ClaudeandClaude Opus 5.5
8064834c37
- test_refined_dfg::test_source_pin runs its mismatch negative controls first (now including window mode) so missing history cannot skip them. - test_re100: renamed to test_manifest_matches_its_pinned_commit, plus an explicit record that today's src differs from the pin. - Structural manifest checks run before any history-dependent check. - Real tamper test: a flipped digest raises via verify_manifest, and a path absent at the commit raises ValueError. - A blobless clone (tree present, blob missing) is reported as unavailable history, not as a missing path. - All history skips go through history_unavailable(): fatal when CELERIS_REQUIRE_HISTORY=1. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
101 lines
3.8 KiB
Python
101 lines
3.8 KiB
Python
"""Explicit experimental repin and historical-source provenance contracts."""
|
|
|
|
import hashlib
|
|
import json
|
|
from pathlib import Path
|
|
import subprocess
|
|
import sys
|
|
|
|
ROOT = Path(__file__).resolve().parents[2]
|
|
sys.path.insert(0, str(ROOT / "tests/validation"))
|
|
import pytest
|
|
from source_provenance import (HistoryUnavailable, checkout_provenance, historical_sha256,
|
|
historical_tree, history_unavailable, verify_manifest)
|
|
import unittest
|
|
import re100_helpers
|
|
import run_jet_circle_cv
|
|
|
|
ORIGIN = "47b05bc8abfc957fe876b0c63bdb386e3c6cbaa5"
|
|
|
|
|
|
def test_manifests_match_their_recorded_source_and_preserve_origin():
|
|
"""Historical evidence integrity: each manifest describes the tree of its own commit."""
|
|
manifests = []
|
|
for group, hashes_key, revision_key in (
|
|
("dfg_re100", "production_sha256", "commit"),
|
|
("jet_circle_cv", "files", "revision"),
|
|
):
|
|
manifest = json.loads(
|
|
(
|
|
ROOT / "tests/validation/data" / group / "source_manifest.json"
|
|
).read_text()
|
|
)
|
|
# Structural checks need no history and always run.
|
|
assert manifest[revision_key] == ORIGIN
|
|
assert manifest["source_origin_commits"][-1] == ORIGIN
|
|
assert (
|
|
manifest["historical_checkpoints"]["production"]
|
|
== "d79e3f6f0a424f5bcc3a1a6d610313e49daf97a2"
|
|
)
|
|
assert "not inherited from d79" in manifest["experiment"]
|
|
manifests.append(manifest[hashes_key])
|
|
try:
|
|
names = historical_tree(ROOT, ORIGIN)
|
|
for digests in manifests:
|
|
assert set(digests) == set(names)
|
|
verify_manifest(ROOT, ORIGIN, digests)
|
|
except HistoryUnavailable as error:
|
|
history_unavailable(unittest.TestCase(), error)
|
|
assert re100_helpers.PIN == run_jet_circle_cv.PIN == ORIGIN
|
|
|
|
|
|
def test_runtime_checkout_identity_is_separate_from_source_origin(tmp_path):
|
|
subprocess.run(["git", "init", "-q", str(tmp_path)], check=True)
|
|
(tmp_path / "test.txt").write_text("tracked\n")
|
|
subprocess.run(["git", "-C", str(tmp_path), "add", "test.txt"], check=True)
|
|
subprocess.run(
|
|
[
|
|
"git",
|
|
"-C",
|
|
str(tmp_path),
|
|
"-c",
|
|
"user.name=Test",
|
|
"-c",
|
|
"user.email=test@example.invalid",
|
|
"commit",
|
|
"-qm",
|
|
"fixture",
|
|
],
|
|
check=True,
|
|
)
|
|
expected = subprocess.check_output(
|
|
["git", "-C", str(tmp_path), "rev-parse", "HEAD"], text=True
|
|
).strip()
|
|
assert checkout_provenance(tmp_path) == dict(
|
|
checkout_revision=expected, checkout_dirty=False
|
|
)
|
|
(tmp_path / "test.txt").write_text("changed\n")
|
|
assert checkout_provenance(tmp_path) == dict(
|
|
checkout_revision=expected, checkout_dirty=True
|
|
)
|
|
|
|
|
|
def test_historical_check_rejects_tampering_and_reports_missing_history(tmp_path):
|
|
with pytest.raises(HistoryUnavailable):
|
|
historical_sha256(tmp_path, ORIGIN, ["src/x.py"]) # not a repository
|
|
with pytest.raises(HistoryUnavailable):
|
|
historical_sha256(ROOT, "0" * 40, ["src/x.py"]) # commit absent
|
|
manifest = json.loads(
|
|
(ROOT / "tests/validation/data/dfg_re100/source_manifest.json").read_text())
|
|
digests = dict(manifest["production_sha256"])
|
|
name = sorted(digests)[0]
|
|
tampered = dict(digests, **{name: ("0" if digests[name][0] != "0" else "1") + digests[name][1:]})
|
|
try:
|
|
historical_tree(ROOT, ORIGIN)
|
|
except HistoryUnavailable as error:
|
|
history_unavailable(unittest.TestCase(), error)
|
|
with pytest.raises(RuntimeError, match="Historical manifest mismatch"):
|
|
verify_manifest(ROOT, ORIGIN, tampered) # one flipped digest is caught
|
|
with pytest.raises(ValueError, match="not part of"):
|
|
historical_sha256(ROOT, ORIGIN, ["src/CelerisLab/does_not_exist.py"])
|