mirror of
https://github.com/norandom/OpenUltraSAST.git
synced 2026-10-16 11:49:28 +00:00
Marius CiepluchandClaude Fable 5.1
d9d6335ab3
hunter-learning-loop groups 1-3 (first increment). The feature table is the contract: one row per touched function with diff, source, semantic and taxonomy columns, provenance per producer, labels by the benchmark's rule. Producers read the repository at head only (a planted advisory id and commit text never reach a feature). The hunt program is a DSPy ReAct loop over a signature whose inputs are the unmasked columns in a fixed order; answers outside the closed family list or citing unseen lines score as unknown. The runner writes records the benchmark scorer reads, resumable by change, program id and mask, and refuses the held-out pilot unless the run is a report run. Research extra [hunt] = dspy, pandas; the core install is unchanged. 43 tests on a scripted model, no network. Co-Authored-By: Claude Fable 5.1 <[email protected]>
48 lines
1.7 KiB
Python
48 lines
1.7 KiB
Python
"""Checks for fix-site and ordinary labels."""
|
|
|
|
import pytest
|
|
|
|
from openultrasast.hunt.labels import label_rows
|
|
from openultrasast.hunt.table import FeatureTable
|
|
|
|
|
|
def positive():
|
|
return {"kind": "vulnerability", "path": "app.py", "function": "handle", "function_lines": [10, 20],
|
|
"family": "injection", "case_id": "verified-case"}
|
|
|
|
|
|
def test_labels_match_name_or_span_on_same_path():
|
|
rows = [{"path": "app.py", "function": "handle", "line": 1},
|
|
{"path": "app.py", "function": "nested", "line": 12},
|
|
{"path": "other.py", "function": "handle", "line": 12},
|
|
{"path": "app.py", "function": "other", "line": 30}]
|
|
labelled = label_rows(rows, positive())
|
|
assert [r["label"] for r in labelled] == [1, 1, None, None]
|
|
for data in labelled[:2]:
|
|
assert data["family"] == "injection"
|
|
assert data["fix_span"] == [10, 20]
|
|
assert data["label_ref"] == "verified-case"
|
|
assert data["verification"] is None
|
|
assert "label" not in rows[0]
|
|
|
|
|
|
def test_feature_function_and_ordinary_rows():
|
|
rows = [{"path": "app.py", "line": 1, "feature.source.function": "handle"}]
|
|
assert label_rows(rows, positive())[0]["label"] == 1
|
|
ordinary = label_rows([{"path": "app.py", "line": 1}], {"kind": "ordinary", "case_id": "draw"})
|
|
assert ordinary[0]["label"] == 0
|
|
assert len(FeatureTable(ordinary).labelled().rows) == 1
|
|
|
|
|
|
@pytest.mark.parametrize("kind", [None, "fix", "unknown"])
|
|
def test_unknown_change_kind_rejected(kind):
|
|
with pytest.raises(ValueError):
|
|
label_rows([], {"kind": kind})
|
|
|
|
|
|
def test_positive_requires_traceable_reference():
|
|
change = positive()
|
|
del change["case_id"]
|
|
with pytest.raises(ValueError, match="case_id"):
|
|
label_rows([], change)
|