Skip to content

"Vibe-Trading: Your Personal Trading Agent"

fix(tests): accept LLM timeout kwargs in loop stubs; fix order-level test encoding

Will Liao86d04222dac7
diff --git a/agent/tests/test_agent_goal_context.py b/agent/tests/test_agent_goal_context.py
index ce7c56074..84b663156 100644
--- a/agent/tests/test_agent_goal_context.py
+++ b/agent/tests/test_agent_goal_context.py
@@ -30,6 +30,8 @@ class _CapturingLLM:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _AnswerResponse:
del tools, on_text_chunk, on_reasoning_chunk
diff --git a/agent/tests/test_agent_grounding.py b/agent/tests/test_agent_grounding.py
index b06e5a843..2c81c5373 100644
--- a/agent/tests/test_agent_grounding.py
+++ b/agent/tests/test_agent_grounding.py
@@ -948,6 +948,8 @@ class _CorrectingLLM:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _Response:
response = self.responses.pop(0)
@@ -2210,8 +2212,8 @@ def test_in_text_decimal_survives_list_marker_mask(
def test_order_level_prices_are_not_observed_quotes(tmp_path: Path) -> None:
"""GTC order limits (100 @ $3.50) are prospective, not observed prices.
- Regression for the RXRX run: the draft "两档 GTC (100 @ $3.50 / 100 @ $4.00)
- 均未触发 (周高 $3.42 < $3.50)" was rejected with numeric_claim_conflict for
+ Regression for the RXRX run: the draft "两档 GTC (100 @ $3.50 / 100 @ $4.00)
+ 均未触发 (周高 $3.42 < $3.50)" was rejected with numeric_claim_conflict for
100, 3.5 and 4.0 against the observed OHLC range. Order levels are levels,
like targets/stops, and share counts are quantities - neither is a claim
about an observed quote.
@@ -2226,7 +2228,7 @@ def test_order_level_prices_are_not_observed_quotes(tmp_path: Path) -> None:
)
ledger = GroundingLedger(
run_dir=tmp_path,
- user_message="请分析 RXRX.US 并更新周报",
+ user_message="请分析 RXRX.US 并更新周报",
)
ledger.ingest_tool_result(
tool_name="search_symbol",
@@ -2245,18 +2247,17 @@ def test_order_level_prices_are_not_observed_quotes(tmp_path: Path) -> None:
)
answer = (
- "RXRX.US(yahoo,USD)本周高 $3.42。订单: 两档 GTC (100 @ $3.50 / 100 @ $4.00) "
- "均未触发 (周高 $3.42 < $3.50) — 价格推断未成交。"
+ "RXRX.US(yahoo,USD)本周高 $3.42。订单: 两档 GTC (100 @ $3.50 / 100 @ $4.00) "
+ "均未触发 (周高 $3.42 < $3.50) — 价格推断未成交。"
)
result = ledger.validate_final_answer(answer)
assert result.valid is True, result.issues
# A genuinely fabricated close is still rejected.
fabricated = ledger.validate_final_answer(
- "RXRX.US(yahoo,USD)本周高 $3.42。另: 收盘 2.50 已确认。"
+ "RXRX.US(yahoo,USD)本周高 $3.42。另: 收盘 2.50 已确认。"
)
assert fabricated.valid is False
assert any(
issue["code"] == "numeric_claim_conflict" for issue in fabricated.issues
)
-
diff --git a/agent/tests/test_agent_loop_content_filter.py b/agent/tests/test_agent_loop_content_filter.py
index c6a05fbfd..e9284ff55 100644
--- a/agent/tests/test_agent_loop_content_filter.py
+++ b/agent/tests/test_agent_loop_content_filter.py
@@ -33,6 +33,8 @@ class _ContentFilterLoopLLM:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> LLMResponse:
self.calls += 1
@@ -63,6 +65,8 @@ class _EmptyResponseLoopLLM:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> LLMResponse:
self.calls += 1
@@ -202,6 +206,8 @@ def test_single_empty_response_recovers_with_nudge(
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> LLMResponse:
self.calls += 1
@@ -256,6 +262,8 @@ class _RatioFilterLoopLLM:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> LLMResponse:
self.calls += 1
diff --git a/agent/tests/test_agent_loop_no_assistant_prefill.py b/agent/tests/test_agent_loop_no_assistant_prefill.py
index 69cf2707d..66f003e6a 100644
--- a/agent/tests/test_agent_loop_no_assistant_prefill.py
+++ b/agent/tests/test_agent_loop_no_assistant_prefill.py
@@ -44,6 +44,8 @@ class _StubLLM:
tools: Any = None,
on_text_chunk: Any = None,
on_reasoning_chunk: Any = None,
+ timeout: Any = None,
+ idle_timeout_s: Any = None,
should_cancel: Any = None,
) -> Any:
messages = [m for m in messages]
diff --git a/agent/tests/test_agent_loop_stream_retry.py b/agent/tests/test_agent_loop_stream_retry.py
index 79cb781f2..df742cbb9 100644
--- a/agent/tests/test_agent_loop_stream_retry.py
+++ b/agent/tests/test_agent_loop_stream_retry.py
@@ -40,6 +40,8 @@ class _FlakyLoopLLM:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> LLMResponse:
"""Raise the next queued error or return the final response.
diff --git a/agent/tests/test_agent_loop_terminal_state.py b/agent/tests/test_agent_loop_terminal_state.py
index f18e6789c..3784f30b2 100644
--- a/agent/tests/test_agent_loop_terminal_state.py
+++ b/agent/tests/test_agent_loop_terminal_state.py
@@ -45,6 +45,8 @@ class _StubLLMNoFinal:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _StubLLMResponse:
return _StubLLMResponse()
@@ -62,6 +64,8 @@ class _StubLLMWithUsage:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _StubLLMResponse:
response = _StubLLMResponse()
@@ -85,6 +89,8 @@ class _StubLLMSuccess:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _StubLLMResponse:
self.calls += 1
@@ -112,6 +118,8 @@ class _StubLLMCancelMidStream:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _StubLLMResponse:
# Set _cancelled on the bound agent so the next loop iteration check
@@ -213,6 +221,8 @@ class _StubLLMCancelWithToolCalls:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _StubLLMResponse:
self._agent_ref[0]._cancel_event.set()
@@ -301,6 +311,8 @@ class _StubLLMAlwaysToolCalls:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _StubLLMResponse:
resp = _StubLLMResponse()
@@ -328,6 +340,8 @@ class _StubLLMIgnoresForcedTextOnly:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _StubLLMResponse:
resp = _StubLLMResponse()
@@ -348,6 +362,8 @@ class _StubLLMStreamFailure:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _StubLLMResponse:
raise ProviderStreamError(
diff --git a/agent/tests/test_agent_loop_trace.py b/agent/tests/test_agent_loop_trace.py
index 7e291de53..8fb2e785d 100644
--- a/agent/tests/test_agent_loop_trace.py
+++ b/agent/tests/test_agent_loop_trace.py
@@ -97,6 +97,8 @@ class _LongAnswerLLM:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _LongAnswerResponse:
del messages, tools, on_reasoning_chunk
diff --git a/agent/tests/test_content_filter_e2e.py b/agent/tests/test_content_filter_e2e.py
index 44d5e7fa3..2ec909cce 100644
--- a/agent/tests/test_content_filter_e2e.py
+++ b/agent/tests/test_content_filter_e2e.py
@@ -49,6 +49,8 @@ class _InterleavedFilterLLM:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> LLMResponse:
self.calls += 1
@@ -92,6 +94,8 @@ class _CleanRunLLM:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> LLMResponse:
self.calls += 1
diff --git a/agent/tests/test_grounding_recovery.py b/agent/tests/test_grounding_recovery.py
index d89a759e9..a1ea60c9a 100644
--- a/agent/tests/test_grounding_recovery.py
+++ b/agent/tests/test_grounding_recovery.py
@@ -333,6 +333,8 @@ class _FailingDraftLLM:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> LLMResponse:
self.calls += 1
diff --git a/agent/tests/test_reasoning_delta_throttle.py b/agent/tests/test_reasoning_delta_throttle.py
index c18d518b0..df581b05a 100644
--- a/agent/tests/test_reasoning_delta_throttle.py
+++ b/agent/tests/test_reasoning_delta_throttle.py
@@ -36,6 +36,8 @@ class _ReasoningBurstLLM:
tools: list[Any] | None = None,
on_text_chunk: Callable[[str], None] | None = None,
on_reasoning_chunk: Callable[[str], None] | None = None,
+ timeout: int | None = None,
+ idle_timeout_s: float | None = None,
should_cancel: Callable[[], bool] | None = None,
) -> _StubLLMResponse:
assert on_reasoning_chunk is not None