Merge branch 'main' into fix/consistent-create-array-signature

Author: d-v-b

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -57,7 +57,22 @@ body:
     id: reproduce
     attributes:
       label: Steps to reproduce
-      description: Minimal, reproducible code sample, a copy-pastable example if possible.
+      description: Minimal, reproducible code sample. Must list dependencies in [inline script metadata](https://packaging.python.org/en/latest/specifications/inline-script-metadata/#example). When put in a file named `issue.py` calling `uv run issue.py` should show the issue.
+      value: |
+        ```python
+        # /// script
+        # requires-python = ">=3.11"
+        # dependencies = [
+        #   "zarr@git+https://github.com/zarr-developers/zarr-python.git@main",
+        # ]
+        # ///
+        #
+        # This script automatically imports the development branch of zarr to check for issues
+
+        import zarr
+        # your reproducer code
+        # zarr.print_debug_info()
+        ```
     validations:
       required: true
   - type: textarea

changes/3081.feature.rst

@@ -0,0 +1 @@
+Adds ``fill_value`` to the list of attributes displayed in the output of the ``AsyncArray.info()`` method.

changes/3082.feature.rst

@@ -0,0 +1 @@
+Use :py:func:`numpy.zeros` instead of :py:func:`np.full` for a performance speedup when creating a `zarr.core.buffer.NDBuffer` with `fill_value=0`.

docs/user-guide/arrays.rst

@@ -183,6 +183,7 @@ which can be used to print useful diagnostics, e.g.::
    Type               : Array
    Zarr format        : 3
    Data type          : DataType.int32
+   Fill value         : 0
    Shape              : (10000, 10000)
    Chunk shape        : (1000, 1000)
    Order              : C
@@ -200,6 +201,7 @@ prints additional diagnostics, e.g.::
    Type               : Array
    Zarr format        : 3
    Data type          : DataType.int32
+   Fill value         : 0
    Shape              : (10000, 10000)
    Chunk shape        : (1000, 1000)
    Order              : C
@@ -287,6 +289,7 @@ Here is an example using a delta filter with the Blosc compressor::
    Type               : Array
    Zarr format        : 3
    Data type          : DataType.int32
+   Fill value         : 0
    Shape              : (10000, 10000)
    Chunk shape        : (1000, 1000)
    Order              : C
@@ -601,6 +604,7 @@ Sharded arrays can be created by providing the ``shards`` parameter to :func:`za
   Type               : Array
   Zarr format        : 3
   Data type          : DataType.uint8
+  Fill value         : 0
   Shape              : (10000, 10000)
   Shard shape        : (1000, 1000)
   Chunk shape        : (100, 100)

docs/user-guide/groups.rst

@@ -129,6 +129,7 @@ property. E.g.::
    Type               : Array
    Zarr format        : 3
    Data type          : DataType.int64
+   Fill value         : 0
    Shape              : (1000000,)
    Chunk shape        : (100000,)
    Order              : C
@@ -145,6 +146,7 @@ property. E.g.::
    Type               : Array
    Zarr format        : 3
    Data type          : DataType.float32
+   Fill value         : 0.0
    Shape              : (1000, 1000)
    Chunk shape        : (100, 100)
    Order              : C

docs/user-guide/performance.rst

@@ -92,6 +92,7 @@ To use sharding, you need to specify the ``shards`` parameter when creating the
    Type               : Array
    Zarr format        : 3
    Data type          : DataType.uint8
+   Fill value         : 0
    Shape              : (10000, 10000, 1000)
    Shard shape        : (1000, 1000, 1000)
    Chunk shape        : (100, 100, 100)
@@ -122,6 +123,7 @@ ratios, depending on the correlation structure within the data. E.g.::
    Type               : Array
    Zarr format        : 3
    Data type          : DataType.int32
+   Fill value         : 0
    Shape              : (10000, 10000)
    Chunk shape        : (1000, 1000)
    Order              : C
@@ -141,6 +143,7 @@ ratios, depending on the correlation structure within the data. E.g.::
    Type               : Array
    Zarr format        : 3
    Data type          : DataType.int32
+   Fill value         : 0
    Shape              : (10000, 10000)
    Chunk shape        : (1000, 1000)
    Order              : F

pyproject.toml

@@ -291,8 +291,8 @@ extend-exclude = [
 extend-select = [
     "ANN",  # flake8-annotations
     "B",    # flake8-bugbear
-    "EXE",  # flake8-executable
     "C4",   # flake8-comprehensions
+    "EXE",  # flake8-executable
     "FA",   # flake8-future-annotations
     "FLY",  # flynt
     "FURB", # refurb
@@ -364,14 +364,14 @@ module = [
     "tests.test_store.test_local",
     "tests.test_store.test_fsspec",
     "tests.test_store.test_memory",
+    "tests.test_codecs.test_codecs",
 ]
 strict = false
 
 # TODO: Move the next modules up to the strict = false section
 # and fix the errors
 [[tool.mypy.overrides]]
 module = [
-    "tests.test_codecs.test_codecs",
     "tests.test_metadata.*",
     "tests.test_store.test_core",
     "tests.test_store.test_logging",

src/zarr/api/asynchronous.py

@@ -329,7 +329,7 @@ async def open(
         try:
             metadata_dict = await get_array_metadata(store_path, zarr_format=zarr_format)
             # TODO: remove this cast when we fix typing for array metadata dicts
-            _metadata_dict = cast(ArrayMetadataDict, metadata_dict)
+            _metadata_dict = cast("ArrayMetadataDict", metadata_dict)
             # for v2, the above would already have raised an exception if not an array
             zarr_format = _metadata_dict["zarr_format"]
             is_v3_array = zarr_format == 3 and _metadata_dict.get("node_type") == "array"

src/zarr/codecs/crc32c_.py

@@ -40,7 +40,9 @@ async def _decode_single(
         inner_bytes = data[:-4]
 
         # Need to do a manual cast until https://github.com/numpy/numpy/issues/26783 is resolved
-        computed_checksum = np.uint32(crc32c(cast(typing_extensions.Buffer, inner_bytes))).tobytes()
+        computed_checksum = np.uint32(
+            crc32c(cast("typing_extensions.Buffer", inner_bytes))
+        ).tobytes()
         stored_checksum = bytes(crc32_bytes)
         if computed_checksum != stored_checksum:
             raise ValueError(
@@ -55,7 +57,7 @@ async def _encode_single(
     ) -> Buffer | None:
         data = chunk_bytes.as_numpy_array()
         # Calculate the checksum and "cast" it to a numpy array
-        checksum = np.array([crc32c(cast(typing_extensions.Buffer, data))], dtype=np.uint32)
+        checksum = np.array([crc32c(cast("typing_extensions.Buffer", data))], dtype=np.uint32)
         # Append the checksum (as bytes) to the data
         return chunk_spec.prototype.buffer.from_array_like(np.append(data, checksum.view("B")))
 

src/zarr/codecs/sharding.py

@@ -115,7 +115,7 @@ class _ShardIndex(NamedTuple):
     def chunks_per_shard(self) -> ChunkCoords:
         result = tuple(self.offsets_and_lengths.shape[0:-1])
         # The cast is required until https://github.com/numpy/numpy/pull/27211 is merged
-        return cast(ChunkCoords, result)
+        return cast("ChunkCoords", result)
 
     def _localize_chunk(self, chunk_coords: ChunkCoords) -> ChunkCoords:
         return tuple(