def test_bytes():
sch1 = '''{
"type": "record",
"name": "Test",
"fields": [
{"name": "bytes_field", "type": "bytes"}
]
}'''
sch2 = '''{
"type": "record",
"name": "Test",
"fields": [
{"name": "bytes_field", "type": "bytes"}
]
}'''
serializer = pyavroc.AvroSerializer(sch1)
deserializer = pyavroc.AvroDeserializer(sch2)
data = {"bytes_field": b'some bytes'}
ser_python_avro = Serializer(sch1).serialize(data)
res_pyavroc = deserializer.deserialize(ser_python_avro, writer_schema=sch1)
res_python_avro = Deserializer(sch2).deserialize(ser_python_avro)
assert res_pyavroc['bytes_field'] == res_python_avro[
'bytes_field'] == b'some bytes'
def test_unicode_map_keys():
schema = """\
{"type": "record",
"name": "foo",
"fields": [{"name": "bar",
"type": ["null", {"type": "map", "values": "string"}]}]}
"""
serializer = pyavroc.AvroSerializer(schema)
rec_bytes = serializer.serialize({"bar": {"k": "v"}})
assert serializer.serialize({"bar": {u"k": "v"}}) == rec_bytes
def test_serialize_utf8_string():
schema = '["string"]'
serializer = pyavroc.AvroSerializer(schema)
deserializer = Deserializer(schema)
if sys.version_info < (3, ):
datum = "barà"
rec_bytes = serializer.serialize(datum)
assert deserializer.deserialize(rec_bytes) == unicode(datum, "utf-8")
datum = u"barà"
rec_bytes = serializer.serialize(datum)
assert deserializer.deserialize(rec_bytes) == datum
def test_serialize_record():
n_recs = 10
avtypes = pyavroc.create_types(SCHEMA)
serializer = pyavroc.AvroSerializer(SCHEMA)
deserializer = Deserializer(SCHEMA)
for i in range(n_recs):
name, office = "name-%d" % i, "office-%d" % i
avro_obj = avtypes.User(name=name, office=office)
rec_bytes = serializer.serialize(avro_obj)
deser_rec = deserializer.deserialize(rec_bytes)
assert set(deser_rec) == set(['name', 'office', 'favorite_number'])
assert deser_rec['name'] == name
assert deser_rec['office'] == office
assert deser_rec['favorite_number'] is None
def test_serialize_reuse_record_type():
schema = """\
{ "type": "record",
"name": "foo",
"namespace": "org.pyavroc",
"fields": [
{ "name": "c1",
"type": { "type": "record", "name": "Contig", "fields": [ { "name": "contigName", "type": "string" } ] }
},
{ "name": "c2",
"type": [ "null", "Contig" ]
}
]
}
"""
ser = pyavroc.AvroSerializer(schema)
datum = {"c1": {"contigName": "contig1"}, "c2": {"contigName": "contig2"}}
obytes = ser.serialize(datum)
assert obytes
def test_write_wrong_type_primitive():
schema = '''{
"type": "record",
"name": "Obj",
"fields": [
{"name": "string", "type": "string"},
{"name": "number", "type": "int"}
]
}'''
avtypes = pyavroc.create_types(schema)
serializer = pyavroc.AvroSerializer(schema)
# this shouldn't raise
serializer.serialize(avtypes.Obj(string="pippo", number=1))
# place an int in place of a str
u = avtypes.Obj(string=1, number=1)
with pytest.raises(TypeError):
serializer.serialize(u)
# string in place of int
u = avtypes.Obj(string="a", number="a")
with pytest.raises(TypeError):
serializer.serialize(u)
def test_serialize_union():
schema = '["string", "null"]'
serializer = pyavroc.AvroSerializer(schema)
deserializer = Deserializer(schema)
for datum in "foo", u"foo", None:
assert deserializer.deserialize(serializer.serialize(datum)) == datum
def test_big():
avtypes = pyavroc.create_types(SCHEMA)
serializer = pyavroc.AvroSerializer(SCHEMA)
long_str = 'X' * (10 * 1024 * 1024)
avro_obj = avtypes.User(name=long_str, office=long_str)
serializer.serialize(avro_obj)
def test_exc():
with pytest.raises(TypeError):
pyavroc.AvroSerializer(1)
with pytest.raises(IOError):
pyavroc.AvroSerializer('NOT_A_VALID_JSON')