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
fix(doctor): leave tombstoned pages out of the store reconciliation (#3907)
Raghav Chamadiya1507ad459028
diff --git a/packages/cli/src/repowise/cli/commands/doctor_cmd/repo_checks.py b/packages/cli/src/repowise/cli/commands/doctor_cmd/repo_checks.pyindex 61990d381..a3a90574d 100644--- a/packages/cli/src/repowise/cli/commands/doctor_cmd/repo_checks.py+++ b/packages/cli/src/repowise/cli/commands/doctor_cmd/repo_checks.py@@ -141,9 +141,10 @@ async def _all_pages_for_reconciliation(session: object, repo_id: str) -> list:8900 of them. ``--repair`` deletes what this reports, so what it removedwas the live index.- Four columns rather than whole rows: the id to match against the stores,+ A few columns rather than whole rows: the id to match against the stores,the content and digest for the information floor, metadata_json for the stub- predicate, and target_path to tell whether the file is excluded. Nothing downstream reads any other field, and hydrating full+ predicate, target_path to tell whether the file is excluded, and+ freshness_status to leave tombstones out. Nothing downstream reads any other field, and hydrating fullORM objects for every page only to discard them is what made a cap looknecessary in the first place."""@@ -152,9 +153,14 @@ async def _all_pages_for_reconciliation(session: object, repo_id: str) -> list:from repowise.core.persistence.models import Pageresult = await session.execute( # type: ignore[attr-defined]- select(Page.id, Page.content, Page.digest, Page.metadata_json, Page.target_path).where(- Page.repository_id == repo_id- )+ select(+ Page.id,+ Page.content,+ Page.digest,+ Page.metadata_json,+ Page.target_path,+ Page.freshness_status,+ ).where(Page.repository_id == repo_id))return list(result.all())@@ -514,6 +520,12 @@ def _run_repo_checks(await engine.dispose()return set(), set(), set(), set(), 0, None, Falsepages = await _all_pages_for_reconciliation(session, repo.id)+ # A tombstone (its file is gone) is taken out of both+ # indexes by update, so it is never missing, and a store+ # row still holding one is an orphan `--repair` deletes.+ # Counting it as live sent `--repair` to re-index and+ # re-embed pages for deleted files.+ pages = [p for p in pages if p.freshness_status != "tombstone"]sql_ids = {p.id for p in pages}# ``Page``'s primary key is the column ``id``; there is no# ``page_id`` attribute. The old spelling raised here ondiff --git a/tests/unit/cli/test_doctor_tombstones.py b/tests/unit/cli/test_doctor_tombstones.pynew file mode 100644index 000000000..be41ef3fb--- /dev/null+++ b/tests/unit/cli/test_doctor_tombstones.py@@ -0,0 +1,99 @@+"""``doctor`` treats a tombstoned page as out of both indexes, as update does.++Update deletes a tombstone's full-text and vector rows, so its absence is the+correct state, and a row still holding one is an orphan. Counted as a live page+it read as missing, and ``--repair`` re-indexed and re-embedded pages for+deleted files.+"""++from __future__ import annotations++import asyncio+from pathlib import Path++from repowise.cli.commands.doctor_cmd import repo_checks++BODY = (+ "## Overview\n\n"+ "Resolves each import against the package manifest before descending, so "+ "a symlinked workspace member is visited once rather than once per alias "+ "that reaches it. Cycles break on the real path, never the alias.\n"+)+LIVE = "file_page:live.py"+GONE = "file_page:gone.py" # tombstone, already out of both indexes+LEFT = "file_page:left.py" # tombstone whose index rows were left behind+++async def _build_repo(tmp_path: Path) -> Path:+ import git as gitpython++ from repowise.core.persistence import (+ FullTextSearch,+ create_engine,+ create_session_factory,+ get_session,+ )+ from repowise.core.persistence.crud import upsert_page, upsert_repository+ from repowise.core.persistence.database import init_db+ from repowise.core.persistence.vector_store import LanceDBVectorStore+ from repowise.core.providers.embedding.base import MockEmbedder++ repo_path = (tmp_path / "repo").resolve()+ repo_path.mkdir()+ gitpython.Repo.init(repo_path)+ repowise_dir = repo_path / ".repowise"+ repowise_dir.mkdir()++ engine = create_engine(f"sqlite+aiosqlite:///{repowise_dir / 'wiki.db'}")+ await init_db(engine)+ sf = create_session_factory(engine)+ async with get_session(sf) as session:+ repo = await upsert_repository(+ session, name="repo", local_path=str(repo_path), url="https://example.test/repo"+ )+ for page_id, status in ((LIVE, "fresh"), (GONE, "tombstone"), (LEFT, "tombstone")):+ path = page_id.split(":", 1)[1]+ await upsert_page(+ session,+ page_id=page_id,+ repository_id=repo.id,+ page_type="file_page",+ title=f"File: {path}",+ content=BODY,+ summary="",+ target_path=path,+ source_hash="",+ model_name="mock",+ provider_name="mock",+ freshness_status=status,+ )+ await session.commit()++ fts = FullTextSearch(engine)+ await fts.ensure_index()+ store = LanceDBVectorStore(str(repowise_dir / "lancedb"), embedder=MockEmbedder())+ for page_id in (LIVE, LEFT):+ path = page_id.split(":", 1)[1]+ await fts.index(page_id, f"File: {path}", BODY, summary="", target_path=path)+ await store.embed_and_upsert(page_id, BODY, {"title": f"File: {path}"})+ await store.close()+ await engine.dispose()+ return repo_path+++def _rows(repo_path: Path, repair: bool = False) -> dict[str, tuple[bool, str]]:+ _all_ok, checks = repo_checks._run_repo_checks(repo_path, repair=repair)+ return {c.name: (c.ok, c.detail) for c in checks}+++def test_tombstones_are_never_missing_and_their_leftovers_are_orphans(tmp_path: Path) -> None:+ repo_path = asyncio.run(_build_repo(tmp_path))++ rows = _rows(repo_path)+ assert rows["SQL ↔ Vector Store"] == (False, "0 missing, 1 orphaned")+ assert rows["SQL ↔ FTS Index"] == (False, "0 missing, 1 orphaned")++ _rows(repo_path, repair=True)+ rows = _rows(repo_path)+ assert rows["SQL ↔ Vector Store"] == (True, "in sync")+ assert rows["SQL ↔ FTS Index"] == (True, "in sync")