Fix chat completions fallback output indexes

Author: Aphroq

src/agents/models/chatcmpl_stream_handler.py

@@ -624,6 +624,7 @@ async def handle_stream(
             )
 
         # Send completion events for function calls
+        fallback_emitted_count = 0
         for index, function_call in state.function_calls.items():
             if state.function_call_streaming.get(index, False):
                 # Function call was streamed, just send the completion event
@@ -668,6 +669,7 @@ async def handle_stream(
                 fallback_starting_index += sum(
                     1 for streaming in state.function_call_streaming.values() if streaming
                 )
+                fallback_output_index = fallback_starting_index + fallback_emitted_count
 
                 # Build function call kwargs, include provider_data if present
                 fallback_func_call_kwargs: dict[str, Any] = {
@@ -690,23 +692,24 @@ async def handle_stream(
                 # Send all events at once (backward compatibility)
                 yield ResponseOutputItemAddedEvent(
                     item=ResponseFunctionToolCall(**fallback_func_call_kwargs),
-                    output_index=fallback_starting_index,
+                    output_index=fallback_output_index,
                     type="response.output_item.added",
                     sequence_number=sequence_number.get_and_increment(),
                 )
                 yield ResponseFunctionCallArgumentsDeltaEvent(
                     delta=function_call.arguments,
                     item_id=FAKE_RESPONSES_ID,
-                    output_index=fallback_starting_index,
+                    output_index=fallback_output_index,
                     type="response.function_call_arguments.delta",
                     sequence_number=sequence_number.get_and_increment(),
                 )
                 yield ResponseOutputItemDoneEvent(
                     item=ResponseFunctionToolCall(**fallback_func_call_kwargs),
-                    output_index=fallback_starting_index,
+                    output_index=fallback_output_index,
                     type="response.output_item.done",
                     sequence_number=sequence_number.get_and_increment(),
                 )
+                fallback_emitted_count += 1
 
         # Finally, send the Response completed event
         outputs: list[ResponseOutputItem] = []

tests/models/test_openai_chatcompletions_stream.py

@@ -553,3 +553,78 @@ async def patched_fetch_response(self, *args, **kwargs):
     assert isinstance(function_call_output, ResponseFunctionToolCall)
     assert function_call_output.name == "write_file"
     assert function_call_output.arguments == '{"filename": "test.py", "content": "print(hello)"}'
+
+
+@pytest.mark.allow_call_model_methods
+@pytest.mark.asyncio
+async def test_stream_response_fallback_tool_calls_use_distinct_output_indexes(monkeypatch) -> None:
+    tool_call_delta1 = ChoiceDeltaToolCall(
+        index=0,
+        function=ChoiceDeltaToolCallFunction(name="first_tool", arguments='{"a": 1}'),
+        type="function",
+    )
+    tool_call_delta2 = ChoiceDeltaToolCall(
+        index=1,
+        function=ChoiceDeltaToolCallFunction(name="second_tool", arguments='{"b": 2}'),
+        type="function",
+    )
+    chunk1 = ChatCompletionChunk(
+        id="chunk-id",
+        created=1,
+        model="fake",
+        object="chat.completion.chunk",
+        choices=[Choice(index=0, delta=ChoiceDelta(tool_calls=[tool_call_delta1]))],
+    )
+    chunk2 = ChatCompletionChunk(
+        id="chunk-id",
+        created=1,
+        model="fake",
+        object="chat.completion.chunk",
+        choices=[Choice(index=0, delta=ChoiceDelta(tool_calls=[tool_call_delta2]))],
+        usage=CompletionUsage(completion_tokens=1, prompt_tokens=1, total_tokens=2),
+    )
+
+    async def fake_stream() -> AsyncIterator[ChatCompletionChunk]:
+        for chunk in (chunk1, chunk2):
+            yield chunk
+
+    async def patched_fetch_response(self, *args, **kwargs):
+        response = Response(
+            id="resp-id",
+            created_at=0,
+            model="fake-model",
+            object="response",
+            output=[],
+            tool_choice="none",
+            tools=[],
+            parallel_tool_calls=False,
+        )
+        return response, fake_stream()
+
+    monkeypatch.setattr(OpenAIChatCompletionsModel, "_fetch_response", patched_fetch_response)
+    model = OpenAIProvider(use_responses=False).get_model("gpt-4")
+    output_events = []
+
+    async for event in model.stream_response(
+        system_instructions=None,
+        input="",
+        model_settings=ModelSettings(),
+        tools=[],
+        output_schema=None,
+        handoffs=[],
+        tracing=ModelTracing.DISABLED,
+        previous_response_id=None,
+        conversation_id=None,
+        prompt=None,
+    ):
+        output_events.append(event)
+
+    added_events = [event for event in output_events if event.type == "response.output_item.added"]
+    delta_events = [
+        event for event in output_events if event.type == "response.function_call_arguments.delta"
+    ]
+    done_events = [event for event in output_events if event.type == "response.output_item.done"]
+
+    assert [event.output_index for event in added_events] == [0, 1]
+    assert [event.output_index for event in delta_events] == [0, 1]
+    assert [event.output_index for event in done_events] == [0, 1]