feat: overhaul video player controller

This commit is contained in:
Media
2026-08-06 20:08:12 -05:00
parent 72cc66df92
commit 5e81403be3
10 changed files with 2019 additions and 122 deletions
+163 -23
View File
@@ -90,6 +90,7 @@ class DownloadManager:
self._lock = threading.Lock()
self._futures: set[Future[Any]] = set()
self._processes: set[subprocess.Popen[str]] = set()
self._deletions_in_flight: set[tuple[int, int]] = set()
self._stopping = False
def enqueue(
@@ -484,21 +485,43 @@ class DownloadManager:
slot_directory = canonical_slot_directory(self.config.storage.media_base, token.slot)
trash_root = self.config.storage.media_base / ".controller-trash"
trash_path: Path | None = None
if self._slot_is_guarded(token.slot):
return False
if os.path.lexists(slot_directory):
if slot_directory.is_symlink():
slot_directory.unlink()
else:
trash_root.mkdir(parents=True, exist_ok=True)
self._fsync_directory(self.config.storage.media_base)
trash_path = trash_root / (
f"slot{token.slot}.generation{token.generation}.{uuid.uuid4().hex}"
)
os.replace(slot_directory, trash_path)
self._fsync_directory(self.config.storage.media_base)
self._fsync_directory(trash_root)
if self._slot_is_guarded(token.slot):
if (
trash_path is not None
and not os.path.lexists(slot_directory)
and os.path.lexists(trash_path)
):
os.replace(trash_path, slot_directory)
self._fsync_directory(self.config.storage.media_base)
self._fsync_directory(trash_root)
return False
if trash_path is not None:
shutil.rmtree(trash_path)
self._fsync_directory(trash_root)
return self.store.finish_delete(token)
def _slot_is_guarded(self, slot: int) -> bool:
return any(
guard.slot == slot for guard in self.store.list_live_guards()
)
def expire_due_slots(self) -> list[int]:
"""Synchronously expire slots for maintenance callers and tests."""
deleted: list[int] = []
for token in self.store.begin_expired_deletions():
try:
@@ -508,47 +531,164 @@ class DownloadManager:
LOG.exception("could not finish expired deletion for slot %s", token.slot)
return deleted
def schedule_delete_slot(
self,
slot: int,
*,
reason: str,
expected_url: str | None = None,
) -> DeleteToken:
"""Atomically fence a manual deletion, then finish it off-thread."""
token = self.store.begin_delete(
slot,
reason=reason,
expected_url=expected_url,
)
self._schedule_deletion(token)
return token
def schedule_expired_deletions(self) -> list[DeleteToken]:
"""Fence all currently due slots and queue filesystem work."""
tokens = self.store.begin_expired_deletions()
for token in tokens:
self._schedule_deletion(token)
return tokens
def recover_pending_deletions(self) -> list[int]:
recovered: list[int] = []
trash_root = self.config.storage.media_base / ".controller-trash"
for token in self.store.pending_deletions():
if any(
guard.slot == token.slot
for guard in self.store.list_live_guards()
):
LOG.warning(
"deferring interrupted deletion for protected slot %s",
token.slot,
)
continue
if trash_root.is_dir():
prefix = f"slot{token.slot}.generation{token.generation}."
for candidate in trash_root.iterdir():
if candidate.name.startswith(prefix):
if candidate.is_symlink() or candidate.is_file():
candidate.unlink()
elif candidate.is_dir():
shutil.rmtree(candidate)
if self.finish_deletion(token):
if self._recover_deletion(token):
recovered.append(token.slot)
return recovered
def schedule_pending_deletion_recovery(self) -> bool:
"""Run recoverable trash cleanup without delaying service startup."""
"""Queue every safe pending deletion without blocking the coordinator."""
with self._lock:
if self._stopping:
return False
scheduled = False
for token in self.store.pending_deletions():
scheduled = self._schedule_deletion(token) or scheduled
return scheduled
def _schedule_deletion(self, token: DeleteToken) -> bool:
key = (token.slot, token.generation)
with self._lock:
if self._stopping or key in self._deletions_in_flight:
return False
self._deletions_in_flight.add(key)
try:
future = self._executor.submit(self.recover_pending_deletions)
except BaseException:
LOG.exception("could not schedule interrupted deletion recovery")
future = self._executor.submit(self._deletion_worker, token)
except BaseException as exc:
with self._lock:
self._deletions_in_flight.discard(key)
self._emit(
DownloadEvent(
slot=token.slot,
state="delete_error",
message=f"could not start deletion worker: {exc}",
generation=token.generation,
)
)
return False
with self._lock:
self._futures.add(future)
future.add_done_callback(self._forget_future)
return True
def _deletion_worker(self, token: DeleteToken) -> None:
key = (token.slot, token.generation)
try:
if self._recover_deletion(token):
self._emit(
DownloadEvent(
slot=token.slot,
state="deleted",
generation=token.generation,
)
)
elif self.store.get_slot(token.slot).media_state == "deleting":
self._emit(
DownloadEvent(
slot=token.slot,
state="delete_deferred",
message="slot deletion was deferred by playback protection",
generation=token.generation,
)
)
except BaseException as exc:
LOG.exception(
"could not finish deletion for slot %s generation %s",
token.slot,
token.generation,
)
self._emit(
DownloadEvent(
slot=token.slot,
state="delete_error",
message=f"deletion remains pending: {exc}",
generation=token.generation,
)
)
finally:
with self._lock:
self._deletions_in_flight.discard(key)
def _recover_deletion(self, token: DeleteToken) -> bool:
trash_root = self.config.storage.media_base / ".controller-trash"
slot_directory = canonical_slot_directory(
self.config.storage.media_base,
token.slot,
)
if self._slot_is_guarded(token.slot):
LOG.warning(
"deferring interrupted deletion for protected slot %s",
token.slot,
)
self._restore_guarded_trash(token, trash_root, slot_directory)
return False
if trash_root.is_dir():
prefix = f"slot{token.slot}.generation{token.generation}."
for candidate in trash_root.iterdir():
if not candidate.name.startswith(prefix):
continue
if self._slot_is_guarded(token.slot):
self._restore_guarded_trash(
token,
trash_root,
slot_directory,
)
return False
if candidate.is_symlink() or candidate.is_file():
candidate.unlink()
elif candidate.is_dir():
shutil.rmtree(candidate)
self._fsync_directory(trash_root)
return self.finish_deletion(token)
def _restore_guarded_trash(
self,
token: DeleteToken,
trash_root: Path,
slot_directory: Path,
) -> None:
if not trash_root.is_dir() or os.path.lexists(slot_directory):
return
prefix = f"slot{token.slot}.generation{token.generation}."
candidates = sorted(
candidate
for candidate in trash_root.iterdir()
if candidate.name.startswith(prefix)
)
if not candidates:
return
os.replace(candidates[0], slot_directory)
self._fsync_directory(self.config.storage.media_base)
self._fsync_directory(trash_root)
def shutdown(self, *, wait: bool = True) -> None:
with self._lock:
self._stopping = True
+110 -33
View File
@@ -18,6 +18,7 @@ from .store import ControllerStore
TERMINAL_REASONS = frozenset(
{"stop", "finished", "mpv_error", "sigterm", "sigint", "shutdown"}
)
TERMINAL_ERROR_REASONS = frozenset({"startup_error", "worker_start_error"})
CONTROL_ACTIONS = frozenset({"play", "stop", "pause", "resume", "toggle"})
@@ -485,13 +486,31 @@ class PlaybackReducer:
return effects
if state == "busy":
if slot is not None or path is not None:
previous = self.store.player_observation()
if (
slot is not None
or path is not None
or previous.active_slot is not None
or previous.active_path is not None
):
self._active_evidence(
slot=slot,
path=path,
paused=None,
status=payload,
)
else:
self.store.set_player_observation(
availability="online",
phase="unknown",
active_slot=None,
active_path=None,
paused=None,
reconciliation_state="unknown",
reconciliation_due_at=None,
reconciliation_id=None,
status_payload=payload,
)
effects.append(PlaybackEffect("player_busy", slot=slot, message=payload.get("message")))
if self.pending is not None and self.pending.action == "play":
effects.append(self._finish_operation("failed", "Player is busy"))
@@ -499,40 +518,87 @@ class PlaybackReducer:
if state == "error":
previous = self.store.player_observation()
message = str(payload.get("message") or "Player error")
reason = payload.get("reason")
if reason in TERMINAL_ERROR_REASONS:
terminal_slot = slot if slot is not None else previous.active_slot
self._idle_evidence(status=payload)
effects.extend(
(
PlaybackEffect(
"player_error",
slot=terminal_slot,
message=message,
),
PlaybackEffect(
"player_idle",
slot=terminal_slot,
message=str(reason),
),
)
)
if self.pending is not None:
effects.append(self._finish_operation("failed", message))
return effects
active_slot = slot if slot is not None else previous.active_slot
active_path = path if path is not None else previous.active_path
due = None
if self.config.storage.uncertain_review_after_seconds > 0:
due = (
self._wall_clock()
+ self.config.storage.uncertain_review_after_seconds
)
if active_slot is not None:
self.store.reconcile_active_guard(
guard = self.store.mark_slot_uncertain(
active_slot,
kind="player_active",
details={"error": payload.get("message")},
review_after=due,
details={"reason": "Player error", "message": message},
)
due = guard.review_after
self.store.set_player_observation(
availability="online",
phase="error",
active_slot=active_slot,
active_path=active_path,
paused=(
previous.paused
if self._same_playback_identity(
previous,
active_slot,
active_path,
)
else None
),
reconciliation_state=(
"operator_review"
if guard.state == "review_required"
else "awaiting_evidence"
),
reconciliation_due_at=due,
reconciliation_id=guard.guard_id,
status_payload=payload,
)
elif active_path is not None:
self.store.reconcile_external_active()
self.store.set_player_observation(
availability="online",
phase="error",
active_slot=active_slot,
active_path=active_path,
paused=(
previous.paused
if self._same_playback_identity(
previous,
active_slot,
active_path,
)
else None
),
reconciliation_state=(
"confirmed"
if active_slot is not None or active_path is not None
else "unknown"
),
reconciliation_due_at=None,
reconciliation_id=None,
status_payload=payload,
)
message = str(payload.get("message") or "Player error")
self.store.mark_external_uncertain(
active_path,
review_after=due,
availability="online",
details={"reason": "Player error", "message": message},
status_payload=payload,
)
else:
self.store.set_player_observation(
availability="online",
phase="unknown",
active_slot=None,
active_path=None,
paused=None,
reconciliation_state="unknown",
reconciliation_due_at=None,
reconciliation_id=None,
status_payload=payload,
)
effects.append(PlaybackEffect("player_error", slot=active_slot, message=message))
if self.pending is not None:
effects.append(self._finish_operation("failed", message))
@@ -645,6 +711,19 @@ class PlaybackReducer:
"reason": "stop timeout",
},
)
elif operation.action in {"pause", "resume", "toggle"}:
previous = self.store.player_observation()
if previous.active_slot is not None or previous.active_path is not None:
self.store.set_player_observation(
availability=previous.availability,
phase="unknown",
active_slot=previous.active_slot,
active_path=previous.active_path,
paused=None,
reconciliation_state=previous.reconciliation_state,
reconciliation_due_at=previous.reconciliation_due_at,
reconciliation_id=previous.reconciliation_id,
)
return self._finish_operation("timed_out", "no authoritative Player confirmation")
def _mark_play_uncertain(self, operation: PlayerOperation, reason: str) -> None:
@@ -753,8 +832,6 @@ class PlaybackReducer:
def projected_slot_state(self, slot: int) -> str:
record = self.store.get_slot(slot)
if record.media_state != "ready":
return record.media_state
observation = self.store.player_observation()
if observation.active_slot == slot:
if observation.reconciliation_state == "operator_review":
@@ -771,9 +848,9 @@ class PlaybackReducer:
)
if guard is not None:
if guard.kind == "pending_play":
return "ready"
return record.media_state
return "operator_review" if guard.state == "review_required" else "unknown"
return "ready"
return record.media_state
def mark_due_reconciliation(self) -> list[PlaybackEffect]:
now = self._wall_clock()
+175 -49
View File
@@ -99,7 +99,7 @@ class ControllerService:
self._processed_packets: dict[tuple[int, str, int], float] = {}
self._player_evidence_ready = False
self._operator_cleanup_override = False
self._recovery_scheduled = False
self._cleanup_review_id: str | None = None
@property
def started(self) -> bool:
@@ -149,9 +149,7 @@ class ControllerService:
)
# Cleanup is deliberately deferred until fresh Player evidence
# (or an explicit administrative reconciliation) makes it safe.
self._player_evidence_ready = False
self._operator_cleanup_override = False
self._recovery_scheduled = False
self._reset_cleanup_evidence()
self._stop_event.clear()
self._next_expiry = (
self._monotonic()
@@ -271,8 +269,13 @@ class ControllerService:
guards = self.store.list_live_guards() if started else []
player_operation_pending = bool(
started
and self.playback is not None
and self.playback.pending is not None
and (
self._pending_play is not None
or (
self.playback is not None
and self.playback.pending is not None
)
)
)
mqtt_connected = bool(started and self.mqtt.connected)
degraded: list[str] = []
@@ -321,6 +324,22 @@ class ControllerService:
"clearable": not player_operation_pending,
}
)
cleanup_reconciliation = None
if (
started
and not self._cleanup_allowed()
and self._cleanup_review_id is not None
and not guards
and player_observation is not None
and player_observation.active_slot is None
and player_observation.active_path is None
):
cleanup_reconciliation = {
"review_id": self._cleanup_review_id,
"kind": "player_evidence",
"state": "awaiting_evidence",
"clearable": not player_operation_pending,
}
return {
"status": (
"stopped"
@@ -346,6 +365,7 @@ class ControllerService:
}
),
"reconciliation_reviews": reviews,
"cleanup_reconciliation": cleanup_reconciliation,
"degraded_reasons": degraded,
}
@@ -416,16 +436,14 @@ class ControllerService:
handled_successfully = True
return
if event.kind == "connected":
self._player_evidence_ready = False
self._operator_cleanup_override = False
self._reset_cleanup_evidence()
self._announce_discovery()
self._sync_slot_states(force=True)
self._announce_reconciliation_reviews()
handled_successfully = True
return
if event.kind == "disconnected":
self._player_evidence_ready = False
self._operator_cleanup_override = False
self._reset_cleanup_evidence()
assert self.playback is not None
assert self.router is not None
playback_effect = self.playback.mark_transport_uncertain(
@@ -441,30 +459,41 @@ class ControllerService:
handled_successfully = True
return
if event.topic == self.config.topics.player_status:
payload = self._decode_json(event.payload, "Player status")
if payload is not None:
assert self.playback is not None
before_generation = self.store.player_observation().generation
self._handle_playback_effects(self.playback.handle_status(payload))
self._refresh_player_evidence(before_generation)
if event.retain:
self._publish_event(
"retained_player_evidence_rejected",
"warning",
"Retained Player status was ignored as historical evidence",
)
else:
payload = self._decode_json(event.payload, "Player status")
if payload is not None:
assert self.playback is not None
before_generation = self.store.player_observation().generation
effects = self.playback.handle_status(payload)
self._handle_playback_effects(effects)
self._refresh_player_evidence(before_generation, effects)
elif event.topic == self.config.topics.player_telemetry:
payload = self._decode_json(event.payload, "Player telemetry")
if payload is not None:
assert self.playback is not None
before_generation = self.store.player_observation().generation
self._handle_playback_effects(self.playback.handle_telemetry(payload))
self._refresh_player_evidence(before_generation)
if event.retain:
self._publish_event(
"retained_player_evidence_rejected",
"warning",
"Retained Player telemetry was ignored as historical evidence",
)
else:
payload = self._decode_json(event.payload, "Player telemetry")
if payload is not None:
assert self.playback is not None
before_generation = self.store.player_observation().generation
effects = self.playback.handle_telemetry(payload)
self._handle_playback_effects(effects)
self._refresh_player_evidence(before_generation, effects)
elif event.topic == self.config.topics.player_state:
try:
flat_state = event.payload.decode("utf-8").strip()
except UnicodeDecodeError:
flat_state = ""
# Retained flat idle is sufficient only as a quiescence gate;
# it never mutates detailed playback state or clears guards.
self._player_evidence_ready = flat_state == "idle"
self._operator_cleanup_override = False
if self._player_evidence_ready:
self._schedule_deletion_recovery()
# The Player owns this retained Companion topic. Its value is
# historical across outages and reconnect deliberately emits
# flat ``idle`` even while MPV is active, so it is never fresh
# cleanup authority.
pass
elif event.topic == self.config.topics.controller_command:
if event.retain:
self._publish_event(
@@ -511,8 +540,9 @@ class ControllerService:
else:
decision = self.router.observe_mode(mode)
self._handle_router_decision(decision)
# Flat state is only a cold-start cleanup gate. Detailed playback
# truth continues to come exclusively from status and telemetry.
# The flat state subscription is compatibility observation only.
# Playback and cleanup truth come exclusively from detailed status
# and telemetry or an explicit administrative reconciliation.
handled_successfully = True
finally:
if handled_successfully:
@@ -873,17 +903,20 @@ class ControllerService:
expected_url = payload.get("url")
if expected_url is not None and not isinstance(expected_url, str):
raise ValueError("delete URL guard must be a string")
self.downloads.delete_slot(
self.downloads.schedule_delete_slot(
slot,
reason="manual",
expected_url=expected_url,
)
self._announce_slot(slot, force=True)
return {"status": "deleted", "slot": slot}
return {"status": "deleting", "slot": slot}
def _request_reconciliation(self, payload: Mapping[str, Any]) -> dict[str, Any]:
if payload.get("decision") != "clear" or payload.get("confirm") is not True:
raise ValueError("reconciliation clear requires decision=clear and confirm=true")
scope = payload.get("scope")
if scope not in {None, "cleanup"}:
raise ValueError("reconciliation scope must be cleanup when provided")
slot = (
self._parse_slot(payload.get("slot"))
if payload.get("slot") is not None
@@ -899,7 +932,36 @@ class ControllerService:
raise ValueError(
"playback reconciliation cannot be cleared while a Player operation is pending"
)
if slot is None:
if scope == "cleanup":
if slot is not None:
raise ValueError("cleanup reconciliation does not accept a slot")
observation = self.store.player_observation()
if (
self._cleanup_allowed()
or self._cleanup_review_id is None
or review_id.strip() != self._cleanup_review_id
):
raise StaleOperationError("cleanup reconciliation review_id is stale")
if (
self.store.list_live_guards()
or observation.active_slot is not None
or observation.active_path is not None
):
raise SlotProtectedError(
"cleanup reconciliation cannot bypass known playback protection"
)
self.store.record_audit_event(
"player_evidence_admin_override",
actor=actor.strip(),
details={
"review_id": review_id.strip(),
"reason": reason.strip(),
},
)
self._operator_cleanup_override = True
self._cleanup_review_id = None
cleared = True
elif slot is None:
cleared = self.store.clear_external_reconciliation(
review_id.strip(),
actor=actor.strip(),
@@ -933,7 +995,9 @@ class ControllerService:
and observation.active_slot is None
and observation.active_path is None
):
self._operator_cleanup_override = True
if scope != "cleanup":
self._operator_cleanup_override = True
self._cleanup_review_id = None
self._schedule_deletion_recovery()
assert self.router is not None
self.router.abandon(
@@ -954,6 +1018,8 @@ class ControllerService:
"status": "cleared",
"review_id": review_id.strip(),
}
if scope == "cleanup":
result["scope"] = "cleanup"
if slot is not None:
result["slot"] = slot
return result
@@ -996,7 +1062,11 @@ class ControllerService:
no_active_player = (
observation.active_slot is None and observation.active_path is None
)
if terminal or (failed_play and no_active_player):
unidentified_busy = (
no_active_player
and any(effect.kind == "player_busy" for effect in effects)
)
if terminal or (failed_play and no_active_player and not unidentified_busy):
self._handle_router_decision(self.router.release(self._session_lease_id))
self._session_lease_id = None
for effect in effects:
@@ -1028,6 +1098,27 @@ class ControllerService:
self._sync_slot_states()
def _handle_download_event(self, event: DownloadEvent) -> None:
if event.state == "deleted":
self._announce_slot(event.slot, force=True)
return
if event.state == "delete_deferred":
self._announce_slot(event.slot, force=True)
return
if event.state == "delete_error":
message = event.message or "slot deletion remains pending"
self._publish_event(
"slot_deletion_pending",
"warning",
message,
slot=event.slot,
)
self._announce_slot(event.slot, force=True)
retry_delay = max(
1.0,
min(30.0, self.config.storage.expiry_interval_seconds),
)
self._next_expiry = self._monotonic() + retry_delay
return
payload: dict[str, Any] = {"slot": event.slot, "state": event.state}
for key in ("title", "count", "message", "url", "submitter", "note"):
value = getattr(event, key)
@@ -1171,21 +1262,55 @@ class ControllerService:
return self._player_evidence_ready or self._operator_cleanup_override
def _schedule_deletion_recovery(self) -> None:
if self._recovery_scheduled or not self._cleanup_allowed():
if not self._cleanup_allowed():
return
assert self.downloads is not None
self._recovery_scheduled = (
self.downloads.schedule_pending_deletion_recovery()
)
self.downloads.schedule_pending_deletion_recovery()
def _refresh_player_evidence(self, before_generation: int) -> None:
def _refresh_player_evidence(
self,
before_generation: int,
effects: list[PlaybackEffect],
) -> None:
observation = self.store.player_observation()
if observation.generation == before_generation:
return
self._operator_cleanup_override = False
self._player_evidence_ready = observation.availability == "online"
if self._player_evidence_ready and observation.phase == "idle":
kinds = {effect.kind for effect in effects}
identity_known = (
observation.active_slot is not None
or observation.active_path is not None
)
safe_evidence = "player_idle" in kinds or (
identity_known
and bool(
kinds
& {
"player_active",
"player_busy",
"player_reconnected_active",
"player_telemetry",
"control_issued",
}
)
)
if safe_evidence and observation.availability == "online":
self._player_evidence_ready = True
self._operator_cleanup_override = False
self._cleanup_review_id = None
self._schedule_deletion_recovery()
return
self._mark_cleanup_unknown()
def _reset_cleanup_evidence(self) -> None:
self._player_evidence_ready = False
self._operator_cleanup_override = False
self._cleanup_review_id = uuid.uuid4().hex
def _mark_cleanup_unknown(self) -> None:
self._player_evidence_ready = False
self._operator_cleanup_override = False
if self._cleanup_review_id is None:
self._cleanup_review_id = uuid.uuid4().hex
def _run_timers(self) -> None:
assert self.playback is not None
@@ -1203,6 +1328,7 @@ class ControllerService:
self._handle_playback_effects(self.playback.mark_due_reconciliation())
self._next_reconciliation = now + 1.0
if now >= self._next_expiry and self._cleanup_allowed():
for slot in self.downloads.expire_due_slots():
self._announce_slot(slot, force=True)
self._schedule_deletion_recovery()
for token in self.downloads.schedule_expired_deletions():
self._announce_slot(token.slot, force=True)
self._next_expiry = now + self.config.storage.expiry_interval_seconds
+28 -5
View File
@@ -704,7 +704,16 @@ class ControllerStore:
title = str(legacy.get("title") or "")
count = _coerce_count(legacy.get("item_count", 0), slot, warnings)
created_at = _utc_timestamp(legacy.get("created_at"), f"slot {slot} created_at", warnings)
expires_at = _utc_timestamp(legacy.get("expires_at"), f"slot {slot} expires_at", warnings)
expires_at = _utc_timestamp(
legacy.get("expires_at"),
f"slot {slot} expires_at",
warnings,
)
if media_state == "error" and expires_at is None:
expires_at = now + self.config.retention_seconds
warnings.append(
f"slot {slot}: missing error expiry defaulted to retention window"
)
generation = 1
connection.execute(
"""
@@ -816,10 +825,17 @@ class ControllerStore:
connection.execute(
"""
UPDATE slots
SET media_state = 'error', last_error = ?, updated_at = ?
SET media_state = 'error', last_error = ?,
expires_at = COALESCE(expires_at, ?), updated_at = ?
WHERE slot_number = ? AND generation = ?
""",
(message, now, row["slot_number"], row["generation"]),
(
message,
now + self.config.retention_seconds,
now,
row["slot_number"],
row["generation"],
),
)
def list_slots(self) -> list[SlotRecord]:
@@ -1060,10 +1076,17 @@ class ControllerStore:
connection.execute(
"""
UPDATE slots
SET media_state = 'error', last_error = ?, updated_at = ?
SET media_state = 'error', last_error = ?,
expires_at = COALESCE(expires_at, ?), updated_at = ?
WHERE slot_number = ? AND generation = ?
""",
(error, now, slot, generation),
(
error,
now + self.config.retention_seconds,
now,
slot,
generation,
),
)
return True