Files
jarca-ledger-mcp/jarca_ledger_mcp/application/monthly_settlement_service.py
T
gitea-impl-agent 53ce20956d Reject malformed persisted settlement header evidence
Preserve safe shared error envelopes and read-only legacy defaults. Add exact/latest evidence-integrity regressions for service, gateway and MCP consumers.
2026-10-14 04:44:28 +00:00

261 lines
16 KiB
Python

from __future__ import annotations
import json
from .shared import read_db, DatabaseNotInitializedError
from .assets_service import get_asset_snapshot_report
from .imports_service import list_ledger_import_batches
from .reconciliation_service import get_balance_reconciliation_report
from .review_decision_service import get_review_operating_indicators
from .shared import Any, fail, normalize_month, normalize_snapshot_at, ok
from .transaction_report_services import get_budget_vs_actual_report, get_monthly_cashflow_report
def _month_end_date(month: str) -> str:
return normalize_snapshot_at(month, default_now=False)[:10]
def _budget_overrun_count(budget_report: dict[str, Any]) -> int:
return sum(1 for item in budget_report.get("items", []) if item.get("status") == "over")
def _unresolved_blocker_summary(review_report: dict[str, Any], reconciliation_report: dict[str, Any]) -> dict[str, Any]:
indicators = review_report.get("indicators", {}) if review_report.get("success") else {}
reconciliation_items = reconciliation_report.get("items", []) if reconciliation_report.get("success") else []
reconciliation_attention_count = sum(1 for item in reconciliation_items if item.get("status") != "OK")
counts = {
"uncategorized_recommendation_backlog": int(indicators.get("uncategorized_recommendation_backlog") or 0),
"pending_review_count": int(indicators.get("pending_review_count") or 0),
"held_recheck_count": int(indicators.get("held_recheck_count") or 0),
"unreconciled_account_count": int(indicators.get("unreconciled_account_count") or 0),
"reconciliation_attention_count": reconciliation_attention_count,
}
return {"count": sum(counts.values()), "counts": counts}
def _source_coverage(
cashflow_report: dict[str, Any],
budget_report: dict[str, Any],
review_report: dict[str, Any],
reconciliation_report: dict[str, Any],
import_report: dict[str, Any],
asset_report: dict[str, Any],
) -> dict[str, Any]:
latest_import = (import_report.get("import_batches") or [None])[0]
return {
"cashflow": {"success": bool(cashflow_report.get("success")), "source": "transactions"},
"budget": {
"success": bool(budget_report.get("success")),
"budget_count": len(budget_report.get("items", [])),
},
"review": {
"success": bool(review_report.get("success")),
"indicators": review_report.get("indicators", {}) if review_report.get("success") else {},
},
"reconciliation": {
"success": bool(reconciliation_report.get("success")),
"item_count": int(reconciliation_report.get("count") or 0),
},
"imports": {
"success": bool(import_report.get("success")),
"recent_batch_count": int(import_report.get("count") or 0),
"latest_imported_at": latest_import.get("imported_at") if isinstance(latest_import, dict) else "",
},
"assets": {
"success": bool(asset_report.get("success")),
"basis_date": asset_report.get("basis_date", ""),
"source": asset_report.get("source", ""),
"is_closed_snapshot": bool(asset_report.get("is_closed_snapshot", False)),
"net_worth": asset_report.get("net_worth"),
"close_status": asset_report.get("close_status", ""),
"approval": asset_report.get("approval", {}),
"correction": asset_report.get("correction", {}),
"warnings": asset_report.get("warnings", []),
"provenance": asset_report.get(
"provenance", {"valuation_date": "", "warnings": [], "exclusions": []}
),
"missing_items": [item for item in asset_report.get("excluded_items", []) if "missing" in str(item.get("reason", ""))],
"excluded_items": asset_report.get("excluded_items", []),
"carried_items": asset_report.get("carried_items", asset_report.get("included_items", [])),
"monthly_source_snapshot_id": asset_report.get("snapshot_run_id", ""),
"draft_id": asset_report.get("close_metadata", {}).get("draft_id", ""),
"draft_version": asset_report.get("close_metadata", {}).get("draft_version", 0),
"review_fingerprint": asset_report.get("close_metadata", {}).get("review_fingerprint", ""),
"contains_substitutions": asset_report.get("close_metadata", {}).get("contains_substitutions", False),
"contains_carried": asset_report.get("close_metadata", {}).get("contains_carried", False),
"has_approved_missing": asset_report.get("close_metadata", {}).get("has_approved_missing", False),
"applied_evidence": asset_report.get("close_metadata", {}).get("applied_evidence", []),
},
}
def get_monthly_settlement_summary(month: str = "", db_path: str | None = None) -> dict[str, Any]:
"""Return the read-only monthly settlement summary foundation for dashboard/MCP consumers."""
try:
selected_month = normalize_month(month)
basis_date = _month_end_date(selected_month)
except Exception as exc:
return fail("validation_error", str(exc))
cashflow_report = get_monthly_cashflow_report(selected_month, db_path=db_path)
if not cashflow_report.get("success"):
return cashflow_report
budget_report = get_budget_vs_actual_report(selected_month, db_path=db_path)
review_report = get_review_operating_indicators(db_path=db_path)
reconciliation_report = get_balance_reconciliation_report(db_path=db_path)
import_report = list_ledger_import_batches(limit=5, db_path=db_path)
asset_report = get_asset_snapshot_report(basis_date=basis_date, db_path=db_path)
unresolved_blockers = _unresolved_blocker_summary(review_report, reconciliation_report)
return ok(
selected_month=selected_month,
month=selected_month,
basis_date=basis_date,
actual_income=int(cashflow_report.get("income") or 0),
actual_expense=int(cashflow_report.get("expenses") or 0),
net_cashflow=int(cashflow_report.get("net_cashflow") or 0),
savings_rate=cashflow_report.get("savings_rate"),
budget_overrun_count=_budget_overrun_count(budget_report),
unresolved_blocker_count=int(unresolved_blockers["count"]),
unresolved_blockers=unresolved_blockers,
source_coverage=_source_coverage(
cashflow_report,
budget_report,
review_report,
reconciliation_report,
import_report,
asset_report,
),
reports={
"cashflow": cashflow_report,
"budget": budget_report,
"assets": asset_report,
},
read_only=True,
)
def get_monthly_settlement_snapshot(month: str = "", run_id: str = "", db_path: str | None = None) -> dict[str, Any]:
"""Read a persisted monthly settlement close without consulting live ledger state."""
if not isinstance(month, str) or not isinstance(run_id, str):
return fail("validation_error", "month and run_id must be strings")
try:
selected_month = normalize_month(month) if month else ""
if selected_month and (len(selected_month) != 7 or selected_month[4] != "-"):
raise ValueError("invalid month")
except (TypeError, ValueError):
return fail("validation_error", "invalid settlement month")
def parsed(value: Any, default: Any):
if value in (None, ""):
return default
return json.loads(value)
try:
with read_db(db_path) as conn:
tables = {r["name"] for r in conn.execute("SELECT name FROM sqlite_master WHERE type='table'")}
if "asset_snapshot_runs" not in tables or "asset_snapshots" not in tables:
return fail("not_found", "settlement snapshot not found")
run = None
rows = []
legacy = False
metadata: dict[str, Any] = {}
if run_id:
run = conn.execute("SELECT rowid AS _run_rowid,* FROM asset_snapshot_runs WHERE id=?", (run_id,)).fetchone()
if run:
from .assets_service import _snapshot_rows_for_run
rows, metadata = _snapshot_rows_for_run(conn, run_id)
else:
cols = {r["name"] for r in conn.execute("PRAGMA table_info(asset_snapshots)")}
if {"snapshot_run_id", "closed_at"} <= cols:
rows = conn.execute("SELECT rowid AS _rowid,* FROM asset_snapshots WHERE snapshot_run_id=? AND closed_at<>'' ORDER BY rowid", (run_id,)).fetchall()
if rows:
legacy = True
metadata = {"snapshot_run_id": run_id, "legacy_closed": True, "close_status": "legacy_read_only"}
else:
if selected_month:
run = conn.execute("SELECT rowid AS _run_rowid,* FROM asset_snapshot_runs WHERE basis_date=? ORDER BY rowid DESC LIMIT 1", (_month_end_date(selected_month),)).fetchone()
if run is None:
basis = _month_end_date(selected_month) if selected_month else ""
if basis:
from .assets_service import _snapshot_rows_for_latest_run
rows, metadata = _snapshot_rows_for_latest_run(conn, basis)
legacy = bool(metadata.get("legacy_closed"))
if legacy:
rows = [r for r in rows if r["snapshot_run_id"] == metadata.get("snapshot_run_id") and r["closed_at"]]
elif not metadata.get("snapshot_run_id"):
rows = []
if metadata.get("snapshot_run_id"):
run = conn.execute("SELECT rowid AS _run_rowid,* FROM asset_snapshot_runs WHERE id=?", (metadata["snapshot_run_id"],)).fetchone()
if not run and not rows:
return fail("not_found", "settlement snapshot not found")
if run:
rid = run["id"]
basis_date = run["basis_date"]
run_month = run["settlement_month"] if "settlement_month" in run.keys() else None
if selected_month and ((run_month and run_month != selected_month) or (not run_month and basis_date[:7] != selected_month)):
return fail("validation_error", "run does not match settlement month")
if not metadata:
from .assets_service import _snapshot_rows_for_run
rows, metadata = _snapshot_rows_for_run(conn, rid)
else:
rid = metadata.get("snapshot_run_id", run_id)
basis_date = rows[0]["basis_date"]
if selected_month and basis_date[:7] != selected_month:
return fail("validation_error", "run does not match settlement month")
legacy = True
if metadata.get("evidence_integrity_error"):
return fail("data_integrity_error", "snapshot evidence is malformed", snapshot_run_id=rid)
keys = set(run.keys()) if run else set()
def field(name): return run[name] if run and name in keys else None
from .settlement_cashflow_service import _AGGREGATION_VERSION
if (field("snapshot_schema_version") is not None or field("aggregation_version") is not None) and (
field("snapshot_schema_version") != 1 or field("aggregation_version") != _AGGREGATION_VERSION
):
raise ValueError("unsupported comprehensive snapshot version")
comprehensive = bool(field("snapshot_schema_version") == 1 and not legacy)
legacy = not comprehensive
cashflows = []
if comprehensive and "settlement_cashflow_snapshots" in tables:
cashflows = [dict(r) for r in conn.execute("SELECT cashflow_kind,category,subcategory,amount,transaction_count,currency FROM settlement_cashflow_snapshots WHERE settlement_run_id=? ORDER BY cashflow_kind,category,subcategory,currency", (rid,))]
totals = {"income_total": field("income_total") if comprehensive else None, "expense_total": field("expense_total") if comprehensive else None, "net_cashflow": field("net_cashflow") if comprehensive else None, "total_assets": field("total_assets") if comprehensive else None, "total_liabilities": field("total_liabilities") if comprehensive else None, "net_worth": field("net_worth") if comprehensive else None}
safe_items = []
for row in rows:
d = dict(row)
safe_items.append({k: d.get(k) for k in ("id","source_type","source_id","name","amount","currency","liquidity_bucket","valuation_source","snapshot_run_id","correction_of_run_id","correction_reason","is_correction","closed_at","valuation_id","selection_policy","verification_status","evidence_kind","exclusion_reason","created_at") if k in d})
source = parsed(field("source_snapshot_json"), {}) if comprehensive else {}
if not isinstance(source, dict): raise ValueError()
if comprehensive:
tx_fields = ("id", "transaction_date", "type", "amount", "currency", "category", "subcategory", "original_cashflow_kind", "originalcashflow_kind", "cashflow_kind", "applied_amount", "sourceprovenance", "category_cashflow_kind", "category_active")
asset_fields = ("id", "source_id", "source_type", "valuation_id", "valuation_date", "basis_date", "amount", "currency", "liquidity_bucket", "selection_mode", "applied_amount", "authority_source", "is_liability", "status", "verified", "verification_status", "evidence_kind", "exclusion_reason", "state_hash")
def project(records, fields):
if not isinstance(records, list) or any(not isinstance(x, dict) for x in records):
raise ValueError()
result = []
for record in records:
item = {k: record[k] for k in fields if k in record}
if any(value is not None and not isinstance(value, (str, int, bool)) for value in item.values()):
raise ValueError()
for key, allowed in (("decision", ("category", "subcategory", "cashflow_kind", "reason")), ("source_classification", ("id", "category", "subcategory", "cashflow_kind", "active"))):
value = record.get(key)
if isinstance(value, dict):
item[key] = {k: value[k] for k in allowed if k in value and (value[k] is None or isinstance(value[k], (str, int, bool)))}
result.append(item)
return result
source = {"transactions": project(source.get("transactions", []), tx_fields), "assets": project(source.get("assets", []), asset_fields)}
policy = parsed(field("cashflow_policy_json"), {}) if comprehensive else None
if comprehensive:
from .settlement_cashflow_service import _normalize_cashflow_policy
if not isinstance(policy, dict):
raise ValueError()
policy = _normalize_cashflow_policy(policy)
from .assets_service import _closed_run_metadata
public_metadata = _closed_run_metadata(metadata)
header_evidence = {key: public_metadata[key] for key in ("excluded_items", "carried_items", "applied_evidence", "approval")}
return ok(**header_evidence, source="settlement_snapshot", snapshot_run_id=rid, run_id=rid, settlement_month=field("settlement_month") or basis_date[:7], basis_date=basis_date, currency=field("currency"), is_closed_snapshot=True, is_comprehensive=comprehensive, legacy=legacy, aggregation_version=field("aggregation_version") if comprehensive else None, snapshot_schema_version=field("snapshot_schema_version") if comprehensive else None, cashflow_policy=policy, cashflow_items=cashflows, items=safe_items, source_snapshot=source, totals=totals, **totals, close_status=(field("close_status") or metadata.get("close_status", "legacy_read_only")), correction={"is_correction": bool(field("correction_of_run_id") or metadata.get("is_correction")), "correction_of_run_id": field("correction_of_run_id") or metadata.get("correction_of_run_id", ""), "correction_reason": field("correction_reason") or metadata.get("correction_reason", "")})
except DatabaseNotInitializedError:
return fail("db_not_initialized", "ledger database is not initialized")
except Exception:
return fail("db_error", "settlement snapshot read failed")