Skip to content

fix(validation): accept empty kernel-fusion fallback findings #941

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
bong-water-water-bong wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix(validation): accept empty kernel-fusion fallback findings #941

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
36 changes: 26 additions & 10 deletions TraceLens/Agent/Analysis/utils/validation_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -129,6 +129,20 @@ def _category_findings_empty(filepath):
return isinstance(cf, list) and len(cf) == 0


def _fusion_no_data(filepath):
"""True when kernel_fusion_metrics.json has no impact estimates (kernel-fusion-analyzer § empty fallback)."""
mp = _metrics_json_for_findings(filepath)
try:
with open(mp) as f:
data = json.load(f)
except (OSError, json.JSONDecodeError):
return False
if data.get("status") == "NO_DATA":
return True
est = data.get("impact_estimates")
return isinstance(est, list) and len(est) == 0


def validate_findings_file(filepath, tier, comparison_scope=None):
"""Validate a single findings file against the sub-agent spec contract.

Expand Down Expand Up @@ -160,22 +174,24 @@ def validate_findings_file(filepath, tier, comparison_scope=None):

errors = []

header_positions = []
for h in _REQUIRED_FINDINGS_HEADERS:
pos = content.find(h)
if pos < 0:
errors.append(f"Missing required section: {h}")
header_positions.append(pos)
relaxed_empty = tier == "compute" and _category_findings_empty(filepath)
relaxed_empty = relaxed_empty or (tier == "fusion" and _fusion_no_data(filepath))

if all(p >= 0 for p in header_positions):
header_positions = []
if not relaxed_empty:
for h in _REQUIRED_FINDINGS_HEADERS:
pos = content.find(h)
if pos < 0:
errors.append(f"Missing required section: {h}")
header_positions.append(pos)

if header_positions and all(p >= 0 for p in header_positions):
if header_positions[0] > header_positions[1]:
errors.append("## Recommendations must appear before ## Detailed Analysis")

rec_start = content.find("## Recommendations")
da_start = content.find("## Detailed Analysis")

relaxed_empty = tier == "compute" and _category_findings_empty(filepath)

p_items = []
if rec_start >= 0:
rec_end = da_start if da_start > rec_start else len(content)
Expand Down Expand Up @@ -999,7 +1015,7 @@ def check_findings_file(cls, path, file_class, *, skip_p_item_required=False):
if file_class == "category_findings" and rel not in cls.COMPUTE_NO_P_ITEM:
if not skip_p_item_required and "p_item" not in seen_kinds:
errors.append(f"{rel}: missing required kind=p_item")
if file_class == "system_findings" and "p_item" not in seen_kinds:
if file_class == "system_findings" and not skip_p_item_required and "p_item" not in seen_kinds:
errors.append(f"{rel}: missing required kind=p_item")
n_headings = len(_P_ITEM_RE.findall(text))
n_markers = sum(
Expand Down