root/Jyotisha
2
0
Fork 0
Code Issues Pull Requests 1 Actions Packages Projects Releases Wiki Activity

feat: complete minute birth-time rectification flow

Browse Source
This commit is contained in:
Jesse_Chen
2026-07-25 01:15:48 +08:00
parent f90aba217f
commit 3ca30ed7de
93 changed files with 8347 additions and 1646 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/test_active_rectification_api.py
+398 -29
View File
@@ -9,6 +9,7 @@ SCRIPTS = Path(__file__).resolve().parents[1] / "scripts"
if str(SCRIPTS) not in sys.path:
sys.path.insert(0, str(SCRIPTS))
import jyotish_api_server as api_server # noqa: E402
from jyotish_api_server import BadRequest, JyotishAPIHandler # noqa: E402
@@ -406,19 +407,115 @@ def test_gochara_conflict_downgrades_without_verified_timing_claim_red() -> None
assert scored["timing_claim_status"] == "exploratory_unvalidated"
def test_high_rigor_event_rectification_queues_vedastro_packet(monkeypatch) -> None:
def test_high_rigor_event_rectification_requires_real_vedastro_candidate_discrimination(monkeypatch) -> None:
original_loader = api_server._load_local_module
class LocalScorer:
@staticmethod
def score_life_events(request):
return {
"result_id": "local-result",
"confidence": "high",
"can_apply": True,
"winning_segment": {
"start_time": "14:30",
"end_time": "14:30",
"representative_time": "14:30",
"width_minutes": 1,
},
"event_count": len(request["events"]),
"domain_count": len({event["domain"] for event in request["events"]}),
"top_score": 30,
"second_score": 20,
"margin_percent": 33.33,
"reasons": [],
"evidence": [],
"algorithm_version": "fixture",
"canonical_input_hash": "canonical-fixture",
"calculation_contract": {"events": request["events"]},
"stability_diagnostics": {
"neighbor_stability": {"all_required_passed": True},
"leave_one_event_out": {"status": "pass"},
},
"missing_layers": [],
"candidate_ranking_summary": [
{"rank": 1, "time": "14:30", "score": 30, "tied_minute_count": 1},
{"rank": 2, "time": "14:31", "score": 20, "tied_minute_count": 1},
],
}
class VedAstroAdapter:
@staticmethod
def run_rectification_minute_snapshot_for_case(case, case_id="user_chart"):
minute = case["minute"]
return {
"available": True,
"status": "ok",
"source": "vedastro_official",
"layers": {
"ascendant_house_boundaries": {
"status": "ok",
"fingerprint": f"asc-{minute}",
"ascendant": {"sign": "Leo", "degree_in_sign": minute / 10},
"houses": {"House1": {}},
},
"D9": {
"status": "ok",
"fingerprint": f"d9-{minute}",
"houses": {"House1": {}},
"planets": {},
},
"D10": {
"status": "ok",
"fingerprint": "d10-same",
"houses": {"House1": {}},
"planets": {},
},
"dasha_boundaries": {
"status": "ok",
"fingerprint": f"dasha-{minute}",
"boundary_count": 3,
},
"kp_cusp_sub_lord": {
"status": "unsupported_by_verified_official_interface",
"reason": "not supported by verified official interface",
},
},
"raw_response": {"must_not": "leak"},
}
@staticmethod
def run_range_scan_for_case(case, _domain, _start, _end, case_id="user_chart"):
return {
"available": True,
"status": "ok",
"event_count": 1,
"top_event": {"event_id": f"event-{case_id}"},
"evidence_ledger": [{"signal_lift": 1}],
"raw_response": {"must_not": "leak"},
}
monkeypatch.setattr(
"scripts.rectification_three_engine_packet._enqueue_vedastro_gateway_job",
api_server,
"_load_local_module",
lambda name: LocalScorer if name == "active_rectification_events" else VedAstroAdapter if name == "vedastro_service_adapter" else original_loader(name),
)
monkeypatch.setattr(
"scripts.rectification_three_engine_packet.build_packet",
lambda _case: {
"engine_status": {"local": "ok", "pyjhora": "ok", "jyotishganit": "ok"},
"match_count": 3,
"mismatch_count": 0,
},
)
monkeypatch.setattr(
JyotishAPIHandler,
"_compute_vedastro_gateway_run",
lambda *_args, **_kwargs: {
"scope": "vedastro_gateway_job_receipt",
"status": "queued",
"job_id": "vgw_rectification",
"poll_path": "/api/vedastro_gateway/jobs/vgw_rectification",
"raw_response_archive": {
"status": "pending",
"official_raw_response_available": False,
},
"boundary": "VedAstro raw response remains server-side; this receipt never returns request data or raw evidence.",
"status": "ok",
"official_closure_state": "official_verified",
"official_closure_reason": "official_raw_response_present",
"official_raw_response": {"must_not": "leak"},
},
)
result = _handler()._compute_active_rectification_events(
@@ -444,30 +541,290 @@ def test_high_rigor_event_rectification_queues_vedastro_packet(monkeypatch) -> N
}
)
receipt = result["three_engine_packet"]["vedastro"]
assert receipt["status"] == "queued"
assert receipt["job_id"] == "vgw_rectification"
assert "1993" not in str(receipt)
assert receipt["status"] == "official_verified"
contract = result["technique_contract"]
assert result["can_apply"] is False
assert contract["confirmation_allowed"] is False
assert contract["decision"] == "continue_rectification"
validation = contract["external_engines"]["validation"]
assert result["can_apply"] is True
assert contract["confirmation_allowed"] is True
assert contract["decision"] == "confirm_minute"
assert contract["canonical_input_hash"]
assert contract["gates"]["public_holdout_release"]["status"] == "blocked"
assert contract["gates"]["vedastro_minute_sensitive_validation"]["status"] == "pass"
assert validation["minute_sensitive_validation"]["discriminated"] is True
assert validation["minute_sensitive_validation"]["discriminated_layers"]
assert validation["event_background_validation"]["used_for_decision"] is False
assert validation["event_background_validation"]["candidates"][0]["metric"] == validation["event_background_validation"]["candidates"][1]["metric"]
assert "must_not" not in str(validation)
assert result["calculation_contract"]["events"][0]["summary"] == "2011 年 9 月离开家乡开始大学生活"
def test_active_rectification_events_accepts_product_limit_of_eight() -> None:
events = [
def test_long_real_conversation_reaches_vedastro_after_local_range_is_narrow(monkeypatch) -> None:
original_loader = api_server._load_local_module
vedastro_calls: list[tuple[str, str, str, str]] = []
class VedAstroAdapter:
@staticmethod
def run_rectification_minute_snapshot_for_case(case, case_id="user_chart"):
candidate_time = f'{case["hour"]:02d}:{case["minute"]:02d}'
return {
"available": True,
"status": "ok",
"source": "vedastro_official",
"layers": {
"ascendant_house_boundaries": {
"status": "ok",
"fingerprint": f"asc-{candidate_time}",
"ascendant": {"sign": "Leo", "degree_in_sign": case["minute"] / 10},
"houses": {"House1": {}},
},
"D9": {
"status": "ok",
"fingerprint": f"d9-{candidate_time}",
"houses": {"House1": {}},
"planets": {},
},
"D10": {
"status": "ok",
"fingerprint": f"d10-{candidate_time}",
"houses": {"House1": {}},
"planets": {},
},
"dasha_boundaries": {
"status": "ok",
"fingerprint": f"dasha-{candidate_time}",
"boundary_count": 3,
},
"kp_cusp_sub_lord": {
"status": "unsupported_by_verified_official_interface",
"reason": "not supported by verified official interface",
},
},
}
@staticmethod
def run_range_scan_for_case(case, domain, start, end, case_id="user_chart"):
candidate_time = f'{case["hour"]:02d}:{case["minute"]:02d}'
vedastro_calls.append((candidate_time, domain, start, end))
return {
"available": True,
"status": "ok",
"event_count": 1,
"top_event": {"event_id": f"event-{case_id}"},
"evidence_ledger": [{"signal_lift": 1}],
}
monkeypatch.setattr(
api_server,
"_load_local_module",
lambda name: VedAstroAdapter if name == "vedastro_service_adapter" else original_loader(name),
)
monkeypatch.setattr(
"scripts.rectification_three_engine_packet.build_packet",
lambda _case: {
"engine_status": {"local": "ok", "pyjhora": "ok", "jyotishganit": "ok"},
"match_count": 3,
"mismatch_count": 0,
},
)
monkeypatch.setattr(
JyotishAPIHandler,
"_compute_vedastro_gateway_run",
lambda *_args, **_kwargs: {
"status": "ok",
"official_closure_state": "official_verified",
"official_closure_reason": "official_raw_response_present",
"official_raw_response": {"status": "ok"},
},
)
result = _handler()._compute_active_rectification_events(
{
"id": f"00000000-0000-4000-a000-{index:012d}",
"domain": "education" if index % 2 == 0 else "career",
"date": f"{2011 + index}",
"precision": "year",
"summary": f"event {index}",
"birth_date": "1997-08-08",
"start_time": "04:00",
"end_time": "07:59",
"lat": 36.420487,
"lon": 114.209936,
"tz": 8,
"high_rigor": True,
"events": [
{"id": "00000000-0000-4000-8000-000000000001", "domain": "education", "date": "2016-09", "precision": "month", "summary": "离家去外地上大学"},
{"id": "00000000-0000-4000-8000-000000000002", "domain": "career", "date": "2020-04", "precision": "month", "summary": "去石油化工研究院实习做研究员"},
{"id": "00000000-0000-4000-8000-000000000003", "domain": "career", "date": "2020-10", "precision": "month", "summary": "从研究院辞职"},
{"id": "00000000-0000-4000-8000-000000000004", "domain": "education", "date": "2020-12", "precision": "month", "summary": "参加研究生考试结果不理想"},
{"id": "00000000-0000-4000-8000-000000000005", "domain": "relocation", "date": "2021-01", "precision": "month", "summary": "回家备考并长期在家"},
{"id": "00000000-0000-4000-8000-000000000006", "domain": "education", "date": "2022-12", "precision": "month", "summary": "考研结束后转向自学前端"},
{"id": "00000000-0000-4000-8000-000000000007", "domain": "career", "date": "2023-04", "precision": "month", "summary": "去北京入职医疗器械公司"},
{"id": "00000000-0000-4000-8000-000000000008", "domain": "relationship", "date": "2024-08-08", "precision": "day", "summary": "恋爱关系发生重大转折"},
{"id": "00000000-0000-4000-8000-000000000009", "domain": "relationship", "date": "2024-10", "precision": "month", "summary": "短暂复联后主动断联"},
{"id": "00000000-0000-4000-8000-000000000010", "domain": "finance", "date": "2026-01", "precision": "month", "summary": "公司无法正常发放工资"},
{"id": "00000000-0000-4000-8000-000000000011", "domain": "career", "date": "2026-07-10", "precision": "day", "summary": "与朋友正式决定创业"},
{"id": "00000000-0000-4000-8000-000000000012", "domain": "career", "date": "2026-07-21", "precision": "day", "summary": "提交公司注册材料"},
],
}
for index in range(8)
)
assert result["winning_segment"] == {
"start_time": "05:07",
"end_time": "05:08",
"representative_time": "05:07",
"width_minutes": 2,
}
assert result["stability_diagnostics"]["neighbor_stability"]["all_required_passed"] is False
assert result["stability_diagnostics"]["leave_one_event_out"]["status"] != "pass"
assert len(vedastro_calls) == 6
assert {call[0] for call in vedastro_calls} == {"05:07", "05:21"}
assert {call[1] for call in vedastro_calls} == {"career", "wealth", "marriage"}
assert all(start == end for _, _, start, end in vedastro_calls)
assert {
(domain, start)
for candidate, domain, start, _ in vedastro_calls
if candidate == "05:07"
} == {
("career", "2026-07-21"),
("wealth", "2026-01-16"),
("marriage", "2024-08-08"),
}
assert {
candidate
for candidate, _, _, _ in vedastro_calls
} == {
item["time"]
for item in result["candidate_ranking_summary"][:2]
}
validation = result["technique_contract"]["external_engines"]["validation"]
event_validation = validation["event_background_validation"]
assert event_validation["eligible_event_count"] == 12
assert event_validation["supported_event_count"] == 3
assert event_validation["used_for_decision"] is False
assert event_validation["candidates"][0]["metric"] == event_validation["candidates"][1]["metric"]
assert "one_strongest_event_per_native_adapter_domain" in event_validation["selection_policy"]
assert result["three_engine_packet"]["vedastro"]["status"] == "official_verified"
assert result["three_engine_packet"]["vedastro"]["search_events_role"] == "background_only"
assert result["technique_contract"]["gates"]["vedastro_minute_sensitive_validation"]["status"] == "pass"
assert validation["minute_sensitive_validation"]["discriminated"] is True
assert result["technique_contract"]["confirmation_allowed"] is True
assert result["technique_contract"]["decision"] == "confirm_minute"
assert result["can_apply"] is True
assert "neighbor_stability_not_passed" not in result["technique_contract"]["hard_blockers"]
assert "leave_one_event_out_not_passed" not in result["technique_contract"]["hard_blockers"]
def test_identical_vedastro_minute_sensitive_snapshots_do_not_discriminate_candidates() -> None:
layers = {
name: {"status": "ok", "fingerprint": f"same-{name}"}
for name in api_server._VEDASTRO_MINUTE_SENSITIVE_LAYERS
}
snapshots = [
{"candidate_time": "05:07", "available": True, "layers": layers},
{"candidate_time": "05:21", "available": True, "layers": layers},
]
comparison = api_server._compare_vedastro_minute_snapshots(snapshots)
assert comparison["comparison_ready"] is True
assert comparison["discriminated"] is False
assert comparison["discriminated_layers"] == []
assert all(item["status"] == "same" for item in comparison["differences"].values())
def test_search_events_difference_cannot_override_identical_minute_snapshots(monkeypatch) -> None:
original_loader = api_server._load_local_module
class LocalScorer:
@staticmethod
def score_life_events(request):
return {
"result_id": "local-result",
"confidence": "high",
"can_apply": True,
"winning_segment": {
"start_time": "14:30",
"end_time": "14:30",
"representative_time": "14:30",
"width_minutes": 1,
},
"event_count": len(request["events"]),
"domain_count": len({event["domain"] for event in request["events"]}),
"top_score": 30,
"second_score": 20,
"margin_percent": 33.33,
"reasons": [],
"evidence": [],
"algorithm_version": "fixture",
"canonical_input_hash": "canonical-fixture",
"calculation_contract": {"events": request["events"]},
"stability_diagnostics": {
"neighbor_stability": {"all_required_passed": True},
"leave_one_event_out": {"status": "pass"},
},
"missing_layers": [],
"candidate_ranking_summary": [
{"rank": 1, "time": "14:30", "score": 30, "tied_minute_count": 1},
{"rank": 2, "time": "14:31", "score": 20, "tied_minute_count": 1},
],
}
class VedAstroAdapter:
@staticmethod
def run_rectification_minute_snapshot_for_case(_case, case_id="user_chart"):
layers = {
name: {
"status": "ok",
"fingerprint": f"same-{name}",
"houses": {"House1": {}},
"planets": {},
"boundary_count": 3,
}
for name in api_server._VEDASTRO_MINUTE_SENSITIVE_LAYERS
}
layers["ascendant_house_boundaries"]["ascendant"] = {
"sign": "Leo",
"degree_in_sign": 12.5,
}
layers["kp_cusp_sub_lord"] = {
"status": "unsupported_by_verified_official_interface",
"reason": "not supported by verified official interface",
}
return {
"available": True,
"status": "ok",
"source": "vedastro_official",
"layers": layers,
}
@staticmethod
def run_range_scan_for_case(case, _domain, _start, _end, case_id="user_chart"):
event_count = 10 if case["minute"] == 30 else 1
return {
"available": True,
"status": "ok",
"event_count": event_count,
"top_event": {"event_id": f"event-{case_id}"},
"evidence_ledger": [{"signal_lift": event_count}],
}
monkeypatch.setattr(
api_server,
"_load_local_module",
lambda name: LocalScorer if name == "active_rectification_events" else VedAstroAdapter if name == "vedastro_service_adapter" else original_loader(name),
)
monkeypatch.setattr(
"scripts.rectification_three_engine_packet.build_packet",
lambda _case: {
"engine_status": {"local": "ok", "pyjhora": "ok", "jyotishganit": "ok"},
"match_count": 3,
"mismatch_count": 0,
},
)
monkeypatch.setattr(
JyotishAPIHandler,
"_compute_vedastro_gateway_run",
lambda *_args, **_kwargs: {
"status": "ok",
"official_closure_state": "official_verified",
"official_closure_reason": "official_raw_response_present",
},
)
result = _handler()._compute_active_rectification_events(
{
"birth_date": "1993-04-17",
@@ -476,9 +833,21 @@ def test_active_rectification_events_accepts_product_limit_of_eight() -> None:
"lat": 36.683333,
"lon": 114.35,
"tz": 8,
"events": events,
"high_rigor": True,
"events": [
{"id": "5cb071d6-6d99-46be-85dc-a9bf59ef6ac5", "domain": "education", "date": "2011-09", "precision": "month"},
{"id": "0790866c-ad5e-4a45-b2b4-a5c73f6be6ea", "domain": "career", "date": "2019-07-01", "precision": "day"},
{"id": "0ef52e51-ab5f-453b-81e5-adb44a929224", "domain": "relationship", "date": "2021", "precision": "year"},
],
}
)
assert result["success"] is True
assert result["event_count"] == 8
validation = result["technique_contract"]["external_engines"]["validation"]
background_candidates = validation["event_background_validation"]["candidates"]
assert background_candidates[0]["metric"] != background_candidates[1]["metric"]
assert validation["event_background_validation"]["used_for_decision"] is False
assert validation["minute_sensitive_validation"]["discriminated"] is False
assert result["can_apply"] is False
assert result["technique_contract"]["confirmation_allowed"] is False
assert "vedastro_minute_sensitive_layers_not_discriminated" in result["reasons"]
assert "vedastro_candidate_not_discriminated" not in result["reasons"]