cleanup[2]

Signed-off-by: Anatoly Myachev <anatoly.myachev@intel.com>

.github/workflows/mkdocs.yml

@@ -19,14 +19,11 @@ jobs:
         with:
           python-version: 3.x
       - run: echo "cache_id=$(date --utc '+%V')" >> $GITHUB_ENV
-
-
       - uses: actions/cache@v3
         with:
           key: mkdocs-material-${{ env.cache_id }}
           path: .cache
           restore-keys: |
             mkdocs-material-
-      - run: pip install -r docs/requirements-docs.txt -e . pandas "polars<0.20.8" # modin[ray] @ git+https://github.com/modin-project/modin.git@abbbd03023f55d4ee9592c4c695308e10a97cb4c
-
+      - run: pip install -r docs/requirements-docs.txt -e . pandas "polars<0.20.8" # modin[ray]
       - run: mkdocs gh-deploy --force

.github/workflows/tox.yml

@@ -28,19 +28,27 @@ jobs:
           key: ${{ runner.os }}-build-${{ matrix.python-version }}
       - name: install-reqs
         run: python -m pip install --upgrade tox virtualenv setuptools pip -r requirements-dev.txt
-      - name: Run pytest
+      - name: Run pytest for pandas and polars
         run: |
           pytest tests --cov=dataframe_api_compat/pandas_standard --cov=tests --cov-append --cov-fail-under=50 --cov-report= --library pandas-numpy
           pytest tests --cov=dataframe_api_compat/pandas_standard --cov=tests --cov-append --cov-fail-under=50 --cov-report= --library pandas-nullable
-          pytest tests --cov=dataframe_api_compat/polars_standard --cov=tests --cov-append --cov-fail-under=100 --library polars-lazy
-      - name: install type-checking reqs
-        run: python -m pip install 'git+https://github.com/data-apis/dataframe-api.git#egg=dataframe_api&subdirectory=spec/API_specification' mypy typing-extensions
-      - name: run mypy
-        run: mypy dataframe_api_compat tests
+          pytest tests --cov=dataframe_api_compat/polars_standard --cov=tests --cov-append --cov-fail-under=95 --library polars-lazy
       - name: run polars integration tests
         run: pip uninstall pandas -y && pytest tests/integration/upstream_test.py::TestPolars && pip install -U pandas
       - name: run pandas integration tests
         run: pip uninstall polars -y && pytest tests/integration/upstream_test.py::TestPandas
+      - name: Update env for modin
+        run: |
+          python -m pip uninstall -r requirements-dev.txt
+          python -m pip install -r requirements-dev-modin.txt
+      - name: Run pytest for modin
+        run: |
+          pytest tests --cov=dataframe_api_compat/modin_standard --cov=tests --cov-append --cov-fail-under=100 --library modin
+      - name: install type-checking reqs
+        run: python -m pip install 'git+https://github.com/data-apis/dataframe-api.git#egg=dataframe_api&subdirectory=spec/API_specification' mypy typing-extensions
+      - name: run mypy
+        run: mypy dataframe_api_compat tests
+
 
   tox-modin:
     strategy:

dataframe_api_compat/modin_standard/dataframe_object.py

@@ -211,7 +211,8 @@ def rename(self, mapping: Mapping[str, str]) -> DataFrame:
             self.dataframe.rename(columns=mapping),
         )
 
-    def get_column_names(self) -> list[str]:
+    def get_column_names(self) -> list[str]:  # pragma: no cover
+        # TODO: add a test after modin implements `__dataframe_consortium_standard__`
         # DO NOT REMOVE
         # This one is used in upstream tests - even if deprecated,
         # just leave it in for backwards compatibility

pyproject.toml

@@ -82,6 +82,7 @@ line-length = 90
 filterwarnings = [
   "error",
   'ignore:distutils Version classes are deprecated:DeprecationWarning',
+  # modin specific
   'ignore:.*pkg_resources:DeprecationWarning',
   'ignore:Ray execution environment not yet initialized:UserWarning',
   "ignore:Distributing <class '.*'> object.:UserWarning",

requirements-dev-modin.txt

@@ -1,6 +1,5 @@
 covdefaults
-modin[ray] @ git+https://github.com/modin-project/modin.git@25d143ff3c6043104bda3c51a204eab626103eb4
-polars
+modin[ray]
 pre-commit
 pytest
 pytest-cov

tests/conftest.py

@@ -60,15 +60,28 @@ def pytest_generate_tests(metafunc: Any) -> None:
         metafunc.parametrize("library", lib_handlers, ids=libraries)
 
 
+ci_skip_ids = [
+    # polars does not allow to create a dataframe with non-unique columns
+    "non_unique_column_names.py::test_repeated_columns[polars-lazy]",
+]
+
+
 ci_xfail_ids = [
     # https://github.com/modin-project/modin/issues/7212
     "join_test.py::test_join_left[modin]",
     "join_test.py::test_join_two_keys[modin]",
     "persistedness_test.py::test_cross_df_propagation[modin]",
+    # https://github.com/modin-project/modin/issues/3602
+    "aggregate_test.py::test_aggregate[modin]",
+    "aggregate_test.py::test_aggregate_only_size[modin]",
 ]
 
 
-def pytest_collection_modifyitems(items: list[pytest.Item]) -> None:
+def pytest_collection_modifyitems(items: list[pytest.Item]) -> None:  # pragma: no cover
     for item in items:
         if any(id_ in item.nodeid for id_ in ci_xfail_ids):
             item.add_marker(pytest.mark.xfail(strict=True))
+        elif any(id_ in item.nodeid for id_ in ci_skip_ids):
+            item.add_marker(
+                pytest.mark.skip("does not make sense for a specific implementation"),
+            )

tests/groupby/aggregate_test.py

@@ -1,13 +1,9 @@
-import pytest
-
 from tests.utils import BaseHandler
 from tests.utils import compare_dataframe_with_reference
 from tests.utils import integer_dataframe_4
 
 
 def test_aggregate(library: BaseHandler) -> None:
-    if library.name == "modin":
-        pytest.skip("https://github.com/modin-project/modin/issues/3602")
     df = integer_dataframe_4(library)
     df = df.assign((df.col("b") > 0).rename("d"))
     ns = df.__dataframe_namespace__()
@@ -62,8 +58,6 @@ def test_aggregate(library: BaseHandler) -> None:
 
 
 def test_aggregate_only_size(library: BaseHandler) -> None:
-    if library.name == "modin":
-        pytest.skip("https://github.com/modin-project/modin/issues/3602")
     df = integer_dataframe_4(library)
     ns = df.__dataframe_namespace__()
     result = (

tests/utils.py

@@ -332,6 +332,8 @@ def temporal_dataframe_1(library: BaseHandler) -> DataFrame:
 
         df = pd.DataFrame(
             {
+                # the data for column "a" differs from other implementations due to pandas 1.5 compat
+                # https://github.com/data-apis/dataframe-api-compat/commit/aeca5cf1a052033b72388e3f87ad8b70d66cedec
                 "a": [
                     datetime(2020, 1, 1, 1, 2, 1, 123000),
                     datetime(2020, 1, 2, 3, 1, 2, 321000),