Files
govoplan-committee/src/govoplan_committee/backend/dsar_provider.py
T

397 lines
14 KiB
Python

from __future__ import annotations
from collections.abc import Sequence
from dataclasses import dataclass
from datetime import datetime, timezone
from sqlalchemy.orm import Session
from govoplan_committee.backend.db.models import (
CommitteeDecisionProjection,
CommitteeWorkspaceEvent,
CommitteeWorkspaceRevision,
)
from govoplan_core.core.dsar import (
DsarErasureActionRef,
DsarExecutionResultRef,
DsarRecordRef,
DsarSubjectRef,
dsar_capability_name,
)
COMMITTEE_DSAR_CAPABILITY = dsar_capability_name("committee")
_MAX_RECORDS = 5_000
_CONFLICT = object()
@dataclass(frozen=True, slots=True)
class _SubjectSelectors:
actor_ids: tuple[str, ...]
object_kind: str | None
object_id: str | None
decision_id: str | None
class CommitteeDsarProvider:
provider_id = "committee"
module_id = "committee"
def search_subject(
self,
session: object,
*,
tenant_id: str,
subject: DsarSubjectRef,
) -> Sequence[DsarRecordRef]:
db = _session(session)
selectors = _subject_selectors(subject)
if selectors is None:
return ()
records: list[DsarRecordRef] = []
revisions = db.query(CommitteeWorkspaceRevision).filter(
CommitteeWorkspaceRevision.tenant_id == tenant_id,
CommitteeWorkspaceRevision.changed_by.in_(selectors.actor_ids),
)
events = db.query(CommitteeWorkspaceEvent).filter(
CommitteeWorkspaceEvent.tenant_id == tenant_id,
CommitteeWorkspaceEvent.actor_id.in_(selectors.actor_ids),
)
decisions = db.query(CommitteeDecisionProjection).filter(
CommitteeDecisionProjection.tenant_id == tenant_id,
CommitteeDecisionProjection.changed_by.in_(selectors.actor_ids),
)
if selectors.object_kind:
revisions = revisions.filter(
CommitteeWorkspaceRevision.object_kind == selectors.object_kind
)
events = events.filter(
CommitteeWorkspaceEvent.object_kind == selectors.object_kind
)
if selectors.object_id:
revisions = revisions.filter(
CommitteeWorkspaceRevision.object_id == selectors.object_id
)
events = events.filter(
CommitteeWorkspaceEvent.object_id == selectors.object_id
)
if selectors.object_kind == "meeting":
decisions = decisions.filter(
CommitteeDecisionProjection.meeting_id == selectors.object_id
)
elif selectors.object_kind in {"agenda", "agenda_item"}:
decisions = decisions.filter(
CommitteeDecisionProjection.agenda_item_id == selectors.object_id
)
elif selectors.object_kind == "decision":
decisions = decisions.filter(
CommitteeDecisionProjection.decision_id == selectors.object_id
)
if selectors.decision_id:
decisions = decisions.filter(
CommitteeDecisionProjection.decision_id == selectors.decision_id
)
records.extend(
_revision_attribution(row)
for row in _limited(
revisions,
CommitteeWorkspaceRevision.recorded_at,
CommitteeWorkspaceRevision.id,
label="workspace revision attribution",
)
)
records.extend(
_event_attribution(row)
for row in _limited(
events,
CommitteeWorkspaceEvent.occurred_at,
CommitteeWorkspaceEvent.id,
label="workspace event attribution",
)
)
records.extend(
_decision_attribution(row)
for row in _limited(
decisions,
CommitteeDecisionProjection.recorded_at,
CommitteeDecisionProjection.id,
label="Decision projection attribution",
)
)
if len(records) > _MAX_RECORDS:
raise ValueError(
"Committee DSAR combined result limit exceeded; narrow the selectors."
)
return tuple(
sorted(records, key=lambda item: (item.resource_type, item.resource_id))
)
def plan_erasure(
self,
session: object,
*,
tenant_id: str,
subject: DsarSubjectRef,
records: Sequence[DsarRecordRef],
) -> Sequence[DsarErasureActionRef]:
del tenant_id
_session(session)
if _subject_selectors(subject) is None:
raise ValueError("Committee DSAR subject selectors conflict.")
actions: list[DsarErasureActionRef] = []
for record in records:
_validate_record(record)
actions.append(
DsarErasureActionRef(
action_id=(
f"committee:retain:{record.resource_type}:{record.resource_id}"
),
provider_id=self.provider_id,
module_id=self.module_id,
kind="retain",
resource_type=record.resource_type,
resource_id=record.resource_id,
title=f"Retain {record.title}",
rationale=record.retention_reason
or "Committee attribution is immutable evidence.",
executable=False,
)
)
return tuple(actions)
def execute_erasure(
self,
session: object,
*,
tenant_id: str,
subject: DsarSubjectRef,
actions: Sequence[DsarErasureActionRef],
request_id: str,
) -> Sequence[DsarExecutionResultRef]:
del tenant_id
_session(session)
if _subject_selectors(subject) is None:
raise ValueError("Committee DSAR subject selectors conflict.")
results: list[DsarExecutionResultRef] = []
for action in actions:
_validate_action(action)
if action.executable or action.kind != "retain":
raise ValueError("Committee DSAR publishes retain-only actions.")
results.append(
DsarExecutionResultRef(
action_id=action.action_id,
status="blocked",
summary=(
"Committee workspace, event, and Decision attribution remains "
"immutable institutional evidence."
),
evidence={"request_id": request_id},
)
)
return tuple(results)
def _subject_selectors(subject: DsarSubjectRef) -> _SubjectSelectors | None:
references = subject.external_references
values = {
"account_id": _coalesce(
subject.account_id,
references.get("committee.account"),
references.get("access.account"),
),
"membership_id": _coalesce(
subject.membership_id,
references.get("committee.membership"),
references.get("tenancy.membership"),
),
"identity_id": _coalesce(
subject.identity_id,
references.get("committee.identity"),
references.get("identity.id"),
),
"actor_id": _coalesce(
references.get("committee.actor"),
references.get("committee.changed_by"),
),
"object_kind": _coalesce(
references.get("committee.object_kind"),
references.get("committee.kind"),
),
"object_id": _coalesce(
references.get("committee.object"),
references.get("committee.object_id"),
),
"decision_id": _coalesce(
references.get("committee.decision"),
references.get("committee.decision_id"),
),
}
if any(value is _CONFLICT for value in values.values()):
return None
actor_ids = tuple(
dict.fromkeys(
value
for value in (
_optional_string(values["account_id"]),
_prefixed("account", values["account_id"]),
_optional_string(values["membership_id"]),
_prefixed("membership", values["membership_id"]),
_optional_string(values["identity_id"]),
_prefixed("identity", values["identity_id"]),
)
if value
)
)
direct_actor = _optional_string(values["actor_id"])
if direct_actor:
if actor_ids and direct_actor not in actor_ids:
return None
if not actor_ids:
actor_ids = (direct_actor,)
if not actor_ids:
return None
return _SubjectSelectors(
actor_ids=actor_ids,
object_kind=_optional_string(values["object_kind"]),
object_id=_optional_string(values["object_id"]),
decision_id=_optional_string(values["decision_id"]),
)
def _revision_attribution(row: CommitteeWorkspaceRevision) -> DsarRecordRef:
return DsarRecordRef(
provider_id="committee",
module_id="committee",
resource_type="committee_workspace_actor_attribution",
resource_id=row.id,
category="committee_workspace_accountability",
title="Committee workspace revision attribution",
data={
"object_kind": row.object_kind,
"object_id": row.object_id,
"revision": row.revision,
"parent_kind": row.parent_kind,
"parent_id": row.parent_id,
"state": row.state,
"recorded_at": _iso(row.recorded_at),
"superseded_at": _iso(row.superseded_at),
"activity": "recorded_workspace_revision",
},
observed_at=_aware(row.recorded_at),
immutable_evidence=True,
retention_reason="Committee workspace attribution is immutable evidence.",
)
def _event_attribution(row: CommitteeWorkspaceEvent) -> DsarRecordRef:
return DsarRecordRef(
provider_id="committee",
module_id="committee",
resource_type="committee_event_actor_attribution",
resource_id=row.id,
category="committee_workspace_accountability",
title="Committee lifecycle event attribution",
data={
"object_kind": row.object_kind,
"object_id": row.object_id,
"object_revision": row.object_revision,
"event_id": row.event_id,
"event_type": row.event_type,
"occurred_at": _iso(row.occurred_at),
},
observed_at=_aware(row.occurred_at),
immutable_evidence=True,
retention_reason="Committee event attribution is immutable evidence.",
)
def _decision_attribution(row: CommitteeDecisionProjection) -> DsarRecordRef:
return DsarRecordRef(
provider_id="committee",
module_id="committee",
resource_type="committee_decision_actor_attribution",
resource_id=row.id,
category="committee_decision_accountability",
title="Committee fallback Decision attribution",
data={
"decision_id": row.decision_id,
"revision": row.revision,
"meeting_id": row.meeting_id,
"agenda_item_id": row.agenda_item_id,
"state": row.state,
"recorded_at": _iso(row.recorded_at),
"superseded_at": _iso(row.superseded_at),
"activity": "recorded_decision_projection",
},
observed_at=_aware(row.recorded_at),
immutable_evidence=True,
retention_reason=(
"Committee fallback Decision attribution is immutable evidence."
),
)
def _limited(query, first, second, *, label: str):
rows = query.order_by(first, second).limit(_MAX_RECORDS + 1).all()
if len(rows) > _MAX_RECORDS:
raise ValueError(f"Committee DSAR {label} limit exceeded; narrow selectors.")
return rows
def _coalesce(*values: str | None) -> str | None | object:
normalized = {str(value).strip() for value in values if str(value or "").strip()}
if len(normalized) > 1:
return _CONFLICT
return next(iter(normalized), None)
def _optional_string(value: object) -> str | None:
return value if isinstance(value, str) and value else None
def _prefixed(prefix: str, value: object) -> str | None:
normalized = _optional_string(value)
return f"{prefix}:{normalized}" if normalized else None
def _iso(value: datetime | None) -> str | None:
aware = _aware(value)
return aware.isoformat() if aware else None
def _aware(value: datetime | None) -> datetime | None:
if value is None or value.tzinfo is not None:
return value
return value.replace(tzinfo=timezone.utc)
def _session(value: object) -> Session:
if not isinstance(value, Session):
raise TypeError("Committee DSAR requires a SQLAlchemy Session.")
return value
_RESOURCE_TYPES = {
"committee_workspace_actor_attribution",
"committee_event_actor_attribution",
"committee_decision_actor_attribution",
}
def _validate_record(record: DsarRecordRef) -> None:
if record.provider_id != "committee" or record.module_id != "committee":
raise ValueError("Committee DSAR cannot plan a foreign provider record.")
if record.resource_type not in _RESOURCE_TYPES or not record.resource_id:
raise ValueError("Committee DSAR record identity is invalid.")
def _validate_action(action: DsarErasureActionRef) -> None:
if action.provider_id != "committee" or action.module_id != "committee":
raise ValueError("Committee DSAR cannot execute a foreign provider action.")
if not action.action_id.startswith("committee:retain:"):
raise ValueError("Committee DSAR action identity is invalid.")
__all__ = ["COMMITTEE_DSAR_CAPABILITY", "CommitteeDsarProvider"]