Merge branch 'main' into refactor/exists-ok-to-overwrite

Author: d-v-b

.pre-commit-config.yaml

@@ -1,10 +1,9 @@
 ci:
   autoupdate_commit_msg: "chore: update pre-commit hooks"
+  autoupdate_schedule: "monthly"
   autofix_commit_msg: "style: pre-commit fixes"
   autofix_prs: false
 default_stages: [pre-commit, pre-push]
-default_language_version:
-  python: python3
 repos:
   - repo: https://github.com/astral-sh/ruff-pre-commit
     rev: v0.8.2
@@ -16,7 +15,7 @@ repos:
     rev: v2.3.0
     hooks:
       - id: codespell
-        args: ["-L", "ba,ihs,kake,nd,noe,nwo,te,fo,zar", "-S", "fixture"]
+        args: ["-L", "fo,ihs,kake,te", "-S", "fixture"]
   - repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v5.0.0
     hooks:

src/zarr/core/array.py

@@ -604,7 +604,7 @@ async def _create_v2(
         dtype: npt.DTypeLike,
         chunks: ChunkCoords,
         dimension_separator: Literal[".", "/"] | None = None,
-        fill_value: None | float = None,
+        fill_value: float | None = None,
         order: MemoryOrder | None = None,
         filters: list[dict[str, JSON]] | None = None,
         compressor: dict[str, JSON] | None = None,

src/zarr/core/common.py

@@ -36,7 +36,7 @@
 ChunkCoordsLike = Iterable[int]
 ZarrFormat = Literal[2, 3]
 NodeType = Literal["array", "group"]
-JSON = None | str | int | float | Mapping[str, "JSON"] | tuple["JSON", ...]
+JSON = str | int | float | Mapping[str, "JSON"] | tuple["JSON", ...] | None
 MemoryOrder = Literal["C", "F"]
 AccessModeLiteral = Literal["r", "r+", "a", "w", "w-"]
 

src/zarr/core/metadata/common.py

@@ -6,7 +6,7 @@
     from zarr.core.common import JSON
 
 
-def parse_attributes(data: None | dict[str, JSON]) -> dict[str, JSON]:
+def parse_attributes(data: dict[str, JSON] | None) -> dict[str, JSON]:
     if data is None:
         return {}
 

src/zarr/core/metadata/v2.py

@@ -44,7 +44,7 @@ class ArrayV2Metadata(Metadata):
     shape: ChunkCoords
     chunks: tuple[int, ...]
     dtype: np.dtype[Any]
-    fill_value: None | int | float | str | bytes = 0
+    fill_value: int | float | str | bytes | None = 0
     order: MemoryOrder = "C"
     filters: tuple[numcodecs.abc.Codec, ...] | None = None
     dimension_separator: Literal[".", "/"] = "."

src/zarr/core/metadata/v3.py

@@ -225,9 +225,9 @@ def __init__(
         chunk_key_encoding: dict[str, JSON] | ChunkKeyEncoding,
         fill_value: Any,
         codecs: Iterable[Codec | dict[str, JSON]],
-        attributes: None | dict[str, JSON],
-        dimension_names: None | Iterable[str],
-        storage_transformers: None | Iterable[dict[str, JSON]] = None,
+        attributes: dict[str, JSON] | None,
+        dimension_names: Iterable[str] | None,
+        storage_transformers: Iterable[dict[str, JSON]] | None = None,
     ) -> None:
         """
         Because the class is a frozen dataclass, we set attributes using object.__setattr__
@@ -540,7 +540,7 @@ class DataType(Enum):
     bytes = "bytes"
 
     @property
-    def byte_count(self) -> None | int:
+    def byte_count(self) -> int | None:
         data_type_byte_counts = {
             DataType.bool: 1,
             DataType.int8: 1,
@@ -626,7 +626,7 @@ def from_numpy(cls, dtype: np.dtype[Any]) -> DataType:
         return DataType[dtype_to_data_type[dtype.str]]
 
     @classmethod
-    def parse(cls, dtype: None | DataType | Any) -> DataType:
+    def parse(cls, dtype: DataType | Any | None) -> DataType:
         if dtype is None:
             return DataType[DEFAULT_DTYPE]
         if isinstance(dtype, DataType):

src/zarr/storage/_utils.py

@@ -45,7 +45,7 @@ def normalize_path(path: str | bytes | Path | None) -> str:
 
 
 def _normalize_interval_index(
-    data: Buffer, interval: None | tuple[int | None, int | None]
+    data: Buffer, interval: tuple[int | None, int | None] | None
 ) -> tuple[int, int]:
     """
     Convert an implicit interval into an explicit start and length

src/zarr/testing/store.py

@@ -106,7 +106,7 @@ def test_store_supports_listing(self, store: S) -> None:
     @pytest.mark.parametrize("data", [b"\x01\x02\x03\x04", b""])
     @pytest.mark.parametrize("byte_range", [None, (0, None), (1, None), (1, 2), (None, 1)])
     async def test_get(
-        self, store: S, key: str, data: bytes, byte_range: None | tuple[int | None, int | None]
+        self, store: S, key: str, data: bytes, byte_range: tuple[int | None, int | None] | None
     ) -> None:
         """
         Ensure that data can be read from the store using the store.get method.

src/zarr/testing/strategies.py

@@ -117,7 +117,7 @@ def arrays(
     shapes: st.SearchStrategy[tuple[int, ...]] = array_shapes,
     compressors: st.SearchStrategy = compressors,
     stores: st.SearchStrategy[StoreLike] = stores,
-    paths: st.SearchStrategy[None | str] = paths,
+    paths: st.SearchStrategy[str | None] = paths,
     array_names: st.SearchStrategy = array_names,
     arrays: st.SearchStrategy | None = None,
     attrs: st.SearchStrategy = attrs,

tests/test_codecs/test_vlen.py

@@ -25,7 +25,7 @@
 @pytest.mark.parametrize("as_object_array", [False, True])
 @pytest.mark.parametrize("codecs", [None, [VLenUTF8Codec()], [VLenUTF8Codec(), ZstdCodec()]])
 def test_vlen_string(
-    store: Store, dtype: None | np.dtype[Any], as_object_array: bool, codecs: None | list[Codec]
+    store: Store, dtype: np.dtype[Any] | None, as_object_array: bool, codecs: list[Codec] | None
 ) -> None:
     strings = ["hello", "world", "this", "is", "a", "test"]
     data = np.array(strings, dtype=dtype).reshape((2, 3))
@@ -62,7 +62,7 @@ def test_vlen_string(
 @pytest.mark.parametrize("store", ["memory", "local"], indirect=["store"])
 @pytest.mark.parametrize("as_object_array", [False, True])
 @pytest.mark.parametrize("codecs", [None, [VLenBytesCodec()], [VLenBytesCodec(), ZstdCodec()]])
-def test_vlen_bytes(store: Store, as_object_array: bool, codecs: None | list[Codec]) -> None:
+def test_vlen_bytes(store: Store, as_object_array: bool, codecs: list[Codec] | None) -> None:
     bstrings = [b"hello", b"world", b"this", b"is", b"a", b"test"]
     data = np.array(bstrings).reshape((2, 3))
     assert data.dtype == "|S5"