gardel/Mofox-Core
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix：修复神秘问题

Browse Source
This commit is contained in:
SengokuCola
2025-04-25 19:15:15 +08:00
parent 4f919ce15c
commit 75924bf499
4 changed files with 132 additions and 96 deletions
Download Patch File Download Diff File Expand all files Collapse all files

115
src/plugins/heartFC_chat/heartFC_chat.py
View File

@@ -343,51 +343,48 @@ class HeartFChatting:
# 初始化周期状态
cycle_timers = {}
loop_cycle_start_time = time.monotonic()
with Timer("Total Cycle", cycle_timers):
# 执行规划和处理阶段
async with self._get_cycle_context() as acquired_lock:
if not acquired_lock:
continue
# 记录规划开始时间点
planner_start_db_time = time.time()
# 执行规划和处理阶段
async with self._get_cycle_context() as acquired_lock:
if not acquired_lock:
continue
# 执行规划阶段
with Timer("Planning Phase", cycle_timers):
action_taken, thinking_id = await self._think_plan_execute(
cycle_timers, planner_start_db_time
)
# 更新循环信息
self._current_cycle.set_thinking_id(thinking_id)
self._current_cycle.timers = cycle_timers
# 防止循环过快消耗资源
with Timer("Cycle Delay", cycle_timers):
await self._handle_cycle_delay(action_taken, loop_cycle_start_time, self.log_prefix)
# 记录规划开始时间点
planner_start_db_time = time.time()
# 等待直到所有消息都发送完成
with Timer("Wait Messages Complete", cycle_timers):
while await self._should_skip_cycle(thinking_id):
await asyncio.sleep(0.2)
# 完成当前循环并保存历史
self._current_cycle.complete_cycle()
self._cycle_history.append(self._current_cycle)
# 记录循环信息和计时器结果
timer_strings = []
for name, elapsed in cycle_timers.items():
formatted_time = f"{elapsed * 1000:.2f}毫秒" if elapsed < 1 else f"{elapsed:.2f}秒"
timer_strings.append(f"{name}: {formatted_time}")
logger.debug(
f"{self.log_prefix} 循环 #{self._current_cycle.cycle_id} 完成, "
f"耗时: {self._current_cycle.end_time - self._current_cycle.start_time:.2f}秒, "
f"动作: {self._current_cycle.action_type}"
+ (f"\n计时器详情: {'; '.join(timer_strings)}" if timer_strings else "")
# 执行规划阶段
action_taken, thinking_id = await self._think_plan_execute_loop(
cycle_timers, planner_start_db_time
)
# 更新循环信息
self._current_cycle.set_thinking_id(thinking_id)
self._current_cycle.timers = cycle_timers
# 防止循环过快消耗资源
await self._handle_cycle_delay(action_taken, loop_cycle_start_time, self.log_prefix)
# 等待直到所有消息都发送完成
with Timer("发送消息", cycle_timers):
while await self._should_skip_cycle(thinking_id):
await asyncio.sleep(0.2)
# 完成当前循环并保存历史
self._current_cycle.complete_cycle()
self._cycle_history.append(self._current_cycle)
# 记录循环信息和计时器结果
timer_strings = []
for name, elapsed in cycle_timers.items():
formatted_time = f"{elapsed * 1000:.2f}毫秒" if elapsed < 1 else f"{elapsed:.2f}秒"
timer_strings.append(f"{name}: {formatted_time}")
logger.debug(
f"{self.log_prefix} 第 #{self._current_cycle.cycle_id}次思考完成,"
f"耗时: {self._current_cycle.end_time - self._current_cycle.start_time:.2f}秒, "
f"动作: {self._current_cycle.action_type}"
+ (f"\n计时器详情: {'; '.join(timer_strings)}" if timer_strings else "")
)
except asyncio.CancelledError:
logger.info(f"{self.log_prefix} HeartFChatting: 麦麦的激情水群(HFC)被取消了")
@@ -434,22 +431,22 @@ class HeartFChatting:
logger.error(f"{self.log_prefix} 检查新消息时出错: {e}")
return False
async def _think_plan_execute(
async def _think_plan_execute_loop(
self, cycle_timers: dict, planner_start_db_time: float
) -> tuple[bool, str]:
"""执行规划阶段"""
try:
# 获取子思维思考结果
current_mind = ""
with Timer("SubMind Thinking", cycle_timers):
with Timer("思考", cycle_timers):
current_mind = await self._get_submind_thinking()
# 记录子思维思考内容
if self._current_cycle:
self._current_cycle.set_response_info(sub_mind_thinking=current_mind)
# 执行规划
with Timer("Planner", cycle_timers):
planner_result = await self._planner(current_mind)
with Timer("决策", cycle_timers):
planner_result = await self._planner(current_mind, cycle_timers)
# 在获取规划结果后检查新消息
if await self._check_new_messages(planner_start_db_time):
@@ -471,7 +468,8 @@ class HeartFChatting:
return False, ""
# 根据动作类型执行对应处理
return await self._handle_action(action, reasoning, planner_result.get("emoji_query", ""), cycle_timers, planner_start_db_time)
with Timer("执行", cycle_timers):
return await self._handle_action(action, reasoning, planner_result.get("emoji_query", ""), cycle_timers, planner_start_db_time)
except PlannerError as e:
logger.error(f"{self.log_prefix} 规划错误: {e}")
@@ -684,8 +682,8 @@ class HeartFChatting:
):
"""处理循环延迟"""
cycle_duration = time.monotonic() - cycle_start_time
if cycle_duration > 0.1:
logger.debug(f"{log_prefix} HeartFChatting: 周期耗时 {cycle_duration:.2f}s.")
# if cycle_duration > 0.1:
# logger.debug(f"{log_prefix} HeartFChatting: 周期耗时 {cycle_duration:.2f}s.")
try:
sleep_duration = 0.0
@@ -718,7 +716,7 @@ class HeartFChatting:
logger.error(traceback.format_exc())
return "[思考时出错]"
async def _planner(self, current_mind: str) -> Dict[str, Any]:
async def _planner(self, current_mind: str, cycle_timers: dict) -> Dict[str, Any]:
"""
规划器 (Planner): 使用LLM根据上下文决定是否和如何回复。
@@ -726,15 +724,12 @@ class HeartFChatting:
current_mind: 子思维的当前思考结果
"""
logger.info(f"{self.log_prefix}[Planner] 开始执行规划器")
planner_timers = {} # 用于存储各阶段计时结果
# 获取观察信息
with Timer("获取观察信息", planner_timers):
observation = self.observations[0]
# await observation.observe()
observed_messages = observation.talking_message
observed_messages_str = observation.talking_message_str
observation = self.observations[0]
# await observation.observe()
observed_messages = observation.talking_message
observed_messages_str = observation.talking_message_str
# --- 使用 LLM 进行决策 --- #
action = "no_reply" # 默认动作
@@ -744,7 +739,7 @@ class HeartFChatting:
try:
# 构建提示词
with Timer("构建提示词", planner_timers):
with Timer("构建提示词", cycle_timers):
prompt = await self._build_planner_prompt(
observed_messages_str, current_mind, self.sub_mind.structured_info
)
@@ -756,7 +751,7 @@ class HeartFChatting:
}
# 执行LLM请求
with Timer("LLM请求", planner_timers):
with Timer("LLM回复", cycle_timers):
try:
response = await self.planner_llm._execute_request(
endpoint="/chat/completions", payload=payload, prompt=prompt
@@ -773,7 +768,7 @@ class HeartFChatting:
}
# 处理LLM响应
with Timer("处理LLM响应", planner_timers):
with Timer("使用工具", cycle_timers):
# 使用辅助函数处理工具调用响应
success, arguments, error_msg = process_llm_tool_response(
response, expected_tool_name="decide_reply_action", log_prefix=f"{self.log_prefix}[Planner] "