site/node_modules/micromark-util-subtokenize/index.js
2024-10-14 08:09:33 +02:00

225 lines
6.2 KiB
JavaScript
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

/**
* @typedef {import('micromark-util-types').Chunk} Chunk
* @typedef {import('micromark-util-types').Event} Event
* @typedef {import('micromark-util-types').Token} Token
*/
import {splice} from 'micromark-util-chunked'
/**
* Tokenize subcontent.
*
* @param {Array<Event>} events
* List of events.
* @returns {boolean}
* Whether subtokens were found.
*/ // eslint-disable-next-line complexity
export function subtokenize(events) {
/** @type {Record<string, number>} */
const jumps = {}
let index = -1
/** @type {Event} */
let event
/** @type {number | undefined} */
let lineIndex
/** @type {number} */
let otherIndex
/** @type {Event} */
let otherEvent
/** @type {Array<Event>} */
let parameters
/** @type {Array<Event>} */
let subevents
/** @type {boolean | undefined} */
let more
while (++index < events.length) {
while (index in jumps) {
index = jumps[index]
}
event = events[index]
// Add a hook for the GFM tasklist extension, which needs to know if text
// is in the first content of a list item.
if (
index &&
event[1].type === 'chunkFlow' &&
events[index - 1][1].type === 'listItemPrefix'
) {
subevents = event[1]._tokenizer.events
otherIndex = 0
if (
otherIndex < subevents.length &&
subevents[otherIndex][1].type === 'lineEndingBlank'
) {
otherIndex += 2
}
if (
otherIndex < subevents.length &&
subevents[otherIndex][1].type === 'content'
) {
while (++otherIndex < subevents.length) {
if (subevents[otherIndex][1].type === 'content') {
break
}
if (subevents[otherIndex][1].type === 'chunkText') {
subevents[otherIndex][1]._isInFirstContentOfListItem = true
otherIndex++
}
}
}
}
// Enter.
if (event[0] === 'enter') {
if (event[1].contentType) {
Object.assign(jumps, subcontent(events, index))
index = jumps[index]
more = true
}
}
// Exit.
else if (event[1]._container) {
otherIndex = index
lineIndex = undefined
while (otherIndex--) {
otherEvent = events[otherIndex]
if (
otherEvent[1].type === 'lineEnding' ||
otherEvent[1].type === 'lineEndingBlank'
) {
if (otherEvent[0] === 'enter') {
if (lineIndex) {
events[lineIndex][1].type = 'lineEndingBlank'
}
otherEvent[1].type = 'lineEnding'
lineIndex = otherIndex
}
} else {
break
}
}
if (lineIndex) {
// Fix position.
event[1].end = Object.assign({}, events[lineIndex][1].start)
// Switch container exit w/ line endings.
parameters = events.slice(lineIndex, index)
parameters.unshift(event)
splice(events, lineIndex, index - lineIndex + 1, parameters)
}
}
}
return !more
}
/**
* Tokenize embedded tokens.
*
* @param {Array<Event>} events
* @param {number} eventIndex
* @returns {Record<string, number>}
*/
function subcontent(events, eventIndex) {
const token = events[eventIndex][1]
const context = events[eventIndex][2]
let startPosition = eventIndex - 1
/** @type {Array<number>} */
const startPositions = []
const tokenizer =
token._tokenizer || context.parser[token.contentType](token.start)
const childEvents = tokenizer.events
/** @type {Array<[number, number]>} */
const jumps = []
/** @type {Record<string, number>} */
const gaps = {}
/** @type {Array<Chunk>} */
let stream
/** @type {Token | undefined} */
let previous
let index = -1
/** @type {Token | undefined} */
let current = token
let adjust = 0
let start = 0
const breaks = [start]
// Loop forward through the linked tokens to pass them in order to the
// subtokenizer.
while (current) {
// Find the position of the event for this token.
while (events[++startPosition][1] !== current) {
// Empty.
}
startPositions.push(startPosition)
if (!current._tokenizer) {
stream = context.sliceStream(current)
if (!current.next) {
stream.push(null)
}
if (previous) {
tokenizer.defineSkip(current.start)
}
if (current._isInFirstContentOfListItem) {
tokenizer._gfmTasklistFirstContentOfListItem = true
}
tokenizer.write(stream)
if (current._isInFirstContentOfListItem) {
tokenizer._gfmTasklistFirstContentOfListItem = undefined
}
}
// Unravel the next token.
previous = current
current = current.next
}
// Now, loop back through all events (and linked tokens), to figure out which
// parts belong where.
current = token
while (++index < childEvents.length) {
if (
// Find a void token that includes a break.
childEvents[index][0] === 'exit' &&
childEvents[index - 1][0] === 'enter' &&
childEvents[index][1].type === childEvents[index - 1][1].type &&
childEvents[index][1].start.line !== childEvents[index][1].end.line
) {
start = index + 1
breaks.push(start)
// Help GC.
current._tokenizer = undefined
current.previous = undefined
current = current.next
}
}
// Help GC.
tokenizer.events = []
// If theres one more token (which is the cases for lines that end in an
// EOF), thats perfect: the last point we found starts it.
// If there isnt then make sure any remaining content is added to it.
if (current) {
// Help GC.
current._tokenizer = undefined
current.previous = undefined
} else {
breaks.pop()
}
// Now splice the events from the subtokenizer into the current events,
// moving back to front so that splice indices arent affected.
index = breaks.length
while (index--) {
const slice = childEvents.slice(breaks[index], breaks[index + 1])
const start = startPositions.pop()
jumps.unshift([start, start + slice.length - 1])
splice(events, start, 2, slice)
}
index = -1
while (++index < jumps.length) {
gaps[adjust + jumps[index][0]] = adjust + jumps[index][1]
adjust += jumps[index][1] - jumps[index][0] - 1
}
return gaps
}