| 912 | // throw invalidToken()
|
---|
| 913 | // }
|
---|
| 914 |
|
---|
| 915 | if (token.type === 'eof') {
|
---|
| 916 | throw invalidEOF()
|
---|
| 917 | }
|
---|
| 918 |
|
---|
| 919 | switch (token.value) {
|
---|
| 920 | case ',':
|
---|
| 921 | parseState = 'beforeArrayValue'
|
---|
| 922 | return
|
---|
| 923 |
|
---|
| 924 | case ']':
|
---|
| 925 | pop()
|
---|
| 926 | }
|
---|
| 927 |
|
---|
| 928 | // This code is unreachable since it's handled by the lexState.
|
---|
| 929 | // throw invalidToken()
|
---|
| 930 | },
|
---|
| 931 |
|
---|
| 932 | end () {
|
---|
| 933 | // This code is unreachable since it's handled by the lexState.
|
---|
| 934 | // if (token.type !== 'eof') {
|
---|
| 935 | // throw invalidToken()
|
---|
| 936 | // }
|
---|
| 937 | },
|
---|
| 938 | }
|
---|
| 939 |
|
---|
| 940 | function push () {
|
---|
| 941 | let value
|
---|
| 942 |
|
---|
| 943 | switch (token.type) {
|
---|
| 944 | case 'punctuator':
|
---|
| 945 | switch (token.value) {
|
---|
| 946 | case '{':
|
---|
| 947 | value = {}
|
---|
| 948 | break
|
---|
| 949 |
|
---|
| 950 | case '[':
|
---|
| 951 | value = []
|
---|
| 952 | break
|
---|
| 953 | }
|
---|
| 954 |
|
---|
| 955 | break
|
---|
| 956 |
|
---|
| 957 | case 'null':
|
---|
| 958 | case 'boolean':
|
---|
| 959 | case 'numeric':
|
---|
| 960 | case 'string':
|
---|
| 961 | value = token.value
|
---|
| 962 | break
|
---|
| 963 |
|
---|
| 964 | // This code is unreachable.
|
---|
| 965 | // default:
|
---|
| 966 | // throw invalidToken()
|
---|
| 967 | }
|
---|
| 968 |
|
---|
| 969 | if (root === undefined) {
|
---|
| 970 | root = value
|
---|
| 971 | } else {
|
---|
| 972 | const parent = stack[stack.length - 1]
|
---|
| 973 | if (Array.isArray(parent)) {
|
---|
| 974 | parent.push(value)
|
---|
| 975 | } else {
|
---|
| 976 | parent[key] = value
|
---|
| 977 | }
|
---|
| 978 | }
|
---|
| 979 |
|
---|
| 980 | if (value !== null && typeof value === 'object') {
|
---|
| 981 | stack.push(value)
|
---|
| 982 |
|
---|
| 983 | if (Array.isArray(value)) {
|
---|
| 984 | parseState = 'beforeArrayValue'
|
---|
| 985 | } else {
|
---|
| 986 | parseState = 'beforePropertyName'
|
---|
| 987 | }
|
---|
| 988 | } else {
|
---|
| 989 | const current = stack[stack.length - 1]
|
---|
| 990 | if (current == null) {
|
---|
| 991 | parseState = 'end'
|
---|
| 992 | } else if (Array.isArray(current)) {
|
---|
| 993 | parseState = 'afterArrayValue'
|
---|
| 994 | } else {
|
---|
| 995 | parseState = 'afterPropertyValue'
|
---|
| 996 | }
|
---|
| 997 | }
|
---|
| 998 | }
|
---|
| 999 |
|
---|
| 1000 | function pop () {
|
---|
| 1001 | stack.pop()
|
---|
| 1002 |
|
---|
| 1003 | const current = stack[stack.length - 1]
|
---|
| 1004 | if (current == null) {
|
---|
| 1005 | parseState = 'end'
|
---|
| 1006 | } else if (Array.isArray(current)) {
|
---|
| 1007 | parseState = 'afterArrayValue'
|
---|
| 1008 | } else {
|
---|
| 1009 | parseState = 'afterPropertyValue'
|
---|
| 1010 | }
|
---|
| 1011 | }
|
---|
| 1012 |
|
---|
| 1013 | // This code is unreachable.
|
---|
| 1014 | // function invalidParseState () {
|
---|
| 1015 | // return new Error(`JSON5: invalid parse state '${parseState}'`)
|
---|
| 1016 | // }
|
---|
| 1017 |
|
---|
| 1018 | // This code is unreachable.
|
---|
| 1019 | // function invalidLexState (state) {
|
---|
| 1020 | // return new Error(`JSON5: invalid lex state '${state}'`)
|
---|
| 1021 | // }
|
---|
| 1022 |
|
---|
| 1023 | function invalidChar (c) {
|
---|
| 1024 | if (c === undefined) {
|
---|
| 1025 | return syntaxError(`JSON5: invalid end of input at ${line}:${column}`)
|
---|
| 1026 | }
|
---|
| 1027 |
|
---|
| 1028 | return syntaxError(`JSON5: invalid character '${formatChar(c)}' at ${line}:${column}`)
|
---|
| 1029 | }
|
---|
| 1030 |
|
---|
| 1031 | function invalidEOF () {
|
---|
| 1032 | return syntaxError(`JSON5: invalid end of input at ${line}:${column}`)
|
---|
| 1033 | }
|
---|
| 1034 |
|
---|
| 1035 | // This code is unreachable.
|
---|
| 1036 | // function invalidToken () {
|
---|
| 1037 | // if (token.type === 'eof') {
|
---|
| 1038 | // return syntaxError(`JSON5: invalid end of input at ${line}:${column}`)
|
---|
| 1039 | // }
|
---|
| 1040 |
|
---|
| 1041 | // const c = String.fromCodePoint(token.value.codePointAt(0))
|
---|
| 1042 | // return syntaxError(`JSON5: invalid character '${formatChar(c)}' at ${line}:${column}`)
|
---|
| 1043 | // }
|
---|
| 1044 |
|
---|
| 1045 | function invalidIdentifier () {
|
---|
| 1046 | column -= 5
|
---|
| 1047 | return syntaxError(`JSON5: invalid identifier character at ${line}:${column}`)
|
---|
| 1048 | }
|
---|
| 1049 |
|
---|
| 1050 | function separatorChar (c) {
|
---|
| 1051 | console.warn(`JSON5: '${formatChar(c)}' in strings is not valid ECMAScript; consider escaping`)
|
---|
| 1052 | }
|
---|
| 1053 |
|
---|
| 1054 | function formatChar (c) {
|
---|
| 1055 | const replacements = {
|
---|
| 1056 | "'": "\\'",
|
---|
| 1057 | '"': '\\"',
|
---|
| 1058 | '\\': '\\\\',
|
---|
| 1059 | '\b': '\\b',
|
---|
| 1060 | '\f': '\\f',
|
---|
| 1061 | '\n': '\\n',
|
---|
| 1062 | '\r': '\\r',
|
---|
| 1063 | '\t': '\\t',
|
---|
| 1064 | '\v': '\\v',
|
---|
| 1065 | '\0': '\\0',
|
---|
| 1066 | '\u2028': '\\u2028',
|
---|
| 1067 | '\u2029': '\\u2029',
|
---|
| 1068 | }
|
---|
| 1069 |
|
---|
| 1070 | if (replacements[c]) {
|
---|
| 1071 | return replacements[c]
|
---|
| 1072 | }
|
---|
| 1073 |
|
---|
| 1074 | if (c < ' ') {
|
---|
| 1075 | const hexString = c.charCodeAt(0).toString(16)
|
---|
| 1076 | return '\\x' + ('00' + hexString).substring(hexString.length)
|
---|
| 1077 | }
|
---|
| 1078 |
|
---|
| 1079 | return c
|
---|
| 1080 | }
|
---|
| 1081 |
|
---|
| 1082 | function syntaxError (message) {
|
---|
| 1083 | const err = new SyntaxError(message)
|
---|
| 1084 | err.lineNumber = line
|
---|
| 1085 | err.columnNumber = column
|
---|
| 1086 | return err
|
---|
| 1087 | }
|
---|