Merge pull request #40 from p2p-ld/bugfix-numpy-str

Support np.str_ dtype annotations, properly check tuple dtypes

docs/changelog.md

@@ -1,6 +1,10 @@
 # Changelog
 
-## Upcoming
+## 1.*
+
+### 1.6.*
+
+#### 1.6.6 - 24-12-13
 
 **Bugfix**
 - [#38](https://github.com/p2p-ld/numpydantic/issues/38), [#39](https://github.com/p2p-ld/numpydantic/pull/39) -
@@ -8,6 +12,11 @@
     An additional check was added for presence of `__name__` when embedding.
   - `NDArray` types were incorrectly cached s.t. pipe-union dtypes were considered equivalent to `Union[]`
     dtypes. An additional tuple with the type of the args was added to the cache key to disambiguate them.
+- [#38](https://github.com/p2p-ld/numpydantic/issues/38), [#40](https://github.com/p2p-ld/numpydantic/pull/40) - 
+  - Tuple dtypes were naively checked by just testing for whether the given dtype was contained by the tuple,
+    ignoring special cases like string type checking. Tuple dtypes are now checked recursively with the same
+    logic as all other type checking.
+  - Zarr treats `dtype=str` as numpy type `O` - added special case when validating from JSON to cast to `np.str_`
 
 **Testing**
 - [#39](https://github.com/p2p-ld/numpydantic/pull/39) - Test that all combinations of shapes, dtypes, and interfaces
@@ -15,10 +24,8 @@
 - [#39](https://github.com/p2p-ld/numpydantic/pull/39) - Add python 3.13 to the testing matrix.
 - [#39](https://github.com/p2p-ld/numpydantic/pull/39) - Add an additional `marks` field to ValidationCase
   for finer-grained control over running tests.
-
+- [#40](https://github.com/p2p-ld/numpydantic/pull/40) - Explicitly test for `np.str_` annotation dtypes alone and 
-## 1.*
+  in tuples.
-
-### 1.6.*
 
 #### 1.6.5 - 24-12-04 - Bump Pydantic Minimum
 

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "numpydantic"
-version = "1.6.5"
+version = "1.6.6"
 description = "Type and shape validation and serialization for arbitrary array types in pydantic models"
 authors = [
     {name = "sneakers-the-rat", email = "sneakers-the-rat@protonmail.com"},

src/numpydantic/interface/zarr.py

@@ -74,7 +74,8 @@ class ZarrJsonDict(JsonDict):
         if self.file:
             array = ZarrArrayPath(file=self.file, path=self.path)
         else:
-            array = zarr.array(self.value, dtype=self.dtype)
+            dtype = np.str_ if self.dtype == "str" else self.dtype
+            array = zarr.array(self.value, dtype=dtype)
         return array
 
 

src/numpydantic/testing/cases.py

@@ -126,6 +126,27 @@ DTYPE_CASES = [
     ValidationCase(annotation_dtype=str, dtype=str, passes=True, id="str-str"),
     ValidationCase(annotation_dtype=str, dtype=int, passes=False, id="str-int"),
     ValidationCase(annotation_dtype=str, dtype=float, passes=False, id="str-float"),
+    ValidationCase(
+        annotation_dtype=np.str_,
+        dtype=str,
+        passes=True,
+        id="np_str-str",
+        marks={"np_str", "str"},
+    ),
+    ValidationCase(
+        annotation_dtype=np.str_,
+        dtype=np.str_,
+        passes=True,
+        id="np_str-np_str",
+        marks={"np_str", "str"},
+    ),
+    ValidationCase(
+        annotation_dtype=(int, np.str_),
+        dtype=str,
+        passes=True,
+        id="tuple_np_str-str",
+        marks={"np_str", "str", "tuple"},
+    ),
     ValidationCase(
         annotation_dtype=BasicModel, dtype=BasicModel, passes=True, id="model-model"
     ),

src/numpydantic/testing/interfaces.py

@@ -75,6 +75,8 @@ class HDF5Case(_HDF5MetaCase):
             data = np.array(array, dtype=dtype)
         elif dtype is str:
             data = generator.random(shape).astype(bytes)
+        elif dtype is np.str_:
+            data = generator.random(shape).astype("S32")
         elif dtype is datetime:
             data = np.empty(shape, dtype="S32")
             data.fill(datetime.now(timezone.utc).isoformat().encode("utf-8"))
@@ -106,7 +108,7 @@ class HDF5CompoundCase(_HDF5MetaCase):
         array_path = "/" + "_".join([str(s) for s in shape]) + "__" + dtype.__name__
         if array is not None:
             data = np.array(array, dtype=dtype)
-        elif dtype is str:
+        elif dtype in (str, np.str_):
             dt = np.dtype([("data", np.dtype("S10")), ("extra", "i8")])
             data = np.array([("hey", 0)] * np.prod(shape), dtype=dt).reshape(shape)
         elif dtype is datetime:

src/numpydantic/validation/dtype.py

@@ -32,7 +32,7 @@ def validate_dtype(dtype: Any, target: DtypeType) -> bool:
         return True
 
     if isinstance(target, tuple):
-        valid = dtype in target
+        valid = any(validate_dtype(dtype, target_dt) for target_dt in target)
     elif is_union(target):
         valid = any(
             [validate_dtype(dtype, target_dt) for target_dt in get_args(target)]