Function generateUsageReport

src/commands/insights.ts:2796–3023

(options?: {
  collectRemote?: boolean
})

// ============================================================================

export async function generateUsageReport(options?: {
  collectRemote?: boolean
}): Promise<{
  insights: InsightResults
  htmlPath: string
  data: AggregatedData
  remoteStats?: { hosts: RemoteHostInfo[]; totalCopied: number }
  facets: Map<string, SessionFacets>
}> {
  let remoteStats: { hosts: RemoteHostInfo[]; totalCopied: number } | undefined

  // Optionally collect data from remote hosts first (ant-only)
  if (process.env.USER_TYPE === 'ant' && options?.collectRemote) {
    const destDir = join(getClaudeConfigHomeDir(), 'projects')
    const { hosts, totalCopied } = await collectAllRemoteHostData(destDir)
    remoteStats = { hosts, totalCopied }
  }

  // Phase 1: Lite scan — filesystem metadata only (no JSONL parsing)
  const allScannedSessions = await scanAllSessions()
  const totalSessionsScanned = allScannedSessions.length

  // Phase 2: Load SessionMeta — use cache where available, parse only uncached
  // Read cached metas in parallel batches to avoid blocking the event loop
  const META_BATCH_SIZE = 50
  const MAX_SESSIONS_TO_LOAD = 200
  let allMetas: SessionMeta[] = []
  const uncachedSessions: LiteSessionInfo[] = []

  for (let i = 0; i < allScannedSessions.length; i += META_BATCH_SIZE) {
    const batch = allScannedSessions.slice(i, i + META_BATCH_SIZE)
    const results = await Promise.all(
      batch.map(async sessionInfo => ({
        sessionInfo,
        cached: await loadCachedSessionMeta(sessionInfo.sessionId),
      })),
    )
    for (const { sessionInfo, cached } of results) {
      if (cached) {
        allMetas.push(cached)
      } else if (uncachedSessions.length < MAX_SESSIONS_TO_LOAD) {
        uncachedSessions.push(sessionInfo)
      }
    }
  }

  // Load full message data only for uncached sessions and compute SessionMeta
  const logsForFacets = new Map<string, LogOption>()

  // Filter out /insights meta-sessions (facet extraction API calls get logged as sessions)
  const isMetaSession = (log: LogOption): boolean => {
    for (const msg of log.messages.slice(0, 5)) {
      if (msg.type === 'user' && msg.message) {
        const content = msg.message.content
        if (typeof content === 'string') {
          if (
            content.includes('RESPOND WITH ONLY A VALID JSON OBJECT') ||
            content.includes('record_facets')