HEL-543: add data hub observability side-channel (provider status, source catalog, lineage)

- New provider_call_log/provider_health tables (additive-only schema),
  wired via a fail-open observability.observe()/record_call() helper.
- Tushare pipeline keeps its existing src_calls record unchanged and now
  also feeds the unified provider_health/provider_call_log side channel.
- Eastmoney/Tencent realtime_serve.py call sites and the iFinD steward
  call site are wrapped with observability.observe() at the call site
  only; no adapter internals, routing, fallback order, or return values
  are touched.
- New read-only admin API endpoints: /admin/api/providers/status,
  /admin/api/source-catalog, /admin/api/lineage,
  /admin/api/lineage/affected.
- New static, read-only source_catalog.py and lineage.py registries
  documenting existing providers/interfaces/datasets and known
  main-site consumers (cited against backend/features/screener and
  backend/features/heaven call sites).
- provider_call_log is purged by the existing pipeline.cleanup() job
  alongside src_calls/job_runs.
- 47 new unit/integration tests covering classification, fail-open
  behavior under DB/log failures, unchanged payloads/exceptions on
  success and failure paths, and the new HTTP endpoints. Full suite:
  173 tests, all green.

Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: multica-agent <github@multica.ai>
This commit is contained in:
总工
2026-09-14 00:59:03 +08:00
co-authored by Cursor multica-agent
parent 4a90c32fcc
commit f014eb11bd
14 changed files with 1568 additions and 13 deletions
@@ -0,0 +1,163 @@
from __future__ import annotations
import tempfile
import unittest
from contextlib import contextmanager
from pathlib import Path
from unittest.mock import patch
from datahub.adapters.base import AdapterError
from datahub.db import HubDB
from datahub.realtime_serve import fetch_index_quotes, fetch_intraday, fetch_quotes
class _WriteBreaksDB:
"""Wraps a real HubDB but breaks only the write path, to prove the
real serving path (reads/caches) is untouched by an observability
failure while still exercising real fetch/cache code around it."""
def __init__(self, real: HubDB) -> None:
self._real = real
def fetchall(self, sql, params=()):
return self._real.fetchall(sql, params)
def fetchone(self, sql, params=()):
return self._real.fetchone(sql, params)
def execute(self, sql, params=()):
return self._real.execute(sql, params)
def executemany(self, sql, rows):
return self._real.executemany(sql, rows)
@contextmanager
def write(self):
raise RuntimeError("db is not writable right now")
yield None # pragma: no cover
class RealtimeObservabilityTests(unittest.TestCase):
def setUp(self) -> None:
self.tmp = tempfile.TemporaryDirectory()
self.db = HubDB(Path(self.tmp.name) / "hub.db")
def tearDown(self) -> None:
self.tmp.cleanup()
def test_eastmoney_success_is_logged_without_changing_payload(self) -> None:
rows = [
{"ts_code": "000001.SH", "code": "000001", "name": "上证指数", "price": 3000.0,
"previous_close": 2990.0, "quote_time_epoch": 0, "source": "eastmoney_push2"},
{"ts_code": "399001.SZ", "code": "399001", "name": "深证成指", "price": 9000.0,
"previous_close": 8990.0, "quote_time_epoch": 0, "source": "eastmoney_push2"},
{"ts_code": "399006.SZ", "code": "399006", "name": "创业板指", "price": 1800.0,
"previous_close": 1790.0, "quote_time_epoch": 0, "source": "eastmoney_push2"},
]
with patch("datahub.realtime_serve.EastmoneyAdapter") as mocked:
mocked.return_value.fetch_indices.return_value = rows
payload = fetch_index_quotes(self.db)
self.assertEqual(payload["data"], rows)
self.assertEqual(payload["meta"]["source"], "eastmoney:ulist")
log = self.db.fetchall("SELECT * FROM provider_call_log WHERE provider = 'eastmoney'")
self.assertEqual(len(log), 1)
self.assertEqual(log[0]["interface"], "indices")
self.assertEqual(log[0]["status"], "ok")
health = self.db.fetchone(
"SELECT * FROM provider_health WHERE provider = 'eastmoney' AND interface = 'indices'"
)
self.assertEqual(health["state"], "ok")
def test_eastmoney_failure_falls_back_to_tencent_and_logs_both(self) -> None:
tencent_rows = [
{"ts_code": "000001.SH", "code": "000001", "name": "上证指数", "price": 3000.0,
"previous_close": 2990.0, "quote_time_epoch": 0, "source": "tencent_qt"},
{"ts_code": "399001.SZ", "code": "399001", "name": "深证成指", "price": 9000.0,
"previous_close": 8990.0, "quote_time_epoch": 0, "source": "tencent_qt"},
{"ts_code": "399006.SZ", "code": "399006", "name": "创业板指", "price": 1800.0,
"previous_close": 1790.0, "quote_time_epoch": 0, "source": "tencent_qt"},
]
with patch("datahub.realtime_serve.EastmoneyAdapter") as eastmoney, patch(
"datahub.realtime_serve.TencentAdapter"
) as tencent:
eastmoney.return_value.fetch_indices.side_effect = AdapterError("Eastmoney returned 0/3 indices")
tencent.return_value.fetch_indices.return_value = tencent_rows
payload = fetch_index_quotes(self.db)
self.assertEqual(payload["meta"]["source"], "tencent:qt")
self.assertEqual(payload["data"], tencent_rows)
east_log = self.db.fetchone("SELECT * FROM provider_call_log WHERE provider = 'eastmoney'")
self.assertEqual(east_log["status"], "empty")
tencent_log = self.db.fetchone("SELECT * FROM provider_call_log WHERE provider = 'tencent'")
self.assertEqual(tencent_log["status"], "ok")
def test_observability_db_failure_never_breaks_a_real_successful_fetch(self) -> None:
rows = [
{"ts_code": "000001.SH", "price": 3000.0, "previous_close": 2990.0, "quote_time_epoch": 0},
{"ts_code": "399001.SZ", "price": 9000.0, "previous_close": 8990.0, "quote_time_epoch": 0},
{"ts_code": "399006.SZ", "price": 1800.0, "previous_close": 1790.0, "quote_time_epoch": 0},
]
broken = _WriteBreaksDB(self.db)
with patch("datahub.realtime_serve.EastmoneyAdapter") as mocked:
mocked.return_value.fetch_indices.return_value = rows
payload = fetch_index_quotes(broken)
self.assertEqual(payload["data"], rows)
self.assertEqual(payload["meta"]["source"], "eastmoney:ulist")
def test_observability_db_failure_never_masks_a_real_source_outage(self) -> None:
broken = _WriteBreaksDB(self.db)
with patch("datahub.realtime_serve.EastmoneyAdapter") as eastmoney, patch(
"datahub.realtime_serve.TencentAdapter"
) as tencent:
eastmoney.return_value.fetch_indices.side_effect = AdapterError("down")
tencent.return_value.fetch_indices.side_effect = AdapterError("also down")
with self.assertRaises(Exception):
fetch_index_quotes(broken)
def test_named_quotes_records_both_providers_on_partial_merge(self) -> None:
with patch("datahub.realtime_serve.EastmoneyAdapter") as eastmoney, patch(
"datahub.realtime_serve.TencentAdapter"
) as tencent:
eastmoney.return_value.fetch_quotes.return_value = [
{"ts_code": "000001.SZ", "close": 10, "pre_close": 9, "quote_date": "20260907"},
]
tencent.return_value.fetch_quotes.return_value = [
{"ts_code": "000002.SZ", "close": 20, "pre_close": 19, "quote_date": "20260907"},
]
payload = fetch_quotes(self.db, ["000001.SZ", "000002.SZ"])
self.assertEqual(payload["meta"]["complete"], True)
east_log = self.db.fetchone(
"SELECT * FROM provider_call_log WHERE provider = 'eastmoney' AND interface = 'named_quotes'"
)
self.assertEqual(east_log["status"], "ok")
tencent_log = self.db.fetchone(
"SELECT * FROM provider_call_log WHERE provider = 'tencent' AND interface = 'named_quotes'"
)
self.assertEqual(tencent_log["status"], "ok")
def test_intraday_success_is_logged_as_ok(self) -> None:
payload_data = {
"entity_type": "stock", "ts_code": "601318.SH", "trade_date": "2026-09-07",
"previous_close": 55.8, "points": [{"date": "2026-09-07", "time": "09:30", "close": 55.9}],
}
with patch("datahub.realtime_serve.EastmoneyAdapter") as mocked:
mocked.return_value.fetch_intraday.return_value = payload_data
payload = fetch_intraday(self.db, "601318.SH")
self.assertEqual(payload["data"], payload_data)
log = self.db.fetchone(
"SELECT * FROM provider_call_log WHERE provider = 'eastmoney' AND interface = 'intraday'"
)
self.assertEqual(log["status"], "ok")
def test_intraday_failure_is_logged_as_empty(self) -> None:
with patch("datahub.realtime_serve.EastmoneyAdapter") as mocked:
mocked.return_value.fetch_intraday.side_effect = AdapterError("No intraday chart data returned")
with self.assertRaises(Exception):
fetch_intraday(self.db, "000001.SZ")
log = self.db.fetchone(
"SELECT * FROM provider_call_log WHERE provider = 'eastmoney' AND interface = 'intraday'"
)
self.assertEqual(log["status"], "empty")
if __name__ == "__main__":
unittest.main()