Merge pull request #138 from d-v-b/fix/deserialize-in-one-place

fix premature model construction

Author: d-v-b

docs/experimental/index.md

@@ -84,6 +84,7 @@ array = ArraySpec(
     chunk_grid={"name": "regular", "configuration": {"chunk_shape": (1,)}},
     chunk_key_encoding = {"name": "default"},
     fill_value = 0,
+    attributes={}
     )
 
 class MyMembers(TypedDict):

src/pydantic_zarr/experimental/v2.py

@@ -695,7 +695,7 @@ def from_zarr(cls, group: zarr.Group, *, depth: int = -1) -> Self:
 
         result: GroupSpec
         attributes = group.attrs.asdict()
-        members: dict[str, object] = {}
+        members: dict[str, dict[str, object]] = {}
 
         if depth < -1:
             msg = (

src/pydantic_zarr/experimental/v3.py

@@ -20,7 +20,7 @@
 import numpy as np
 import numpy.typing as npt
 from packaging.version import Version
-from pydantic import BeforeValidator, Field, field_validator, model_validator
+from pydantic import BeforeValidator, field_validator, model_validator
 from typing_extensions import TypedDict
 
 from pydantic_zarr.experimental.core import (
@@ -189,7 +189,7 @@ class ArraySpec(NodeSpec):
     """
 
     node_type: Literal["array"] = "array"
-    attributes: BaseAttributes = Field(default_factory=dict)  # type: ignore[arg-type]
+    attributes: BaseAttributes
     shape: tuple[int, ...]
     data_type: DTypeLike
     chunk_grid: RegularChunking  # todo: validate this against shape
@@ -799,7 +799,7 @@ def from_zarr(cls, group: zarr.Group, *, depth: int = -1) -> Self:
 
         result: GroupSpec
         attributes = group.attrs.asdict()
-        members: dict[str, ArraySpec | GroupSpec] = {}
+        members: dict[str, dict[str, object]] = {}
 
         if depth < -1:
             msg = (
@@ -812,9 +812,9 @@ def from_zarr(cls, group: zarr.Group, *, depth: int = -1) -> Self:
         new_depth = max(depth - 1, -1)
         for name, item in group.members():
             if isinstance(item, zarr.Array):
-                members[name] = ArraySpec.from_zarr(item)
+                members[name] = ArraySpec.from_zarr(item).model_dump()
             elif isinstance(item, zarr.Group):
-                members[name] = GroupSpec.from_zarr(item, depth=new_depth)
+                members[name] = GroupSpec.from_zarr(item, depth=new_depth).model_dump()
             else:
                 msg = (  # type: ignore[unreachable]
                     f"Unparsable object encountered: {type(item)}. Expected zarr.Array"

tests/test_pydantic_zarr/conftest.py

@@ -2,13 +2,16 @@
 
 import warnings
 from dataclasses import dataclass
-from importlib.metadata import PackageNotFoundError, version
+from importlib.metadata import version
+from importlib.util import find_spec
 
 from packaging.version import Version
 
-try:
+ZARR_AVAILABLE = find_spec("zarr") is not None
+
+if ZARR_AVAILABLE:
     ZARR_PYTHON_VERSION = Version(version("zarr"))
-except PackageNotFoundError:
+else:
     ZARR_PYTHON_VERSION = Version("0.0.0")
 
 DTYPE_EXAMPLES_V2: tuple[DTypeExample, ...]

tests/test_pydantic_zarr/test_experimental/test_v2.py

@@ -19,7 +19,7 @@
 from pydantic_zarr.core import tuplify_json
 from pydantic_zarr.experimental.core import json_eq
 
-from ..conftest import DTYPE_EXAMPLES_V2, ZARR_PYTHON_VERSION, DTypeExample
+from ..conftest import DTYPE_EXAMPLES_V2, ZARR_AVAILABLE, ZARR_PYTHON_VERSION, DTypeExample
 
 if TYPE_CHECKING:
     from numcodecs.abc import Codec
@@ -598,6 +598,48 @@ def test_mix_v3_v2_fails() -> None:
         GroupSpec.from_flat(members_flat)  # type: ignore[arg-type]
 
 
+@pytest.mark.skipif(not ZARR_AVAILABLE, reason="zarr-python is not installed")
+def test_typed_members() -> None:
+    """
+    Test GroupSpec creation with typed members
+    """
+    array1d = ArraySpec(
+        shape=(1,),
+        dtype="uint8",
+        chunks=(1,),
+        fill_value=0,
+        compressor=None,
+        attributes={},
+    )
+
+    class DatasetMembers(TypedDict):
+        x: ArraySpec
+        y: ArraySpec
+
+    class DatasetGroup(GroupSpec):
+        members: DatasetMembers
+
+    class ExpectedMembers(TypedDict):
+        r10m: DatasetGroup
+        r20m: DatasetGroup
+
+    class ExpectedGroup(GroupSpec):
+        members: ExpectedMembers
+
+    flat = {
+        "": BaseGroupSpec(attributes={}),
+        "/r10m": BaseGroupSpec(attributes={}),
+        "/r20m": BaseGroupSpec(attributes={}),
+        "/r10m/x": array1d,
+        "/r10m/y": array1d,
+        "/r20m/x": array1d,
+        "/r20m/y": array1d,
+    }
+
+    zg = GroupSpec.from_flat(flat).to_zarr({}, path="")
+    ExpectedGroup.from_zarr(zg)
+
+
 def test_arrayspec_with_methods() -> None:
     """
     Test that ArraySpec with_* methods create new validated copies

tests/test_pydantic_zarr/test_experimental/test_v3.py

@@ -1,14 +1,13 @@
 from __future__ import annotations
 
-import importlib
-import importlib.util
 import json
 import re
 from dataclasses import asdict
 
 import numpy as np
 import pytest
 from pydantic import ValidationError
+from typing_extensions import TypedDict
 
 from pydantic_zarr.experimental.core import json_eq
 from pydantic_zarr.experimental.v3 import (
@@ -23,9 +22,7 @@
     auto_codecs,
 )
 
-from ..conftest import DTYPE_EXAMPLES_V3, DTypeExample
-
-ZARR_AVAILABLE = importlib.util.find_spec("zarr") is not None
+from ..conftest import DTYPE_EXAMPLES_V3, ZARR_AVAILABLE, DTypeExample
 
 
 @pytest.mark.parametrize("invalid_dimension_names", [[], "hi", ["1", 2, None]], ids=str)
@@ -304,6 +301,50 @@ def test_dim_names_from_zarr_array(
     assert spec.dimension_names == expected_names
 
 
+@pytest.mark.skipif(not ZARR_AVAILABLE, reason="zarr-python is not installed")
+def test_typed_members() -> None:
+    """
+    Test GroupSpec creation with typed members
+    """
+
+    array1d = ArraySpec(
+        shape=(1,),
+        data_type="uint8",
+        chunk_grid={"name": "regular", "configuration": {"chunk_shape": (1,)}},
+        chunk_key_encoding={"name": "default", "configuration": {"separator": "/"}},
+        fill_value=0,
+        codecs=({"name": "bytes"},),
+        attributes={},
+    )
+
+    class DatasetMembers(TypedDict):
+        x: ArraySpec
+        y: ArraySpec
+
+    class DatasetGroup(GroupSpec):
+        members: DatasetMembers
+
+    class ExpectedMembers(TypedDict):
+        r10m: DatasetGroup
+        r20m: DatasetGroup
+
+    class ExpectedGroup(GroupSpec):
+        members: ExpectedMembers
+
+    flat = {
+        "": BaseGroupSpec(attributes={}),
+        "/r10m": BaseGroupSpec(attributes={}),
+        "/r20m": BaseGroupSpec(attributes={}),
+        "/r10m/x": array1d,
+        "/r10m/y": array1d,
+        "/r20m/x": array1d,
+        "/r20m/y": array1d,
+    }
+
+    zg = GroupSpec.from_flat(flat).to_zarr({}, path="")
+    ExpectedGroup.from_zarr(zg)
+
+
 def test_arrayspec_with_methods() -> None:
     """
     Test that ArraySpec with_* methods create new validated copies