Merge branch 'main' into dependabot/pip/numpy-1.26.1

.github/workflows/codeql-analysis.yml

@@ -42,7 +42,7 @@ jobs:
 
     # Initializes the CodeQL tools for scanning.
     - name: Initialize CodeQL
-      uses: github/codeql-action/init@v2
+      uses: github/codeql-action/init@v3
       with:
         languages: ${{ matrix.language }}
         # If you wish to specify custom queries, you can do so here or in a config file.
@@ -56,7 +56,7 @@ jobs:
     # Autobuild attempts to build any compiled languages  (C/C++, C#, or Java).
     # If this step fails, then you should remove it and run the build manually (see below)
     - name: Autobuild
-      uses: github/codeql-action/autobuild@v2
+      uses: github/codeql-action/autobuild@v3
 
     # ℹ️ Command-line programs to run using the OS shell.
     # 📚 See https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#jobsjob_idstepsrun
@@ -69,4 +69,4 @@ jobs:
     #   ./location_of_script_within_repo/buildscript.sh
 
     - name: Perform CodeQL Analysis
-      uses: github/codeql-action/analyze@v2
+      uses: github/codeql-action/analyze@v3

.github/workflows/releases.yml

@@ -16,7 +16,7 @@ jobs:
           submodules: true
           fetch-depth: 0
 
-      - uses: actions/setup-python@v4.7.1
+      - uses: actions/setup-python@v5.0.0
         name: Install Python
         with:
           python-version: '3.8'

.pre-commit-config.yaml

@@ -6,28 +6,25 @@ default_stages: [commit, push]
 default_language_version:
   python: python3
 repos:
-  - repo: https://github.com/charliermarsh/ruff-pre-commit
+  - repo: https://github.com/astral-sh/ruff-pre-commit
     # Ruff version.
-    rev: 'v0.0.224'
+    rev: 'v0.1.8'
     hooks:
     - id: ruff
-      # Respect `exclude` and `extend-exclude` settings.
-      args: ["--force-exclude"]
   - repo: https://github.com/psf/black
-    rev: 23.10.1
+    rev: 23.12.0
     hooks:
     - id: black
   - repo: https://github.com/codespell-project/codespell
-    rev: v2.2.5
+    rev: v2.2.6
     hooks:
       - id: codespell
-        args: ["-L", "ba,ihs,kake,nd,noe,nwo,te,fo,zar", "-S", "fixture"]
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
+    rev: v4.5.0
     hooks:
     - id: check-yaml
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.3.0
+    rev: v1.7.1
     hooks:
       - id: mypy
         files: zarr

docs/release.rst

@@ -18,6 +18,12 @@ Release notes
 Unreleased
 ----------
 
+Enhancements
+~~~~~~~~~~~~
+
+* Added type hints to ``zarr.creation.create()``.
+  By :user:`David Stansby <dstansby>` :issue:`1536`.
+
 Docs
 ~~~~
 

pyproject.toml

@@ -65,6 +65,7 @@ Homepage = "https://github.com/zarr-developers/zarr-python"
 exclude_lines = [
     "pragma: no cover",
     "pragma: ${PY_MAJOR_VERSION} no cover",
+    '.*\.\.\.' # Ignore "..." lines
 ]
 
 [tool.coverage.run]
@@ -119,9 +120,10 @@ exclude = '''
 '''
 
 [tool.mypy]
-python_version = "3.8"
 ignore_missing_imports = true
-follow_imports = "silent"
+warn_unused_configs = true
+warn_redundant_casts = true
+warn_unused_ignores = true
 
 [tool.pytest.ini_options]
 doctest_optionflags = [
@@ -137,3 +139,8 @@ filterwarnings = [
     "ignore:PY_SSIZE_T_CLEAN will be required.*:DeprecationWarning",
     "ignore:The loop argument is deprecated since Python 3.8.*:DeprecationWarning",
 ]
+
+
+[tool.codespell]
+ignore-words-list = "ba,ihs,kake,nd,noe,nwo,te,fo,zar"
+skip = 'fixture,.git'

requirements_dev_optional.txt

@@ -18,6 +18,6 @@ pytest-cov==4.1.0
 pytest-doctestplus==1.0.0
 pytest-timeout==2.2.0
 h5py==3.10.0
-fsspec==2023.10.0
-s3fs==2023.10.0
+fsspec==2023.12.1
+s3fs==2023.12.1
 moto[server]>=4.0.8

zarr/_storage/store.py

@@ -9,6 +9,7 @@
 from zarr.meta import Metadata2, Metadata3
 from zarr.util import normalize_storage_path
 from zarr.context import Context
+from zarr.types import ZARR_VERSION
 
 # v2 store keys
 array_meta_key = ".zarray"
@@ -19,7 +20,7 @@
 meta_root = "meta/root/"
 data_root = "data/root/"
 
-DEFAULT_ZARR_VERSION = 2
+DEFAULT_ZARR_VERSION: ZARR_VERSION = 2
 
 v3_api_available = os.environ.get("ZARR_V3_EXPERIMENTAL_API", "0").lower() not in ["0", "false"]
 
@@ -641,10 +642,10 @@ def _rmdir_from_keys_v3(store: StoreV3, path: str = "") -> None:
     sfx = _get_metadata_suffix(store)
     array_meta_file = meta_dir + ".array" + sfx
     if array_meta_file in store:
-        store.erase(array_meta_file)  # type: ignore
+        store.erase(array_meta_file)
     group_meta_file = meta_dir + ".group" + sfx
     if group_meta_file in store:
-        store.erase(group_meta_file)  # type: ignore
+        store.erase(group_meta_file)
 
 
 def _listdir_from_keys(store: BaseStore, path: Optional[str] = None) -> List[str]:

zarr/_storage/v3_storage_transformers.py

@@ -351,7 +351,7 @@ def erase_prefix(self, prefix):
 
     def rmdir(self, path=None):
         path = normalize_storage_path(path)
-        _rmdir_from_keys_v3(self, path)  # type: ignore
+        _rmdir_from_keys_v3(self, path)
 
     def __contains__(self, key):
         if self._is_data_key(key):

zarr/core.py

@@ -2536,7 +2536,7 @@ def hexdigest(self, hashname="sha1"):
         checksum = binascii.hexlify(self.digest(hashname=hashname))
 
         # This is a bytes object on Python 3 and we want a str.
-        if type(checksum) is not str:
+        if not isinstance(checksum, str):
             checksum = checksum.decode("utf8")
 
         return checksum

zarr/creation.py

@@ -1,7 +1,10 @@
-from typing import Optional
+from collections.abc import MutableMapping
+from typing import Optional, Tuple, Union, Sequence
 from warnings import warn
 
 import numpy as np
+import numpy.typing as npt
+from numcodecs.abc import Codec
 from numcodecs.registry import codec_registry
 
 from zarr._storage.store import DEFAULT_ZARR_VERSION
@@ -19,32 +22,35 @@
     normalize_storage_path,
     normalize_store_arg,
 )
+from zarr._storage.store import StorageTransformer
+from zarr.sync import Synchronizer
+from zarr.types import ZARR_VERSION, DIMENSION_SEPARATOR, MEMORY_ORDER, MetaArray, PathLike
 from zarr.util import normalize_dimension_separator
 
 
 def create(
-    shape,
-    chunks=True,
-    dtype=None,
+    shape: Union[int, Tuple[int, ...]],
+    chunks: Union[int, Tuple[int, ...], bool] = True,
+    dtype: Optional[npt.DTypeLike] = None,
     compressor="default",
     fill_value: Optional[int] = 0,
-    order="C",
-    store=None,
-    synchronizer=None,
-    overwrite=False,
-    path=None,
-    chunk_store=None,
-    filters=None,
-    cache_metadata=True,
-    cache_attrs=True,
-    read_only=False,
-    object_codec=None,
-    dimension_separator=None,
-    write_empty_chunks=True,
+    order: MEMORY_ORDER = "C",
+    store: Optional[Union[str, MutableMapping]] = None,
+    synchronizer: Optional[Synchronizer] = None,
+    overwrite: bool = False,
+    path: Optional[PathLike] = None,
+    chunk_store: Optional[MutableMapping] = None,
+    filters: Optional[Sequence[Codec]] = None,
+    cache_metadata: bool = True,
+    cache_attrs: bool = True,
+    read_only: bool = False,
+    object_codec: Optional[Codec] = None,
+    dimension_separator: Optional[DIMENSION_SEPARATOR] = None,
+    write_empty_chunks: bool = True,
     *,
-    zarr_version=None,
-    meta_array=None,
-    storage_transformers=(),
+    zarr_version: Optional[ZARR_VERSION] = None,
+    meta_array: Optional[MetaArray] = None,
+    storage_transformers: Sequence[StorageTransformer] = (),
     **kwargs,
 ):
     """Create an array.

zarr/meta.py

@@ -234,8 +234,8 @@ def decode_fill_value(cls, v: Any, dtype: np.dtype, object_codec: Any = None) ->
                 return np.array(v, dtype=dtype)[()]
         elif dtype.kind in "c":
             v = (
-                cls.decode_fill_value(v[0], dtype.type().real.dtype),  # type: ignore
-                cls.decode_fill_value(v[1], dtype.type().imag.dtype),  # type: ignore
+                cls.decode_fill_value(v[0], dtype.type().real.dtype),
+                cls.decode_fill_value(v[1], dtype.type().imag.dtype),
             )
             v = v[0] + 1j * v[1]
             return np.array(v, dtype=dtype)[()]

zarr/storage.py

@@ -40,6 +40,7 @@
 from numcodecs.compat import ensure_bytes, ensure_text, ensure_contiguous_ndarray_like
 from numcodecs.registry import codec_registry
 from zarr.context import Context
+from zarr.types import PathLike as Path
 
 from zarr.errors import (
     MetadataError,
@@ -105,7 +106,6 @@
     default_compressor = Zlib()
 
 
-Path = Union[str, bytes, None]
 # allow MutableMapping for backwards compatibility
 StoreLike = Union[BaseStore, MutableMapping]
 
@@ -206,7 +206,7 @@ def rmdir(store: StoreLike, path: Path = None):
     store_version = getattr(store, "_store_version", 2)
     if hasattr(store, "rmdir") and store.is_erasable():  # type: ignore
         # pass through
-        store.rmdir(path)  # type: ignore
+        store.rmdir(path)
     else:
         # slow version, delete one key at a time
         if store_version == 2:
@@ -236,7 +236,7 @@ def listdir(store: BaseStore, path: Path = None):
     path = normalize_storage_path(path)
     if hasattr(store, "listdir"):
         # pass through
-        return store.listdir(path)  # type: ignore
+        return store.listdir(path)
     else:
         # slow version, iterate through all keys
         warnings.warn(
@@ -289,7 +289,7 @@ def getsize(store: BaseStore, path: Path = None) -> int:
     if hasattr(store, "getsize"):
         # pass through
         path = normalize_storage_path(path)
-        return store.getsize(path)  # type: ignore
+        return store.getsize(path)
     elif isinstance(store, MutableMapping):
         return _getsize(store, path)
     else:
@@ -627,7 +627,7 @@ def _init_array_metadata(
 
     key = _prefix_to_array_key(store, _path_to_prefix(path))
     if hasattr(store, "_metadata_class"):
-        store[key] = store._metadata_class.encode_array_metadata(meta)  # type: ignore
+        store[key] = store._metadata_class.encode_array_metadata(meta)
     else:
         store[key] = encode_array_metadata(meta)
 
@@ -730,10 +730,10 @@ def _init_group_metadata(
     if store_version == 3:
         meta = {"attributes": {}}  # type: ignore
     else:
-        meta = {}  # type: ignore
+        meta = {}
     key = _prefix_to_group_key(store, _path_to_prefix(path))
     if hasattr(store, "_metadata_class"):
-        store[key] = store._metadata_class.encode_group_metadata(meta)  # type: ignore
+        store[key] = store._metadata_class.encode_group_metadata(meta)
     else:
         store[key] = encode_group_metadata(meta)
 

zarr/sync.py

@@ -1,11 +1,19 @@
 import os
 from collections import defaultdict
 from threading import Lock
+from typing import Protocol
 
 import fasteners
 
 
-class ThreadSynchronizer:
+class Synchronizer(Protocol):
+    """Base class for synchronizers."""
+
+    def __getitem__(self, item):
+        ...
+
+
+class ThreadSynchronizer(Synchronizer):
     """Provides synchronization using thread locks."""
 
     def __init__(self):
@@ -24,7 +32,7 @@ def __setstate__(self, *args):
         self.__init__()
 
 
-class ProcessSynchronizer:
+class ProcessSynchronizer(Synchronizer):
     """Provides synchronization using file locks via the
     `fasteners <https://fasteners.readthedocs.io/en/latest/api/inter_process/>`_
     package.

zarr/tests/test_core.py

@@ -3,7 +3,7 @@
 import sys
 import pickle
 import shutil
-from typing import Any, Literal, Optional, Tuple, Union
+from typing import Any, Literal, Optional, Tuple, Union, Sequence
 import unittest
 from itertools import zip_longest
 from tempfile import mkdtemp
@@ -26,6 +26,7 @@
     VLenUTF8,
     Zlib,
 )
+from numcodecs.abc import Codec
 from numcodecs.compat import ensure_bytes, ensure_ndarray
 from numcodecs.tests.common import greetings
 from numpy.testing import assert_array_almost_equal, assert_array_equal
@@ -73,6 +74,7 @@
 from zarr.tests.test_storage_v3 import DummyStorageTransfomer
 from zarr.util import buffer_size
 from zarr.tests.util import abs_container, skip_test_env_var, have_fsspec, mktemp
+from zarr.types import DIMENSION_SEPARATOR
 
 # noinspection PyMethodMayBeStatic
 
@@ -82,8 +84,8 @@ class TestArray:
     root = ""
     path = ""
     compressor = Zlib(level=1)
-    filters = None
-    dimension_separator: Literal["/", ".", None] = None
+    filters: Optional[Sequence[Codec]] = None
+    dimension_separator: Optional[DIMENSION_SEPARATOR] = None
     cache_metadata = True
     cache_attrs = True
     partial_decompress: bool = False

zarr/types.py

@@ -0,0 +1,13 @@
+from typing import Literal, Protocol, Union
+
+ZARR_VERSION = Literal[2, 3]
+DIMENSION_SEPARATOR = Literal[".", "/"]
+MEMORY_ORDER = Literal["C", "F"]
+
+
+PathLike = Union[str, bytes, None]
+
+
+class MetaArray(Protocol):
+    def __array_function__(self, func, types, args, kwargs):
+        ...