codespine 1.0.1__tar.gz → 1.0.2__tar.gz

{codespine-1.0.1 → codespine-1.0.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: codespine
-Version: 1.0.1
+Version: 1.0.2
 Summary: Local Java code intelligence indexer backed by a graph database
 Author: CodeSpine contributors
 License: MIT License

{codespine-1.0.1 → codespine-1.0.2}/codespine/__init__.py

@@ -1,4 +1,4 @@
 """CodeSpine package."""
 
 __all__ = ["__version__"]
-__version__ = "1.0.1"
+__version__ = "1.0.2"

{codespine-1.0.1 → codespine-1.0.2}/codespine/db/duckdb_store.py

@@ -41,6 +41,17 @@ LOGGER = logging.getLogger(__name__)
 _VECTOR_DIM = SETTINGS.vector_dim  # 384
 
 
+def _remove_path(path: str) -> None:
+    """Remove *path* whether it is a file, symlink, or directory tree."""
+    try:
+        if os.path.isdir(path) and not os.path.islink(path):
+            shutil.rmtree(path)
+        elif os.path.exists(path) or os.path.islink(path):
+            os.remove(path)
+    except OSError as exc:
+        LOGGER.warning("Could not remove %s: %s", path, exc)
+
+
 # ---------------------------------------------------------------------------
 # Schema DDL
 # ---------------------------------------------------------------------------
@@ -206,36 +217,41 @@ class DuckDBStore:
         db_file = self._snapshot_path if read_only and snap_exists else self._db_path
 
         # ----------------------------------------------------------------
-        # Legacy KùzuDB migration guard
-        # KùzuDB stores its database as a *directory*; DuckDB uses a single
-        # file.  When CODESPINE_BACKEND is changed to "duckdb" an existing
-        # KùzuDB shard directory sits at the same path and DuckDB refuses to
-        # open it.  Detect this case, wipe the directory, and start fresh.
+        # Robust open: handle legacy KùzuDB artifacts at the target path.
+        # KùzuDB may leave a directory, a partial file, or a 0-byte sentinel
+        # at the same path DuckDB expects.  Rather than guessing the type,
+        # we attempt to connect and on any IOException we wipe whatever is
+        # there and retry once with a clean slate.
         # ----------------------------------------------------------------
-        for legacy_path in (self._db_path, self._snapshot_path):
-            if os.path.isdir(legacy_path):
+        os.makedirs(os.path.dirname(db_file) or ".", exist_ok=True)
+        for attempt in range(2):
+            # If read-only and the target file doesn't exist, we have nothing
+            # to read — use an in-memory DB so callers get [] instead of crash.
+            if read_only and not os.path.exists(db_file):
+                self._conn = duckdb.connect(":memory:")
+                self._ensure_schema()
+                return
+
+            try:
+                self._conn: duckdb.DuckDBPyConnection = duckdb.connect(
+                    db_file, read_only=read_only
+                )
+                break
+            except duckdb.IOException as exc:
+                if attempt > 0:
+                    raise  # second attempt also failed — give up
                 LOGGER.info(
-                    "Removing legacy KùzuDB directory at %s — "
-                    "re-index with 'codespine analyse' to rebuild.",
-                    legacy_path,
+                    "Cannot open DB at %s (%s) — removing stale artifact "
+                    "and starting fresh. Re-index with 'codespine analyse'.",
+                    db_file,
+                    exc,
                 )
-                shutil.rmtree(legacy_path)
+                _remove_path(db_file)
+                # If the bad path was the snapshot, fall back to the write DB.
+                if db_file == self._snapshot_path:
+                    db_file = self._db_path
+                    os.makedirs(os.path.dirname(db_file) or ".", exist_ok=True)
 
-        # Re-evaluate db_file after possible cleanup.
-        snap_exists = os.path.exists(self._snapshot_path)
-        db_file = self._snapshot_path if read_only and snap_exists else self._db_path
-
-        # When opening read-only and no snapshot exists yet, open the write
-        # DB in read-only mode so callers get an empty-but-valid store rather
-        # than an error.
-        if read_only and not snap_exists and not os.path.exists(self._db_path):
-            # No data at all — open an in-memory DB so queries return [] cleanly.
-            self._conn = duckdb.connect(":memory:")
-            self._ensure_schema()
-            return
-
-        os.makedirs(os.path.dirname(db_file) or ".", exist_ok=True)
-        self._conn: duckdb.DuckDBPyConnection = duckdb.connect(db_file, read_only=read_only)
         if not read_only:
             self._ensure_schema()
 

{codespine-1.0.1 → codespine-1.0.2}/codespine.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: codespine
-Version: 1.0.1
+Version: 1.0.2
 Summary: Local Java code intelligence indexer backed by a graph database
 Author: CodeSpine contributors
 License: MIT License

{codespine-1.0.1 → codespine-1.0.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "codespine"
-version = "1.0.1"
+version = "1.0.2"
 description = "Local Java code intelligence indexer backed by a graph database"
 readme = "README.md"
 requires-python = ">=3.10"

{codespine-1.0.1 → codespine-1.0.2}/tests/test_duckdb_store.py

@@ -399,3 +399,50 @@ def test_sharded_duckdb_multi_project_isolation(tmp_path: Path):
     all_ids = {p["id"] for p in sg.list_project_metadata()}
     assert pid_a in all_ids
     assert pid_b in all_ids
+
+
+# ---------------------------------------------------------------------------
+# Legacy KùzuDB artifact recovery
+# ---------------------------------------------------------------------------
+
+
+def test_legacy_kuzu_directory_at_db_path_is_removed(tmp_path: Path):
+    """DuckDBStore auto-removes a KùzuDB directory left at the DB path."""
+    db_path = str(tmp_path / "db")
+    snap_path = str(tmp_path / "db_read")
+
+    # Simulate a KùzuDB directory at the write path
+    os.makedirs(db_path)
+    (Path(db_path) / "nodes.index").write_bytes(b"\x00" * 16)
+
+    store = DuckDBStore(db_path_override=db_path, snapshot_path_override=snap_path)
+    rows = store.query_records("SELECT * FROM projects")
+    assert rows == []  # fresh empty DB, no crash
+
+
+def test_legacy_kuzu_file_at_snapshot_path_is_removed(tmp_path: Path):
+    """DuckDBStore auto-removes a non-DuckDB file left at the snapshot path."""
+    db_path = str(tmp_path / "db")
+    snap_path = str(tmp_path / "db_read")
+
+    # Write a KùzuDB-style snapshot *directory* at the snap path
+    os.makedirs(snap_path)
+    (Path(snap_path) / "catalog.json").write_bytes(b"{}")
+
+    # Open read-only — should clear the bad snapshot and open fresh
+    store = DuckDBStore(read_only=True, db_path_override=db_path, snapshot_path_override=snap_path)
+    rows = store.query_records("SELECT * FROM projects")
+    assert rows == []
+
+
+def test_corrupt_file_at_db_path_is_replaced(tmp_path: Path):
+    """DuckDBStore replaces a corrupt (non-DuckDB) regular file at the DB path."""
+    db_path = str(tmp_path / "db")
+    snap_path = str(tmp_path / "db_read")
+
+    # Write garbage that DuckDB cannot open
+    Path(db_path).write_bytes(b"NOT A DUCKDB FILE\x00\x01\x02")
+
+    store = DuckDBStore(db_path_override=db_path, snapshot_path_override=snap_path)
+    rows = store.query_records("SELECT * FROM projects")
+    assert rows == []