mirror of
https://github.com/HKUDS/nanobot.git
synced 2026-05-04 00:35:58 +00:00
422 lines
15 KiB
Python
422 lines
15 KiB
Python
from nanobot.session.manager import Session
|
|
|
|
|
|
def _assert_no_orphans(history: list[dict]) -> None:
|
|
"""Assert every tool result in history has a matching assistant tool_call."""
|
|
declared = {
|
|
tc["id"]
|
|
for m in history if m.get("role") == "assistant"
|
|
for tc in (m.get("tool_calls") or [])
|
|
}
|
|
orphans = [
|
|
m.get("tool_call_id") for m in history
|
|
if m.get("role") == "tool" and m.get("tool_call_id") not in declared
|
|
]
|
|
assert orphans == [], f"orphan tool_call_ids: {orphans}"
|
|
|
|
|
|
def _tool_turn(prefix: str, idx: int) -> list[dict]:
|
|
"""Helper: one assistant with 2 tool_calls + 2 tool results."""
|
|
return [
|
|
{
|
|
"role": "assistant",
|
|
"content": None,
|
|
"tool_calls": [
|
|
{"id": f"{prefix}_{idx}_a", "type": "function", "function": {"name": "x", "arguments": "{}"}},
|
|
{"id": f"{prefix}_{idx}_b", "type": "function", "function": {"name": "y", "arguments": "{}"}},
|
|
],
|
|
},
|
|
{"role": "tool", "tool_call_id": f"{prefix}_{idx}_a", "name": "x", "content": "ok"},
|
|
{"role": "tool", "tool_call_id": f"{prefix}_{idx}_b", "name": "y", "content": "ok"},
|
|
]
|
|
|
|
|
|
# --- Original regression test (from PR 2075) ---
|
|
|
|
def test_get_history_drops_orphan_tool_results_when_window_cuts_tool_calls():
|
|
session = Session(key="telegram:test")
|
|
session.messages.append({"role": "user", "content": "old turn"})
|
|
for i in range(20):
|
|
session.messages.extend(_tool_turn("old", i))
|
|
session.messages.append({"role": "user", "content": "problem turn"})
|
|
for i in range(25):
|
|
session.messages.extend(_tool_turn("cur", i))
|
|
session.messages.append({"role": "user", "content": "new telegram question"})
|
|
|
|
history = session.get_history(max_messages=100)
|
|
_assert_no_orphans(history)
|
|
|
|
|
|
# --- Positive test: legitimate pairs survive trimming ---
|
|
|
|
def test_legitimate_tool_pairs_preserved_after_trim():
|
|
"""Complete tool-call groups within the window must not be dropped."""
|
|
session = Session(key="test:positive")
|
|
session.messages.append({"role": "user", "content": "hello"})
|
|
for i in range(5):
|
|
session.messages.extend(_tool_turn("ok", i))
|
|
session.messages.append({"role": "assistant", "content": "done"})
|
|
|
|
history = session.get_history(max_messages=500)
|
|
_assert_no_orphans(history)
|
|
tool_ids = [m["tool_call_id"] for m in history if m.get("role") == "tool"]
|
|
assert len(tool_ids) == 10
|
|
assert history[0]["role"] == "user"
|
|
|
|
|
|
def test_retain_recent_legal_suffix_keeps_recent_messages():
|
|
session = Session(key="test:trim")
|
|
for i in range(10):
|
|
session.messages.append({"role": "user", "content": f"msg{i}"})
|
|
|
|
session.retain_recent_legal_suffix(4)
|
|
|
|
assert len(session.messages) == 4
|
|
assert session.messages[0]["content"] == "msg6"
|
|
assert session.messages[-1]["content"] == "msg9"
|
|
|
|
|
|
def test_retain_recent_legal_suffix_adjusts_last_consolidated():
|
|
session = Session(key="test:trim-cons")
|
|
for i in range(10):
|
|
session.messages.append({"role": "user", "content": f"msg{i}"})
|
|
session.last_consolidated = 7
|
|
|
|
session.retain_recent_legal_suffix(4)
|
|
|
|
assert len(session.messages) == 4
|
|
assert session.last_consolidated == 1
|
|
|
|
|
|
def test_retain_recent_legal_suffix_zero_clears_session():
|
|
session = Session(key="test:trim-zero")
|
|
for i in range(10):
|
|
session.messages.append({"role": "user", "content": f"msg{i}"})
|
|
session.last_consolidated = 5
|
|
|
|
session.retain_recent_legal_suffix(0)
|
|
|
|
assert session.messages == []
|
|
assert session.last_consolidated == 0
|
|
|
|
|
|
def test_retain_recent_legal_suffix_keeps_legal_tool_boundary():
|
|
session = Session(key="test:trim-tools")
|
|
session.messages.append({"role": "user", "content": "old"})
|
|
session.messages.extend(_tool_turn("old", 0))
|
|
session.messages.append({"role": "user", "content": "keep"})
|
|
session.messages.extend(_tool_turn("keep", 0))
|
|
session.messages.append({"role": "assistant", "content": "done"})
|
|
|
|
session.retain_recent_legal_suffix(4)
|
|
|
|
history = session.get_history(max_messages=500)
|
|
_assert_no_orphans(history)
|
|
assert history[0]["role"] == "user"
|
|
assert history[0]["content"] == "keep"
|
|
|
|
|
|
# --- last_consolidated > 0 ---
|
|
|
|
def test_orphan_trim_with_last_consolidated():
|
|
"""Orphan trimming works correctly when session is partially consolidated."""
|
|
session = Session(key="test:consolidated")
|
|
for i in range(10):
|
|
session.messages.append({"role": "user", "content": f"old {i}"})
|
|
session.messages.extend(_tool_turn("cons", i))
|
|
session.last_consolidated = 30
|
|
|
|
session.messages.append({"role": "user", "content": "recent"})
|
|
for i in range(15):
|
|
session.messages.extend(_tool_turn("new", i))
|
|
session.messages.append({"role": "user", "content": "latest"})
|
|
|
|
history = session.get_history(max_messages=20)
|
|
_assert_no_orphans(history)
|
|
assert all(m.get("role") != "tool" or m["tool_call_id"].startswith("new_") for m in history)
|
|
|
|
|
|
# --- Edge: no tool messages at all ---
|
|
|
|
def test_no_tool_messages_unchanged():
|
|
session = Session(key="test:plain")
|
|
for i in range(5):
|
|
session.messages.append({"role": "user", "content": f"q{i}"})
|
|
session.messages.append({"role": "assistant", "content": f"a{i}"})
|
|
|
|
history = session.get_history(max_messages=6)
|
|
assert len(history) == 6
|
|
_assert_no_orphans(history)
|
|
|
|
|
|
# --- Edge: all leading messages are orphan tool results ---
|
|
|
|
def test_all_orphan_prefix_stripped():
|
|
"""If the window starts with orphan tool results and nothing else, they're all dropped."""
|
|
session = Session(key="test:all-orphan")
|
|
session.messages.append({"role": "tool", "tool_call_id": "gone_1", "name": "x", "content": "ok"})
|
|
session.messages.append({"role": "tool", "tool_call_id": "gone_2", "name": "y", "content": "ok"})
|
|
session.messages.append({"role": "user", "content": "fresh start"})
|
|
session.messages.append({"role": "assistant", "content": "hi"})
|
|
|
|
history = session.get_history(max_messages=500)
|
|
_assert_no_orphans(history)
|
|
assert history[0]["role"] == "user"
|
|
assert len(history) == 2
|
|
|
|
|
|
# --- Edge: empty session ---
|
|
|
|
def test_empty_session_history():
|
|
session = Session(key="test:empty")
|
|
history = session.get_history(max_messages=500)
|
|
assert history == []
|
|
|
|
|
|
def test_get_history_preserves_reasoning_content():
|
|
session = Session(key="test:reasoning")
|
|
session.messages.append({"role": "user", "content": "hi"})
|
|
session.messages.append({
|
|
"role": "assistant",
|
|
"content": "done",
|
|
"reasoning_content": "hidden chain of thought",
|
|
"thinking_blocks": [{"type": "thinking", "thinking": "hidden chain of thought", "signature": "sig"}],
|
|
})
|
|
|
|
history = session.get_history(max_messages=500)
|
|
|
|
assert history == [
|
|
{"role": "user", "content": "hi"},
|
|
{
|
|
"role": "assistant",
|
|
"content": "done",
|
|
"reasoning_content": "hidden chain of thought",
|
|
"thinking_blocks": [{
|
|
"type": "thinking",
|
|
"thinking": "hidden chain of thought",
|
|
"signature": "sig",
|
|
}],
|
|
},
|
|
]
|
|
|
|
|
|
def test_get_history_annotates_user_turns_but_not_assistant_turns():
|
|
"""Only user turns carry the timestamp prefix.
|
|
|
|
Annotating assistant turns trains the model (via in-context examples) to
|
|
start its own replies with ``[Message Time: ...]``. User-side stamps are
|
|
enough to pin adjacent assistant replies for relative-time reasoning.
|
|
"""
|
|
session = Session(key="test:timestamps")
|
|
session.messages.append({
|
|
"role": "user",
|
|
"content": "10 点提醒是昨天发生的",
|
|
"timestamp": "2026-04-26T22:00:00",
|
|
})
|
|
session.messages.append({
|
|
"role": "assistant",
|
|
"content": "记下来了",
|
|
"timestamp": "2026-04-26T22:00:05",
|
|
})
|
|
|
|
history = session.get_history(max_messages=500, include_timestamps=True)
|
|
|
|
assert history == [
|
|
{
|
|
"role": "user",
|
|
"content": "[Message Time: 2026-04-26T22:00:00]\n10 点提醒是昨天发生的",
|
|
},
|
|
{
|
|
"role": "assistant",
|
|
"content": "记下来了",
|
|
},
|
|
]
|
|
|
|
|
|
def test_get_history_annotates_proactive_assistant_deliveries_with_timestamps():
|
|
"""Cron / heartbeat assistant pushes still carry a timestamp prefix.
|
|
|
|
These proactive deliveries can sit hours away from the next user reply,
|
|
so the model needs to know when they fired. They are rare enough that
|
|
they don't act as in-context demonstrations encouraging the model to
|
|
prefix its own normal replies with ``[Message Time: ...]``.
|
|
"""
|
|
session = Session(key="test:proactive-timestamps")
|
|
session.messages.append({
|
|
"role": "assistant",
|
|
"content": "记得喝水",
|
|
"timestamp": "2026-04-26T15:00:00",
|
|
"_channel_delivery": True,
|
|
})
|
|
session.messages.append({
|
|
"role": "user",
|
|
"content": "好",
|
|
"timestamp": "2026-04-26T18:00:00",
|
|
})
|
|
|
|
history = session.get_history(max_messages=500, include_timestamps=True)
|
|
|
|
assert history == [
|
|
{
|
|
"role": "assistant",
|
|
"content": "[Message Time: 2026-04-26T15:00:00]\n记得喝水",
|
|
},
|
|
{
|
|
"role": "user",
|
|
"content": "[Message Time: 2026-04-26T18:00:00]\n好",
|
|
},
|
|
]
|
|
|
|
|
|
def test_get_history_does_not_annotate_tool_results_with_timestamps():
|
|
session = Session(key="test:tool-timestamps")
|
|
session.messages.append({"role": "user", "content": "run tool"})
|
|
session.messages.extend(_tool_turn("ts", 0))
|
|
session.messages[-1]["timestamp"] = "2026-04-26T22:00:10"
|
|
|
|
history = session.get_history(max_messages=500, include_timestamps=True)
|
|
|
|
tool_result = history[-1]
|
|
assert tool_result["role"] == "tool"
|
|
assert tool_result["content"] == "ok"
|
|
|
|
|
|
# --- Window cuts mid-group: assistant present but some tool results orphaned ---
|
|
|
|
def test_window_cuts_mid_tool_group():
|
|
"""If the window starts between an assistant's tool results, the partial group is trimmed."""
|
|
session = Session(key="test:mid-cut")
|
|
session.messages.append({"role": "user", "content": "setup"})
|
|
session.messages.append({
|
|
"role": "assistant", "content": None,
|
|
"tool_calls": [
|
|
{"id": "split_a", "type": "function", "function": {"name": "x", "arguments": "{}"}},
|
|
{"id": "split_b", "type": "function", "function": {"name": "y", "arguments": "{}"}},
|
|
],
|
|
})
|
|
session.messages.append({"role": "tool", "tool_call_id": "split_a", "name": "x", "content": "ok"})
|
|
session.messages.append({"role": "tool", "tool_call_id": "split_b", "name": "y", "content": "ok"})
|
|
session.messages.append({"role": "user", "content": "next"})
|
|
session.messages.extend(_tool_turn("intact", 0))
|
|
session.messages.append({"role": "assistant", "content": "final"})
|
|
|
|
# Window of 6 should cut off the "setup" user msg and the assistant with split_a/split_b,
|
|
# leaving orphan tool results for split_a at the front.
|
|
history = session.get_history(max_messages=6)
|
|
_assert_no_orphans(history)
|
|
|
|
|
|
# --- Image breadcrumbs: media kwarg is synthesized into content for replay ---
|
|
|
|
|
|
def test_get_history_synthesizes_image_breadcrumb_from_media_kwarg():
|
|
"""Persisted user turns carry image paths as a ``media`` kwarg; LLM
|
|
replay must still see an ``[image: path]`` breadcrumb so the assistant's
|
|
follow-up reply has a referent instead of trailing an empty user row."""
|
|
session = Session(key="test:media")
|
|
session.messages.append(
|
|
{"role": "user", "content": "look", "media": ["/m/a.png", "/m/b.png"]}
|
|
)
|
|
session.messages.append({"role": "assistant", "content": "nice"})
|
|
|
|
history = session.get_history(max_messages=500)
|
|
|
|
assert history == [
|
|
{"role": "user", "content": "look\n[image: /m/a.png]\n[image: /m/b.png]"},
|
|
{"role": "assistant", "content": "nice"},
|
|
]
|
|
|
|
|
|
def test_get_history_synthesizes_breadcrumb_for_image_only_turn():
|
|
"""Turns with no text but attached images must not replay as empty
|
|
strings — the LLM would otherwise see a bare user turn followed by an
|
|
unexplained assistant answer."""
|
|
session = Session(key="test:image-only")
|
|
session.messages.append({"role": "user", "content": "", "media": ["/m/pic.png"]})
|
|
session.messages.append({"role": "assistant", "content": "I see a cat"})
|
|
|
|
history = session.get_history(max_messages=500)
|
|
|
|
assert history[0] == {"role": "user", "content": "[image: /m/pic.png]"}
|
|
|
|
|
|
def test_get_history_ignores_media_kwarg_on_non_user_rows():
|
|
"""``media`` only ever appears on user entries in practice, but the
|
|
synthesizer must be defensive: assistants / tools with list content
|
|
don't get the breadcrumb pasted on top."""
|
|
session = Session(key="test:defensive")
|
|
session.messages.append(
|
|
{
|
|
"role": "assistant",
|
|
"content": [{"type": "text", "text": "structured"}],
|
|
"media": ["/m/x.png"], # nonsense but shouldn't crash
|
|
}
|
|
)
|
|
history = session.get_history(max_messages=500)
|
|
# List content is passed through verbatim — the synthesizer only
|
|
# rewrites plain-string content.
|
|
assert history[0]["content"] == [{"type": "text", "text": "structured"}]
|
|
|
|
|
|
def test_get_history_respects_max_tokens(monkeypatch):
|
|
session = Session(key="test:token-cap")
|
|
session.messages.extend(
|
|
[
|
|
{"role": "user", "content": "u1"},
|
|
{"role": "assistant", "content": "a1"},
|
|
{"role": "user", "content": "u2"},
|
|
{"role": "assistant", "content": "a2"},
|
|
{"role": "user", "content": "u3"},
|
|
{"role": "assistant", "content": "a3"},
|
|
]
|
|
)
|
|
|
|
token_map = {"u1": 50, "a1": 50, "u2": 50, "a2": 50, "u3": 50, "a3": 50}
|
|
monkeypatch.setattr(
|
|
"nanobot.session.manager.estimate_message_tokens",
|
|
lambda message: token_map.get(message.get("content"), 0),
|
|
)
|
|
|
|
history = session.get_history(max_messages=500, max_tokens=120)
|
|
assert [m["content"] for m in history] == ["u3", "a3"]
|
|
|
|
|
|
def test_get_history_recovers_user_when_token_slice_would_be_assistant_only(monkeypatch):
|
|
session = Session(key="test:assistant-only-slice")
|
|
session.messages.extend(
|
|
[
|
|
{"role": "user", "content": "u1"},
|
|
{"role": "assistant", "content": "a1"},
|
|
{"role": "user", "content": "u2"},
|
|
{"role": "assistant", "content": "a2"},
|
|
]
|
|
)
|
|
token_map = {"u1": 100, "a1": 100, "u2": 100, "a2": 100}
|
|
monkeypatch.setattr(
|
|
"nanobot.session.manager.estimate_message_tokens",
|
|
lambda message: token_map.get(message.get("content"), 0),
|
|
)
|
|
|
|
history = session.get_history(max_messages=500, max_tokens=100)
|
|
assert [m["content"] for m in history] == ["u2", "a2"]
|
|
|
|
|
|
def test_retain_recent_legal_suffix_hard_cap_with_long_non_user_chain():
|
|
session = Session(key="test:hard-cap-chain")
|
|
session.messages.append({"role": "user", "content": "u0"})
|
|
session.messages.append(
|
|
{
|
|
"role": "assistant",
|
|
"content": None,
|
|
"tool_calls": [
|
|
{"id": "c1", "type": "function", "function": {"name": "x", "arguments": "{}"}}
|
|
],
|
|
}
|
|
)
|
|
for i in range(12):
|
|
session.messages.append({"role": "assistant", "content": f"a{i}"})
|
|
|
|
session.retain_recent_legal_suffix(6)
|
|
|
|
assert len(session.messages) <= 6
|