ci: Test optional dependencies separately (#135)

Author: AndreasAlbertQC

.github/workflows/ci.yml

@@ -41,6 +41,14 @@ jobs:
       matrix:
         os: [ubuntu-latest, windows-latest]
         environment: [py310, py311, py312, py313]
+        with_optionals: [false]
+        include:
+          - os: ubuntu-latest
+            environment: py313-optionals
+            with_optionals: true
+          - os: windows-latest
+            environment: py313-optionals
+            with_optionals: true
     steps:
       - name: Checkout branch
         uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
@@ -51,7 +59,7 @@ jobs:
       - name: Install repository
         run: pixi run -e ${{ matrix.environment }} postinstall
       - name: Run pytest
-        run: pixi run -e ${{ matrix.environment }} test-coverage --color=yes
+        run: pixi run -e ${{ matrix.environment }} test-coverage --color=yes ${{ matrix.with_optionals && '-m with_optionals' || '-m "not with_optionals"'}} --cov=dataframely --cov-report=xml
       - name: Upload codecov
         uses: codecov/codecov-action@18283e04ce6e62d37312384ff67231eb8fd56d24 # v5.4.3
         with:

dataframely/_compat.py

@@ -18,6 +18,9 @@ def __getattr__(self, name: str) -> Any:
 try:
     import sqlalchemy as sa
     import sqlalchemy.dialects.mssql as sa_mssql
+    from sqlalchemy import Dialect
+    from sqlalchemy.dialects.mssql.pyodbc import MSDialect_pyodbc
+    from sqlalchemy.dialects.postgresql.psycopg2 import PGDialect_psycopg2
     from sqlalchemy.sql.type_api import TypeEngine as sa_TypeEngine
 except ImportError:  # pragma: no cover
     sa = _DummyModule("sqlalchemy")  # type: ignore
@@ -26,7 +29,14 @@ def __getattr__(self, name: str) -> Any:
     class sa_TypeEngine:  # type: ignore # noqa: N801
         pass
 
+    class MSDialect_pyodbc:  # type: ignore # noqa: N801
+        pass
+
+    class PGDialect_psycopg2:  # type: ignore # noqa: N801
+        pass
 
+    class Dialect:  # type: ignore # noqa: N801
+        pass
 # -------------------------------------- PYARROW ------------------------------------- #
 
 try:
@@ -36,4 +46,11 @@ class sa_TypeEngine:  # type: ignore # noqa: N801
 
 # ------------------------------------------------------------------------------------ #
 
-__all__ = ["sa", "sa_mssql", "sa_TypeEngine", "pa"]
+__all__ = [
+    "sa",
+    "sa_mssql",
+    "sa_TypeEngine",
+    "pa",
+    "MSDialect_pyodbc",
+    "PGDialect_psycopg2",
+]

pixi.lock

@@ -43,15 +43,18 @@ readthedocs = "rm -rf $READTHEDOCS_OUTPUT/html && cp -r docs/_build/html $READTH
 mypy = ">=1.13"
 pandas = "*"
 pandas-stubs = "*"
-pyarrow = "*"
 pyodbc = "*"
 pytest = ">=6"
 pytest-benchmark = "*"
 pytest-cov = "*"
 pytest-md = "*"
 scikit-learn = "*"
+
+[feature.optionals.dependencies]
+pyarrow = "*"
 sqlalchemy = ">=2"
 
+
 [feature.test.tasks]
 test = "pytest"
 test-bench = "pytest tests/benches --benchmark-only --benchmark-sort mean --benchmark-columns min,max,mean,rounds"
@@ -102,11 +105,12 @@ install-polars-nightly = "pip install --pre --no-deps --upgrade --only-binary :a
 
 [environments]
 build = ["build"]
-default = ["dev", "lint", "test"]
+default = ["dev", "lint", "test", "optionals"]
 docs = ["docs"]
 lint = { features = ["lint"], no-default-feature = true }
 nightly = ["nightly", "test"]
 py310 = ["py310", "test"]
 py311 = ["py311", "test"]
 py312 = ["py312", "test"]
 py313 = ["py313", "test"]
+py313-optionals = ["py313", "test", "optionals"]

pixi.toml

@@ -27,6 +27,10 @@ readme = "README.md"
 requires-python = ">=3.10"
 version = "0.0.0"
 
+[project.optional-dependencies]
+sqlalchemy = ["sqlalchemy"]
+pyarrow = ["pyarrow"]
+
 [project.urls]
 Documentation = "https://dataframely.readthedocs.io/"
 Repository = "https://github.com/quantco/dataframely"
@@ -94,6 +98,9 @@ filterwarnings = [
   "ignore:The 'nullable' argument was not explicitly set:FutureWarning",
 ]
 testpaths = ["tests"]
+markers = [
+  "with_optionals: Tests that require optional dependencies to be installed",
+]
 
 [tool.coverage.run]
 omit = ["dataframely/mypy.py", "dataframely/testing/generate.py"]

pyproject.toml

@@ -14,6 +14,8 @@
     create_schema,
 )
 
+pytestmark = pytest.mark.with_optionals
+
 
 @pytest.mark.parametrize("column_type", ALL_COLUMN_TYPES)
 def test_equal_to_polars_schema(column_type: type[Column]) -> None:

tests/columns/test_pyarrow.py

@@ -2,14 +2,14 @@
 # SPDX-License-Identifier: BSD-3-Clause
 
 import pytest
-import sqlalchemy as sa
-from sqlalchemy.dialects.mssql.pyodbc import MSDialect_pyodbc
-from sqlalchemy.dialects.postgresql.psycopg2 import PGDialect_psycopg2
 
 import dataframely as dy
+from dataframely._compat import Dialect, MSDialect_pyodbc, PGDialect_psycopg2
 from dataframely.columns import Column
 from dataframely.testing import COLUMN_TYPES, create_schema
 
+pytestmark = pytest.mark.with_optionals
+
 
 @pytest.mark.parametrize(
     ("column", "datatype"),
@@ -107,7 +107,7 @@ def test_postgres_datatype(column: Column, datatype: str) -> None:
 @pytest.mark.parametrize("nullable", [True, False])
 @pytest.mark.parametrize("dialect", [MSDialect_pyodbc()])
 def test_sql_nullability(
-    column_type: type[Column], nullable: bool, dialect: sa.Dialect
+    column_type: type[Column], nullable: bool, dialect: Dialect
 ) -> None:
     schema = create_schema("test", {"a": column_type(nullable=nullable)})
     columns = schema.sql_schema(dialect)
@@ -119,7 +119,7 @@ def test_sql_nullability(
 @pytest.mark.parametrize("primary_key", [True, False])
 @pytest.mark.parametrize("dialect", [MSDialect_pyodbc(), PGDialect_psycopg2()])
 def test_sql_primary_key(
-    column_type: type[Column], primary_key: bool, dialect: sa.Dialect
+    column_type: type[Column], primary_key: bool, dialect: Dialect
 ) -> None:
     schema = create_schema("test", {"a": column_type(primary_key=primary_key)})
     columns = schema.sql_schema(dialect)
@@ -129,37 +129,37 @@ def test_sql_primary_key(
 
 
 @pytest.mark.parametrize("dialect", [MSDialect_pyodbc(), PGDialect_psycopg2()])
-def test_sql_multiple_columns(dialect: sa.Dialect) -> None:
+def test_sql_multiple_columns(dialect: Dialect) -> None:
     schema = create_schema("test", {"a": dy.Int32(nullable=False), "b": dy.Integer()})
     assert len(schema.sql_schema(dialect)) == 2
 
 
 @pytest.mark.parametrize("dialect", [MSDialect_pyodbc(), PGDialect_psycopg2()])
-def test_raise_for_list_column(dialect: sa.Dialect) -> None:
+def test_raise_for_list_column(dialect: Dialect) -> None:
     with pytest.raises(
         NotImplementedError, match="SQL column cannot have 'List' type."
     ):
         dy.List(dy.String()).sqlalchemy_dtype(dialect)
 
 
 @pytest.mark.parametrize("dialect", [MSDialect_pyodbc(), PGDialect_psycopg2()])
-def test_raise_for_array_column(dialect: sa.Dialect) -> None:
+def test_raise_for_array_column(dialect: Dialect) -> None:
     with pytest.raises(
         NotImplementedError, match="SQL column cannot have 'Array' type."
     ):
         dy.Array(dy.String(), 1).sqlalchemy_dtype(dialect)
 
 
 @pytest.mark.parametrize("dialect", [MSDialect_pyodbc(), PGDialect_psycopg2()])
-def test_raise_for_struct_column(dialect: sa.Dialect) -> None:
+def test_raise_for_struct_column(dialect: Dialect) -> None:
     with pytest.raises(
         NotImplementedError, match="SQL column cannot have 'Struct' type."
     ):
         dy.Struct({"a": dy.String()}).sqlalchemy_dtype(dialect)
 
 
 @pytest.mark.parametrize("dialect", [MSDialect_pyodbc(), PGDialect_psycopg2()])
-def test_raise_for_object_column(dialect: sa.Dialect) -> None:
+def test_raise_for_object_column(dialect: Dialect) -> None:
     with pytest.raises(
         NotImplementedError, match="SQL column cannot have 'Object' type."
     ):