mirror of
https://github.com/headroomlabs-ai/headroom.git
synced 2026-08-27 14:17:10 -04:00
## Description Refs #1877. `handle_openai_responses` (the `/v1/responses` HTTP handler) had zero CCR / `headroom_retrieve` wiring, so a retrieve `function_call` in a Responses API reply passed straight through to the client instead of being resolved server-side, unlike the parallel chat-completions backend path (`handle_openai_chat`, ~2775-2848), which already intercepts `headroom_retrieve` tool calls via `ccr_response_handler.has_ccr_tool_calls()` / `handle_response()`. This PR is scoped to the core interception gap only. The issue's proposals A (egress scrubber) and B/C (event-level SSE parsing/splicing for true mid-stream interception) are out of scope here; the streaming case is instead handled by forcing a buffered (non-streaming) upstream call when `headroom_retrieve` is offered, matching the existing buffered-CCR pattern in the Anthropic handler. ## Type of Change - [ ] Bug fix (non-breaking change that fixes an issue) - [x] New feature (non-breaking change that adds functionality) - [ ] Breaking change (fix or feature that would cause existing functionality to change) - [ ] Documentation update - [ ] Performance improvement - [ ] Code refactoring (no functional changes) ## Changes Made - `headroom/ccr/response_handler.py`: added an `"openai_responses"` provider branch to `CCRResponseHandler`; `_extract_tool_calls` reads flat `function_call` items from the top-level `output[]` array, tool-call IDs key off `call_id`, and `_extract_assistant_message` / `_create_tool_result_message` return sentinel-keyed item lists that `handle_response()` extends into the running item history. - `headroom/ccr/tool_injection.py`: added a `parse_tool_call` branch for `"openai_responses"` where name and arguments are flat on the item. - `headroom/proxy/handlers/openai.py`: detects non-streaming `headroom_retrieve` function calls and runs `ccr_response_handler.handle_response()` with a stateless continuation that resends the full `input[]` item history. - `headroom/proxy/handlers/openai.py`: forces `stream:true` requests with `headroom_retrieve` available through a buffered `stream:false` upstream call, resolves retrieval server-side, then reconstructs a minimal Responses SSE stream for the client. - `headroom/proxy/handlers/openai.py`: treats `ccr_response_handler` as optional on `OpenAIHandlerMixin` consumers, so handlers without CCR support keep the existing Codex routing, streaming, header stripping, memory timeout, and compression fail-open behavior. - Non-CCR streaming requests are unaffected; they still go through `_stream_response()` as before. ## Testing - [x] Unit tests pass (`uv run pytest tests/test_ccr_response_handler_openai_responses.py -q`) - [x] Integration tests pass (`uv run pytest tests/test_proxy/test_openai_responses_ccr.py -q`) - [x] Regression tests pass (`uv run pytest tests/test_openai_codex_routing.py -q`) - [x] Linting passes (`uv run ruff check headroom/proxy/handlers/openai.py tests/test_openai_codex_routing.py tests/test_proxy/test_openai_responses_ccr.py tests/test_ccr_response_handler_openai_responses.py`) - [x] New tests added for new functionality when applicable - [x] Manual testing performed ### Test Output ```text $ uv run pytest tests/test_openai_codex_routing.py -q tests\test_openai_codex_routing.py .................... [100%] 20 passed in 0.51s $ uv run pytest tests/test_proxy/test_openai_responses_ccr.py tests/test_ccr_response_handler_openai_responses.py -q tests\test_proxy\test_openai_responses_ccr.py .... [ 25%] tests\test_ccr_response_handler_openai_responses.py ............ [100%] 16 passed, 1 warning in 27.51s $ uv run ruff check headroom/proxy/handlers/openai.py tests/test_openai_codex_routing.py tests/test_proxy/test_openai_responses_ccr.py tests/test_ccr_response_handler_openai_responses.py All checks passed! ``` ## Real Behavior Proof - Environment: Windows, Python 3.12 via uv-managed venv, local PR worktree on branch `pr/1877-ccr-responses-interception`, no live LLM provider needed because the tests stub upstream HTTP and CCR continuation behavior. - Exact command / steps: Ran `uv run pytest tests/test_openai_codex_routing.py -q` to reproduce the CI-failing Codex routing surface after the optional-handler fix; ran `uv run pytest tests/test_proxy/test_openai_responses_ccr.py tests/test_ccr_response_handler_openai_responses.py -q` to cover the positive Responses CCR interception path; ran targeted Ruff on the touched handler and related tests. - Observed result: Codex routing tests that previously failed with `AttributeError: '_DummyOpenAIHandler' object has no attribute 'ccr_response_handler'` now pass; Responses CCR still detects and resolves `headroom_retrieve` when a real proxy installs `ccr_response_handler`; non-CCR streaming requests still route through `_stream_response()`. - Not tested: true event-level mid-stream Responses SSE splicing and client-bound egress marker scrubbing are out of scope for this PR and remain future work from issue #1877's broader proposals. ## Review Readiness - [x] I have performed a self-review - [x] This PR is ready for human review ## Checklist - [x] My code follows the project's style guidelines - [x] I have performed a self-review of my code - [x] I have commented my code, particularly in hard-to-understand areas - [ ] I have made corresponding changes to the documentation - [x] My changes generate no new warnings - [x] I have added tests that prove my fix is effective or that my feature works - [x] New and existing unit tests pass locally with my changes - [ ] I have updated the CHANGELOG.md if applicable ## Additional Notes No documentation or changelog update was made because this is internal proxy CCR behavior, not a user-facing command or configuration change.
245 lines
8.4 KiB
Python
245 lines
8.4 KiB
Python
"""Tests for CCR response handling of the OpenAI Responses API shape.
|
|
|
|
Covers #1877: `CCRResponseHandler` previously only understood "anthropic",
|
|
"openai" (chat completions), and "google" response shapes. Responses API
|
|
function calls are flat `function_call` items in a top-level `output[]`
|
|
array (not nested under `choices[].message.tool_calls`), and results are
|
|
`function_call_output` items appended to `input[]` rather than a single
|
|
role/content message — these tests exercise the new "openai_responses"
|
|
provider branch end to end.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import json
|
|
|
|
import pytest
|
|
|
|
from headroom.cache.compression_store import (
|
|
get_compression_store,
|
|
reset_compression_store,
|
|
)
|
|
from headroom.ccr.response_handler import CCRResponseHandler, CCRToolResult
|
|
from headroom.ccr.tool_injection import CCR_TOOL_NAME, parse_tool_call
|
|
|
|
|
|
@pytest.fixture(autouse=True)
|
|
def reset_store():
|
|
reset_compression_store()
|
|
yield
|
|
reset_compression_store()
|
|
|
|
|
|
def _function_call_response(hash_key: str, call_id: str = "call_abc") -> dict:
|
|
return {
|
|
"id": "resp_1",
|
|
"object": "response",
|
|
"status": "completed",
|
|
"output": [
|
|
{
|
|
"type": "reasoning",
|
|
"id": "rs_1",
|
|
"summary": [],
|
|
},
|
|
{
|
|
"type": "function_call",
|
|
"id": "fc_1",
|
|
"call_id": call_id,
|
|
"name": CCR_TOOL_NAME,
|
|
"arguments": json.dumps({"hash": hash_key}),
|
|
},
|
|
],
|
|
"usage": {"input_tokens": 50, "output_tokens": 10},
|
|
}
|
|
|
|
|
|
class TestOpenAIResponsesDetection:
|
|
def test_detect_function_call_tool_call(self) -> None:
|
|
handler = CCRResponseHandler()
|
|
response = _function_call_response("abc123def456abc123def456")
|
|
|
|
assert handler.has_ccr_tool_calls(response, "openai_responses")
|
|
|
|
def test_no_false_positive_for_other_function_call(self) -> None:
|
|
handler = CCRResponseHandler()
|
|
response = {
|
|
"output": [
|
|
{
|
|
"type": "function_call",
|
|
"id": "fc_1",
|
|
"call_id": "call_1",
|
|
"name": "some_other_tool",
|
|
"arguments": "{}",
|
|
}
|
|
]
|
|
}
|
|
|
|
assert not handler.has_ccr_tool_calls(response, "openai_responses")
|
|
|
|
def test_no_false_positive_for_message_only_output(self) -> None:
|
|
handler = CCRResponseHandler()
|
|
response = {
|
|
"output": [
|
|
{
|
|
"type": "message",
|
|
"role": "assistant",
|
|
"content": [{"type": "output_text", "text": "hi"}],
|
|
}
|
|
]
|
|
}
|
|
|
|
assert not handler.has_ccr_tool_calls(response, "openai_responses")
|
|
|
|
def test_empty_output(self) -> None:
|
|
handler = CCRResponseHandler()
|
|
assert not handler.has_ccr_tool_calls({"output": []}, "openai_responses")
|
|
assert not handler.has_ccr_tool_calls({}, "openai_responses")
|
|
|
|
|
|
class TestOpenAIResponsesParsing:
|
|
def test_parse_extracts_call_id_not_item_id(self) -> None:
|
|
"""`call_id` (not the function_call item's own `id`) matches the
|
|
`function_call_output.call_id` the continuation must echo back."""
|
|
handler = CCRResponseHandler()
|
|
response = _function_call_response("abc123def456abc123def456", call_id="call_xyz")
|
|
|
|
ccr_calls, other_calls = handler._parse_ccr_tool_calls(response, "openai_responses")
|
|
|
|
assert len(ccr_calls) == 1
|
|
assert ccr_calls[0].tool_call_id == "call_xyz"
|
|
assert ccr_calls[0].hash_key == "abc123def456abc123def456"
|
|
assert not other_calls
|
|
|
|
def test_parse_tool_call_direct(self) -> None:
|
|
"""`parse_tool_call` reads flat name/arguments, not a nested `function` key."""
|
|
tool_call = {
|
|
"type": "function_call",
|
|
"call_id": "call_1",
|
|
"name": CCR_TOOL_NAME,
|
|
"arguments": '{"hash": "abc123def456abc123def456"}',
|
|
}
|
|
|
|
assert parse_tool_call(tool_call, "openai_responses") == "abc123def456abc123def456"
|
|
|
|
def test_parse_tool_call_rejects_other_names(self) -> None:
|
|
tool_call = {
|
|
"type": "function_call",
|
|
"call_id": "call_1",
|
|
"name": "read_file",
|
|
"arguments": '{"path": "/etc/config"}',
|
|
}
|
|
|
|
assert parse_tool_call(tool_call, "openai_responses") is None
|
|
|
|
def test_parse_tool_call_malformed_arguments(self) -> None:
|
|
tool_call = {
|
|
"type": "function_call",
|
|
"call_id": "call_1",
|
|
"name": CCR_TOOL_NAME,
|
|
"arguments": "not json",
|
|
}
|
|
|
|
assert parse_tool_call(tool_call, "openai_responses") is None
|
|
|
|
|
|
class TestOpenAIResponsesMessageShaping:
|
|
def test_extract_assistant_message_echoes_full_output_array(self) -> None:
|
|
handler = CCRResponseHandler()
|
|
response = _function_call_response("abc123def456abc123def456")
|
|
|
|
result = handler._extract_assistant_message(response, "openai_responses")
|
|
|
|
assert result == {"_openai_responses_output_items": response["output"]}
|
|
|
|
def test_create_tool_result_message_uses_call_id(self) -> None:
|
|
handler = CCRResponseHandler()
|
|
results = [
|
|
CCRToolResult(tool_call_id="call_xyz", content='{"data": "x"}', success=True),
|
|
CCRToolResult(tool_call_id="call_abc", content='{"data": "y"}', success=True),
|
|
]
|
|
|
|
message = handler._create_tool_result_message(results, "openai_responses")
|
|
|
|
assert "_openai_responses_tool_results" in message
|
|
items = message["_openai_responses_tool_results"]
|
|
assert len(items) == 2
|
|
assert items[0] == {
|
|
"type": "function_call_output",
|
|
"call_id": "call_xyz",
|
|
"output": '{"data": "x"}',
|
|
}
|
|
|
|
|
|
class TestOpenAIResponsesHandleResponse:
|
|
@pytest.mark.asyncio
|
|
async def test_handle_response_resolves_retrieve_and_extends_input(self) -> None:
|
|
store = get_compression_store()
|
|
original = json.dumps([{"id": i} for i in range(30)])
|
|
hash_key = store.store(original=original, compressed="[]", original_item_count=30)
|
|
|
|
handler = CCRResponseHandler()
|
|
initial_response = _function_call_response(hash_key, call_id="call_1")
|
|
final_response = {
|
|
"id": "resp_2",
|
|
"object": "response",
|
|
"status": "completed",
|
|
"output": [
|
|
{
|
|
"type": "message",
|
|
"role": "assistant",
|
|
"content": [{"type": "output_text", "text": "Here are all 30 items."}],
|
|
}
|
|
],
|
|
}
|
|
|
|
captured_calls: list[list[dict]] = []
|
|
|
|
async def mock_api_call(items, tools):
|
|
captured_calls.append(items)
|
|
return final_response
|
|
|
|
result = await handler.handle_response(
|
|
initial_response,
|
|
[{"role": "user", "content": "get the data"}],
|
|
None,
|
|
mock_api_call,
|
|
"openai_responses",
|
|
)
|
|
|
|
assert result == final_response
|
|
assert len(captured_calls) == 1
|
|
# Original input item + the two echoed output items (reasoning +
|
|
# function_call) + the function_call_output — extended, not
|
|
# appended as a single blob.
|
|
sent_items = captured_calls[0]
|
|
assert sent_items[0] == {"role": "user", "content": "get the data"}
|
|
assert {"type": "function_call", "name": CCR_TOOL_NAME} in [
|
|
{"type": i.get("type"), "name": i.get("name")}
|
|
for i in sent_items
|
|
if i.get("type") == "function_call"
|
|
]
|
|
tool_outputs = [i for i in sent_items if i.get("type") == "function_call_output"]
|
|
assert len(tool_outputs) == 1
|
|
assert tool_outputs[0]["call_id"] == "call_1"
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_handle_response_no_ccr_passthrough(self) -> None:
|
|
handler = CCRResponseHandler()
|
|
response = {
|
|
"output": [
|
|
{
|
|
"type": "message",
|
|
"role": "assistant",
|
|
"content": [{"type": "output_text", "text": "no tool call here"}],
|
|
}
|
|
]
|
|
}
|
|
|
|
async def mock_api_call(items, tools):
|
|
raise AssertionError("should not be called")
|
|
|
|
result = await handler.handle_response(
|
|
response, [], None, mock_api_call, "openai_responses"
|
|
)
|
|
|
|
assert result == response
|