Function extractIncludePathsFromTokens

src/utils/claudemd.ts:451–535

(
  tokens: ReturnType<Lexer['lex']>,
  basePath: string,
)

// absolute paths. Skips html tokens so @paths inside block comments are
// ignored — the caller may pass pre-strip tokens.
function extractIncludePathsFromTokens(
  tokens: ReturnType<Lexer['lex']>,
  basePath: string,
): string[] {
  const absolutePaths = new Set<string>()

  // Extract @paths from a text string and add resolved paths to absolutePaths.
  function extractPathsFromText(textContent: string) {
    const includeRegex = /(?:^|\s)@((?:[^\s\\]|\\ )+)/g
    let match
    while ((match = includeRegex.exec(textContent)) !== null) {
      let path = match[1]
      if (!path) continue

      // Strip fragment identifiers (#heading, #section-name, etc.)
      const hashIndex = path.indexOf('#')
      if (hashIndex !== -1) {
        path = path.substring(0, hashIndex)
      }
      if (!path) continue

      // Unescape the spaces in the path
      path = path.replace(/\\ /g, ' ')

      // Accept @path, @./path, @~/path, or @/path
      if (path) {
        const isValidPath =
          path.startsWith('./') ||
          path.startsWith('~/') ||
          (path.startsWith('/') && path !== '/') ||
          (!path.startsWith('@') &&
            !path.match(/^[#%^&*()]+/) &&
            path.match(/^[a-zA-Z0-9._-]/))

        if (isValidPath) {
          const resolvedPath = expandPath(path, dirname(basePath))
          absolutePaths.add(resolvedPath)
        }
      }
    }
  }

  // Recursively process elements to find text nodes
  function processElements(elements: MarkdownToken[]) {
    for (const element of elements) {
      if (element.type === 'code' || element.type === 'codespan') {
        continue
      }

      // For html tokens that contain comments, strip the comment spans and
      // check the residual for @paths (e.g. `<!-- note --> @./file.md`).
      // Other html tokens (non-comment tags) are skipped entirely.
      if (element.type === 'html') {
        const raw = element.raw || ''
        const trimmed = raw.trimStart()
        if (trimmed.startsWith('<!--') && trimmed.includes('-->')) {
          const commentSpan = /<!--[\s\S]*?-->/g
          const residue = raw.replace(commentSpan, '')