Skip to content

fix(verify): preserve original input_dump and previous_manifest_path in VerificationReport #183

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
rishiiicreates wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix(verify): preserve original input_dump and previous_manifest_path in VerificationReport #183

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
39 changes: 22 additions & 17 deletions Legacy/openverifiablellm/verify.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -232,21 +232,26 @@ def verify_preprocessing(
VerificationReport
Structured report with per-check pass/fail results.
"""
input_dump = Path(input_dump).resolve()
original_input_dump = str(input_dump)
resolved_input_dump = Path(input_dump).resolve()
root = project_root or Path.cwd()

if manifest_path is None:
manifest_path = root / "data" / "dataset_manifest.json"
else:
manifest_path = Path(manifest_path)

if previous_manifest_path is not None:
previous_manifest_path = Path(previous_manifest_path).resolve()
original_previous_manifest_path = (
str(previous_manifest_path) if previous_manifest_path is not None else None
)
resolved_previous_manifest_path = (
Path(previous_manifest_path).resolve() if previous_manifest_path is not None else None
)

report = VerificationReport(
input_dump=str(input_dump),
input_dump=original_input_dump,
manifest_path=str(manifest_path),
previous_manifest_path=str(previous_manifest_path) if previous_manifest_path else None,
previous_manifest_path=original_previous_manifest_path,
)

# 1. Load existing manifest
Expand Down Expand Up @@ -280,24 +285,24 @@ def verify_preprocessing(
)

# ===== Verify manifest chain if previous manifest is provided =====
if previous_manifest_path is not None:
if not previous_manifest_path.exists():
if resolved_previous_manifest_path is not None:
if not resolved_previous_manifest_path.exists():
report.add(
CheckResult(
name="manifest_chain_link",
status=CheckStatus.FAIL,
detail=f"Previous manifest not found: {previous_manifest_path}",
detail=f"Previous manifest not found: {resolved_previous_manifest_path}",
)
)
else:
try:
chain_valid = verify_manifest_chain_link(previous_manifest_path, manifest)
chain_valid = verify_manifest_chain_link(resolved_previous_manifest_path, manifest)
status = CheckStatus.PASS if chain_valid else CheckStatus.FAIL
report.add(
CheckResult(
name="manifest_chain_link",
status=status,
expected=previous_manifest_path.name,
expected=resolved_previous_manifest_path.name,
actual="✓ linked" if chain_valid else "✗ broken",
detail="Verifies parent_manifest_hash matches previous manifest hash",
)
Expand Down Expand Up @@ -333,20 +338,20 @@ def verify_preprocessing(
# ========================================================================

# 2. Validate raw file integrity BEFORE re-processing
if not input_dump.exists():
if not resolved_input_dump.exists():
report.add(
CheckResult(
name="raw_file_exists",
status=CheckStatus.FAIL,
detail=f"Input dump not found: {input_dump}",
detail=f"Input dump not found: {resolved_input_dump}",
)
)
return report

report.add(CheckResult(name="raw_file_exists", status=CheckStatus.PASS))

# SHA256 of raw file
raw_sha256_actual = utils.compute_sha256(file_path=input_dump)
raw_sha256_actual = utils.compute_sha256(file_path=resolved_input_dump)
_check_field(
report,
"raw_sha256",
Expand All @@ -373,7 +378,7 @@ def verify_preprocessing(

# Merkle root of raw file
if "raw_merkle_root" in manifest:
raw_merkle_actual = utils.compute_merkle_root(input_dump, chunk_size=chunk_size)
raw_merkle_actual = utils.compute_merkle_root(resolved_input_dump, chunk_size=chunk_size)
_check_field(
report,
"raw_merkle_root",
Expand Down Expand Up @@ -406,15 +411,15 @@ def verify_preprocessing(
report,
"dump_date",
expected=manifest.get("dump_date"),
actual=utils.extract_dump_date(input_dump.name),
actual=utils.extract_dump_date(resolved_input_dump.name),
detail="Dump date parsed from filename",
)

_check_field(
report,
"wikipedia_dump_name",
expected=manifest.get("wikipedia_dump"),
actual=input_dump.name,
actual=resolved_input_dump.name,
detail="Raw filename recorded in manifest",
)

Expand Down Expand Up @@ -488,7 +493,7 @@ def verify_preprocessing(
[
sys.executable,
str(script_path),
str(input_dump),
str(resolved_input_dump),
],
cwd=tmp_dir,
check=True,
Expand Down
17 changes: 17 additions & 0 deletions Legacy/tests/test_verify.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -258,6 +258,23 @@ def test_report_stores_input_dump_path(self):
r = verify_preprocessing(self.dump, project_root=self.tmp)
self.assertEqual(r.input_dump, str(self.dump))

def test_report_stores_relative_input_dump_path(self):
original = os.getcwd()
os.chdir(self.tmp)
try:
rel_path = "simplewiki-20260201-pages-articles.xml.bz2"
r = verify_preprocessing(rel_path, project_root=self.tmp)
self.assertEqual(r.input_dump, rel_path)
finally:
os.chdir(original)

def test_report_stores_previous_manifest_path(self):
prev_manifest = self.tmp / "prev_manifest.json"
prev_manifest.write_text(json.dumps({"manifest_hash": "abc"}))
r = verify_preprocessing(self.dump, previous_manifest_path=prev_manifest, project_root=self.tmp)
self.assertEqual(r.previous_manifest_path, str(prev_manifest))



# Integration: failure scenarios

Expand Down