1.workflow has reached a state of basic functionality.
2. implemented AI message bus infrastructure,
This commit is contained in:
Liu Zhicong
+230
-120
@@ -6,146 +6,217 @@ from typing import Optional,Tuple
|
||||
from abc import ABC, abstractmethod
|
||||
|
||||
from .environment import Environment,EnvironmentEvent
|
||||
from .agent import AgentPrompt,AgentMsg,AIChatSession
|
||||
from .role import AIRole
|
||||
from .agent_message import AgentMsg,AgentMsgState
|
||||
from .agent import AgentPrompt,AgentMsg
|
||||
from .chatsession import AIChatSession
|
||||
from .role import AIRole,AIRoleGroup
|
||||
from .ai_function import CallChain
|
||||
from .compute_kernel import ComputeKernel
|
||||
from .bus import AIBus
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
class MessageFilter:
|
||||
def __init__(self) -> None:
|
||||
pass
|
||||
def select(self,msg:AgentMsg) -> AIRole:
|
||||
pass
|
||||
self.filters = {}
|
||||
|
||||
def select(self,msg:AgentMsg) -> str:
|
||||
star_target = self.filters.get("*")
|
||||
if star_target is not None:
|
||||
return star_target
|
||||
|
||||
# TODO: add more filter
|
||||
return None
|
||||
|
||||
def load_from_config(self,config:dict) -> bool:
|
||||
self.filters = config
|
||||
return True
|
||||
|
||||
|
||||
class Workflow:
|
||||
def __init__(self) -> None:
|
||||
self.workflow_name : str = None
|
||||
self.rule_prompt : AgentPrompt = None
|
||||
self.workflow_config = None
|
||||
self.role_group = None
|
||||
self.role_group : dict = None
|
||||
self.input_filter : MessageFilter= None
|
||||
self.msg_queue = Queue()
|
||||
self.connected_environment = {}
|
||||
self.sub_workflows = {}
|
||||
self.owner_workflow = None
|
||||
self.db_file = None
|
||||
|
||||
self.is_start = False
|
||||
self.msg_queue = Queue()
|
||||
|
||||
def get_bus(self) -> AIBus:
|
||||
return AIBus.get_default_bus()
|
||||
|
||||
def set_owner(self,owner):
|
||||
self.owner_workflow = owner
|
||||
|
||||
def load_from_config(self,config:dict) -> bool:
|
||||
if config is None:
|
||||
return False
|
||||
|
||||
def load_from_disk(self,config_path:str,context_dir_path) -> int:
|
||||
pass
|
||||
if config.get("name") is None:
|
||||
logger.error("workflow config must have name")
|
||||
return False
|
||||
self.workflow_name = config.get("name")
|
||||
|
||||
#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:AgentMsg) -> None:
|
||||
self.msg_queue.put_nowait(msg)
|
||||
return
|
||||
|
||||
async def send_msg(self,msg:AgentMsg) -> str:
|
||||
pass
|
||||
|
||||
async def run(self):
|
||||
# TODO add tracking design of msg processing
|
||||
while True:
|
||||
the_msg = await self._pop_msg()
|
||||
chatsession:AIChatSession = 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
|
||||
#if config.get("rule_prompt") is None:
|
||||
# logger.error("workflow config must have rule_prompt")
|
||||
# return False
|
||||
#self.rule_prompt = AgentPrompt()
|
||||
#if self.rule_prompt.load_from_config(config.get("rule_prompt")) is False:
|
||||
# logger.error("Workflow load rule_prompt failed")
|
||||
# return False
|
||||
if config.get("roles") is None:
|
||||
logger.error("workflow config must have roles")
|
||||
return False
|
||||
self.role_group = AIRoleGroup()
|
||||
if self.role_group.load_from_config(config.get("roles")) is False:
|
||||
logger.error("Workflow load role_group failed")
|
||||
return False
|
||||
|
||||
chatsession.append_recv(the_msg)
|
||||
if config.get("input_filter") is not None:
|
||||
self.input_filter = MessageFilter()
|
||||
if self.input_filter.load_from_config(config.get("input_filter")) is False:
|
||||
logger.error("Workflow load input_filter failed")
|
||||
return False
|
||||
|
||||
sub_workflows = config.get("sub_workflows")
|
||||
if sub_workflows is not None:
|
||||
if self._load_sub_workflows(sub_workflows) is False:
|
||||
logger.error("Workflow load sub workflows failed")
|
||||
return False
|
||||
|
||||
#TODO: load env
|
||||
|
||||
async def _process_msg(msg:AgentMsg,the_role) -> None:
|
||||
# prompt generat progress is most important part of workflow(app) develope
|
||||
prompt = AgentPrompt()
|
||||
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
|
||||
return True
|
||||
|
||||
result = await ComputeKernel().do_llm_completion(prompt,the_role.agent.get_llm_model_name(),the_role.agent.get_max_token_size())
|
||||
def _load_sub_workflows(self,config:dict) -> bool:
|
||||
for k,v in config.items():
|
||||
sub_workflow = Workflow()
|
||||
sub_workflow.set_owner(self)
|
||||
if sub_workflow.load_from_config(v) is False:
|
||||
logger.error(f"load sub workflow {k} failed!")
|
||||
return False
|
||||
self.sub_workflows[k] = sub_workflow
|
||||
return True
|
||||
|
||||
async def _process_msg(self,msg:AgentMsg):
|
||||
final_result = None
|
||||
chatsession = None
|
||||
if self.input_filter is not None:
|
||||
select_role_id = self.input_filter.select(msg)
|
||||
if select_role_id is not None:
|
||||
select_role = self.role_group.get(select_role_id)
|
||||
if select_role is None:
|
||||
logger.error(f"input_filter return invalid role id:{select_role_id}, role not found in role_group")
|
||||
return None
|
||||
|
||||
result = await self._role_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
|
||||
result_type : str = self._get_llm_result_type(result)
|
||||
is_ignore = False
|
||||
match result_type:
|
||||
case "function":
|
||||
callchain:CallChain = self._parse_function_call_chain(result)
|
||||
resp = await callchain.exec()
|
||||
if callchain.have_result():
|
||||
# generator proc resp prompt with WAITING state
|
||||
proc_resp_prompt:AgentPrompt = self._get_resp_prompt(resp,msg,the_role,prompt,chatsession)
|
||||
final_result = await ComputeKernel().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:AgentMsg = self._parse_to_msg(result)
|
||||
if next_msg is not None:
|
||||
# TODO: Next Target can be another role in workflow
|
||||
next_workflow: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())
|
||||
else:
|
||||
logger.error(f"input_filter return None for :{msg}")
|
||||
return
|
||||
|
||||
else:
|
||||
results = {}
|
||||
for this_role in self.role_group.roles.values():
|
||||
# TODO : we would do this in parallel
|
||||
a_result = await self._role_process_msg(msg,this_role)
|
||||
results[this_role.get_name()] = a_result
|
||||
|
||||
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:AgentPrompt = self._get_resp_prompt(resp,msg,the_role,prompt,chatsession)
|
||||
final_result = await ComputeKernel().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:AgentMsg = self._parse_to_msg(result)
|
||||
if next_msg is not None:
|
||||
next_workflow: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)
|
||||
# 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:AgentMsg = self._merge_msg_result(results)
|
||||
if chatsession is not None:
|
||||
chatsession.append_post(final_result)
|
||||
|
||||
return final_result
|
||||
|
||||
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:AgentMsg) -> 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:AgentMsg = 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 _role_process_msg(self,msg:AgentMsg,the_role:AIRole) -> None:
|
||||
# TODO : we just record role's chatsession, but in future, we would record workflow's chatsession(like a groupo chat)
|
||||
session_topic = f"{the_role.get_name()}#{msg.sender}#{msg.topic}"
|
||||
chatsession = AIChatSession.get_session(self.workflow_name,session_topic,self.db_file)
|
||||
if chatsession is None:
|
||||
logger.error(f"get session {session_topic}@{self.workflow_name} failed!")
|
||||
return None
|
||||
|
||||
# prompt generat progress is most important part of workflow(app) develope
|
||||
prompt = AgentPrompt()
|
||||
prompt.append(the_role.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))
|
||||
#prompt.append(await self._get_prompt_from_session(chatsession,the_role.get_name())) # chat context
|
||||
|
||||
msg_prompt = AgentPrompt()
|
||||
msg_prompt.messages = [{"role":"user","content":msg.body}]
|
||||
prompt.append(msg_prompt)
|
||||
|
||||
result = await ComputeKernel().do_llm_completion(prompt,the_role.agent.get_llm_model_name(),the_role.agent.get_max_token_size())
|
||||
chatsession.append_recv(msg)
|
||||
final_result = result
|
||||
|
||||
result_type : str = self._get_llm_result_type(result)
|
||||
is_ignore = False
|
||||
match result_type:
|
||||
case "function":
|
||||
callchain:CallChain = self._parse_function_call_chain(result)
|
||||
resp = await callchain.exec()
|
||||
if callchain.have_result():
|
||||
# generator proc resp prompt with WAITING state
|
||||
proc_resp_prompt:AgentPrompt = self._get_resp_prompt(resp,msg,the_role,prompt,chatsession)
|
||||
final_result = await ComputeKernel().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:AgentMsg = self._parse_to_msg(result)
|
||||
if next_msg is not None:
|
||||
next_msg.sender = self.workflow_name
|
||||
logger.info(f"W#{self.workflow_name} send message to {next_msg.get_target()}")
|
||||
resp_msg = await self.get_bus().send_message(next_msg.get_target(),next_msg)
|
||||
if resp_msg is not None:
|
||||
msg_prompt = AgentPrompt()
|
||||
msg_prompt.messages = [{"role":"assistant","content":result},{"role":"user","content":f"{next_msg.get_target()}:{resp_msg.body}"}]
|
||||
|
||||
final_result = await ComputeKernel().do_llm_completion(proc_resp_prompt,the_role.agent.get_llm_model_name(),the_role.agent.get_max_token_size())
|
||||
|
||||
|
||||
case "post_message":
|
||||
# post message to other / sub workflow
|
||||
next_msg:AgentMsg = self._parse_to_msg(result)
|
||||
if next_msg is not None:
|
||||
next_msg.sender = self.workflow_name
|
||||
logger.info(f"W#{self.workflow_name} post message to {next_msg.get_target()}")
|
||||
self.get_bus().post_message(next_msg.get_target(),next_msg)
|
||||
|
||||
case "ignore":
|
||||
is_ignore = True
|
||||
|
||||
if is_ignore:
|
||||
return None
|
||||
|
||||
resp_msg = AgentMsg()
|
||||
resp_msg.set(self.workflow_name,msg.sender,final_result)
|
||||
chatsession.append_post(resp_msg)
|
||||
return resp_msg
|
||||
|
||||
async def _pop_msg(self) -> AgentMsg:
|
||||
pass
|
||||
@@ -153,14 +224,25 @@ class Workflow:
|
||||
def _get_chat_session_for_msg(self,msg:AgentMsg) -> AIChatSession:
|
||||
pass
|
||||
|
||||
async def _get_prompt_from_session(self,chatsession:AIChatSession,role_name:str) -> AgentPrompt:
|
||||
pass
|
||||
async def _get_prompt_from_session(self,chatsession:AIChatSession) -> AgentPrompt:
|
||||
messages = chatsession.read_history() # read last 10 message
|
||||
result_prompt = AgentPrompt()
|
||||
for msg in reversed(messages):
|
||||
if msg.target == chatsession.owner_id:
|
||||
result_prompt.messages.append({"role":"user","content":f"{msg.sender}:{msg.body}"})
|
||||
if msg.sender == chatsession.owner_id:
|
||||
result_prompt.messages.append({"role":"assistant","content":msg.body})
|
||||
|
||||
return result_prompt
|
||||
|
||||
def _get_msg_queue(self,session_id:str):
|
||||
pass
|
||||
|
||||
def _merge_msg_result(self,results:dict) -> AgentMsg:
|
||||
pass
|
||||
# TODO: one input msg can have multiple result msg, at this while ,we only support one result msg
|
||||
for k,v in results.items():
|
||||
if v is not None:
|
||||
return v
|
||||
|
||||
def _get_function_prompt(self,role_name:str) -> AgentPrompt:
|
||||
pass
|
||||
@@ -168,20 +250,48 @@ class Workflow:
|
||||
def _get_knowlege_prompt(self,role_name:str) -> AgentPrompt:
|
||||
pass
|
||||
|
||||
def _get_resp_prompt(self,resp:str,msg:AgentMsg,role:AIRole,prompt:AgentPrompt,chatsession:AIChatSession) -> AgentPrompt:
|
||||
def _get_resp_prompt(self,resp:str,msg:AgentMsg,role:AIRole,prompt:AgentPrompt) -> AgentPrompt:
|
||||
pass
|
||||
|
||||
def get_workflow_rule_prompt(self) -> AgentPrompt:
|
||||
return self.rule_prompt
|
||||
|
||||
def _get_llm_result_type(self,llm_resp_str:str) -> str:
|
||||
pass
|
||||
if llm_resp_str == "ignore":
|
||||
return "ignore"
|
||||
|
||||
if llm_resp_str.find("sendmsg(") != -1:
|
||||
return "send_message"
|
||||
|
||||
if llm_resp_str.find("postmsg(") != -1:
|
||||
return "post_message"
|
||||
|
||||
if llm_resp_str.find("call(") != -1:
|
||||
return "function"
|
||||
|
||||
return "text"
|
||||
|
||||
def _parse_function_call_chain(self,llm_resp_str) -> CallChain:
|
||||
pass
|
||||
|
||||
def _parse_to_msg(self,llm_resp_str) -> AgentMsg:
|
||||
pass
|
||||
lines = llm_resp_str.splitlines()
|
||||
for line in lines:
|
||||
if line.startswith("sendmsg("):
|
||||
line = line[8:]
|
||||
_index = line.find(",")
|
||||
msg = AgentMsg()
|
||||
msg.set("",line[:_index],line[_index+1:])
|
||||
return msg
|
||||
|
||||
if line.startswith("postmsg("):
|
||||
line = line[8:]
|
||||
_index = line.find(",")
|
||||
msg = AgentMsg()
|
||||
msg.set("",line[:_index],line[_index+1:])
|
||||
return msg
|
||||
|
||||
return None
|
||||
|
||||
def get_workflow(self,workflow_name:str):
|
||||
"""get workflow from known workflow list or sub workflow list"""
|
||||
|
||||
Reference in New Issue
Block a user