localstack-py-avro-schema 3.9.4__tar.gz → 3.9.6__tar.gz

{localstack_py_avro_schema-3.9.4 → localstack_py_avro_schema-3.9.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: localstack-py-avro-schema
-Version: 3.9.4
+Version: 3.9.6
 Summary: Generate Apache Avro schemas for Python types including standard library data-classes and Pydantic data models.
 Author-email: LocalStack Contributors <info@localstack.cloud>, "J.P. Morgan Chase & Co." <open_source@jpmorgan.com>
 License:                                  Apache License

{localstack_py_avro_schema-3.9.4 → localstack_py_avro_schema-3.9.6}/src/localstack_py_avro_schema.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: localstack-py-avro-schema
-Version: 3.9.4
+Version: 3.9.6
 Summary: Generate Apache Avro schemas for Python types including standard library data-classes and Pydantic data models.
 Author-email: LocalStack Contributors <info@localstack.cloud>, "J.P. Morgan Chase & Co." <open_source@jpmorgan.com>
 License:                                  Apache License

{localstack_py_avro_schema-3.9.4 → localstack_py_avro_schema-3.9.6}/src/py_avro_schema/_schemas.py

@@ -671,6 +671,33 @@ class SequenceSchema(Schema):
         return [self.items_schema.make_default(item) for item in py_default]
 
 
+@register_schema
+class SetSchema(SequenceSchema):
+    """An Avro array schema for a given Python set"""
+
+    @classmethod
+    def handles_type(cls, py_type: type) -> bool:
+        """Whether this schema class can represent a given Python class"""
+        py_type = _type_from_annotated(py_type)
+        origin = get_origin(py_type)
+        return _is_class(origin, collections.abc.MutableSet)
+
+    def __init__(
+        self,
+        py_type: type[collections.abc.MutableSet],
+        namespace: str | None = None,
+        options: Option = Option(0),
+    ):
+        """
+        An Avro array schema for a given Python sequence
+
+        :param py_type:   The Python class to generate a schema for.
+        :param namespace: The Avro namespace to add to schemas.
+        :param options:   Schema generation options.
+        """
+        super().__init__(py_type, namespace=namespace, options=options)  # type: ignore
+
+
 @register_schema
 class DictSchema(Schema):
     """An Avro map schema for a given Python mapping"""
@@ -860,6 +887,13 @@ class EnumSchema(NamedSchema):
             return False
         return True
 
+    def data(self, names: NamesType) -> JSONType:
+        """Return the schema data"""
+        # For invalid enums we don't want to deduplicate the data
+        if not self._is_valid_enum():
+            return self.data_before_deduplication(names=names)
+        return super().data(names)
+
     def data_before_deduplication(self, names: NamesType) -> JSONObj:
         """Return the schema data"""
         if not self._is_valid_enum():
@@ -1141,7 +1175,7 @@ class PlainClassSchema(RecordSchema):
         if field := self.signature_fields.get(name):
             _annotation, _default = field
             if actual_type is _annotation:
-                default = _default
+                default = _default or dataclasses.MISSING
         field_obj = RecordField(
             py_type=actual_type,
             name=name,

{localstack_py_avro_schema-3.9.4 → localstack_py_avro_schema-3.9.6}/tests/test_primitives.py

@@ -8,7 +8,7 @@
 # Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an
 # "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
 # specific language governing permissions and limitations under the License.
-
+import dataclasses
 import enum
 import re
 import sys
@@ -174,6 +174,12 @@ def test_string_list():
     assert_schema(py_type, expected)
 
 
+def test_string_set():
+    py_type = set[str]
+    expected = {"type": "array", "items": "string"}
+    assert_schema(py_type, expected)
+
+
 def test_string_list_annotated():
     py_type = Annotated[List[str], ...]
     expected = {"type": "array", "items": "string"}
@@ -192,6 +198,12 @@ def test_int_list():
     assert_schema(py_type, expected)
 
 
+def test_int_set():
+    py_type = set[int]
+    expected = {"type": "array", "items": "long"}
+    assert_schema(py_type, expected)
+
+
 def test_string_tuple():
     py_type = Tuple[str]
     expected = {"type": "array", "items": "string"}
@@ -214,6 +226,18 @@ def test_string_mutable_sequence():
     assert_schema(py_type, expected)
 
 
+def test_string_set_of_set():
+    py_type = set[set[str]]
+    expected = {
+        "type": "array",
+        "items": {
+            "type": "array",
+            "items": "string",
+        },
+    }
+    assert_schema(py_type, expected)
+
+
 def test_string_list_of_lists():
     py_type = List[List[str]]
     expected = {
@@ -519,3 +543,25 @@ def test_str_enum_invalid_name():
     expected = {"namedString": "OriginProtocolPolicy", "type": "string"}
 
     assert_schema(OriginProtocolPolicy, expected)
+
+
+def test_duplicated_invalid_enum():
+    class OriginProtocolPolicy(str, enum.Enum):
+        http_only = "http-only"
+        match_viewer = "match-viewer"
+        https_only = "https-only"
+
+    @dataclasses.dataclass
+    class PyType:
+        enum_one: OriginProtocolPolicy
+        enum_two: OriginProtocolPolicy
+
+    expected = {
+        "type": "record",
+        "name": "PyType",
+        "fields": [
+            {"type": {"type": "string", "namedString": "OriginProtocolPolicy"}, "name": "enum_one"},
+            {"type": {"type": "string", "namedString": "OriginProtocolPolicy"}, "name": "enum_two"},
+        ],
+    }
+    assert_schema(PyType, expected)