refactor: extract item validators from _preflight_codex_input_items#23952
refactor: extract item validators from _preflight_codex_input_items#23952kshitijk4poor wants to merge 2 commits into
Conversation
…nput Split the 192-line dispatcher into three focused functions: - _convert_assistant_msg_to_responses_items() — reasoning replay, message item replay, content emission, and tool call conversion - _convert_tool_msg_to_responses_item() — tool result formatting - _chat_messages_to_responses_input() — thin loop dispatcher Complexity: 45 → <10 on the main dispatcher (C901 cleared). Zero logic changes — pure extraction.
Split the 210-line validator into five focused functions, one per item type: - _validate_function_call_item() — call_id, name, arguments - _validate_function_call_output_item() — output (string or multimodal array) - _validate_reasoning_item() — encrypted content with dedup - _validate_message_item() — role=assistant with content parts - _validate_role_content_item() — bare role+content items Main dispatcher complexity: 51 → <10 (C901 cleared). Zero logic changes — pure extraction. Depends on: refactor/codex-chat-to-responses (same file)
|
Closing — re-evaluated against the original tracker goal in #23972 ("ruff complexity reduction"). The actual LOC accounting here is net positive (+26 / +66 / +85 / +139 across the stack), not negative. The C901 number drops (45/51/39/37 → all <22) come at a real cost: each extracted helper carries ~3-6 lines of decomposition tax (signature + docstring + return-plumbing for shared locals that the original monolithic function held in scope). The helpers themselves are real single-responsibility units, but Python's function-scope semantics don't give us a free way to split without that tax. For complexity reductions that trade LOC for testability, the bar is "is the function so untestable that the tax is worth it?" — and for these four codex functions, they're already exercised by the agent-loop integration tests in What's still moving forward from this tracker:
If a future refactor of these codex functions ends up net-negative on LOC (e.g. by collapsing repeated patterns or eliminating dead-code branches alongside the extraction), I'll reopen. |
Summary
Split the 210-line
_preflight_codex_input_items(C901 complexity 51) into five focused validators:_preflight_codex_input_items_validate_function_call_item_validate_function_call_output_item_validate_reasoning_item_validate_message_item_validate_role_content_itemZero logic changes — pure extraction. Each validator handles one item type's validation + normalization + append.
Part 2 of 4 for
codex_responses_adapter.py.Depends on: #23951 (refactor/codex-chat-to-responses) — same file.