Merge branch 'main' of github.com:zarr-developers/zarr-python into chore/handle-numcodecs-codecs

Author: d-v-b

changes/3367.bugfix.rst

@@ -0,0 +1 @@
+Added `zarr.errors.ArrayNotFoundError`, which is raised when attempting to open a zarr array that does not exist, and `zarr.errors.NodeNotFoundError`, which is raised when failing to open an array or a group in a context where either an array or a group was expected.

pyproject.toml

@@ -115,11 +115,11 @@ docs = [
 
 
 [project.urls]
-"Bug Tracker" = "https://github.com/zarr-developers/zarr-python/issues"
-Changelog = "https://zarr.readthedocs.io/en/stable/release-notes.html"
+issues = "https://github.com/zarr-developers/zarr-python/issues"
+changelog = "https://zarr.readthedocs.io/en/stable/release-notes.html"
 Discussions = "https://github.com/zarr-developers/zarr-python/discussions"
-Documentation = "https://zarr.readthedocs.io/"
-Homepage = "https://github.com/zarr-developers/zarr-python"
+documentation = "https://zarr.readthedocs.io/"
+homepage = "https://github.com/zarr-developers/zarr-python"
 
 [dependency-groups]
 dev = [
@@ -352,14 +352,14 @@ module = [
     "tests.test_store.test_fsspec",
     "tests.test_store.test_memory",
     "tests.test_codecs.test_codecs",
+    "tests.test_metadata.*",
 ]
 strict = false
 
 # TODO: Move the next modules up to the strict = false section
 # and fix the errors
 [[tool.mypy.overrides]]
 module = [
-    "tests.test_metadata.*",
     "tests.test_store.test_core",
     "tests.test_store.test_logging",
     "tests.test_store.test_object",

src/zarr/abc/codec.py

@@ -138,7 +138,7 @@ def validate(
         """
 
     async def _decode_single(self, chunk_data: CodecOutput, chunk_spec: ArraySpec) -> CodecInput:
-        raise NotImplementedError
+        raise NotImplementedError  # pragma: no cover
 
     async def decode(
         self,
@@ -161,7 +161,7 @@ async def decode(
     async def _encode_single(
         self, chunk_data: CodecInput, chunk_spec: ArraySpec
     ) -> CodecOutput | None:
-        raise NotImplementedError
+        raise NotImplementedError  # pragma: no cover
 
     async def encode(
         self,
@@ -242,7 +242,7 @@ async def _encode_partial_single(
         selection: SelectorTuple,
         chunk_spec: ArraySpec,
     ) -> None:
-        raise NotImplementedError
+        raise NotImplementedError  # pragma: no cover
 
     async def encode_partial(
         self,

src/zarr/api/asynchronous.py

@@ -39,6 +39,7 @@
 )
 from zarr.core.metadata import ArrayMetadataDict, ArrayV2Metadata, ArrayV3Metadata
 from zarr.errors import (
+    ArrayNotFoundError,
     GroupNotFoundError,
     NodeTypeValidationError,
     ZarrDeprecationWarning,
@@ -1257,7 +1258,7 @@ async def open_array(
 
     try:
         return await AsyncArray.open(store_path, zarr_format=zarr_format)
-    except FileNotFoundError:
+    except FileNotFoundError as err:
         if not store_path.read_only and mode in _CREATE_MODES:
             overwrite = _infer_overwrite(mode)
             _zarr_format = zarr_format or _default_zarr_format()
@@ -1267,7 +1268,7 @@ async def open_array(
                 overwrite=overwrite,
                 **kwargs,
             )
-        raise
+        raise ArrayNotFoundError(store_path.store, store_path.path) from err
 
 
 async def open_like(

src/zarr/core/array.py

@@ -3,7 +3,7 @@
 import json
 import warnings
 from asyncio import gather
-from collections.abc import Iterable
+from collections.abc import Iterable, Mapping
 from dataclasses import dataclass, field, replace
 from itertools import starmap
 from logging import getLogger
@@ -117,7 +117,12 @@
 )
 from zarr.core.metadata.v3 import parse_node_type_array
 from zarr.core.sync import sync
-from zarr.errors import MetadataValidationError, ZarrDeprecationWarning, ZarrUserWarning
+from zarr.errors import (
+    ArrayNotFoundError,
+    MetadataValidationError,
+    ZarrDeprecationWarning,
+    ZarrUserWarning,
+)
 from zarr.registry import (
     _parse_array_array_codec,
     _parse_array_bytes_codec,
@@ -216,11 +221,19 @@ async def get_array_metadata(
             (store_path / ZATTRS_JSON).get(prototype=cpu_buffer_prototype),
         )
         if zarray_bytes is None:
-            raise FileNotFoundError(store_path)
+            msg = (
+                "A Zarr V2 array metadata document was not found in store "
+                f"{store_path.store!r} at path {store_path.path!r}."
+            )
+            raise ArrayNotFoundError(msg)
     elif zarr_format == 3:
         zarr_json_bytes = await (store_path / ZARR_JSON).get(prototype=cpu_buffer_prototype)
         if zarr_json_bytes is None:
-            raise FileNotFoundError(store_path)
+            msg = (
+                "A Zarr V3 array metadata document was not found in store "
+                f"{store_path.store!r} at path {store_path.path!r}."
+            )
+            raise ArrayNotFoundError(msg)
     elif zarr_format is None:
         zarr_json_bytes, zarray_bytes, zattrs_bytes = await gather(
             (store_path / ZARR_JSON).get(prototype=cpu_buffer_prototype),
@@ -232,7 +245,11 @@ async def get_array_metadata(
             msg = f"Both zarr.json (Zarr format 3) and .zarray (Zarr format 2) metadata objects exist at {store_path}. Zarr v3 will be used."
             warnings.warn(msg, category=ZarrUserWarning, stacklevel=1)
         if zarr_json_bytes is None and zarray_bytes is None:
-            raise FileNotFoundError(store_path)
+            msg = (
+                f"Neither Zarr V3 nor Zarr V2 array metadata documents "
+                f"were found in store {store_path.store!r} at path {store_path.path!r}."
+            )
+            raise ArrayNotFoundError(msg)
         # set zarr_format based on which keys were found
         if zarr_json_bytes is not None:
             zarr_format = 3
@@ -3907,7 +3924,7 @@ def _build_parents(
 
 CompressorsLike: TypeAlias = (
     Iterable[dict[str, JSON] | BytesBytesCodec | Numcodec]
-    | dict[str, JSON]
+    | Mapping[str, JSON]
     | BytesBytesCodec
     | Numcodec
     | Literal["auto"]

src/zarr/errors.py

@@ -1,6 +1,7 @@
 from typing import Any
 
 __all__ = [
+    "ArrayNotFoundError",
     "BaseZarrError",
     "ContainsArrayAndGroupError",
     "ContainsArrayError",
@@ -26,12 +27,49 @@ def __init__(self, *args: Any) -> None:
         super().__init__(self._msg.format(*args))
 
 
-class GroupNotFoundError(BaseZarrError, FileNotFoundError):
+class NodeNotFoundError(BaseZarrError, FileNotFoundError):
+    """
+    Raised when a node (array or group) is not found at a certain path.
+    """
+
+    def __init__(self, *args: Any) -> None:
+        if len(args) == 1:
+            # Pre-formatted message
+            super(BaseZarrError, self).__init__(args[0])
+        else:
+            # Store and path arguments - format them
+            _msg = "No node found in store {!r} at path {!r}"
+            super(BaseZarrError, self).__init__(_msg.format(*args))
+
+
+class ArrayNotFoundError(NodeNotFoundError):
+    """
+    Raised when an array isn't found at a certain path.
+    """
+
+    def __init__(self, *args: Any) -> None:
+        if len(args) == 1:
+            # Pre-formatted message
+            super(BaseZarrError, self).__init__(args[0])
+        else:
+            # Store and path arguments - format them
+            _msg = "No array found in store {!r} at path {!r}"
+            super(BaseZarrError, self).__init__(_msg.format(*args))
+
+
+class GroupNotFoundError(NodeNotFoundError):
     """
     Raised when a group isn't found at a certain path.
     """
 
-    _msg = "No group found in store {!r} at path {!r}"
+    def __init__(self, *args: Any) -> None:
+        if len(args) == 1:
+            # Pre-formatted message
+            super(BaseZarrError, self).__init__(args[0])
+        else:
+            # Store and path arguments - format them
+            _msg = "No group found in store {!r} at path {!r}"
+            super(BaseZarrError, self).__init__(_msg.format(*args))
 
 
 class ContainsGroupError(BaseZarrError):

tests/test_api.py

@@ -6,7 +6,7 @@
 
 import zarr.codecs
 import zarr.storage
-from zarr.core.array import init_array
+from zarr.core.array import AsyncArray, init_array
 from zarr.storage import LocalStore, ZipStore
 from zarr.storage._common import StorePath
 
@@ -42,7 +42,13 @@
     save_group,
 )
 from zarr.core.buffer import NDArrayLike
-from zarr.errors import MetadataValidationError, ZarrDeprecationWarning, ZarrUserWarning
+from zarr.errors import (
+    ArrayNotFoundError,
+    MetadataValidationError,
+    NodeNotFoundError,
+    ZarrDeprecationWarning,
+    ZarrUserWarning,
+)
 from zarr.storage import MemoryStore
 from zarr.storage._utils import normalize_path
 from zarr.testing.utils import gpu_test
@@ -70,11 +76,11 @@ def test_create(memory_store: Store) -> None:
 
     # create array with float shape
     with pytest.raises(TypeError):
-        z = create(shape=(400.5, 100), store=store, overwrite=True)  # type: ignore [arg-type]
+        z = create(shape=(400.5, 100), store=store, overwrite=True)  # type: ignore[arg-type]
 
     # create array with float chunk shape
     with pytest.raises(TypeError):
-        z = create(shape=(400, 100), chunks=(16, 16.5), store=store, overwrite=True)  # type: ignore [arg-type]
+        z = create(shape=(400, 100), chunks=(16, 16.5), store=store, overwrite=True)  # type: ignore[arg-type]
 
 
 # TODO: parametrize over everything this function takes
@@ -185,10 +191,27 @@ async def test_open_array(memory_store: MemoryStore, zarr_format: ZarrFormat) ->
     assert z.read_only
 
     # path not found
-    with pytest.raises(FileNotFoundError):
+    with pytest.raises(NodeNotFoundError):
         zarr.api.synchronous.open(store="doesnotexist", mode="r", zarr_format=zarr_format)
 
 
+@pytest.mark.asyncio
+async def test_async_array_open_array_not_found() -> None:
+    """Test that AsyncArray.open raises ArrayNotFoundError when array doesn't exist"""
+    store = MemoryStore()
+    # Try to open an array that does not exist
+    with pytest.raises(ArrayNotFoundError):
+        await AsyncArray.open(store, zarr_format=2)
+
+
+def test_array_open_array_not_found_sync() -> None:
+    """Test that Array.open raises ArrayNotFoundError when array doesn't exist"""
+    store = MemoryStore()
+    # Try to open an array that does not exist
+    with pytest.raises(ArrayNotFoundError):
+        Array.open(store)
+
+
 @pytest.mark.parametrize("store", ["memory", "local", "zip"], indirect=True)
 def test_v2_and_v3_exist_at_same_path(store: Store) -> None:
     zarr.create_array(store, shape=(10,), dtype="uint8", zarr_format=3)
@@ -266,7 +289,7 @@ def test_save(store: Store, n_args: int, n_kwargs: int, path: None | str) -> Non
         assert isinstance(array, Array)
         assert_array_equal(array[:], data)
     else:
-        save(store, *args, path=path, **kwargs)  # type: ignore [arg-type]
+        save(store, *args, path=path, **kwargs)  # type: ignore[arg-type]
         group = zarr.api.synchronous.open(store, path=path)
         assert isinstance(group, Group)
         for array in group.array_values():
@@ -1208,13 +1231,13 @@ async def test_metadata_validation_error() -> None:
         MetadataValidationError,
         match="Invalid value for 'zarr_format'. Expected '2, 3, or None'. Got '3.0'.",
     ):
-        await zarr.api.asynchronous.open_group(zarr_format="3.0")  # type: ignore [arg-type]
+        await zarr.api.asynchronous.open_group(zarr_format="3.0")  # type: ignore[arg-type]
 
     with pytest.raises(
         MetadataValidationError,
         match="Invalid value for 'zarr_format'. Expected '2, 3, or None'. Got '3.0'.",
     ):
-        await zarr.api.asynchronous.open_array(shape=(1,), zarr_format="3.0")  # type: ignore [arg-type]
+        await zarr.api.asynchronous.open_array(shape=(1,), zarr_format="3.0")  # type: ignore[arg-type]
 
 
 @pytest.mark.parametrize(
@@ -1224,7 +1247,7 @@ async def test_metadata_validation_error() -> None:
 )
 def test_open_array_with_mode_r_plus(store: Store, zarr_format: ZarrFormat) -> None:
     # 'r+' means read/write (must exist)
-    with pytest.raises(FileNotFoundError):
+    with pytest.raises(ArrayNotFoundError):
         zarr.open_array(store=store, mode="r+", zarr_format=zarr_format)
     zarr.ones(store=store, shape=(3, 3), zarr_format=zarr_format)
     z2 = zarr.open_array(store=store, mode="r+")