Fix broken cicd that has new mypy deps. (#256)

Context:
[mm](https://chat.kaggle.net/kaggle/pl/fcxzntt8efdijjp8kc8uyuoz6a)

**Root cause**
There seem to be 2 failures for cicd:
1. Updated `ruff` and other checking tools for non-top level imports
(fixed with inline comments.)
2. Our lint script `mypy --install-types --non-interactive src/kagglehub
tests` will need to install a new dep `scipy-stubs`, which is NOT
available for Python 3.9. This dependence is likely to be introduced in
[this PR](#196)
  - Long term solution: retire Python 3.9 since it EOL is 2025/Oct.
- Short term: added exceptions based on [this
doc](https://mypy.readthedocs.io/en/stable/running_mypy.html#missing-imports)

I propose we go for the short term solution to unblock, and keep track
the long term solution in other tasks.

Author: dolaameng

README.md

@@ -523,16 +523,16 @@ The following shows how to run `hatch run lint:all` but this also works for any
 ./docker-hatch run lint:all
 
 # Use specific Python version (Must be a valid tag from: https://hub.docker.com/_/python)
-./docker-hatch -v 3.9 run lint:all
+./docker-hatch -v 3.10 run lint:all
 
 # Run test in docker with specific Python version
-./docker-hatch -v 3.9 test
+./docker-hatch -v 3.10 test
 
 # Run python from specific environment (e.g. one with optional dependencies installed)
 ./docker-hatch run extra-deps-env:python -c "print('hello world')"
 
 # Run commands with other root-level hatch options (everything after -- gets passed to hatch)
-./docker-hatch -v 3.9 -- -v env create debug-env-with-verbose-logging
+./docker-hatch -v 3.10 -- -v env create debug-env-with-verbose-logging
 ```
 
 ## VS Code setup

pyproject.toml

@@ -84,7 +84,7 @@ dependencies = [
   "types-requests",
   "types-tqdm",
   "pandas_stubs",
-  "polars"
+  "polars",
 ]
 
 [tool.hatch.envs.lint.scripts]
@@ -237,3 +237,7 @@ exclude_lines = [
 
 [tool.mypy]
 mypy_path = "./stubs"
+
+[[tool.mypy.overrides]]
+module = ["yaml.*"]
+ignore_missing_imports = true

src/kagglehub/auth.py

@@ -51,7 +51,7 @@ def _is_in_notebook() -> bool:
     Adapted to make it work with Google colab as well.
     """
     try:
-        from IPython import get_ipython  # type: ignore
+        from IPython import get_ipython  # type: ignore # noqa: PLC0415
 
         shell_class = get_ipython().__class__
         for parent_class in shell_class.__mro__:  # e.g. "is subclass of"
@@ -66,8 +66,8 @@ def _notebook_login(validate_credentials: bool) -> None:  # noqa: FBT001
     """Prompt the user for their Kaggle token and save it in a widget (Jupyter or Colab)."""
     library_error = "You need the `ipywidgets` module: `pip install ipywidgets`."
     try:
-        from IPython.display import display  # type: ignore
-        from ipywidgets import widgets  # type: ignore
+        from IPython.display import display  # type: ignore # noqa: PLC0415
+        from ipywidgets import widgets  # type: ignore # noqa: PLC0415
     except ImportError:
         raise ImportError(library_error)  # noqa: B904
 

src/kagglehub/config.py

@@ -162,7 +162,7 @@ def _normalize_whitespace(s: str) -> str:
 def get_colab_credentials() -> Optional[KaggleApiCredentials]:
     # userdata access is already thread-safe after b/318732641
     try:
-        from google.colab import userdata  # type: ignore[import]
+        from google.colab import userdata  # type: ignore[import] # noqa: PLC0415
     except ImportError:
         return None
 

src/kagglehub/datasets.py

@@ -126,7 +126,7 @@ def dataset_load(
     polars_frame_type = polars_frame_type if polars_frame_type is not None else PolarsFrameType.LAZY_FRAME
     try:
         if adapter is KaggleDatasetAdapter.HUGGING_FACE:
-            import kagglehub.hf_datasets
+            import kagglehub.hf_datasets  # noqa: PLC0415
 
             return kagglehub.hf_datasets.load_hf_dataset(
                 handle,
@@ -136,13 +136,13 @@ def dataset_load(
                 hf_kwargs=hf_kwargs,
             )
         elif adapter is KaggleDatasetAdapter.PANDAS:
-            import kagglehub.pandas_datasets
+            import kagglehub.pandas_datasets  # noqa: PLC0415
 
             return kagglehub.pandas_datasets.load_pandas_dataset(
                 handle, path, pandas_kwargs=pandas_kwargs, sql_query=sql_query
             )
         elif adapter is KaggleDatasetAdapter.POLARS:
-            import kagglehub.polars_datasets
+            import kagglehub.polars_datasets  # noqa: PLC0415
 
             return kagglehub.polars_datasets.load_polars_dataset(
                 handle, path, polars_frame_type=polars_frame_type, polars_kwargs=polars_kwargs, sql_query=sql_query

src/kagglehub/signing.py

@@ -8,7 +8,7 @@
 
 
 def sign_with_sigstore(local_model_dir: str, handle: ModelHandle) -> bool:
-    from model_signing.signing import Config
+    from model_signing.signing import Config  # noqa: PLC0415
 
     try:
         token = signing_token(handle.owner, handle.model)

tools/cicd/Dockerfile

@@ -6,6 +6,8 @@ ARG PYTHON_VERSION
 
 FROM python:${PYTHON_VERSION}
 
+RUN python -m pip install --upgrade pip
+
 RUN python -m pip install hatch==1.14.0 twine==6.0.1
 
 ENTRYPOINT ["hatch"]