make eval functions async

PiperOrigin-RevId: 756106627

tests/integration/test_evalute_agent_in_fixture.py

@@ -51,12 +51,13 @@ def agent_eval_artifacts_in_fixture():
   return agent_eval_artifacts
 
 
+@pytest.mark.asyncio
 @pytest.mark.parametrize(
     'agent_name, evalfile, initial_session_file',
     agent_eval_artifacts_in_fixture(),
     ids=[agent_name for agent_name, _, _ in agent_eval_artifacts_in_fixture()],
 )
-def test_evaluate_agents_long_running_4_runs_per_eval_item(
+async def test_evaluate_agents_long_running_4_runs_per_eval_item(
     agent_name, evalfile, initial_session_file
 ):
   """Test agents evaluation in fixture folder.
@@ -66,7 +67,7 @@ def test_evaluate_agents_long_running_4_runs_per_eval_item(
 
   A single eval item is a session that can have multiple queries in it.
   """
-  AgentEvaluator.evaluate(
+  await AgentEvaluator.evaluate(
       agent_module=agent_name,
       eval_dataset_file_path_or_dir=evalfile,
       initial_session_file=initial_session_file,

tests/integration/test_multi_agent.py

@@ -15,7 +15,8 @@
 from google.adk.evaluation import AgentEvaluator
 
 
-def test_eval_agent():
+@pytest.mark.asyncio
+async def test_eval_agent():
   AgentEvaluator.evaluate(
       agent_module="tests.integration.fixture.trip_planner_agent",
       eval_dataset_file_path_or_dir=(

tests/integration/test_multi_turn.py

@@ -15,7 +15,8 @@
 from google.adk.evaluation import AgentEvaluator
 
 
-def test_simple_multi_turn_conversation():
+@pytest.mark.asyncio
+async def test_simple_multi_turn_conversation():
   """Test a simple multi-turn conversation."""
   AgentEvaluator.evaluate(
       agent_module="tests.integration.fixture.home_automation_agent",
@@ -24,7 +25,8 @@ def test_simple_multi_turn_conversation():
   )
 
 
-def test_dependent_tool_calls():
+@pytest.mark.asyncio
+async def test_dependent_tool_calls():
   """Test subsequent tool calls that are dependent on previous tool calls."""
   AgentEvaluator.evaluate(
       agent_module="tests.integration.fixture.home_automation_agent",
@@ -33,8 +35,10 @@ def test_dependent_tool_calls():
   )
 
 
-def test_memorizing_past_events():
+@pytest.mark.asyncio
+async def test_memorizing_past_events():
   """Test memorizing past events."""
+
   AgentEvaluator.evaluate(
       agent_module="tests.integration.fixture.home_automation_agent",
       eval_dataset_file_path_or_dir="tests/integration/fixture/home_automation_agent/test_files/memorizing_past_events/eval_data.test.json",

tests/integration/test_single_agent.py

@@ -15,7 +15,8 @@
 from google.adk.evaluation import AgentEvaluator
 
 
-def test_eval_agent():
+@pytest.mark.asyncio
+async def test_eval_agent():
   AgentEvaluator.evaluate(
       agent_module="tests.integration.fixture.home_automation_agent",
       eval_dataset_file_path_or_dir="tests/integration/fixture/home_automation_agent/simple_test.test.json",

tests/integration/test_sub_agent.py

@@ -15,7 +15,8 @@
 from google.adk.evaluation import AgentEvaluator
 
 
-def test_eval_agent():
+@pytest.mark.asyncio
+async def test_eval_agent():
   """Test hotel sub agent in a multi-agent system."""
   AgentEvaluator.evaluate(
       agent_module="tests.integration.fixture.trip_planner_agent",

tests/integration/test_with_test_file.py

@@ -15,7 +15,8 @@
 from google.adk.evaluation import AgentEvaluator
 
 
-def test_with_single_test_file():
+@pytest.mark.asyncio
+async def test_with_single_test_file():
   """Test the agent's basic ability via session file."""
   AgentEvaluator.evaluate(
       agent_module="tests.integration.fixture.home_automation_agent",
@@ -23,7 +24,8 @@ def test_with_single_test_file():
   )
 
 
-def test_with_folder_of_test_files_long_running():
+@pytest.mark.asyncio
+async def test_with_folder_of_test_files_long_running():
   """Test the agent's basic ability via a folder of session files."""
   AgentEvaluator.evaluate(
       agent_module="tests.integration.fixture.home_automation_agent",