From aba98f49dbdbfa4c87d9522d00a3ee0f945320f8 Mon Sep 17 00:00:00 2001
From: Ashwin Bharambe <ashwin.bharambe@gmail.com>
Date: Tue, 28 Oct 2025 14:48:01 -0700
Subject: [PATCH] fix(mypy): complete openai_responses.py type fixes (76 errors
 resolved)

Fixed all 76 type errors in openai_responses.py through proper type-safe refactoring:

Changes made:
- Fixed return type signature for _process_input_with_previous_response to include ToolContext
- Replaced .copy() with list() for Sequence types (line 94)
- Added assertions for text and max_infer_iters to narrow types from None
- Properly typed input_items_data as list[OpenAIResponseInput] to avoid list invariance
- Properly typed conversation_items as list[ConversationItem] to avoid list invariance
- Properly typed output_items as list[ConversationItem] to avoid list invariance
- Fixed StreamingResponseOrchestrator signature to accept str | None for instructions

All fixes use proper type narrowing and union typing without any type: ignore comments.
---
 .../responses/openai_responses.py             | 20 ++++++++++++++-----
 .../meta_reference/responses/streaming.py     |  2 +-
 2 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/src/llama_stack/providers/inline/agents/meta_reference/responses/openai_responses.py b/src/llama_stack/providers/inline/agents/meta_reference/responses/openai_responses.py
index 44749aeb2..f6769e838 100644
--- a/src/llama_stack/providers/inline/agents/meta_reference/responses/openai_responses.py
+++ b/src/llama_stack/providers/inline/agents/meta_reference/responses/openai_responses.py
@@ -91,7 +91,8 @@ class OpenAIResponsesImpl:
         input: str | list[OpenAIResponseInput],
         previous_response: _OpenAIResponseObjectWithInputAndMessages,
     ):
-        new_input_items = previous_response.input.copy()
+        # Convert Sequence to list for mutation
+        new_input_items = list(previous_response.input)
         new_input_items.extend(previous_response.output)
 
         if isinstance(input, str):
@@ -107,7 +108,7 @@ class OpenAIResponsesImpl:
         tools: list[OpenAIResponseInputTool] | None,
         previous_response_id: str | None,
         conversation: str | None,
-    ) -> tuple[str | list[OpenAIResponseInput], list[OpenAIMessageParam]]:
+    ) -> tuple[str | list[OpenAIResponseInput], list[OpenAIMessageParam], ToolContext]:
         """Process input with optional previous response context.
 
         Returns:
@@ -208,6 +209,9 @@ class OpenAIResponsesImpl:
         messages: list[OpenAIMessageParam],
     ) -> None:
         new_input_id = f"msg_{uuid.uuid4()}"
+        # Type input_items_data as the full OpenAIResponseInput union to avoid list invariance issues
+        input_items_data: list[OpenAIResponseInput] = []
+
         if isinstance(input, str):
             # synthesize a message from the input string
             input_content = OpenAIResponseInputMessageContentText(text=input)
@@ -219,7 +223,6 @@ class OpenAIResponsesImpl:
             input_items_data = [input_content_item]
         else:
             # we already have a list of messages
-            input_items_data = []
             for input_item in input:
                 if isinstance(input_item, OpenAIResponseMessage):
                     # These may or may not already have an id, so dump to dict, check for id, and add if missing
@@ -329,6 +332,11 @@ class OpenAIResponsesImpl:
         max_infer_iters: int | None = 10,
         guardrail_ids: list[str] | None = None,
     ) -> AsyncIterator[OpenAIResponseObjectStream]:
+        # These should never be None when called from create_openai_response (which sets defaults)
+        # but we assert here to help mypy understand the types
+        assert text is not None, "text must not be None"
+        assert max_infer_iters is not None, "max_infer_iters must not be None"
+
         # Input preprocessing
         all_input, messages, tool_context = await self._process_input_with_previous_response(
             input, tools, previous_response_id, conversation
@@ -371,7 +379,8 @@ class OpenAIResponsesImpl:
         final_response = None
         failed_response = None
 
-        output_items = []
+        # Type as ConversationItem to avoid list invariance issues
+        output_items: list[ConversationItem] = []
         async for stream_chunk in orchestrator.create_response():
             match stream_chunk.type:
                 case "response.completed" | "response.incomplete":
@@ -415,7 +424,8 @@ class OpenAIResponsesImpl:
         self, conversation_id: str, input: str | list[OpenAIResponseInput] | None, output_items: list[ConversationItem]
     ) -> None:
         """Sync content and response messages to the conversation."""
-        conversation_items = []
+        # Type as ConversationItem union to avoid list invariance issues
+        conversation_items: list[ConversationItem] = []
 
         if isinstance(input, str):
             conversation_items.append(
diff --git a/src/llama_stack/providers/inline/agents/meta_reference/responses/streaming.py b/src/llama_stack/providers/inline/agents/meta_reference/responses/streaming.py
index 2cbfead40..bba5e60af 100644
--- a/src/llama_stack/providers/inline/agents/meta_reference/responses/streaming.py
+++ b/src/llama_stack/providers/inline/agents/meta_reference/responses/streaming.py
@@ -111,7 +111,7 @@ class StreamingResponseOrchestrator:
         text: OpenAIResponseText,
         max_infer_iters: int,
         tool_executor,  # Will be the tool execution logic from the main class
-        instructions: str,
+        instructions: str | None,
         safety_api,
         guardrail_ids: list[str] | None = None,
         prompt: OpenAIResponsePrompt | None = None,