196 lines
5.5 KiB
JavaScript
Raw Normal View History

import { decodeErrPrefix } from './common.js'
import { Type } from './token.js'
import { jump, quick } from './jump.js'
/**
* @typedef {import('./token.js').Token} Token
* @typedef {import('../interface').DecodeOptions} DecodeOptions
* @typedef {import('../interface').DecodeTokenizer} DecodeTokenizer
*/
const defaultDecodeOptions = {
strict: false,
allowIndefinite: true,
allowUndefined: true,
allowBigInt: true
}
/**
* @implements {DecodeTokenizer}
*/
class Tokeniser {
/**
* @param {Uint8Array} data
* @param {DecodeOptions} options
*/
constructor (data, options = {}) {
this.pos = 0
this.data = data
this.options = options
}
done () {
return this.pos >= this.data.length
}
next () {
const byt = this.data[this.pos]
let token = quick[byt]
if (token === undefined) {
const decoder = jump[byt]
/* c8 ignore next 4 */
// if we're here then there's something wrong with our jump or quick lists!
if (!decoder) {
throw new Error(`${decodeErrPrefix} no decoder for major type ${byt >>> 5} (byte 0x${byt.toString(16).padStart(2, '0')})`)
}
const minor = byt & 31
token = decoder(this.data, this.pos, minor, this.options)
}
// @ts-ignore we get to assume encodedLength is set (crossing fingers slightly)
this.pos += token.encodedLength
return token
}
}
const DONE = Symbol.for('DONE')
const BREAK = Symbol.for('BREAK')
/**
* @param {Token} token
* @param {DecodeTokenizer} tokeniser
* @param {DecodeOptions} options
* @returns {any|BREAK|DONE}
*/
function tokenToArray (token, tokeniser, options) {
const arr = []
for (let i = 0; i < token.value; i++) {
const value = tokensToObject(tokeniser, options)
if (value === BREAK) {
if (token.value === Infinity) {
// normal end to indefinite length array
break
}
throw new Error(`${decodeErrPrefix} got unexpected break to lengthed array`)
}
if (value === DONE) {
throw new Error(`${decodeErrPrefix} found array but not enough entries (got ${i}, expected ${token.value})`)
}
arr[i] = value
}
return arr
}
/**
* @param {Token} token
* @param {DecodeTokenizer} tokeniser
* @param {DecodeOptions} options
* @returns {any|BREAK|DONE}
*/
function tokenToMap (token, tokeniser, options) {
const useMaps = options.useMaps === true
const obj = useMaps ? undefined : {}
const m = useMaps ? new Map() : undefined
for (let i = 0; i < token.value; i++) {
const key = tokensToObject(tokeniser, options)
if (key === BREAK) {
if (token.value === Infinity) {
// normal end to indefinite length map
break
}
throw new Error(`${decodeErrPrefix} got unexpected break to lengthed map`)
}
if (key === DONE) {
throw new Error(`${decodeErrPrefix} found map but not enough entries (got ${i} [no key], expected ${token.value})`)
}
if (useMaps !== true && typeof key !== 'string') {
throw new Error(`${decodeErrPrefix} non-string keys not supported (got ${typeof key})`)
}
if (options.rejectDuplicateMapKeys === true) {
// @ts-ignore
if ((useMaps && m.has(key)) || (!useMaps && (key in obj))) {
throw new Error(`${decodeErrPrefix} found repeat map key "${key}"`)
}
}
const value = tokensToObject(tokeniser, options)
if (value === DONE) {
throw new Error(`${decodeErrPrefix} found map but not enough entries (got ${i} [no value], expected ${token.value})`)
}
if (useMaps) {
// @ts-ignore TODO reconsider this .. maybe needs to be strict about key types
m.set(key, value)
} else {
// @ts-ignore TODO reconsider this .. maybe needs to be strict about key types
obj[key] = value
}
}
// @ts-ignore c'mon man
return useMaps ? m : obj
}
/**
* @param {DecodeTokenizer} tokeniser
* @param {DecodeOptions} options
* @returns {any|BREAK|DONE}
*/
function tokensToObject (tokeniser, options) {
// should we support array as an argument?
// check for tokenIter[Symbol.iterator] and replace tokenIter with what that returns?
if (tokeniser.done()) {
return DONE
}
const token = tokeniser.next()
if (token.type === Type.break) {
return BREAK
}
if (token.type.terminal) {
return token.value
}
if (token.type === Type.array) {
return tokenToArray(token, tokeniser, options)
}
if (token.type === Type.map) {
return tokenToMap(token, tokeniser, options)
}
if (token.type === Type.tag) {
if (options.tags && typeof options.tags[token.value] === 'function') {
const tagged = tokensToObject(tokeniser, options)
return options.tags[token.value](tagged)
}
throw new Error(`${decodeErrPrefix} tag not supported (${token.value})`)
}
/* c8 ignore next */
throw new Error('unsupported')
}
/**
* @param {Uint8Array} data
* @param {DecodeOptions} [options]
* @returns {any}
*/
function decode (data, options) {
if (!(data instanceof Uint8Array)) {
throw new Error(`${decodeErrPrefix} data to decode must be a Uint8Array`)
}
options = Object.assign({}, defaultDecodeOptions, options)
const tokeniser = options.tokenizer || new Tokeniser(data, options)
const decoded = tokensToObject(tokeniser, options)
if (decoded === DONE) {
throw new Error(`${decodeErrPrefix} did not find any content to decode`)
}
if (decoded === BREAK) {
throw new Error(`${decodeErrPrefix} got unexpected break`)
}
if (!tokeniser.done()) {
throw new Error(`${decodeErrPrefix} too many terminals, data makes no sense`)
}
return decoded
}
export { Tokeniser, tokensToObject, decode }