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
"""The idle-decay refresh never mints fragment rows, and heals stores built before
non-code files had a history tier."""
from __future__ import annotations
from sqlalchemy import select
from repowise.core.ingestion.git_indexer import GitIndexer
from repowise.core.ingestion.git_indexer.tiers import GitIndexTier
from repowise.core.persistence.crud import upsert_git_metadata_bulk
from repowise.core.persistence.models import GitMetadata
from repowise.core.pipeline.persist import persist_git_refresh
from tests.unit.ingestion.test_git_history_tier import _build
from tests.unit.persistence.helpers import insert_repo
async def _rows(session, repo_id: str) -> dict[str, GitMetadata]:
result = await session.execute(select(GitMetadata).where(GitMetadata.repository_id == repo_id))
return {row.file_path: row for row in result.scalars().all()}
async def test_decay_rows_refresh_stored_rows_and_never_insert(async_session) -> None:
repo = await insert_repo(async_session)
await upsert_git_metadata_bulk(
async_session,
repo.id,
[{"file_path": "a.py", "commit_count_total": 5, "primary_owner_name": "Alice"}],
)
decay = {
"a.py": {"file_path": "a.py", "commit_count_90d": 3},
# No stored row: inserting this partial is the fragment bug.
"b.py": {"file_path": "b.py", "commit_count_90d": 2},
}
await persist_git_refresh(async_session, repo.id, {}, decay, None)
rows = await _rows(async_session, repo.id)
assert set(rows) == {"a.py"}
assert rows["a.py"].commit_count_90d == 3
assert rows["a.py"].commit_count_total == 5
assert rows["a.py"].primary_owner_name == "Alice"
async def test_update_gives_a_pre_tier_store_the_native_history_rows(
async_session, tmp_path, monkeypatch
) -> None:
monkeypatch.setenv("REPOWISE_GIT_WINDOW_ANCHOR", "head")
_build(tmp_path)
_summary, native = await GitIndexer(tmp_path, tier=GitIndexTier.FULL).index_repo("r")
native_history = {row["file_path"]: row for row in native if row["history_only"]}
assert native_history
# A store indexed before the history tier: code rows only.
repo = await insert_repo(async_session)
await upsert_git_metadata_bulk(
async_session, repo.id, [row for row in native if not row["history_only"]]
)
sink: dict[str, dict] = {}
changed = await GitIndexer(tmp_path, tier=GitIndexTier.FULL).index_changed_files(
["a.py"],
# The parsed-file set the update passes holds no markup or config.
all_files={"a.py", "b.py"},
co_change_sink={},
idle_decay_sink=sink,
)
await persist_git_refresh(
async_session, repo.id, {m["file_path"]: m for m in changed}, sink, None
)
rows = await _rows(async_session, repo.id)
assert set(rows) == {row["file_path"] for row in native}
for path, expected in native_history.items():
row = rows[path]
assert row.history_only is True
assert row.churn_percentile == 0.0 and row.is_hotspot is False
for key, value in expected.items():
stored = getattr(row, key)
if hasattr(value, "tzinfo"):
value, stored = value.replace(tzinfo=None), stored.replace(tzinfo=None)
assert stored == value, (path, key)