chore: add mypy typecheck baseline

Author: LuciferYang

.github/workflows/pr.yaml

@@ -39,6 +39,21 @@ jobs:
       - name: Run ruff check
         run: uv run ruff check
 
+  typecheck:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - name: Set up Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: '3.10'
+      - name: Install uv
+        uses: astral-sh/setup-uv@v3
+      - name: Install dev dependencies
+        run: uv sync --extra dev
+      - name: Run mypy
+        run: uv run mypy
+
   tests:
     runs-on: ubuntu-latest
     steps:

Makefile

@@ -5,6 +5,7 @@ help:
 	@echo "Targets:"
 	@echo "  help   Show this help message"
 	@echo "  lint   Run linter and format checker (ruff check + ruff format --check)"
+	@echo "  typecheck Run mypy on the current typed baseline"
 	@echo "  fix    Auto-fix lint issues and format code (ruff check --fix + ruff format)"
 	@echo "  lock   Update uv.lock lockfile"
 	@echo "  build       Install the project with dev and docs dependencies"
@@ -17,6 +18,10 @@ lint: lock
 	uv run ruff check
 	uv run ruff format --check .
 
+.PHONY: typecheck
+typecheck: lock
+	uv run mypy
+
 .PHONY: fix
 fix: lock
 	uv run ruff check --fix --unsafe-fixes
@@ -45,4 +50,4 @@ clean:
 
 .PHONY: test
 test: lock
-	uv run pytest -vv -s
+	uv run pytest -vv -s

lance_ray/pandas.py

@@ -3,7 +3,8 @@
 
 """Utility functions for lance-ray."""
 
-from typing import TYPE_CHECKING, Optional, Union
+from collections.abc import Mapping
+from typing import TYPE_CHECKING, Any, Optional, Union, cast
 
 import pyarrow as pa
 
@@ -12,13 +13,14 @@
 
 
 def pd_to_arrow(
-    df: Union[pa.Table, "pd.DataFrame", dict], schema: Optional[pa.Schema]
+    df: Union[pa.Table, "pd.DataFrame", Mapping[str, Any]],
+    schema: Optional[pa.Schema],
 ) -> pa.Table:
     """Convert a pandas DataFrame to pyarrow Table."""
     from lance.dependencies import _PANDAS_AVAILABLE
     from lance.dependencies import pandas as pd
 
-    if isinstance(df, dict):
+    if isinstance(df, Mapping):
         return pa.Table.from_pydict(df, schema=schema)
     elif _PANDAS_AVAILABLE and isinstance(df, pd.DataFrame):
         tbl = pa.Table.from_pandas(df, schema=schema)
@@ -31,4 +33,4 @@ def pd_to_arrow(
         return new_table
     elif isinstance(df, pa.Table) and df.num_rows > 0 and schema is not None:
         return df.cast(schema)
-    return df
+    return cast(pa.Table, df)

lance_ray/utils.py

@@ -155,7 +155,7 @@ def validate_uri_or_namespace(
 def _get_cached_namespace(
     namespace_impl: str,
     namespace_properties_tuple: Optional[tuple[tuple[str, str], ...]],
-):
+) -> Any:
     """Internal cached namespace loader. Use get_or_create_namespace() instead."""
     import lance_namespace as ln
 
@@ -168,7 +168,7 @@ def _get_cached_namespace(
 def get_or_create_namespace(
     namespace_impl: Optional[str],
     namespace_properties: Optional[dict[str, str]],
-):
+) -> Any | None:
     """Get or create a cached namespace client.
 
     This function loads a namespace client from cache or creates a new one.
@@ -197,7 +197,7 @@ def _create_storage_options_provider(
     namespace_impl: Optional[str],
     namespace_properties: Optional[dict[str, str]],
     table_id: Optional[list[str]],
-):
+) -> Any | None:
     """Create a LanceNamespaceStorageOptionsProvider (pylance 4.x only)."""
     if not has_namespace_params(namespace_impl, table_id):
         return None

pyproject.toml

@@ -35,6 +35,10 @@ dependencies = [
 
 [project.optional-dependencies]
 dev = [
+    "mypy>=1.10,<2",
+    "pandas-stubs",
+    "pathspec>=1.1.1",
+    "pyarrow-stubs",
     "pytest-asyncio>=1.0.0",
     "pytest-xdist>=3.6.0",
     "ruff>=0.8.0",
@@ -65,6 +69,16 @@ ignore = ["E501", "UP045"]
 [tool.ruff.format]
 quote-style = "double"
 
+[tool.mypy]
+python_version = "3.10"
+strict = true
+follow_imports = "skip"
+files = [
+    "lance_ray/pandas.py",
+    "lance_ray/pool.py",
+    "lance_ray/utils.py",
+]
+
 
 [tool.pytest.ini_options]
 testpaths = ["tests"]