Error in task_handler 和 NoneType 错误

Issue Description / 问题描述

用GTP3.5 显示的事Erro in task....
用GPT4 则出现 NoneType错误。 / 请详细描述您遇到的错误或问题。

Steps to Reproduce / 复现步骤

在codebase, 严格按照示例走。 用的命令行，提示词是:
Search the rank of Tsinghua University in computer science field.

Please provide the specific steps to reproduce the error. / 请提供复现错误的具体步骤。

Expected Behavior / 预期行为

Describe the behavior you expected to see. / 请描述您期望的正确行为。

Environment / 环境信息

github上的 codespace
conda
docker compose

• Operating System / 操作系统：
• Python Version / Python 版本：
• Other Relevant Information / 其他相关信息：

Error Screenshots or Logs / 错误截图或日志

GPT 4:
-=-=-=-=-=-=-= GENERATE INITIAL_PLAN -=-=-=-=-=-=-=
Constructing an Agent: PlanGenerateAgent
chatcompletion: using gpt-4
Traceback (most recent call last):
File "/workspaces/XAgent/XAgentServer/server.py", line 104, in interact
task_handler.outer_loop()
File "/workspaces/XAgent/XAgent/workflow/task_handler.py", line 82, in outer_loop
self.plan_agent.initial_plan_generation(
File "/workspaces/XAgent/XAgent/workflow/plan_exec.py", line 164, in initial_plan_generation
new_message , _ = agent.parse(
^^^^^^^^^^^^
File "/workspaces/XAgent/XAgent/agent/plan_generate_agent/agent.py", line 49, in parse
return self.generate(
^^^^^^^^^^^^^^
File "/workspaces/XAgent/XAgent/agent/base_agent.py", line 121, in generate
response = objgenerator.chatcompletion(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/tenacity/init.py", line 289, in wrapped_f
return self(f, *args, **kw)
^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/tenacity/init.py", line 379, in call
do = self.iter(retry_state=retry_state)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/tenacity/init.py", line 314, in iter
return fut.result()
^^^^^^^^^^^^
File "/opt/conda/envs/xagent/lib/python3.11/concurrent/futures/_base.py", line 449, in result
return self.__get_result()
^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/xagent/lib/python3.11/concurrent/futures/_base.py", line 401, in __get_result
raise self._exception
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/tenacity/init.py", line 382, in call
result = fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/workspaces/XAgent/XAgent/ai_functions/request/obj_generator.py", line 72, in chatcompletion
response = self.function_call_refine(kwargs,response)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/workspaces/XAgent/XAgent/ai_functions/request/obj_generator.py", line 194, in function_call_refine
function_schema = list(filter(lambda x: x['name'] == response['choices'][0]['message']['function_call']['name'],req_kwargs['functions']))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/workspaces/XAgent/XAgent/ai_functions/request/obj_generator.py", line 194, in
function_schema = list(filter(lambda x: x['name'] == response['choices'][0]['message']['function_call']['name'],req_kwargs['functions']))
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^
TypeError: 'NoneType' object is not subscriptable

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/workspaces/XAgent/command.py", line 273, in task_handler
server.interact(interaction=interaction)
File "/workspaces/XAgent/XAgentServer/server.py", line 106, in interact
raise XAgentRunningError(str(e)) from e
XAgentServer.exts.exception_ext.XAgentRunningError: 'NoneType' object is not subscriptable
Error in task_handler of 1efc760dc4ea43c2a93b0f1a0bbda3b7: 'NoneType' object is not subscriptable

=================
GPT 3.5
-=-=-=-=-=-=-= GENERATE INITIAL_PLAN -=-=-=-=-=-=-=
Constructing an Agent: PlanGenerateAgent
chatcompletion: using gpt-3.5-turbo-16k
Traceback (most recent call last):
File "/workspaces/XAgent/XAgentServer/server.py", line 104, in interact
task_handler.outer_loop()
File "/workspaces/XAgent/XAgent/workflow/task_handler.py", line 82, in outer_loop
self.plan_agent.initial_plan_generation(
File "/workspaces/XAgent/XAgent/workflow/plan_exec.py", line 178, in initial_plan_generation
subtasks = json5.loads(new_message["function_call"]["arguments"])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/json5/lib.py", line 106, in loads
ast, err, _ = parser.parse()
^^^^^^^^^^^^^^
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/json5/parser.py", line 19, in parse
self.grammar()
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/json5/parser.py", line 160, in grammar
self._seq(
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/json5/parser.py", line 102, in _seq
rule()
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/json5/parser.py", line 172, in sp
self._star(self.ws)
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/json5/parser.py", line 93, in _star
rule()
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/json5/parser.py", line 175, in ws
self._choose(
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/json5/parser.py", line 109, in _choose
rule()
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/json5/parser.py", line 190, in ws__c0
self._ch(' ')
File "/opt/conda/envs/xagent/lib/python3.11/site-packages/json5/parser.py", line 117, in _ch
if p < self.end and self.msg[p] == ch:
~~~~~~~~^^^
KeyError: 0

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/workspaces/XAgent/command.py", line 273, in task_handler
server.interact(interaction=interaction)
File "/workspaces/XAgent/XAgentServer/server.py", line 106, in interact
raise XAgentRunningError(str(e)) from e
XAgentServer.exts.exception_ext.XAgentRunningError: 0
Error in task_handler of 5cd5ce8d237245ee8d67164618d20d18: 0

If possible, please provide relevant screenshots or logs of the error. / 如果可能，请提供相关的错误截图或日志文件。

Additional Notes / 其他备注

If you have any additional information or notes, please add them here. / 如果有其他补充信息，请在此处添加。

Start with XAgent/ai_functions/request/obj_generator.py, especially chatcompletion and function_call_refine, then trace the callers in XAgent/workflow/plan_exec.py and task_handler.py. Reproduce the prompt with GPT-3.5 and GPT-4 in the reported Codespace environment, comparing responses with missing function_call data or invalid arguments. Done should mean the task no longer fails with the reported NoneType or parsing errors, with the expected behavior confirmed.