fuquan/nonebot-plugin-llmchat
1
0
Fork 0
mirror of https://github.com/FuQuan233/nonebot-plugin-llmchat.git synced 2025-09-06 19:40:44 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

🐛 修复assistant消息没有正确添加到历史记录的问题

Browse source
This commit is contained in:
FuQuan233 2025-04-27 11:56:38 +08:00
parent fd18f41f17
commit e3973baa37
1 changed files with 13 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

22
nonebot_plugin_llmchat/__init__.py
View file

@ -268,10 +268,6 @@ async def process_messages(group_id: int):
logger.debug( logger.debug(
f"发送API请求 模型：{preset.model_name} 历史消息数：{len(messages)}" f"发送API请求 模型：{preset.model_name} 历史消息数：{len(messages)}"
) )
mcp_client = MCPClient(plugin_config.mcp_servers)
await mcp_client.connect_to_servers()
available_tools = await mcp_client.get_available_tools()
client_config = { client_config = {
"model": preset.model_name, "model": preset.model_name,
@ -280,7 +276,10 @@ async def process_messages(group_id: int):
"timeout": 60, "timeout": 60,
} }
mcp_client = MCPClient(plugin_config.mcp_servers)
if preset.support_mcp: if preset.support_mcp:
await mcp_client.connect_to_servers()
available_tools = await mcp_client.get_available_tools()
client_config["tools"] = available_tools client_config["tools"] = available_tools
response = await client.chat.completions.create( response = await client.chat.completions.create(
@ -291,10 +290,7 @@ async def process_messages(group_id: int):
if response.usage is not None: if response.usage is not None:
logger.debug(f"收到API响应 使用token数：{response.usage.total_tokens}") logger.debug(f"收到API响应 使用token数：{response.usage.total_tokens}")
final_message = []
message = response.choices[0].message message = response.choices[0].message
if message.content:
final_message.append(message.content)
# 处理响应并处理工具调用 # 处理响应并处理工具调用
while preset.support_mcp and message.tool_calls: while preset.support_mcp and message.tool_calls:
@ -302,6 +298,11 @@ async def process_messages(group_id: int):
"role": "assistant", "role": "assistant",
"tool_calls": [tool_call.model_dump() for tool_call in message.tool_calls] "tool_calls": [tool_call.model_dump() for tool_call in message.tool_calls]
}) })
# 发送LLM调用工具时的回复，一般没有
if message.content:
await handler.send(Message(message.content))
# 处理每个工具调用 # 处理每个工具调用
for tool_call in message.tool_calls: for tool_call in message.tool_calls:
tool_name = tool_call.function.name tool_name = tool_call.function.name
@ -326,8 +327,6 @@ async def process_messages(group_id: int):
) )
message = response.choices[0].message message = response.choices[0].message
if message.content:
final_message.append(message.content)
await mcp_client.cleanup() await mcp_client.cleanup()
@ -339,6 +338,11 @@ async def process_messages(group_id: int):
or matched_reasoning_content or matched_reasoning_content
) )
new_messages.append({
"role": "assistant",
"content": reply,
})
# 请求成功后再保存历史记录，保证user和assistant穿插，防止R1模型报错 # 请求成功后再保存历史记录，保证user和assistant穿插，防止R1模型报错
for message in new_messages: for message in new_messages:
state.history.append(message) state.history.append(message)