Codebase intelligence for AI and humans: code health scores, auto-generated docs, git analytics, dead code detection, and architectural decisions via MCP.
Python aicode-complexitycode-healthcode-intelligencecode-qualitydead-codedeveloper-toolsdocumentationgit-analyticsmcpopen-sourcepythonrefactoringstatic-analysistechnical-debt
"""``_grade_answer`` pinned over a table of inputs.
Each row was graded once and its full ``_Grade`` recorded. The rows cover every
outcome shape and every gate that demotes a grade, so a reordered or rewritten
gate changes a recorded verdict.
"""
from __future__ import annotations
import json
from pathlib import Path
import pytest
from repowise.server.mcp_server.tool_answer.confidence import _grade_answer
_CASES = json.loads(
(Path(__file__).parents[3] / "fixtures" / "mcp" / "grade_answer_cases.json").read_text(
encoding="utf-8"
)
)
@pytest.mark.parametrize("row", _CASES, ids=[f"case{i}" for i in range(len(_CASES))])
def test_grade_matches_recorded_verdict(row: dict, monkeypatch: pytest.MonkeyPatch) -> None:
for name, value in row["env"].items():
monkeypatch.setenv(name, value)
kwargs = dict(row["kwargs"], question_ids=set(row["kwargs"]["question_ids"]))
got = json.loads(json.dumps(_grade_answer(**kwargs)._asdict()))
assert got == row["expected"]
def test_table_reaches_every_grade() -> None:
assert {row["expected"]["confidence"] for row in _CASES} == {"high", "medium", "low"}