def test_basic_sr_avro_encoder_decode_readers_schema(self, **kwargs):
schemaregistry_fully_qualified_namespace = kwargs.pop("schemaregistry_fully_qualified_namespace")
sr_client = self.create_client(fully_qualified_namespace=schemaregistry_fully_qualified_namespace)
schemaregistry_group = kwargs.pop("schemaregistry_group")
sr_avro_encoder = AvroEncoder(client=sr_client, group_name=schemaregistry_group, auto_register=True)
schema_str = """{"namespace":"example.avro","type":"record","name":"User","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]},{"name":"favorite_color","type":["string","null"]}]}"""
dict_content = {"name": u"Ben", "favorite_number": 7, "favorite_color": u"red"}
encoded_message_content = sr_avro_encoder.encode(dict_content, schema=schema_str)
content_type = encoded_message_content["content_type"]
encoded_content = encoded_message_content["content"]
# readers_schema with removed field
readers_schema_remove_field = """{"namespace":"example.avro","type":"record","name":"User","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]}]}"""
encoded_content_dict = {"content": encoded_content, "content_type": content_type}
decoded_content = sr_avro_encoder.decode(encoded_content_dict, readers_schema=readers_schema_remove_field)
assert decoded_content["name"] == u"Ben"
assert decoded_content["favorite_number"] == 7
# readers_schema with extra field with default
readers_schema_extra_field = """{"namespace":"example.avro","type":"record","name":"User","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]},{"name":"favorite_color","type":["string","null"]}, {"name":"favorite_city","type":["string","null"], "default": "Redmond"}]}"""
decoded_content = sr_avro_encoder.decode(encoded_content_dict, readers_schema=readers_schema_extra_field)
assert decoded_content["name"] == u"Ben"
assert decoded_content["favorite_number"] == 7
assert decoded_content["favorite_color"] == "red"
assert decoded_content["favorite_city"] == "Redmond"
# readers_schema with changed name results in error
readers_schema_change_name = """{"namespace":"fakeexample.avro","type":"record","name":"fake_user","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]},{"name":"favorite_color","type":["string","null"]}]}"""
with pytest.raises(InvalidSchemaError) as e:
decoded_content = sr_avro_encoder.decode(encoded_content_dict, readers_schema=readers_schema_change_name)
assert "Incompatible schemas" in e.value.message
assert "schema_id" in e.value.details
assert "schema_definition" in e.value.details
# invalid readers_schema
invalid_schema = {
"name":"User",
"type":"record",
"namespace":"example.avro",
"fields":[{"name":"name","type":"string"}]
}
invalid_schema_string = "{}".format(invalid_schema)
with pytest.raises(InvalidSchemaError) as e:
decoded_content = sr_avro_encoder.decode(encoded_content_dict, readers_schema=invalid_schema_string)
assert "Invalid schema" in e.value.message
assert "schema_id" in e.value.details
assert "schema_definition" in e.value.details
def test_encode_record(self, **kwargs):
schemaregistry_fully_qualified_namespace = kwargs.pop("schemaregistry_fully_qualified_namespace")
schemaregistry_group = kwargs.pop("schemaregistry_group")
sr_client = self.create_client(fully_qualified_namespace=schemaregistry_fully_qualified_namespace)
sr_avro_encoder = AvroEncoder(client=sr_client, group_name=schemaregistry_group, auto_register=True)
# add below to schema fields later if needed
# {"name":"example.innerrec","type":"record","fields":[{"name":"a","type":"int"}]},
# {"name":"innerenum","type":"enum","symbols":["FOO", "BAR"]},
# {"name":"innerarray","type":"array","items":"int"},
# {"name":"innermap","type":"map","values":"int"},
# {"name":"innerfixed","type":"fixed","size":74}
schema_record = """{
"name":"User",
"namespace":"example.avro.populatedrecord",
"type":"record",
"fields":[
{"name":"name","type":"string"},
{"name":"age","type":"int"},
{"name":"married","type":"boolean"},
{"name":"height","type":"float"},
{"name":"randb","type":"bytes"}
]
}"""
content = {
"name": u"Ben",
"age": 3,
"married": False,
"height": 13.5,
"randb": b"\u00FF"
}
encoded_message_content = sr_avro_encoder.encode(content, schema=schema_record)
decoded_content = sr_avro_encoder.decode(encoded_message_content)
assert decoded_content == content
def test_basic_sr_avro_encoder_without_auto_register_schemas(self, **kwargs):
schemaregistry_fully_qualified_namespace = kwargs.pop("schemaregistry_fully_qualified_namespace")
schemaregistry_group = kwargs.pop("schemaregistry_group")
sr_client = self.create_client(fully_qualified_namespace=schemaregistry_fully_qualified_namespace)
sr_avro_encoder = AvroEncoder(client=sr_client, group_name=schemaregistry_group)
schema_str = """{"namespace":"example.avro","type":"record","name":"User","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]},{"name":"favorite_color","type":["string","null"]}]}"""
schema = avro.schema.parse(schema_str)
dict_content = {"name": u"Ben", "favorite_number": 7, "favorite_color": u"red"}
encoded_message_content = sr_avro_encoder.encode(dict_content, schema=schema_str)
content_type = encoded_message_content["content_type"]
encoded_content = encoded_message_content["content"]
assert content_type.split("+")[0] == 'avro/binary'
schema_id = sr_client.get_schema_properties(schemaregistry_group, schema.fullname, str(schema), "Avro").id
assert content_type.split("+")[1] == schema_id
encoded_content_dict = {"content": encoded_content, "content_type": content_type}
decoded_content = sr_avro_encoder.decode(encoded_content_dict)
assert decoded_content["name"] == u"Ben"
assert decoded_content["favorite_number"] == 7
assert decoded_content["favorite_color"] == u"red"
sr_avro_encoder.close()
def test_basic_sr_avro_encoder_decode_readers_schema(
self, schemaregistry_fully_qualified_namespace,
schemaregistry_group, **kwargs):
sr_client = self.create_basic_client(
SchemaRegistryClient,
fully_qualified_namespace=schemaregistry_fully_qualified_namespace)
sr_avro_encoder = AvroEncoder(client=sr_client,
group_name=schemaregistry_group,
auto_register_schemas=True)
schema_str = """{"namespace":"example.avro","type":"record","name":"User","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]},{"name":"favorite_color","type":["string","null"]}]}"""
dict_data = {
"name": u"Ben",
"favorite_number": 7,
"favorite_color": u"red"
}
encoded_metadata = sr_avro_encoder.encode(dict_data, schema=schema_str)
content_type = encoded_metadata["content_type"]
encoded_data = encoded_metadata["data"]
# readers_schema with removed field
readers_schema_remove_field = """{"namespace":"example.avro","type":"record","name":"User","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]}]}"""
encoded_data_dict = {
"data": encoded_data,
"content_type": content_type
}
decoded_data = sr_avro_encoder.decode(
encoded_data_dict, readers_schema=readers_schema_remove_field)
assert decoded_data["name"] == u"Ben"
assert decoded_data["favorite_number"] == 7
# readers_schema with extra field with default
readers_schema_extra_field = """{"namespace":"example.avro","type":"record","name":"User","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]},{"name":"favorite_color","type":["string","null"]}, {"name":"favorite_city","type":["string","null"], "default": "Redmond"}]}"""
decoded_data = sr_avro_encoder.decode(
encoded_data_dict, readers_schema=readers_schema_extra_field)
assert decoded_data["name"] == u"Ben"
assert decoded_data["favorite_number"] == 7
assert decoded_data["favorite_color"] == "red"
assert decoded_data["favorite_city"] == "Redmond"
# readers_schema with changed name results in error
readers_schema_change_name = """{"namespace":"fakeexample.avro","type":"record","name":"fake_user","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]},{"name":"favorite_color","type":["string","null"]}]}"""
with pytest.raises(SchemaDecodeError):
decoded_data = sr_avro_encoder.decode(
encoded_data_dict, readers_schema=readers_schema_change_name)
def test_basic_sr_avro_encoder_with_request_options(self, **kwargs):
schemaregistry_fully_qualified_namespace = kwargs.pop("schemaregistry_fully_qualified_namespace")
schemaregistry_group = kwargs.pop("schemaregistry_group")
sr_client = self.create_client(fully_qualified_namespace=schemaregistry_fully_qualified_namespace)
sr_avro_encoder = AvroEncoder(client=sr_client, group_name=schemaregistry_group, auto_register=True)
schema_str = """{"namespace":"example.avro","type":"record","name":"User","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]},{"name":"favorite_color","type":["string","null"]}]}"""
dict_content = {"name": u"Ben", "favorite_number": 7, "favorite_color": u"red"}
with pytest.raises(TypeError) as e:
encoded_message_content = sr_avro_encoder.encode(dict_content, schema=schema_str, request_options={"fake_kwarg": True})
assert 'request() got an unexpected keyword' in str(e.value)
encoded_message_content = sr_avro_encoder.encode(dict_content, schema=schema_str)
content_type = encoded_message_content["content_type"]
encoded_content = encoded_message_content["content"]
encoded_content_dict = {"content": encoded_content, "content_type": content_type}
with pytest.raises(TypeError) as e:
decoded_content = sr_avro_encoder.decode(encoded_content_dict, request_options={"fake_kwarg": True})
assert 'request() got an unexpected keyword' in str(e.value)
def test_basic_sr_avro_encoder_with_auto_register_schemas(
self, schemaregistry_fully_qualified_namespace,
schemaregistry_group, **kwargs):
sr_client = self.create_basic_client(
SchemaRegistryClient,
fully_qualified_namespace=schemaregistry_fully_qualified_namespace)
sr_avro_encoder = AvroEncoder(client=sr_client,
group_name=schemaregistry_group,
auto_register_schemas=True)
schema_str = """{"namespace":"example.avro","type":"record","name":"User","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]},{"name":"favorite_color","type":["string","null"]}]}"""
schema = avro.schema.parse(schema_str)
dict_data = {
"name": u"Ben",
"favorite_number": 7,
"favorite_color": u"red"
}
encoded_metadata = sr_avro_encoder.encode(dict_data, schema=schema_str)
content_type = encoded_metadata["content_type"]
encoded_data = encoded_metadata["data"]
assert content_type.split("+")[0] == 'avro/binary'
schema_id = sr_client.get_schema_properties(schemaregistry_group,
schema.fullname,
str(schema), "Avro").id
assert content_type.split("+")[1] == schema_id
encoded_data_dict = {
"data": encoded_data,
"content_type": content_type
}
decoded_data = sr_avro_encoder.decode(encoded_data_dict)
assert decoded_data["name"] == u"Ben"
assert decoded_data["favorite_number"] == 7
assert decoded_data["favorite_color"] == u"red"
# check that AvroEncoder won't work with message types that don't follow protocols
class BadExample:
def __init__(self, not_data):
self.not_data = not_data
with pytest.raises(
SchemaEncodeError) as e: # caught avro SchemaParseError
sr_avro_encoder.encode({"name": u"Ben"},
schema=schema_str,
message_type=BadExample)
assert "subtype of the MessageType" in (str(e.value))
bad_ex = BadExample('fake')
with pytest.raises(
SchemaDecodeError) as e: # caught avro SchemaParseError
sr_avro_encoder.decode(message=bad_ex)
assert "subtype of the MessageType" in (str(e.value))
# check that AvroEncoder will work with message types that follow protocols
class GoodExample:
def __init__(self, data: bytes, content_type: str, **kwargs):
self.data = data
self.content_type = content_type
self.extra = kwargs.pop('extra', None)
def __message_data__(self):
return {"data": self.data, "content_type": self.content_type}
def good_callback(data: bytes, content_type: str, **kwargs):
return GoodExample(data, content_type, **kwargs)
good_ex_obj = sr_avro_encoder.encode(dict_data,
schema=schema_str,
message_type=GoodExample,
extra='val')
good_ex_callback = sr_avro_encoder.encode(dict_data,
schema=schema_str,
message_type=good_callback,
extra='val')
decoded_data_obj = sr_avro_encoder.decode(message=good_ex_obj)
decoded_data_callback = sr_avro_encoder.decode(
message=good_ex_callback)
assert decoded_data_obj["name"] == u"Ben"
assert decoded_data_obj["favorite_number"] == 7
assert decoded_data_obj["favorite_color"] == u"red"
assert decoded_data_callback["name"] == u"Ben"
assert decoded_data_callback["favorite_number"] == 7
assert decoded_data_callback["favorite_color"] == u"red"
sr_avro_encoder.close()
def test_basic_sr_avro_encoder_with_auto_register_schemas(self, **kwargs):
schemaregistry_fully_qualified_namespace = kwargs.pop("schemaregistry_fully_qualified_namespace")
schemaregistry_group = kwargs.pop("schemaregistry_group")
sr_client = self.create_client(fully_qualified_namespace=schemaregistry_fully_qualified_namespace)
sr_avro_encoder = AvroEncoder(client=sr_client, group_name=schemaregistry_group, auto_register=True)
schema_str = """{"namespace":"example.avro","type":"record","name":"User","fields":[{"name":"name","type":"string"},{"name":"favorite_number","type":["int","null"]},{"name":"favorite_color","type":["string","null"]}]}"""
schema = avro.schema.parse(schema_str)
dict_content = {"name": u"Ben", "favorite_number": 7, "favorite_color": u"red"}
encoded_message_content = sr_avro_encoder.encode(dict_content, schema=schema_str)
content_type = encoded_message_content["content_type"]
encoded_content = encoded_message_content["content"]
# wrong data type
dict_content_bad = {"name": u"Ben", "favorite_number": 7, "favorite_color": 7}
with pytest.raises(InvalidContentError) as e:
encoded_message_content = sr_avro_encoder.encode(dict_content_bad, schema=schema_str)
assert "schema_id" in e.value.details
assert content_type.split("+")[0] == 'avro/binary'
schema_id = sr_client.get_schema_properties(schemaregistry_group, schema.fullname, str(schema), "Avro").id
assert content_type.split("+")[1] == schema_id
encoded_content_dict = {"content": encoded_content, "content_type": content_type}
decoded_content = sr_avro_encoder.decode(encoded_content_dict)
assert decoded_content["name"] == u"Ben"
assert decoded_content["favorite_number"] == 7
assert decoded_content["favorite_color"] == u"red"
# bad content type
encoded_content_dict["content_type"] = 'a+b+c'
with pytest.raises(InvalidContentError) as e:
decoded_content = sr_avro_encoder.decode(encoded_content_dict)
# check that AvroEncoder won't work with message types that don't follow protocols
class BadExample:
def __init__(self, not_content):
self.not_content = not_content
with pytest.raises(TypeError) as e:
sr_avro_encoder.encode({"name": u"Ben"}, schema=schema_str, message_type=BadExample)
assert "subtype of the MessageType" in (str(e.value))
bad_ex = BadExample('fake')
with pytest.raises(TypeError) as e: # caught TypeError
sr_avro_encoder.decode(message=bad_ex)
assert "subtype of the MessageType" in (str(e.value))
# check that AvroEncoder will work with message types that follow protocols
class GoodExample:
def __init__(self, content, **kwargs):
self.content = content
self.content_type = None
self.extra = kwargs.pop('extra', None)
@classmethod
def from_message_content(cls, content: bytes, content_type: str, **kwargs):
ge = cls(content)
ge.content_type = content_type
return ge
def __message_content__(self):
return {"content": self.content, "content_type": self.content_type}
good_ex_obj = sr_avro_encoder.encode(dict_content, schema=schema_str, message_type=GoodExample, extra='val')
decoded_content_obj = sr_avro_encoder.decode(message=good_ex_obj)
assert decoded_content_obj["name"] == u"Ben"
assert decoded_content_obj["favorite_number"] == 7
assert decoded_content_obj["favorite_color"] == u"red"
sr_avro_encoder.close()