site/node_modules/micromark-core-commonmark/dev/lib/code-text.js

/**
 * @typedef {import('micromark-util-types').Construct} Construct
 * @typedef {import('micromark-util-types').Previous} Previous
 * @typedef {import('micromark-util-types').Resolver} Resolver
 * @typedef {import('micromark-util-types').State} State
 * @typedef {import('micromark-util-types').Token} Token
 * @typedef {import('micromark-util-types').TokenizeContext} TokenizeContext
 * @typedef {import('micromark-util-types').Tokenizer} Tokenizer
 */

import {markdownLineEnding} from 'micromark-util-character'
import {codes, types} from 'micromark-util-symbol'
import {ok as assert} from 'devlop'

/** @type {Construct} */
export const codeText = {
  name: 'codeText',
  tokenize: tokenizeCodeText,
  resolve: resolveCodeText,
  previous
}

// To do: next major: don’t resolve, like `markdown-rs`.
/** @type {Resolver} */
function resolveCodeText(events) {
  let tailExitIndex = events.length - 4
  let headEnterIndex = 3
  /** @type {number} */
  let index
  /** @type {number | undefined} */
  let enter

  // If we start and end with an EOL or a space.
  if (
    (events[headEnterIndex][1].type === types.lineEnding ||
      events[headEnterIndex][1].type === 'space') &&
    (events[tailExitIndex][1].type === types.lineEnding ||
      events[tailExitIndex][1].type === 'space')
  ) {
    index = headEnterIndex

    // And we have data.
    while (++index < tailExitIndex) {
      if (events[index][1].type === types.codeTextData) {
        // Then we have padding.
        events[headEnterIndex][1].type = types.codeTextPadding
        events[tailExitIndex][1].type = types.codeTextPadding
        headEnterIndex += 2
        tailExitIndex -= 2
        break
      }
    }
  }

  // Merge adjacent spaces and data.
  index = headEnterIndex - 1
  tailExitIndex++

  while (++index <= tailExitIndex) {
    if (enter === undefined) {
      if (
        index !== tailExitIndex &&
        events[index][1].type !== types.lineEnding
      ) {
        enter = index
      }
    } else if (
      index === tailExitIndex ||
      events[index][1].type === types.lineEnding
    ) {
      events[enter][1].type = types.codeTextData

      if (index !== enter + 2) {
        events[enter][1].end = events[index - 1][1].end
        events.splice(enter + 2, index - enter - 2)
        tailExitIndex -= index - enter - 2
        index = enter + 2
      }

      enter = undefined
    }
  }

  return events
}

/**
 * @this {TokenizeContext}
 * @type {Previous}
 */
function previous(code) {
  // If there is a previous code, there will always be a tail.
  return (
    code !== codes.graveAccent ||
    this.events[this.events.length - 1][1].type === types.characterEscape
  )
}

/**
 * @this {TokenizeContext}
 * @type {Tokenizer}
 */
function tokenizeCodeText(effects, ok, nok) {
  const self = this
  let sizeOpen = 0
  /** @type {number} */
  let size
  /** @type {Token} */
  let token

  return start

  /**
   * Start of code (text).
   *
   * ```markdown
   * > | `a`
   *     ^
   * > | \`a`
   *      ^
   * ```
   *
   * @type {State}
   */
  function start(code) {
    assert(code === codes.graveAccent, 'expected `` ` ``')
    assert(previous.call(self, self.previous), 'expected correct previous')
    effects.enter(types.codeText)
    effects.enter(types.codeTextSequence)
    return sequenceOpen(code)
  }

  /**
   * In opening sequence.
   *
   * ```markdown
   * > | `a`
   *     ^
   * ```
   *
   * @type {State}
   */
  function sequenceOpen(code) {
    if (code === codes.graveAccent) {
      effects.consume(code)
      sizeOpen++
      return sequenceOpen
    }

    effects.exit(types.codeTextSequence)
    return between(code)
  }

  /**
   * Between something and something else.
   *
   * ```markdown
   * > | `a`
   *      ^^
   * ```
   *
   * @type {State}
   */
  function between(code) {
    // EOF.
    if (code === codes.eof) {
      return nok(code)
    }

    // To do: next major: don’t do spaces in resolve, but when compiling,
    // like `markdown-rs`.
    // Tabs don’t work, and virtual spaces don’t make sense.
    if (code === codes.space) {
      effects.enter('space')
      effects.consume(code)
      effects.exit('space')
      return between
    }

    // Closing fence? Could also be data.
    if (code === codes.graveAccent) {
      token = effects.enter(types.codeTextSequence)
      size = 0
      return sequenceClose(code)
    }

    if (markdownLineEnding(code)) {
      effects.enter(types.lineEnding)
      effects.consume(code)
      effects.exit(types.lineEnding)
      return between
    }

    // Data.
    effects.enter(types.codeTextData)
    return data(code)
  }

  /**
   * In data.
   *
   * ```markdown
   * > | `a`
   *      ^
   * ```
   *
   * @type {State}
   */
  function data(code) {
    if (
      code === codes.eof ||
      code === codes.space ||
      code === codes.graveAccent ||
      markdownLineEnding(code)
    ) {
      effects.exit(types.codeTextData)
      return between(code)
    }

    effects.consume(code)
    return data
  }

  /**
   * In closing sequence.
   *
   * ```markdown
   * > | `a`
   *       ^
   * ```
   *
   * @type {State}
   */
  function sequenceClose(code) {
    // More.
    if (code === codes.graveAccent) {
      effects.consume(code)
      size++
      return sequenceClose
    }

    // Done!
    if (size === sizeOpen) {
      effects.exit(types.codeTextSequence)
      effects.exit(types.codeText)
      return ok(code)
    }

    // More or less accents: mark as data.
    token.type = types.codeTextData
    return data(code)
  }
}