dataframely 1.0.0__tar.gz → 1.1.0__tar.gz

{dataframely-1.0.0 → dataframely-1.1.0}/.github/workflows/build.yml

@@ -15,7 +15,7 @@ jobs:
         with:
           fetch-depth: 0
       - name: Set up pixi
-        uses: prefix-dev/setup-pixi@0f64e482e3d251f735019b1bc7fb0413ead75b2c # v0.8.2
+        uses: prefix-dev/setup-pixi@19eac09b398e3d0c747adc7921926a6d802df4da # v0.8.8
         with:
           environments: build
       - name: Set version
@@ -23,7 +23,7 @@ jobs:
       - name: Build project
         run: pixi run -e build build-sdist
       - name: Upload package
-        uses: actions/upload-artifact@65c4c4a1ddee5b72f698fdd19549f0f0fb45cf08 # v4.6.0
+        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
         with:
           name: sdist
           path: dist/*
@@ -50,7 +50,7 @@ jobs:
         with:
           fetch-depth: 0
       - name: Set up pixi
-        uses: prefix-dev/setup-pixi@0f64e482e3d251f735019b1bc7fb0413ead75b2c # v0.8.2
+        uses: prefix-dev/setup-pixi@19eac09b398e3d0c747adc7921926a6d802df4da # v0.8.8
         with:
           environments: build
       - name: Set version
@@ -64,7 +64,7 @@ jobs:
       - name: Check package
         run: pixi run -e build check-wheel
       - name: Upload package
-        uses: actions/upload-artifact@65c4c4a1ddee5b72f698fdd19549f0f0fb45cf08 # v4.6.0
+        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
         with:
           name: wheel-${{ matrix.target-platform }}
           path: dist/*
@@ -78,7 +78,7 @@ jobs:
       id-token: write
     environment: pypi
     steps:
-      - uses: actions/download-artifact@fa0a91b85d4f404e444e00e005971372dc801d16 # v4.1.8
+      - uses: actions/download-artifact@95815c38cf2ff2164869cbab79da8d1f422bc89e # v4.2.1
         with:
           path: dist
           merge-multiple: true

{dataframely-1.0.0 → dataframely-1.1.0}/.github/workflows/chore.yml

@@ -1,6 +1,6 @@
 name: Chore
 on:
-  pull_request:
+  pull_request_target:
     branches: [main]
     types: [opened, reopened, edited, synchronize]
   push:
@@ -13,7 +13,7 @@ concurrency:
 jobs:
   check-pr-title:
     name: Check PR Title
-    if: github.event_name == 'pull_request'
+    if: github.event_name == 'pull_request_target'
     runs-on: ubuntu-latest
     permissions:
       contents: read
@@ -51,16 +51,16 @@ jobs:
           delete: true
 
   release-drafter:
-    name: ${{ github.event_name == 'pull_request' && 'Assign Labels' || 'Draft Release' }}
+    name: ${{ github.event_name == 'pull_request_target' && 'Assign Labels' || 'Draft Release' }}
     runs-on: ubuntu-latest
     permissions:
       contents: write
       pull-requests: write
     steps:
-      - name: ${{ github.event_name == 'pull_request' && 'Assign labels' || 'Update release draft' }}
+      - name: ${{ github.event_name == 'pull_request_target' && 'Assign labels' || 'Update release draft' }}
         uses: release-drafter/release-drafter@v6
         with:
-          disable-releaser: ${{ github.event_name == 'pull_request' }}
+          disable-releaser: ${{ github.event_name == 'pull_request_target' }}
           disable-autolabeler: ${{ github.event_name == 'push' }}
         env:
           GITHUB_TOKEN: ${{ github.token }}

{dataframely-1.0.0 → dataframely-1.1.0}/.github/workflows/ci.yml

@@ -21,7 +21,7 @@ jobs:
           # needed for 'pre-commit-mirrors-insert-license'
           fetch-depth: 0
       - name: Set up pixi
-        uses: prefix-dev/setup-pixi@0f64e482e3d251f735019b1bc7fb0413ead75b2c # v0.8.2
+        uses: prefix-dev/setup-pixi@19eac09b398e3d0c747adc7921926a6d802df4da # v0.8.8
         with:
           environments: default lint
       - name: Install repository
@@ -42,7 +42,7 @@ jobs:
       - name: Checkout branch
         uses: actions/checkout@v4
       - name: Set up pixi
-        uses: prefix-dev/setup-pixi@8eaba7c61d661f73d558b0b477156b7b62667fa4
+        uses: prefix-dev/setup-pixi@19eac09b398e3d0c747adc7921926a6d802df4da # v0.8.8
         with:
           environments: ${{ matrix.environment }}
       - name: Install repository
@@ -53,3 +53,4 @@ jobs:
         uses: codecov/codecov-action@v5
         with:
           files: ./coverage.xml
+          token: ${{ secrets.CODECOV_TOKEN }}

{dataframely-1.0.0 → dataframely-1.1.0}/.github/workflows/update-lockfiles.yml

@@ -15,13 +15,13 @@ jobs:
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
       - name: Set up pixi
-        uses: prefix-dev/setup-pixi@0f64e482e3d251f735019b1bc7fb0413ead75b2c # v0.8.2
+        uses: prefix-dev/setup-pixi@19eac09b398e3d0c747adc7921926a6d802df4da # v0.8.8
         with:
           run-install: false
       - name: Update lockfiles
         run: pixi update --json --no-install | pixi exec pixi-diff-to-markdown >> diff.md
       - name: Create pull request
-        uses: peter-evans/create-pull-request@67ccf781d68cd99b580ae25a5c18a1cc84ffff1f # v7.0.6
+        uses: peter-evans/create-pull-request@271a8d0340265f705b14b6d32b9829c1cb33d45e # v7.0.8
         with:
           token: ${{ github.token }}
           commit-message: Update pixi lockfile

{dataframely-1.0.0 → dataframely-1.1.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dataframely
-Version: 1.0.0
+Version: 1.1.0
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
@@ -12,7 +12,8 @@ Summary: A declarative, polars-native data frame validation library
 Author-email: Andreas Albert <andreas.albert@quantco.com>, Daniel Elsner <daniel.elsner@quantco.com>, Oliver Borchert <oliver.borchert@quantco.com>
 Requires-Python: >=3.11
 Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
-Project-URL: repository, https://github.com/quantco/dataframely
+Project-URL: Repository, https://github.com/quantco/dataframely
+Project-URL: Documentation, https://dataframely.readthedocs.io/
 
 <!-- LOGO -->
 <br />

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/_base_collection.py

@@ -48,6 +48,12 @@ class CollectionMember:
 
     #: Whether the member should be ignored in the filter method.
     ignored_in_filters: bool = False
+    #: Whether the member's non-primary key columns should be inlined for sampling.
+    #: This means that value overrides are supplied on the top-level rather than in
+    #: a subkey with the member's name. Only valid if the member's primary key matches
+    #: the collection's common primary key. Two members that share common column names
+    #: may not both be inlined for sampling.
+    inline_for_sampling: bool = False
 
 
 # --------------------------------------- UTILS -------------------------------------- #
@@ -79,7 +85,7 @@ class Metadata:
     members: dict[str, MemberInfo] = field(default_factory=dict)
     filters: dict[str, Filter] = field(default_factory=dict)
 
-    def update(self, other: Self):
+    def update(self, other: Self) -> None:
         self.members.update(other.members)
         self.filters.update(other.filters)
 
@@ -92,7 +98,7 @@ class CollectionMeta(ABCMeta):
         namespace: dict[str, Any],
         *args: Any,
         **kwargs: Any,
-    ):
+    ) -> CollectionMeta:
         result = Metadata()
         for base in bases:
             result.update(mcs._get_metadata_recursively(base))
@@ -136,6 +142,30 @@ class CollectionMeta(ABCMeta):
                     f"{len(intersection)} such filters: {sorted(intersection)}."
                 )
 
+        # 3) Check that inlining for sampling is configured correctly.
+        if len(non_ignored_member_schemas) > 0:
+            common_primary_keys = _common_primary_keys(non_ignored_member_schemas)
+            inlined_columns: set[str] = set()
+            for member, info in result.members.items():
+                if info.inline_for_sampling:
+                    if set(info.schema.primary_keys()) != common_primary_keys:
+                        raise ImplementationError(
+                            f"Member '{member}' is inlined for sampling but its primary "
+                            "key is a superset of the common primary key. Such a member "
+                            "must not be inlined to be able to provide multiple values "
+                            "for a single combination of the common primary key."
+                        )
+                    non_primary_key_columns = (
+                        set(info.schema.column_names()) - common_primary_keys
+                    )
+                    if len(inlined_columns & non_primary_key_columns):
+                        raise ImplementationError(
+                            f"At least one column name of member '{member}' clashes "
+                            "with a column name of another member that is inlined for "
+                            "sampling."
+                        )
+                    inlined_columns.update(non_primary_key_columns)
+
         return super().__new__(mcs, name, bases, namespace, *args, **kwargs)
 
     @staticmethod
@@ -201,6 +231,7 @@ class CollectionMeta(ABCMeta):
                         schema=get_args(kls)[0],
                         is_optional=False,
                         ignored_in_filters=collection_member.ignored_in_filters,
+                        inline_for_sampling=collection_member.inline_for_sampling,
                     )
                 else:
                     # Some other unknown annotation

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/_base_schema.py

@@ -58,7 +58,7 @@ class Metadata:
     columns: dict[str, Column] = field(default_factory=dict)
     rules: dict[str, Rule] = field(default_factory=dict)
 
-    def update(self, other: Self):
+    def update(self, other: Self) -> None:
         self.columns.update(other.columns)
         self.rules.update(other.rules)
 
@@ -71,7 +71,7 @@ class SchemaMeta(ABCMeta):
         namespace: dict[str, Any],
         *args: Any,
         **kwargs: Any,
-    ):
+    ) -> SchemaMeta:
         result = Metadata()
         for base in bases:
             result.update(mcs._get_metadata_recursively(base))

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/_compat.py

@@ -6,7 +6,7 @@ from typing import Any
 
 
 class _DummyModule:  # pragma: no cover
-    def __init__(self, module: str):
+    def __init__(self, module: str) -> None:
         self.module = module
 
     def __getattr__(self, name: str) -> Any:

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/_filter.py

@@ -12,7 +12,7 @@ C = TypeVar("C")
 class Filter(Generic[C]):
     """Internal class representing logic for filtering members of a collection."""
 
-    def __init__(self, logic: Callable[[C], pl.LazyFrame]):
+    def __init__(self, logic: Callable[[C], pl.LazyFrame]) -> None:
         self.logic = logic
 
 

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/_rule.py

@@ -12,14 +12,14 @@ ValidationFunction = Callable[[], pl.Expr]
 class Rule:
     """Internal class representing validation rules."""
 
-    def __init__(self, expr: pl.Expr):
+    def __init__(self, expr: pl.Expr) -> None:
         self.expr = expr
 
 
 class GroupRule(Rule):
     """Rule that is evaluated on a group of columns."""
 
-    def __init__(self, expr: pl.Expr, group_columns: list[str]):
+    def __init__(self, expr: pl.Expr, group_columns: list[str]) -> None:
         super().__init__(expr)
         self.group_columns = group_columns
 

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/collection.py

@@ -123,7 +123,10 @@ class Collection(BaseCollection, ABC, Generic[SamplingType]):
                         ...
                     }
 
-                _Any_ member/value can be left out and will be sampled automatically.
+                *Any* member/value can be left out and will be sampled automatically.
+                Note that overrides for columns of members that are annotated with
+                ``inline_for_sampling=True`` can be supplied on the top-level instead
+                of in a nested dictionary.
             generator: The (seeded) generator to use for sampling data. If ``None``, a
                 generator with random seed is automatically created.
 
@@ -198,7 +201,11 @@ class Collection(BaseCollection, ABC, Generic[SamplingType]):
                             else _extract_keys_if_exist(sample, primary_keys)
                         ),
                         **_extract_keys_if_exist(
-                            sample[member] if member in sample else {},
+                            (
+                                sample
+                                if member_infos[member].inline_for_sampling
+                                else (sample[member] if member in sample else {})
+                            ),
                             schema.column_names(),
                         ),
                     }
@@ -498,7 +505,7 @@ class Collection(BaseCollection, ABC, Generic[SamplingType]):
 
     # ---------------------------------- PERSISTENCE --------------------------------- #
 
-    def write_parquet(self, directory: Path):
+    def write_parquet(self, directory: Path) -> None:
         """Write the members of this collection to Parquet files in a directory.
 
         This method writes one Parquet file per member into the provided directory.
@@ -590,7 +597,7 @@ class Collection(BaseCollection, ABC, Generic[SamplingType]):
         return out
 
     @classmethod
-    def _validate_input_keys(cls, data: Mapping[str, FrameType], /):
+    def _validate_input_keys(cls, data: Mapping[str, FrameType], /) -> None:
         actual = set(data)
 
         missing = cls.required_members() - actual

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/columns/_mixins.py

@@ -83,7 +83,7 @@ U = TypeVar("U")
 class IsInMixin(Generic[U], Base):
     """Mixin to use for types implementing "is in"."""
 
-    def __init__(self, *, is_in: Sequence[U] | None = None, **kwargs: Any):
+    def __init__(self, *, is_in: Sequence[U] | None = None, **kwargs: Any) -> None:
         super().__init__(**kwargs)
         self.is_in = is_in
 

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/columns/decimal.py

@@ -148,7 +148,9 @@ class Decimal(OrdinalMixin[decimal.Decimal], Column):
 # --------------------------------------- UTILS -------------------------------------- #
 
 
-def _validate(value: decimal.Decimal, precision: int | None, scale: int, name: str):
+def _validate(
+    value: decimal.Decimal, precision: int | None, scale: int, name: str
+) -> None:
     exponent = value.as_tuple().exponent
     if not isinstance(exponent, int):
         raise ValueError(f"Encountered 'inf' or 'NaN' for `{name}`.")

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/config.py

@@ -25,23 +25,23 @@ class Config(contextlib.ContextDecorator):
     #: Singleton stack to track where to go back after exiting a context.
     _stack: list[Options] = []
 
-    def __init__(self, **options: Unpack[Options]):
+    def __init__(self, **options: Unpack[Options]) -> None:
         self._local_options: Options = {**default_options(), **options}
 
     @staticmethod
-    def set_max_sampling_iterations(iterations: int):
+    def set_max_sampling_iterations(iterations: int) -> None:
         """Set the maximum number of sampling iterations to use on
         :meth:`Schema.sample`."""
         Config.options["max_sampling_iterations"] = iterations
 
     @staticmethod
-    def restore_defaults():
+    def restore_defaults() -> None:
         """Restore the defaults of the configuration."""
         Config.options = default_options()
 
     # ------------------------------------ CONTEXT ----------------------------------- #
 
-    def __enter__(self):
+    def __enter__(self) -> None:
         Config._stack.append(Config.options)
         Config.options = self._local_options
 
@@ -50,5 +50,5 @@ class Config(contextlib.ContextDecorator):
         exc_type: type[BaseException] | None,
         exc_val: BaseException | None,
         exc_tb: TracebackType | None,
-    ):
+    ) -> None:
         Config.options = Config._stack.pop()

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/exc.py

@@ -11,7 +11,7 @@ from ._polars import PolarsDataType
 class ValidationError(Exception):
     """Error raised when :mod:`dataframely` validation encounters an issue."""
 
-    def __init__(self, message: str):
+    def __init__(self, message: str) -> None:
         super().__init__()
         self.message = message
 
@@ -22,7 +22,9 @@ class ValidationError(Exception):
 class DtypeValidationError(ValidationError):
     """Validation error raised when column dtypes are wrong."""
 
-    def __init__(self, errors: dict[str, tuple[PolarsDataType, PolarsDataType]]):
+    def __init__(
+        self, errors: dict[str, tuple[PolarsDataType, PolarsDataType]]
+    ) -> None:
         super().__init__(f"{len(errors)} columns have an invalid dtype")
         self.errors = errors
 
@@ -37,7 +39,7 @@ class DtypeValidationError(ValidationError):
 class RuleValidationError(ValidationError):
     """Complex validation error raised when rule validation fails."""
 
-    def __init__(self, errors: dict[str, int]):
+    def __init__(self, errors: dict[str, int]) -> None:
         super().__init__(f"{len(errors)} rules failed validation")
 
         # Split into schema errors and column errors
@@ -75,11 +77,11 @@ class RuleValidationError(ValidationError):
 class MemberValidationError(ValidationError):
     """Validation error raised when multiple members of a collection fail validation."""
 
-    def __init__(self, errors: dict[str, ValidationError]):
+    def __init__(self, errors: dict[str, ValidationError]) -> None:
         super().__init__(f"{len(errors)} members failed validation")
         self.errors = errors
 
-    def __str__(self):
+    def __str__(self) -> str:
         details = [
             f" > Member '{name}' failed validation:\n"
             + "\n".join("   " + line for line in str(error).split("\n"))
@@ -95,7 +97,7 @@ class ImplementationError(Exception):
 class AnnotationImplementationError(ImplementationError):
     """Error raised when the annotations of a collection are invalid."""
 
-    def __init__(self, attr: str, kls: type):
+    def __init__(self, attr: str, kls: type) -> None:
         message = (
             "Annotations of a 'dy.Collection' may only be an (optional) "
             f"'dy.LazyFrame', but \"{attr}\" has type '{kls}'."
@@ -106,7 +108,9 @@ class AnnotationImplementationError(ImplementationError):
 class RuleImplementationError(ImplementationError):
     """Error raised when a rule is implemented incorrectly."""
 
-    def __init__(self, name: str, return_dtype: pl.DataType, is_group_rule: bool):
+    def __init__(
+        self, name: str, return_dtype: pl.DataType, is_group_rule: bool
+    ) -> None:
         if is_group_rule:
             details = (
                 " When implementing a group rule (i.e. when using the `group_by` "

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/failure.py

@@ -28,7 +28,9 @@ class FailureInfo(Generic[S]):
     #: The schema used to create the input data frame.
     schema: type[S]
 
-    def __init__(self, lf: pl.LazyFrame, rule_columns: list[str], schema: type[S]):
+    def __init__(
+        self, lf: pl.LazyFrame, rule_columns: list[str], schema: type[S]
+    ) -> None:
         self._lf = lf
         self._rule_columns = rule_columns
         self.schema = schema
@@ -71,7 +73,7 @@ class FailureInfo(Generic[S]):
 
     # ---------------------------------- PERSISTENCE --------------------------------- #
 
-    def write_parquet(self, file: str | Path | IO[bytes]):
+    def write_parquet(self, file: str | Path | IO[bytes]) -> None:
         """Write the failure info to a Parquet file.
 
         Args:

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/mypy.py

@@ -50,7 +50,7 @@ TYPED_LAZYFRAME_FULLNAME = "dataframely._typing.LazyFrame"
 # --------------------------------------- RULES -------------------------------------- #
 
 
-def mark_rules_as_staticmethod(ctx: ClassDefContext):
+def mark_rules_as_staticmethod(ctx: ClassDefContext) -> None:
     """Mark all methods decorated with `@rule` as `staticmethod`s."""
     info = ctx.cls.info
     for sym in info.names.values():
@@ -199,7 +199,7 @@ def _convert_dy_column_to_dtype(
 def store_typed_dict_type_for_schema(
     ctx: ClassDefContext,
     schema_registry: dict[str, TypedDictType],
-):
+) -> None:
     """Add `TypedDictType` inferred from the schema's columns to a given registry."""
 
     schema_type = ctx.cls.info
@@ -336,11 +336,13 @@ def alter_dataframe_iter_rows_return_type(
 
 
 class DataframelyPlugin(Plugin):
-    def __init__(self, options: Options):
+    def __init__(self, options: Options) -> None:
         super().__init__(options)
         self.schema_registry: dict[str, TypedDictType] = {}
 
-    def get_base_class_hook(self, fullname: str):
+    def get_base_class_hook(
+        self, fullname: str
+    ) -> Callable[[ClassDefContext], None] | None:
         # Given a class, check whether it is a subclass of `dy.Schema`. If so, mark
         # all methods decorated with `@rule` as staticmethods.
         # Also, store the `TypedDictType` for the schema in a registry to allow downstream
@@ -349,13 +351,12 @@ class DataframelyPlugin(Plugin):
         if sym and isinstance(sym.node, TypeInfo):
             if any(base.fullname == SCHEMA_FULLNAME for base in sym.node.mro):
 
-                def _hook(ctx: ClassDefContext) -> bool:
+                def _hook(ctx: ClassDefContext) -> None:
                     mark_rules_as_staticmethod(ctx)
                     store_typed_dict_type_for_schema(
                         ctx,
                         self.schema_registry,
                     )
-                    return True
 
                 return _hook
         return None

{dataframely-1.0.0 → dataframely-1.1.0}/dataframely/random.py

@@ -33,7 +33,7 @@ class Generator:
     seeding.
     """
 
-    def __init__(self, seed: int | None = None):
+    def __init__(self, seed: int | None = None) -> None:
         """
         Args:
             seed: The seed to use for initializing the random number generator used

{dataframely-1.0.0 → dataframely-1.1.0}/docs/conf.py

@@ -68,7 +68,7 @@ html_favicon = "_static/favicon.ico"
 # Copied and adapted from
 # https://github.com/pandas-dev/pandas/blob/4a14d064187367cacab3ff4652a12a0e45d0711b/doc/source/conf.py#L613-L659
 # Required configuration function to use sphinx.ext.linkcode
-def linkcode_resolve(domain, info):
+def linkcode_resolve(domain: str, info: dict[str, str]) -> str | None:
     """Determine the URL corresponding to a given Python object."""
     if domain != "py":
         return None

{dataframely-1.0.0 → dataframely-1.1.0}/docs/sites/quickstart.rst

@@ -162,7 +162,7 @@ expectations on the schema of the data frame, e.g.:
 
 ::
 
-    def train_model(df: dy.DataFrame[HouseSchema]):
+    def train_model(df: dy.DataFrame[HouseSchema]) -> None:
         ...
 
 The type checker (typically ``mypy``) then ensures that it is actually a

{dataframely-1.0.0 → dataframely-1.1.0}/pyproject.toml

@@ -19,10 +19,11 @@ description = "A declarative, polars-native data frame validation library"
 name = "dataframely"
 readme = "README.md"
 requires-python = ">=3.11"
-version = "1.0.0"
+version = "1.1.0"
 
 [project.urls]
-repository = "https://github.com/quantco/dataframely"
+Repository = "https://github.com/quantco/dataframely"
+Documentation = "https://dataframely.readthedocs.io/"
 
 [tool.maturin]
 module-name = "dataframely._extre"
@@ -66,6 +67,7 @@ quote-style = "double"
 
 [tool.mypy]
 check_untyped_defs = true
+disallow_untyped_defs = true
 exclude = ["docs/"]
 explicit_package_bases = true
 no_implicit_optional = true

{dataframely-1.0.0 → dataframely-1.1.0}/tests/collection/test_base.py

@@ -25,32 +25,32 @@ class MyCollection(dy.Collection):
     second: dy.LazyFrame[MySecondSchema] | None
 
 
-def test_common_primary_keys():
+def test_common_primary_keys() -> None:
     assert MyCollection.common_primary_keys() == ["a"]
 
 
-def test_members():
+def test_members() -> None:
     members = MyCollection.members()
     assert not members["first"].is_optional
     assert members["second"].is_optional
 
 
-def test_member_schemas():
+def test_member_schemas() -> None:
     schemas = MyCollection.member_schemas()
     assert schemas == {"first": MyFirstSchema, "second": MySecondSchema}
 
 
-def test_required_members():
+def test_required_members() -> None:
     required_members = MyCollection.required_members()
     assert required_members == {"first"}
 
 
-def test_optional_members():
+def test_optional_members() -> None:
     optional_members = MyCollection.optional_members()
     assert optional_members == {"second"}
 
 
-def test_cast():
+def test_cast() -> None:
     collection = MyCollection.cast(
         {
             "first": pl.LazyFrame({"a": [1, 2, 3]}),
@@ -74,7 +74,7 @@ def test_cast():
         {"first": pl.LazyFrame({"a": [1, 2, 3]}, schema={"a": pl.UInt8})},
     ],
 )
-def test_to_dict(expected: dict[str, pl.LazyFrame]):
+def test_to_dict(expected: dict[str, pl.LazyFrame]) -> None:
     collection = MyCollection.validate(expected)
 
     # Check that export looks as expected
@@ -87,7 +87,7 @@ def test_to_dict(expected: dict[str, pl.LazyFrame]):
     assert MyCollection.is_valid(observed)
 
 
-def test_collect_all():
+def test_collect_all() -> None:
     collection = MyCollection.cast(
         {
             "first": pl.LazyFrame({"a": [1, 2, 3]}).filter(pl.col("a") < 3),
@@ -106,7 +106,7 @@ def test_collect_all():
     assert len(out.second.collect()) == 2
 
 
-def test_collect_all_optional():
+def test_collect_all_optional() -> None:
     collection = MyCollection.cast({"first": pl.LazyFrame({"a": [1, 2, 3]})})
     out = collection.collect_all()
 
@@ -118,7 +118,9 @@ def test_collect_all_optional():
 @pytest.mark.parametrize(
     "read_fn", [MyCollection.scan_parquet, MyCollection.read_parquet]
 )
-def test_read_write_parquet(tmp_path: Path, read_fn: Callable[[Path], MyCollection]):
+def test_read_write_parquet(
+    tmp_path: Path, read_fn: Callable[[Path], MyCollection]
+) -> None:
     collection = MyCollection.cast(
         {
             "first": pl.LazyFrame({"a": [1, 2, 3]}),
@@ -139,7 +141,7 @@ def test_read_write_parquet(tmp_path: Path, read_fn: Callable[[Path], MyCollecti
 )
 def test_read_write_parquet_optional(
     tmp_path: Path, read_fn: Callable[[Path], MyCollection]
-):
+) -> None:
     collection = MyCollection.cast({"first": pl.LazyFrame({"a": [1, 2, 3]})})
     collection.write_parquet(tmp_path)