P23-03: 生活の観察

This commit is contained in:
Agent Control Runner
2026-10-15 10:03:36 +09:00
parent 55c29f1192
commit 0b7727a66c
4 changed files with 177 additions and 8 deletions
@@ -0,0 +1,72 @@
# Offline checker repair on 55c29f1
Source HEAD: 55c29f119234dd6f993c604c4edc6f377fe3951c. Registration base remains
860136f7099f5e3aee99da5798a627a77f2d8eea. No git writes or engine execution.
Read access: MIRROR_INDEX.json and all 2,622 named files were successfully read
under `/Users/korizatou114/Projects/TheTree-managed-worktrees/p23-03-code/.runtime/P23-03/offline-evidence-55c29f1-r06`.
No mirror bytes were changed. The initial shell heredoc could not create its
temporary file; the read check was executed using Python `-c` instead.
## First defects and production cause
Both layouts first failed at frame-030, checker line 328, with KeyError:
`shaft_id`. State objects:
- A: `layout-A/objects/ac3ed64a6e6162918e4d44c45b2f78d09e50d473b607f3469350a06074a7efb3.json`
- B: `layout-B/objects/97ea43afe3d1b21602ebe2a602e7b7d796ff05f03983cd4d1b3edc9157e39a08.json`
The exception came from `shaft['shaft_id']`, not `t['shaft_id']`. Every trip in
these failing frames has shaft_id. Saved shaft rows have it only in config.
There is no evidence of shaft-less QUEUED trips to support accepting that shape.
`autoload/simulation_runtime.gd:541` writes elevators.create_snapshot().
`scripts/transport/elevator_service.gd:362-367` copies the saved cars without
flattening config; get_report at 345-359 merges config into each report row before
counting queued trips. request at 169-171 assigns shaft_id on creation, and the
restore validator at 412-454 requires the exact trip keys and shaft/connector
agreement. The connector spelling is defined at 617-618; trips are ordered at
707-710. `scenes/main/world_preview.gd:973-975` consumes get_report and sorts
trips by enqueue time/actor. The checker now counts using the merged row identity.
## Reviewable changes
- `tools/check_p23_life_observation_evidence.py`: replace
`t['shaft_id'] == shaft['shaft_id']` with `t['shaft_id'] == row['shaft_id']`;
validate exact trip keys, companion identity, shaft/stops/connector attribution
and QUEUED/BOARDED state before recomputing. Whole report equality remains strict.
- `tests/fixtures/p23-03/queued_elevator.json`: retain one actual frame-030
queued trip and its shaft, without a cabin seat. Distilled unit data only.
- `tools/test_p23_life_observation_evidence.py`: queued shape positive and
shaft attribution, waiting_count, missing/extra key negatives; update two
pre-existing abbreviated synthetic fixtures to the actual shaft/trip shape.
- This repair report. The complete implementation diff is available through
`git diff -- tools/check_p23_life_observation_evidence.py tools/test_p23_life_observation_evidence.py`;
the new fixture and this file are untracked submission files.
## Local results
`inspect_stream` on each exact absolute mirror directory:
```
layout-A inspect_stream PASS 17 frames
layout-B inspect_stream PASS 17 frames
```
For both, labels are frame-000, frame-030, frame-060, frame-090, frame-120,
frame-150, frame-180, frame-210, frame-240, frame-270, frame-300, frame-330,
frame-360, frame-390, frame-420, frame-450, frame-480. Each returns fields:
guests, staff, tasks, elevators, room_service_fields, queues, traffic, goods,
cabin_geometry. Every frame also passes the existing two-motion-history-frame
check, unchanged. No later schema/semantic mismatch was observed.
Python syntax AST checks pass. Local checker tests: 32 tests, OK.
The initial test run exposed two abbreviated synthetic fixtures; subsequent
queue expectations were corrected for actual lift landing exclusion. Tests
were preserved. git diff --check passes; changed files are within allowed_paths.
completion_marker_contracts retains its one required P23 strict template entry;
no marker changes. Seven conditions, six categories and 30/88 remain unchanged.
Root native results (319 tests, A/B PASS), matched metric values (48 metrics,
7153/2585), production data and services were not modified. Root OS evidence
remains pending as expected at worker submission; this repair grants no acceptance.
+27
View File
@@ -0,0 +1,27 @@
{
"provenance": "Distilled layout-A frame-030 state ac3ed64a6e6162918e4d44c45b2f78d09e50d473b607f3469350a06074a7efb3; retain first real queued trip, no cabin seat. Unit fixture only.",
"saved": {
"version": 1,
"current_minute": 691,
"request_sequence": 8,
"shafts": [{
"config": {
"shaft_id": "transport.shaft.0.0.0.2",
"module_ids": ["facility_00000003"],
"stops": [[0, 0, 0], [0, 1, 0]],
"width": 2, "capacity": 4, "travel_minutes_per_floor": 10, "access_id": ""
},
"operational": true, "operator_open": true, "powered": true,
"current_floor": 1, "target_floor": 1, "boarding_floor": 0,
"phase": "PICKUP", "remaining_minutes": 9, "phase_total_minutes": 10,
"phase_started_at": 690, "phase_origin_floor": 1, "phase_passenger_count": 0,
"boarded": []
}],
"trips": [{
"actor_id": "guest:p23-000", "request_id": "transport.trip.000000000002",
"shaft_id": "transport.shaft.0.0.0.2",
"connector_id": "transport.shaft.0.0.0.2.link.0.1",
"from": [0, 0, 0], "to": [0, 1, 0], "enqueued_at": 674, "state": "QUEUED"
}]
}
}
+21 -1
View File
@@ -314,6 +314,25 @@ def compare_actor_rows(snapshot, report, root=ROOT):
def elevator_report(snapshot):
saved = snapshot['systems']['runtime']['elevators']
# ElevatorService.restore_snapshot validates this exact trip shape (412-438).
# QUEUED means no cabin seat yet, not an absent shaft assignment.
configs = {shaft['config']['shaft_id']: shaft['config'] for shaft in saved['shafts']}
for trip in saved['trips']:
keys = {'actor_id', 'request_id', 'shaft_id', 'connector_id', 'from', 'to',
'enqueued_at', 'state'}
if 'companion_id' in trip:
keys.add('companion_id')
require(isinstance(trip['companion_id'], str) and trip['companion_id'],
'invalid elevator companion identity')
require(set(trip) == keys, 'exact elevator trip keys')
require(trip['shaft_id'] in configs, 'elevator trip shaft attribution')
config = configs[trip['shaft_id']]
require(trip['from'] != trip['to'] and trip['from'] in config['stops']
and trip['to'] in config['stops'], 'elevator trip stops')
low, high = sorted((trip['from'][1], trip['to'][1]))
require(trip['connector_id'] == f"{trip['shaft_id']}.link.{low}.{high}",
'elevator trip connector attribution')
require(trip['state'] in ('QUEUED', 'BOARDED'), 'elevator trip state')
rows = []
for shaft in saved['shafts']:
row = copy.deepcopy(shaft)
@@ -325,7 +344,8 @@ def elevator_report(snapshot):
floor = shaft['phase_origin_floor'] + (destination - shaft['phase_origin_floor']) * progress
row.update(visual_floor=floor, passenger_count=len(shaft['boarded']),
passenger_ids=copy.deepcopy(shaft['boarded']),
waiting_count=sum(t['shaft_id'] == shaft['shaft_id'] and t['state'] == 'QUEUED'
# get_report() merges config before counting (345-359).
waiting_count=sum(t['shaft_id'] == row['shaft_id'] and t['state'] == 'QUEUED'
for t in saved['trips']))
rows.append(row)
# WorldPreview sorts its report's trip copy for queue placement.
+57 -7
View File
@@ -18,6 +18,54 @@ sys.dont_write_bytecode = True
import check_p23_life_observation_evidence as check
class QueuedElevatorTests(unittest.TestCase):
def setUp(self):
fixture = json.loads((check.ROOT / 'tests/fixtures/p23-03/queued_elevator.json').read_text())
self.state = {'systems': {'runtime': {'elevators': fixture['saved']}}}
def test_queued_trip_without_cabin_seat_uses_config_shaft(self):
saved = self.state['systems']['runtime']['elevators']
self.assertNotIn('shaft_id', saved['shafts'][0])
self.assertEqual(saved['shafts'][0]['boarded'], [])
report = check.elevator_report(self.state)
self.assertEqual(report['shafts'][0]['waiting_count'], 1)
self.assertEqual(report['shafts'][0]['passenger_count'], 0)
self.assertAlmostEqual(report['shafts'][0]['visual_floor'], .9)
self.assertEqual(report['trips'], saved['trips'])
def test_tampered_shaft_attribution_fails(self):
self.state['systems']['runtime']['elevators']['trips'][0]['shaft_id'] = 'wrong.shaft'
with self.assertRaisesRegex(ValueError, 'shaft attribution'):
check.elevator_report(self.state)
def test_wrong_waiting_count_fails_whole_report_comparison(self):
expected = check.elevator_report(self.state)
tampered = copy.deepcopy(expected)
tampered['shafts'][0]['waiting_count'] = 0
directory = mock.MagicMock()
manifest = dict(label='queued-fixture', pid=123, state='state',
state_after='state', report='report')
directory.__truediv__.return_value.open.return_value = io.StringIO(json.dumps(manifest)+'\n')
# Exercise the actual stream comparison; isolate unrelated actor fields.
with mock.patch.object(check, 'read_object', side_effect=[self.state, self.state,
{'elevators': tampered}]), \
mock.patch.object(check, 'compare_actor_rows', return_value=[]):
with self.assertRaisesRegex(ValueError, 'typed elevator report mismatch'):
check.inspect_stream(directory)
def test_extra_and_missing_trip_keys_fail(self):
for mutation in ('extra', 'missing'):
with self.subTest(mutation=mutation):
state = copy.deepcopy(self.state)
trip = state['systems']['runtime']['elevators']['trips'][0]
if mutation == 'extra':
trip['unexpected'] = True
else:
del trip['shaft_id']
with self.assertRaisesRegex(ValueError, 'exact elevator trip keys'):
check.elevator_report(state)
class SnapshotCellTests(unittest.TestCase):
def test_layout_arrival_compares_serialized_cell_arrays(self):
source = (check.ROOT / 'tests/fixtures/p23-03/layout_comparison.gd').read_text()
@@ -431,12 +479,14 @@ class SpatialTests(unittest.TestCase):
check.compare_traffic(self.snapshot, self.report)
def test_missing_queue_overflow_rejected(self):
self.runtime['elevators']['trips'] = [dict(actor_id='guest:g'+str(i), enqueued_at=i,
state='QUEUED', **{'from': [0, 0, 0]}) for i in range(3)]
self.report.update(queue_positions={'guest/g0': [84, 82], 'guest/g1': [100, 82]},
queue_hidden={'guest/g2': True}, landing_badges=[dict(at=[76, 54],
text='待ち3人(他1人)', cell=[0, 0, 0],
members=['guest/g0', 'guest/g1', 'guest/g2'], visible_count=2, overflow=1)])
saved = json.loads((check.ROOT / 'tests/fixtures/p23-03/queued_elevator.json').read_text())['saved']
self.runtime['elevators'] = saved
saved['trips'] = [dict(copy.deepcopy(saved['trips'][0]), actor_id='guest:g'+str(i),
request_id='trip'+str(i), enqueued_at=i) for i in range(3)]
self.report.update(queue_positions={'guest/g0': [100, 82]},
queue_hidden={'guest/g1': True, 'guest/g2': True}, landing_badges=[dict(at=[-4, 54],
text='待ち3人(他2人)', cell=[0, 0, 0],
members=['guest/g0', 'guest/g1', 'guest/g2'], visible_count=1, overflow=2)])
check.compare_queues(self.snapshot, self.report)
self.report['landing_badges'][0]['overflow'] = 0
with self.assertRaises(ValueError):
@@ -470,7 +520,7 @@ class SpatialTests(unittest.TestCase):
check.compare_room_services(self.snapshot, changed)
def test_cabin_geometry_and_phase(self):
shaft = dict(shaft_id='lift', config=dict(stops=[[0, 0, 0]], width=2, capacity=4),
shaft = dict(config=dict(shaft_id='lift', stops=[[0, 0, 0]], width=2, capacity=4),
current_floor=0, phase='IDLE', operational=True, boarded=[])
self.runtime['elevators']['shafts'] = [shaft]
self.report.update(paused=True, animation_time=0, elevator_tracks={'lift': dict(floor=0, at=0)},