Function parseTxt

packages/parser/src/formats/qq-native-txt.ts:145–328

(options: ParseOptions)

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

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

  const totalBytes = getFileSize(filePath)
  let bytesRead = 0
  let messagesProcessed = 0
  let skippedLines = 0 // 跳过的无效行计数

  // 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 QQ TXT 文件，大小: ${(totalBytes / 1024 / 1024).toFixed(2)} MB`)

  // 收集数据
  let groupName = '未知群聊'
  const memberMap = new Map<string, MemberInfo>()
  const messages: ParsedMessage[] = []

  // 当前正在解析的消息
  let currentMessage: {
    timestamp: number
    platformId: string
    nickname: string
    contentLines: string[]
  } | null = null

  // 保存当前消息
  const saveCurrentMessage = () => {
    if (currentMessage) {
      const content = currentMessage.contentLines.join('\n').trim()
      const type = detectMessageType(content)

      messages.push({
        senderPlatformId: currentMessage.platformId,
        senderAccountName: currentMessage.nickname, // QQ TXT 格式只有一个昵称，作为账号名称追踪历史
        // 不设置 senderGroupNickname，避免同一昵称被重复追踪
        timestamp: currentMessage.timestamp,
        type,
        content: content || null,
      })

      // 更新成员信息（保留最新昵称）
      memberMap.set(currentMessage.platformId, {
        platformId: currentMessage.platformId,
        nickname: currentMessage.nickname,
      })

      messagesProcessed++
    }
  }

  // 逐行读取文件
  const fileStream = fs.createReadStream(filePath, { encoding: 'utf-8' })
  const rl = readline.createInterface({
    input: fileStream,