fix: red-team round 3 hardening across desktop, mobile, core and server

Batch of red-team r3 fixes that were in the working tree before the
2026-09-28 design overhaul, committed as one unit with their tests.

- desktop main: STT timeouts and sidecar, voice recording store, sync
  (credentials, audio, knowledge reindex, push gates), runtime
  provisioner, update policy, AltGr keybindings, voice-command policy,
  dictionary file codec/limits, meeting transcript condensing and a
  local recording ledger so interrupted-session recovery only closes
  meetings this device recorded (a phone's live meeting is left alone).
- mobile: login CSRF via implicit token callbacks rejected, account
  deletion/retention, durable queue retention, knowledge realtime
  without unfiltered DELETE, meeting re-record failure paths, cloud STT
  client, preferences store/resync.
- core: text chunking splits long unbroken transcripts to fit, template
  field policy, dictionary limits, meeting markdown inline handling.
- server: payple webhook policy and cancellation order scope, meeting
  document generation quota, team RPC null-role guard, unified LLM
  quota in-flight accounting, knowledge chunk vector index, meeting
  re-record failure paths (migrations 20260929*).
- ci: portable/runtime feed gates, update-policy schema, Forgejo file
  delete and alias planning.

Four older tests are updated to the new contracts rather than the old
behavior: token-pair auth callbacks are rejected, knowledge realtime no
longer subscribes to DELETE, long transcript lines are split, and
meeting recovery requires the local recording ledger for empty rows.
This commit is contained in:
Yun Chan 2026-09-28 20:45:52 +09:00
parent 2428ede03d
commit ba9ef9741e
161 changed files with 17056 additions and 2379 deletions

View file

@ -0,0 +1,57 @@
import { RecorderBusyError } from './recorder/recorder-errors'
import type {
RecorderOwner,
RecorderSession,
RecordingRuntimeSnapshot,
} from './recorder/recorder-types'
/** The part of the shared recorder facade this policy needs. */
export interface CaptureRecorderPort {
acquire(owner: RecorderOwner): Promise<RecorderSession>
}
function isLive(snapshot: RecordingRuntimeSnapshot): boolean {
return snapshot.state === 'recording' || snapshot.state === 'paused'
}
function isReattachable(snapshot: RecordingRuntimeSnapshot): boolean {
return (snapshot.state === 'stopped' || snapshot.state === 'recoverable')
&& snapshot.recording !== null
}
/**
* Stops the microphone when the signed-in session is lost involuntarily, but
* keeps the meeting capture's file so its owner can process it after signing
* back in (Record restores a stopped capture as a recoverable recording).
*
* - A Talk turn is transient conversation input, not durable work: discarded.
* - A capture the backend cannot hand back after the screen unmounts (the
* in-process iOS recorder reports no reattachable snapshot) is discarded as
* before, because a stranded stopped capture would block every new
* recording.
*/
export async function stopLiveCaptureKeepingFile(recorder: CaptureRecorderPort): Promise<void> {
let session: RecorderSession
try {
session = await recorder.acquire('record')
} catch (error) {
if (!(error instanceof RecorderBusyError)) throw error
const holder = await recorder.acquire(error.holder)
await holder.cancel()
return
}
try {
await session.stop()
} catch {
// Nothing is driven by this JS process (idle, already stopped, or a
// foreground-service capture that outlived a restart): inspect it below.
}
let snapshot = await session.restore()
if (isLive(snapshot)) {
await session.stop()
snapshot = await session.restore()
}
if (isReattachable(snapshot)) return
await session.cancel()
}