Added simple benchmarks

.travis.yml

@@ -23,7 +23,7 @@ before_install:
 install:
 - |
   if [ "$TEST_TYPE" = build ]; then
-    pip install pytest pytest-cov coveralls six
+    pip install pytest pytest-cov pytest-benchmark coveralls six
     pip install -e .
     python setup.py develop
   elif [ "$TEST_TYPE" = lint ]; then

graphene/types/tests/test_objecttype.py

@@ -153,3 +153,9 @@ def test_objecttype_as_container_invalid_kwargs():
         Container(unexisting_field="3")
 
     assert "'unexisting_field' is an invalid keyword argument for Container" == str(excinfo.value)
+
+
+def test_objecttype_container_benchmark(benchmark):
+    @benchmark
+    def create_objecttype():
+        Container(field1='field1', field2='field2')

graphene/types/tests/test_query.py

@@ -1,8 +1,11 @@
+from functools import partial
 
+from graphql import execute, Source, parse
 
 from ..objecttype import ObjectType
-from ..scalars import String
+from ..scalars import String, Int
 from ..schema import Schema
+from ..structures import List
 
 
 def test_query():
@@ -50,3 +53,117 @@ def test_query_middlewares():
     executed = hello_schema.execute('{ hello, other }')
     assert not executed.errors
     assert executed.data == {'hello': 'dlroW', 'other': 'rehto'}
+
+
+def test_big_list_query_benchmark(benchmark):
+    big_list = range(10000)
+
+    class Query(ObjectType):
+        all_ints = List(Int)
+
+        def resolve_all_ints(self, args, context, info):
+            return big_list
+
+    hello_schema = Schema(Query)
+
+    big_list_query = partial(hello_schema.execute, '{ allInts }')
+    result = benchmark(big_list_query)
+    assert not result.errors
+    assert result.data == {'allInts': big_list}
+
+
+def test_big_list_query_compiled_query_benchmark(benchmark):
+    big_list = range(10000)
+
+    class Query(ObjectType):
+        all_ints = List(Int)
+
+        def resolve_all_ints(self, args, context, info):
+            return big_list
+
+    hello_schema = Schema(Query)
+    source = Source('{ allInts }')
+    query_ast = parse(source)
+
+    big_list_query = partial(execute, hello_schema, query_ast)
+    result = benchmark(big_list_query)
+    assert not result.errors
+    assert result.data == {'allInts': big_list}
+
+
+def test_big_list_of_containers_query_benchmark(benchmark):
+    big_list = range(10000)
+
+    class Container(ObjectType):
+        x = Int()
+
+    class Query(ObjectType):
+        all_containers = List(Container)
+
+        def resolve_all_containers(self, args, context, info):
+            return (Container(x=x) for x in big_list)
+
+    hello_schema = Schema(Query)
+
+    big_list_query = partial(hello_schema.execute, '{ allContainers { x } }')
+    result = benchmark(big_list_query)
+    assert not result.errors
+    assert result.data == {'allContainers': [{'x': x} for x in big_list]}
+
+
+def test_big_list_of_containers_multiple_fields_query_benchmark(benchmark):
+    big_list = range(10000)
+
+    class Container(ObjectType):
+        x = Int()
+        y = Int()
+        z = Int()
+        o = Int()
+
+    class Query(ObjectType):
+        all_containers = List(Container)
+
+        def resolve_all_containers(self, args, context, info):
+            return (Container(x=x, y=x, z=x, o=x) for x in big_list)
+
+    hello_schema = Schema(Query)
+
+    big_list_query = partial(hello_schema.execute, '{ allContainers { x, y, z, o } }')
+    result = benchmark(big_list_query)
+    assert not result.errors
+    assert result.data == {'allContainers': [{'x': x, 'y':x, 'z':x, 'o': x} for x in big_list]}
+
+
+def test_big_list_of_containers_multiple_fields_custom_resolvers_query_benchmark(benchmark):
+    big_list = range(10000)
+
+    class Container(ObjectType):
+        x = Int()
+        y = Int()
+        z = Int()
+        o = Int()
+
+        def resolve_x(self, args, context, info):
+            return self.x
+
+        def resolve_y(self, args, context, info):
+            return self.y
+
+        def resolve_z(self, args, context, info):
+            return self.z
+
+        def resolve_o(self, args, context, info):
+            return self.o
+
+    class Query(ObjectType):
+        all_containers = List(Container)
+
+        def resolve_all_containers(self, args, context, info):
+            return (Container(x=x, y=x, z=x, o=x) for x in big_list)
+
+    hello_schema = Schema(Query)
+
+    big_list_query = partial(hello_schema.execute, '{ allContainers { x, y, z, o } }')
+    result = benchmark(big_list_query)
+    assert not result.errors
+    assert result.data == {'allContainers': [{'x': x, 'y':x, 'z':x, 'o': x} for x in big_list]}

setup.py

@@ -76,6 +76,7 @@ setup(
     ],
     tests_require=[
         'pytest>=2.7.2',
+        'pytest-benchmark',
         'mock',
     ],
     extras_require={