More robust tests (#2575)

Author: vblagoje

integrations/openrouter/tests/test_openrouter_chat_generator.py

@@ -339,7 +339,7 @@ def test_live_run_with_tools(self, tools):
         tool_call = message.tool_call
         assert isinstance(tool_call, ToolCall)
         assert tool_call.tool_name == "weather"
-        assert tool_call.arguments == {"city": "Paris"}
+        assert "paris" in tool_call.arguments["city"].lower(), f"Expected 'paris' in city: {tool_call.arguments}"
         assert message.meta["finish_reason"] == "tool_calls"
 
     @pytest.mark.skipif(
@@ -371,7 +371,12 @@ def test_live_run_with_tools_and_response(self, tools):
             assert tool_call.tool_name == "weather"
 
         arguments = [tool_call.arguments for tool_call in tool_calls]
-        assert sorted(arguments, key=lambda x: x["city"]) == [{"city": "Berlin"}, {"city": "Paris"}]
+        # Extract city names and check they contain the expected cities
+        # (LLM may return "Paris, France" or "Berlin, Germany" instead of just city names)
+        cities = [arg["city"].lower() for arg in arguments]
+        assert len(cities) == 2
+        assert any("berlin" in city for city in cities), f"Expected 'berlin' in one of {cities}"
+        assert any("paris" in city for city in cities), f"Expected 'paris' in one of {cities}"
         assert tool_message.meta["finish_reason"] == "tool_calls"
 
         new_messages = [
@@ -453,10 +458,9 @@ def test_pipeline_with_openrouter_chat_generator(self, tools):
             }
         )
 
-        assert (
-            "The weather in Paris is sunny and 32°C"
-            == results["tool_invoker"]["tool_messages"][0].tool_call_result.result
-        )
+        result = results["tool_invoker"]["tool_messages"][0].tool_call_result.result
+        assert "paris" in result.lower(), f"Expected 'paris' in result: {result}"
+        assert "sunny and 32°c" in result.lower(), f"Expected 'sunny and 32°c' in result: {result}"
 
     @pytest.mark.skipif(
         not os.environ.get("OPENROUTER_API_KEY", None),

integrations/openrouter/tests/test_openrouter_chat_generator_async.py

@@ -198,7 +198,7 @@ async def test_live_run_with_tools_and_response_async(self, tools):
         tool_call = tool_message.tool_call
         assert tool_call.id, "Tool call does not contain value for 'id' key"
         assert tool_call.tool_name == "weather"
-        assert tool_call.arguments == {"city": "Paris"}
+        assert "paris" in tool_call.arguments["city"].lower(), f"Expected 'paris' in city: {tool_call.arguments}"
         assert tool_message.meta["finish_reason"] == "tool_calls"
 
         new_messages = [
@@ -260,7 +260,7 @@ async def callback(chunk: StreamingChunk):
         tool_call = tool_message.tool_call
         assert tool_call.id, "Tool call does not contain value for 'id' key"
         assert tool_call.tool_name == "weather"
-        assert tool_call.arguments == {"city": "Paris"}
+        assert "paris" in tool_call.arguments["city"].lower(), f"Expected 'paris' in city: {tool_call.arguments}"
         assert tool_message.meta["finish_reason"] == "tool_calls"
 
     @pytest.mark.skipif(
@@ -314,7 +314,7 @@ def echo_function(text: str) -> str:
 
         # Pass mixed list: echo_tool (individual) and toolset (weather + time) at runtime
         # This tests that both individual tools and toolsets can be combined
-        messages = [ChatMessage.from_user("Echo this: Hello World")]
+        messages = [ChatMessage.from_user("Echo this via tool: Hello World")]
         results = await component.run_async(messages, tools=[echo_tool, toolset])
 
         assert len(results["replies"]) == 1