Digital Drosophila Archive
254 lines
14 KiB
Python
254 lines
14 KiB
Python
"""Temporary synthetic seals only; never publish/register any real evidence."""
|
|
import hashlib
|
|
import json
|
|
from pathlib import Path
|
|
import tempfile
|
|
import unittest
|
|
from unittest import mock
|
|
|
|
import publish_phase4al as publication
|
|
|
|
|
|
HELPER = Path(__file__).resolve().parents[1] / 'pn-recruitment-audit-v1/publication.py'
|
|
if not HELPER.is_file():
|
|
HELPER = publication.ROOT / publication.HELPER_RELATIVE
|
|
|
|
|
|
def sha(path):
|
|
return hashlib.sha256(path.read_bytes()).hexdigest()
|
|
|
|
|
|
def write(path, value):
|
|
path.parent.mkdir(parents=True, exist_ok=True)
|
|
path.write_text(json.dumps(value, indent=2), encoding='utf-8')
|
|
|
|
|
|
def seal(root):
|
|
rows = [dict(path=path.relative_to(root).as_posix(), bytes=path.stat().st_size, sha256=sha(path))
|
|
for path in sorted(root.rglob('*')) if path.is_file() and
|
|
path.name not in ('manifest.json', 'checksums.sha256') and '__pycache__' not in path.parts]
|
|
write(root / 'manifest.json', dict(root=str(root), files=rows))
|
|
lines = [row['sha256'] + ' ' + row['path'] for row in rows]
|
|
lines.append(sha(root / 'manifest.json') + ' manifest.json')
|
|
(root / 'checksums.sha256').write_text('\n'.join(lines) + '\n', encoding='utf-8')
|
|
|
|
|
|
class PublicationTests(unittest.TestCase):
|
|
def setUp(self):
|
|
temporary = tempfile.TemporaryDirectory(prefix='phase4al-publication-synthetic-')
|
|
self.addCleanup(temporary.cleanup)
|
|
self.root = Path(temporary.name).resolve()
|
|
self.model = self.root / publication.MODEL_RELATIVE
|
|
self.experiment = self.root / publication.EXPERIMENT_RELATIVE
|
|
self.registry = self.root / publication.REGISTRY_RELATIVE
|
|
self.model.mkdir(parents=True)
|
|
self.experiment.mkdir(parents=True)
|
|
(self.model / 'module.py').write_text('# synthetic, not a model\n', encoding='utf-8')
|
|
self.config = dict(phase='4AL', neurons=166700, edges=25582938, candidate_edges=14232,
|
|
primary_trials=120, technical_repeats=24, steps=1000, dt_ms=.1, virtual_gate_tick=900,
|
|
kc_trace_tau_ms=100., trace_cap=1., stable_seed_count=3, conditional_depression_fraction=.25,
|
|
groups=['plastic_KC', 'nonplastic_KC', 'DAN', 'MBON', 'APL', 'other'],
|
|
training=False, candidate_generation=False, no_new_overall_credit_threshold=True)
|
|
write(self.model / 'config.json', self.config)
|
|
self.science = dict(arithmetic_and_coverage_pass=True, primary_trials=120, exact_repeats=24,
|
|
pairs=60, candidate_generated=False, learning_achieved=False, conditioning_executed=False,
|
|
scientific_overall_PASS_threshold_defined=False,
|
|
blocks={c + '/' + m: dict(trials=30, pairs=15, pairs_with_both_specific_KC_and_edge_sets=0,
|
|
conditional25_event_changed_trials=0, conditional25_count_or_first_latency_changed_trials=0)
|
|
for c in publication.COHORTS for m in publication.MODES})
|
|
write(self.experiment / 'audit/scientific-summary.json', self.science)
|
|
self.status = dict(status=publication.FINAL_STATUS, **self.science, old_integrity_pass=True,
|
|
frozen_five_reference_pass=True, frozen_eligibility_and_mask_unchanged=True,
|
|
base_weights_unchanged=True, summary_all_144_baseline_raw_states_and_repeat_pair_metrics_reverified=True,
|
|
scientific_summary_sha256=sha(self.experiment / 'audit/scientific-summary.json'))
|
|
write(self.experiment / 'audit/final-status.json', self.status)
|
|
write(self.experiment / 'regression/frozen-five.json', dict(pass_all=True,
|
|
trials=[dict(trial=str(i), exact={k: True for k in ('step', 'neuron_idx', 'counts', 'v', 'g')}) for i in range(5)]))
|
|
write(self.experiment / 'regression/old-integrity-after.json', {'synthetic-old-tree': {'pass': True}})
|
|
write(self.experiment / 'nested/case/summary.json', {'synthetic': True})
|
|
(self.experiment / 'REPORT.md').write_text('Synthetic zero-effect diagnostic.\n', encoding='utf-8')
|
|
(self.experiment / 'tests.log').write_text('Synthetic log\n', encoding='utf-8')
|
|
for suffix in ('.npy', '.npz', '.parquet'):
|
|
(self.experiment / ('bulk' + suffix)).write_bytes(b'synthetic bulk bytes, not scientific data')
|
|
seal(self.model)
|
|
seal(self.experiment)
|
|
|
|
def publish(self):
|
|
return publication.publish(self.root, HELPER)
|
|
|
|
def test_completed_zero_effect_diagnostic_is_not_regraded_as_learning(self):
|
|
before = {str(p): sha(p) for base in (self.model, self.experiment) for p in base.rglob('*') if p.is_file()}
|
|
result = self.publish()
|
|
self.assertTrue(result['publication_complete'])
|
|
receipt = json.loads((self.registry / 'publication-receipt.json').read_text(encoding='utf-8'))
|
|
self.assertEqual(receipt['declared_audit_status'], publication.FINAL_STATUS)
|
|
self.assertFalse(receipt['scientific_overall_PASS_threshold_defined'])
|
|
self.assertFalse(receipt['learning_achieved'])
|
|
self.assertTrue(receipt['publication_success_is_not_scientific_pass'])
|
|
self.assertEqual(receipt['scientific_summary_sha256'], self.status['scientific_summary_sha256'])
|
|
for suffix in ('.npy', '.npz', '.parquet'):
|
|
self.assertFalse((self.registry / ('evidence/bulk' + suffix)).exists())
|
|
for row in receipt['files']:
|
|
self.assertEqual(sha(self.registry / row['path']), row['sha256'])
|
|
self.assertEqual(before, {str(p): sha(p) for base in (self.model, self.experiment) for p in base.rglob('*') if p.is_file()})
|
|
|
|
def test_changed_config_blocks_before_destination(self):
|
|
for key, value in (('phase', '4AK'), ('primary_trials', 119), ('candidate_edges', 14231),
|
|
('conditional_depression_fraction', .5), ('training', True), ('virtual_gate_tick', 901)):
|
|
with self.subTest(key=key):
|
|
write(self.model / 'config.json', dict(self.config, **{key: value}))
|
|
seal(self.model)
|
|
with self.assertRaisesRegex(ValueError, 'configuration/design'):
|
|
self.publish()
|
|
self.assertFalse(self.registry.exists())
|
|
|
|
def test_summary_hash_counts_and_scope_must_match(self):
|
|
for changed in ({'scientific_summary_sha256': '0' * 64}, {'primary_trials': 119},
|
|
{'pairs': 59}, {'learning_achieved': True}, {'arithmetic_and_coverage_pass': 'true'},
|
|
{'old_integrity_pass': False}, {'base_weights_unchanged': False}):
|
|
with self.subTest(changed=changed):
|
|
write(self.experiment / 'audit/final-status.json', dict(self.status, **changed))
|
|
seal(self.experiment)
|
|
with self.assertRaises(ValueError):
|
|
self.publish()
|
|
self.assertFalse(self.registry.exists())
|
|
|
|
def test_missing_or_corrupt_scientific_summary_is_rejected(self):
|
|
path = self.experiment / 'audit/scientific-summary.json'
|
|
path.unlink()
|
|
seal(self.experiment)
|
|
with self.assertRaisesRegex(ValueError, 'Required finalized evidence absent'):
|
|
self.publish()
|
|
self.assertFalse(self.registry.exists())
|
|
|
|
def test_complete_status_requires_all_four_complete_blocks(self):
|
|
value = dict(self.science, blocks={})
|
|
write(self.experiment / 'audit/scientific-summary.json', value)
|
|
write(self.experiment / 'audit/final-status.json', dict(self.status, **value,
|
|
scientific_summary_sha256=sha(self.experiment / 'audit/scientific-summary.json')))
|
|
seal(self.experiment)
|
|
with self.assertRaisesRegex(ValueError, 'cohort-mode blocks'):
|
|
self.publish()
|
|
|
|
def test_false_reference_exactness_is_not_hidden_by_pass_all(self):
|
|
write(self.experiment / 'regression/frozen-five.json', dict(pass_all=True,
|
|
trials=[dict(exact={'step': False})] * 5))
|
|
seal(self.experiment)
|
|
with self.assertRaisesRegex(ValueError, 'five exact trial'):
|
|
self.publish()
|
|
self.assertFalse(self.registry.exists())
|
|
|
|
def test_unknown_or_unfinalized_status_is_not_promoted(self):
|
|
for status in ('INCOMPLETE-NOT-SEALED', 'PASS-FROZEN-D18-INDEPENDENT-VALIDATION'):
|
|
write(self.experiment / 'audit/final-status.json', dict(self.status, status=status))
|
|
seal(self.experiment)
|
|
with self.assertRaisesRegex(ValueError, 'Unrecognized Phase 4AL'):
|
|
self.publish()
|
|
self.assertFalse(self.registry.exists())
|
|
|
|
def test_manifest_path_escape_rejected_before_destination(self):
|
|
path = self.experiment / 'manifest.json'
|
|
manifest = json.loads(path.read_text(encoding='utf-8'))
|
|
manifest['files'][0]['path'] = '../escape.json'
|
|
write(path, manifest)
|
|
controls = self.experiment / 'checksums.sha256'
|
|
controls.write_text('\n'.join([r['sha256'] + ' ' + r['path'] for r in manifest['files']] +
|
|
[sha(path) + ' manifest.json']) + '\n', encoding='utf-8')
|
|
with self.assertRaisesRegex(ValueError, 'escaping relative path'):
|
|
self.publish()
|
|
self.assertFalse(self.registry.exists())
|
|
|
|
def test_completed_status_without_frozen_reference_is_not_publishable(self):
|
|
(self.experiment / 'regression/frozen-five.json').unlink()
|
|
seal(self.experiment)
|
|
with self.assertRaisesRegex(ValueError, 'frozen-reference report/presence'):
|
|
self.publish()
|
|
self.assertFalse(self.registry.exists())
|
|
|
|
def test_existing_registry_never_overwritten_or_resumed(self):
|
|
self.registry.mkdir(parents=True)
|
|
sentinel = self.registry / 'keep.txt'
|
|
sentinel.write_bytes(b'keep these bytes')
|
|
with self.assertRaisesRegex(ValueError, 'Registry already exists'):
|
|
self.publish()
|
|
self.assertEqual(sentinel.read_bytes(), b'keep these bytes')
|
|
self.assertEqual(list(self.registry.iterdir()), [sentinel])
|
|
|
|
def test_missing_integrity_report_blocks_before_destination(self):
|
|
(self.experiment / 'regression/old-integrity-after.json').unlink()
|
|
seal(self.experiment)
|
|
with self.assertRaisesRegex(ValueError, 'Required finalized evidence absent'):
|
|
self.publish()
|
|
self.assertFalse(self.registry.exists())
|
|
|
|
def test_bulk_corruption_blocks_even_though_bulk_is_not_copied(self):
|
|
path = self.experiment / 'bulk.npy'
|
|
path.write_bytes(b'x' * path.stat().st_size)
|
|
with self.assertRaisesRegex(ValueError, 'Payload SHA-256 mismatch'):
|
|
self.publish()
|
|
self.assertFalse(self.registry.exists())
|
|
|
|
def test_unlisted_payload_and_bad_manifest_self_hash_fail(self):
|
|
extra = self.experiment / 'unlisted.csv'
|
|
extra.write_text('unlisted\n')
|
|
with self.assertRaisesRegex(ValueError, 'unlisted/missing'):
|
|
self.publish()
|
|
extra.unlink()
|
|
controls = self.experiment / 'checksums.sha256'
|
|
lines = controls.read_text().splitlines()
|
|
lines[-1] = '0' * 64 + ' manifest.json'
|
|
controls.write_text('\n'.join(lines) + '\n')
|
|
with self.assertRaisesRegex(ValueError, 'manifest self-checksum'):
|
|
self.publish()
|
|
self.assertFalse(self.registry.exists())
|
|
|
|
def test_large_json_verified_but_not_copied(self):
|
|
path = self.experiment / 'large.json'
|
|
path.write_bytes(b' ' * (publication.MAX_EVIDENCE_BYTES + 1))
|
|
seal(self.experiment)
|
|
self.publish()
|
|
self.assertFalse((self.registry / 'evidence/large.json').exists())
|
|
receipt = json.loads((self.registry / 'publication-receipt.json').read_text(encoding='utf-8'))
|
|
self.assertEqual(next(r for r in receipt['experiment_payload_not_copied'] if r['path'] == 'large.json')['reason'],
|
|
'exceeds_4_MiB')
|
|
|
|
def test_interrupted_copy_retained_and_cannot_resume(self):
|
|
with mock.patch.object(publication, '_copy', side_effect=OSError('synthetic interrupted copy')):
|
|
with self.assertRaisesRegex(OSError, 'synthetic interrupted copy'):
|
|
self.publish()
|
|
self.assertTrue((self.registry / 'publication-failure.json').is_file())
|
|
self.assertFalse((self.registry / 'publication-receipt.json').exists())
|
|
with self.assertRaisesRegex(ValueError, 'Registry already exists'):
|
|
self.publish()
|
|
|
|
def test_after_copy_full_verification_detects_bulk_change(self):
|
|
original = publication._copy
|
|
changed = False
|
|
|
|
def copy_then_change(helper, source, destination, size, checksum):
|
|
nonlocal changed
|
|
original(helper, source, destination, size, checksum)
|
|
if not changed:
|
|
changed = True
|
|
(self.experiment / 'bulk.npy').write_bytes(b'changed after initial seal check')
|
|
|
|
with mock.patch.object(publication, '_copy', side_effect=copy_then_change):
|
|
with self.assertRaises(ValueError):
|
|
self.publish()
|
|
self.assertTrue((self.registry / 'publication-failure.json').is_file())
|
|
self.assertFalse((self.registry / 'publication-receipt.json').exists())
|
|
|
|
def test_helper_pin_checked_without_modifying_old_globals_or_cache(self):
|
|
helper = publication.load_helpers(HELPER)
|
|
self.assertEqual(helper.MODEL_RELATIVE, Path('models/male-cns-lif/pn-recruitment-audit-v1'))
|
|
self.assertEqual(helper.FINAL_STATUS, 'PASS-CONDITIONAL-PN-APL-SOURCE-AUDIT')
|
|
wrong = self.root / 'wrong-helper.py'
|
|
wrong.write_bytes(b'raise RuntimeError("must never execute")\n')
|
|
with self.assertRaisesRegex(ValueError, 'helper hash mismatch'):
|
|
publication.publish(self.root, wrong)
|
|
self.assertFalse(self.registry.exists())
|
|
|
|
|
|
if __name__ == '__main__':
|
|
unittest.main()
|