19d54c0611
_build_gateway_agent_history rebuilds a session's own transcript for its next turn. Rows that pass through whole (tool-calling assistants, tool results) kept message_uid and the merge witness, but _build_replay_entry rebuilt plain user and assistant rows from an allowlist that left both out. A context engine therefore saw the store's uids on some roles and none on others in the same replayed conversation. The replay entry now copies message_uid and the merge witness on every role. It leaves out the tool-call uid maps: a plain row whose unanswered tool_calls repair pruned can still hold a map, and replaying it would name calls the row no longer carries.
184 lines
8.0 KiB
Python
184 lines
8.0 KiB
Python
"""Tests for ``gateway.run._build_replay_entry``.
|
|
|
|
The gateway rebuilds ``agent_history`` from the persisted transcript on every
|
|
turn (unlike the CLI, which keeps the live in-memory message list). When a
|
|
pure-text assistant turn (no ``tool_calls``) is replayed, the simple-text
|
|
branch in ``run_sync`` used to whitelist only three reasoning fields:
|
|
``reasoning``, ``reasoning_details``, ``codex_reasoning_items``.
|
|
|
|
That whitelist predated three fields the DB now persists:
|
|
``reasoning_content``, ``codex_message_items``, and ``finish_reason``. The
|
|
unrecovered drop of ``codex_message_items`` in particular kills prefix-cache
|
|
hits for OpenAI Codex Responses API users — OpenAI's docs require the
|
|
``phase`` field be replayed on every assistant message.
|
|
|
|
These tests pin the expanded whitelist so it doesn't regress.
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
|
|
from gateway.run import _build_replay_entry
|
|
|
|
|
|
class TestBuildReplayEntry:
|
|
def test_user_message_has_only_role_and_content(self):
|
|
entry = _build_replay_entry(
|
|
"user",
|
|
"hello",
|
|
{"role": "user", "content": "hello", "reasoning": "leak", "extra": "drop"},
|
|
)
|
|
assert entry == {"role": "user", "content": "hello"}
|
|
|
|
def test_tool_message_has_only_role_and_content(self):
|
|
# Tool messages aren't routed through this helper in production
|
|
# (they take the rich-passthrough branch), but the helper itself
|
|
# must not leak reasoning fields onto non-assistant roles even if
|
|
# someone calls it incorrectly.
|
|
entry = _build_replay_entry(
|
|
"tool",
|
|
"result",
|
|
{"role": "tool", "content": "result", "reasoning": "leak"},
|
|
)
|
|
assert entry == {"role": "tool", "content": "result"}
|
|
|
|
|
|
def test_assistant_drops_falsy_reasoning(self):
|
|
"""Empty/None reasoning fields stay dropped (matching PR #2974
|
|
behaviour) — empty strings/lists for these fields carry no info."""
|
|
msg = {
|
|
"role": "assistant",
|
|
"content": "answer",
|
|
"reasoning": "",
|
|
"reasoning_details": [],
|
|
"codex_reasoning_items": [],
|
|
"codex_message_items": [],
|
|
"finish_reason": "",
|
|
}
|
|
entry = _build_replay_entry("assistant", "answer", msg)
|
|
assert entry == {"role": "assistant", "content": "answer"}
|
|
|
|
|
|
def test_assistant_preserves_all_six_fields_together(self):
|
|
details = [{"type": "reasoning.summary", "summary": "s"}]
|
|
codex_items = [{"type": "reasoning", "encrypted_content": "b"}]
|
|
msg_items = [
|
|
{
|
|
"type": "message",
|
|
"role": "assistant",
|
|
"phase": "final_answer",
|
|
"content": [{"type": "output_text", "text": "x"}],
|
|
}
|
|
]
|
|
msg = {
|
|
"role": "assistant",
|
|
"content": "answer",
|
|
"reasoning": "thinking",
|
|
"reasoning_content": "structured",
|
|
"reasoning_details": details,
|
|
"codex_reasoning_items": codex_items,
|
|
"codex_message_items": msg_items,
|
|
"finish_reason": "stop",
|
|
}
|
|
entry = _build_replay_entry("assistant", "answer", msg)
|
|
assert entry["reasoning"] == "thinking"
|
|
assert entry["reasoning_content"] == "structured"
|
|
assert entry["reasoning_details"] == details
|
|
assert entry["codex_reasoning_items"] == codex_items
|
|
assert entry["codex_message_items"] == msg_items
|
|
assert entry["finish_reason"] == "stop"
|
|
|
|
|
|
|
|
def test_unrelated_keys_are_ignored(self):
|
|
"""Random keys on the message must not leak into the replay entry."""
|
|
msg = {
|
|
"role": "assistant",
|
|
"content": "answer",
|
|
"timestamp": 12345.6,
|
|
"internal_marker": "should not flow",
|
|
"tool_call_id": "should not be set on simple-text branch",
|
|
}
|
|
entry = _build_replay_entry("assistant", "answer", msg)
|
|
assert "timestamp" not in entry
|
|
assert "internal_marker" not in entry
|
|
assert "tool_call_id" not in entry
|
|
|
|
|
|
class TestReplayEntryApiContentSidecar:
|
|
"""The api_content sidecar (persist-what-you-send) must survive the
|
|
gateway's transcript→agent_history rebuild, or the whole prompt-cache
|
|
fix is inert on gateway platforms — but only when this pipeline did not
|
|
rewrite the content (a rewrite means different bytes must replay)."""
|
|
|
|
def test_user_forwards_api_content_when_content_unchanged(self):
|
|
msg = {"role": "user", "content": "hi", "api_content": "hi\n\nCTX"}
|
|
entry = _build_replay_entry("user", "hi", msg)
|
|
assert entry["api_content"] == "hi\n\nCTX"
|
|
assert entry["content"] == "hi"
|
|
|
|
def test_assistant_forwards_api_content_when_content_unchanged(self):
|
|
msg = {"role": "assistant", "content": "a", "api_content": "a <memory-context>"}
|
|
entry = _build_replay_entry("assistant", "a", msg)
|
|
assert entry["api_content"] == "a <memory-context>"
|
|
|
|
|
|
class TestGatewayHistoryBuildForwardsSidecar:
|
|
def test_end_to_end_history_build_keeps_sidecar(self):
|
|
from gateway.run import _build_gateway_agent_history
|
|
|
|
history = [
|
|
{"role": "user", "content": "hi", "api_content": "hi\n\nCTX", "timestamp": 123.0},
|
|
{"role": "assistant", "content": "hello"},
|
|
]
|
|
agent_history, _obs = _build_gateway_agent_history(history)
|
|
assert agent_history[0]["api_content"] == "hi\n\nCTX"
|
|
|
|
|
|
|
|
def test_gateway_history_keeps_sidecar_only_assistant_row():
|
|
"""A reasoning-only clean stop persists content="" with the promoted reply in ``api_content``
|
|
(agent/turn_final_response.py); the gateway rebuild must replay it, not drop it (user->user)."""
|
|
from gateway.run import _build_gateway_agent_history
|
|
|
|
history = [
|
|
{"role": "user", "content": "2+2?"},
|
|
{"role": "assistant", "content": "", "reasoning": "The answer is 4.", "api_content": "The answer is 4."},
|
|
{"role": "user", "content": "thanks"},
|
|
]
|
|
agent_history, _ = _build_gateway_agent_history(history)
|
|
assert [m["role"] for m in agent_history] == ["user", "assistant", "user"]
|
|
assert agent_history[1]["api_content"] == "The answer is 4."
|
|
assert agent_history[1]["reasoning"] == "The answer is 4."
|
|
|
|
|
|
def test_replay_rebuilds_the_same_conversation_so_every_role_keeps_its_identity():
|
|
"""Gateway replay turns the stored transcript back into the SAME session's history for its next turn:
|
|
user, assistant, tool-calling and tool rows all keep ``message_uid`` and the merge witness."""
|
|
from gateway.run import _build_gateway_agent_history
|
|
|
|
call = {"id": "call_1", "type": "function", "function": {"name": "t", "arguments": "{}"}}
|
|
history = [
|
|
{"role": "user", "content": "a\n\nb", "message_uid": "1" * 32, "_absorbed_message_uids": ["b" * 32],
|
|
"timestamp": 1_700_000_000.0},
|
|
{"role": "assistant", "content": "", "tool_calls": [call], "message_uid": "2" * 32},
|
|
{"role": "tool", "content": "r", "tool_call_id": "call_1", "message_uid": "3" * 32},
|
|
{"role": "assistant", "content": "done", "message_uid": "4" * 32, "_absorbed_message_uids": ["c" * 32]},
|
|
]
|
|
replayed, _ = _build_gateway_agent_history(history)
|
|
assert [(m["role"], m.get("message_uid"), m.get("_absorbed_message_uids")) for m in replayed] == [
|
|
(m["role"], m["message_uid"], m.get("_absorbed_message_uids")) for m in history]
|
|
|
|
|
|
def test_a_plain_replay_row_does_not_carry_a_tool_call_uid_map_without_its_calls():
|
|
"""Repair can prune an assistant's unanswered tool_calls and leave its ``_tool_call_uids`` behind: the
|
|
replayed plain row keeps its own uid but not a map naming calls it no longer carries."""
|
|
from gateway.run import _build_gateway_agent_history
|
|
|
|
history = [
|
|
{"role": "user", "content": "q", "message_uid": "1" * 32},
|
|
{"role": "assistant", "content": "done", "message_uid": "2" * 32, "_tool_call_uids": {"call_1": "5" * 32}},
|
|
]
|
|
replayed, _ = _build_gateway_agent_history(history)
|
|
assert replayed[1]["message_uid"] == "2" * 32
|
|
assert "_tool_call_uids" not in replayed[1] and "_tool_call_uid" not in replayed[1]
|