The framework design of the aios kernel has been basically completed, as well as the key logic code centered on workflow.
This commit is contained in:
Liu Zhicong
+182
-14
@@ -1,38 +1,206 @@
|
||||
import environment
|
||||
import agent_prompt,agent_msg
|
||||
|
||||
import logging
|
||||
import asyncio
|
||||
from typing import Optional,Tuple
|
||||
|
||||
from .environment import environment,environment_event
|
||||
from .agent import agent_prompt,agent_msg,ai_chat_session
|
||||
from .role import ai_role
|
||||
from .ai_function import call_chain
|
||||
from .compute_kernel import compute_kernel
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
class ai_message_filter:
|
||||
def __init__(self) -> None:
|
||||
pass
|
||||
def select(self,msg:agent_msg) -> ai_role:
|
||||
pass
|
||||
|
||||
class ai_workflow:
|
||||
def __init__(self) -> None:
|
||||
self.rule_prompt : agent_prompt = None
|
||||
self.workflow_config = None
|
||||
self.context = None
|
||||
|
||||
self.role_group = None
|
||||
self.input_filter : ai_message_filter= None
|
||||
self.msg_queue = []
|
||||
self.connected_environment = {}
|
||||
|
||||
def load_from_disk(self,config_path:str,context_dir_path) -> int:
|
||||
pass
|
||||
|
||||
def send_msg(self,msg:agent_msg,target_group:str = None) -> None:
|
||||
if target_group is None:
|
||||
target_group = self.get_default_group()
|
||||
#workflow is asynchronous.
|
||||
# When processing one message, it can process another message at the same time.
|
||||
# chatsession is synchronous, it has to wait for the previous message to finish processing before it can process the next message.
|
||||
# Therefore, post a message needs to specify the session_id explicitly, if not specified it will be automatically created by workflow.
|
||||
def post_msg(self,msg:agent_msg) -> None:
|
||||
self.msg_queue.append(msg)
|
||||
return
|
||||
|
||||
async def send_msg(self,msg:agent_msg) -> str:
|
||||
pass
|
||||
|
||||
async def run(self):
|
||||
# TODO add tracking design of msg processing
|
||||
while True:
|
||||
the_msg = await self._pop_msg()
|
||||
chatsession:ai_chat_session = self._get_chat_session_for_msg(the_msg)
|
||||
if chatsession is None:
|
||||
logger.error(f"get_chat_session_for_msg return None for :{the_msg}")
|
||||
continue
|
||||
|
||||
chatsession.append_recv(the_msg)
|
||||
|
||||
async def _process_msg(msg:agent_msg,the_role) -> None:
|
||||
# prompt generat progress is most important part of workflow(app) develope
|
||||
prompt = agent_prompt()
|
||||
prompt.append(the_role.get_prompt())
|
||||
prompt.append(self.get_workflow_rule_prompt())
|
||||
prompt.append(self._get_function_prompt(the_role.get_name()))
|
||||
prompt.append(self._get_knowlege_prompt(the_role.get_name()))
|
||||
prompt.append(await self._get_prompt_from_session(chatsession,the_role.get_name())) # chat context
|
||||
|
||||
result = await compute_kernel().do_llm_completion(prompt,the_role.agent.get_llm_model_name(),the_role.agent.get_max_token_size())
|
||||
final_result = result
|
||||
result_type : str = self._get_llm_result_type(result)
|
||||
is_ignore = False
|
||||
match result_type:
|
||||
case "function":
|
||||
callchain:call_chain = self._parse_function_call_chain(result)
|
||||
resp = await callchain.exec()
|
||||
if callchain.have_result():
|
||||
# generator proc resp prompt with WAITING state
|
||||
proc_resp_prompt:agent_prompt = self._get_resp_prompt(resp,msg,the_role,prompt,chatsession)
|
||||
final_result = await compute_kernel().do_llm_completion(proc_resp_prompt,the_role.agent.get_llm_model_name(),the_role.agent.get_max_token_size())
|
||||
return final_result
|
||||
|
||||
|
||||
case "send_message":
|
||||
# send message to other / sub workflow
|
||||
next_msg:agent_msg = self._parse_to_msg(result)
|
||||
if next_msg is not None:
|
||||
# TODO: Next Target can be another role in workflow
|
||||
next_workflow:ai_workflow = self.get_workflow(next_msg.get_target())
|
||||
inner_chat_session = the_role.agent.get_chat_session(next_msg.get_target(),next_msg.get_session_id())
|
||||
|
||||
inner_chat_session.append_post(next_msg)
|
||||
resp = await next_workflow.send_msg(next_msg)
|
||||
inner_chat_session.append_recv(resp)
|
||||
# generator proc resp prompt with WAITING state
|
||||
proc_resp_prompt:agent_prompt = self._get_resp_prompt(resp,msg,the_role,prompt,chatsession)
|
||||
final_result = await compute_kernel().do_llm_completion(proc_resp_prompt,the_role.agent.get_llm_model_name(),the_role.agent.get_max_token_size())
|
||||
|
||||
return final_result
|
||||
|
||||
case "post_message":
|
||||
# post message to other / sub workflow
|
||||
next_msg:agent_msg = self._parse_to_msg(result)
|
||||
if next_msg is not None:
|
||||
next_workflow:ai_workflow = self.get_workflow(next_msg.get_target())
|
||||
inner_chat_session = the_role.agent.get_chat_session(next_msg.get_target(),next_msg.get_session_id())
|
||||
inner_chat_session.append_post(next_msg)
|
||||
next_workflow.post_msg(next_msg)
|
||||
|
||||
case "ignore":
|
||||
is_ignore = True
|
||||
|
||||
if is_ignore is not True:
|
||||
# TODO : how to get inner chat session?
|
||||
inner_chat_session = the_role.agent.get_chat_session_for_msg(msg)
|
||||
if inner_chat_session is not None:
|
||||
inner_chat_session.append_input(msg)
|
||||
inner_chat_session.append_result(final_result)
|
||||
|
||||
return result
|
||||
|
||||
async def _workflow_process_msg(msg:agent_msg) -> None:
|
||||
final_result = None
|
||||
if self.input_filter is not None:
|
||||
select_role = self.input_filter.select(msg)
|
||||
if select_role is not None:
|
||||
result = await _process_msg(msg,select_role)
|
||||
if result is None:
|
||||
logger.error(f"_process_msg return None for :{msg}")
|
||||
return
|
||||
if chatsession is not None:
|
||||
chatsession.append_post(result)
|
||||
final_result = result
|
||||
|
||||
else:
|
||||
results = {}
|
||||
for this_role in self.role_group.roles:
|
||||
a_result = asyncio.create_task(_process_msg(msg,this_role))
|
||||
results[this_role.get_name()] = a_result
|
||||
|
||||
# merge result from all roles
|
||||
# TODO: one input msg can have multiple result msg, at this while ,we only support one result msg
|
||||
final_result:agent_msg = self._merge_msg_result(results)
|
||||
if chatsession is not None:
|
||||
chatsession.append_post(final_result)
|
||||
|
||||
if final_result is not None:
|
||||
# TODO post message to source
|
||||
pass
|
||||
|
||||
asyncio.create_task(_workflow_process_msg(the_msg))
|
||||
|
||||
async def _pop_msg(self) -> agent_msg:
|
||||
pass
|
||||
|
||||
def run(self):
|
||||
|
||||
def _get_chat_session_for_msg(self,msg:agent_msg) -> ai_chat_session:
|
||||
pass
|
||||
|
||||
def _pop_msg(self) -> Tuple[agent_msg,str]:
|
||||
async def _get_prompt_from_session(self,chatsession:ai_chat_session,role_name:str) -> agent_prompt:
|
||||
pass
|
||||
|
||||
def _get_msg_queue(self,session_id:str):
|
||||
pass
|
||||
|
||||
def get_default_group(self) -> agent_group:
|
||||
def _merge_msg_result(self,results:dict) -> agent_msg:
|
||||
pass
|
||||
|
||||
def get_group(self,group_name:str) -> agent_group:
|
||||
def _get_function_prompt(self,role_name:str) -> agent_prompt:
|
||||
pass
|
||||
|
||||
def _get_knowlege_prompt(self,role_name:str) -> agent_prompt:
|
||||
pass
|
||||
|
||||
def _get_resp_prompt(self,resp:str,msg:agent_msg,role:ai_role,prompt:agent_prompt,chatsession:ai_chat_session) -> agent_prompt:
|
||||
pass
|
||||
|
||||
def get_workflow_rule_prompt(self) -> agent_prompt:
|
||||
return self.rule_prompt
|
||||
|
||||
def _get_llm_result_type(self,llm_resp_str:str) -> str:
|
||||
pass
|
||||
|
||||
def get_inner_environment(self) -> environment:
|
||||
def _parse_function_call_chain(self,llm_resp_str) -> call_chain:
|
||||
pass
|
||||
|
||||
def _parse_to_msg(self,llm_resp_str) -> agent_msg:
|
||||
pass
|
||||
|
||||
def get_workflow(self,workflow_name:str) -> ai_workflow:
|
||||
"""get workflow from known workflow list or sub workflow list"""
|
||||
pass
|
||||
|
||||
|
||||
def _env_event_to_msg(self,env_event:environment_event) -> agent_msg:
|
||||
pass
|
||||
|
||||
def get_inner_environment(self,env_id:str) -> environment:
|
||||
pass
|
||||
|
||||
def connect_to_environment(self,env:environment) -> None:
|
||||
pass
|
||||
the_env = self.connected_environment.get(env.get_id())
|
||||
if the_env is None:
|
||||
self.connected_environment[env.get_id()] = env
|
||||
def _env_msg_handler(env_event:environment_event) -> None:
|
||||
the_msg:agent_msg= self._env_event_to_msg(env_event)
|
||||
self.post_msg(the_msg)
|
||||
|
||||
# register all event handler
|
||||
the_env.attach_event_handler(None,_env_msg_handler)
|
||||
else:
|
||||
logger.warn(f"environment {env.get_id()} already connected!")
|
||||
|
||||
|
||||
Reference in New Issue
Block a user