Function parseChatLab

packages/parser/src/formats/chatlab.ts:87–275

(options: ParseOptions)

// ==================== 解析器实现 ====================

async function* parseChatLab(options: ParseOptions): AsyncGenerator<ParseEvent, void, unknown> {
  const { filePath, batchSize = 5000, onProgress, onLog } = options

  const totalBytes = getFileSize(filePath)
  let bytesRead = 0
  let messagesProcessed = 0

  // 发送初始进度
  const initialProgress = createProgress('parsing', 0, totalBytes, 0, '')
  yield { type: 'progress', data: initialProgress }
  onProgress?.(initialProgress)

  // 记录解析开始
  onLog?.('info', `开始解析 ChatLab 格式文件，大小: ${(totalBytes / 1024 / 1024).toFixed(2)} MB`)

  // 读取文件头获取 meta 和 members 信息
  const headContent = readFileHeadBytes(filePath, 200000)

  // 解析 meta
  let meta: ParsedMeta = {
    name: '未知群聊',
    platform: KNOWN_PLATFORMS.UNKNOWN,
    type: ChatType.GROUP,
  }
  try {
    // 使用更健壮的方式解析嵌套 JSON 对象
    // 因为 meta 可能包含 sources 数组（嵌套对象），简单的正则无法正确匹配
    const metaStartMatch = headContent.match(/"meta"\s*:\s*\{/)
    if (metaStartMatch && metaStartMatch.index !== undefined) {
      const startIndex = metaStartMatch.index + metaStartMatch[0].length - 1 // 指向 {
      let depth = 0
      let endIndex = startIndex

      // 遍历字符找到匹配的闭合 }
      for (let i = startIndex; i < headContent.length; i++) {
        const char = headContent[i]
        if (char === '{') {
          depth++
        } else if (char === '}') {
          depth--
          if (depth === 0) {
            endIndex = i
            break
          }
        }
      }

      if (endIndex > startIndex) {
        const metaJson = headContent.slice(startIndex, endIndex + 1)
        const metaObj = JSON.parse(metaJson)
        meta = {
          name: metaObj.name || '未知群聊',
          platform: metaObj.platform || KNOWN_PLATFORMS.UNKNOWN,
          type: (metaObj.type as ChatType) || ChatType.GROUP,
          groupId: metaObj.groupId,
          groupAvatar: metaObj.groupAvatar,
        }
      }