phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-08-12 04:50:39 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

deprecate allow_turn_resume

Browse source
This commit is contained in:
Xi Yan 2025-03-03 17:27:43 -08:00
parent c3155cb1bc
commit 79d23dab1b
3 changed files with 14 additions and 23 deletions
Download patch file Download diff file Expand all files Collapse all files

4
llama_stack/apis/agents/agents.py
View file

@ -296,9 +296,6 @@ class AgentTurnCreateRequest(AgentConfigOverridablePerTurn):
stream: Optional[bool] = False
tool_config: Optional[ToolConfig] = None
# TODO (xiyan): temporary flag, will remove for 0.1.5
allow_turn_resume: Optional[bool] = False
@json_schema_type
class AgentTurnResumeRequest(BaseModel):
@ -355,7 +352,6 @@ class Agents(Protocol):
documents: Optional[List[Document]] = None,
toolgroups: Optional[List[AgentToolGroup]] = None,
tool_config: Optional[ToolConfig] = None,
allow_turn_resume: Optional[bool] = False,
) -> Union[Turn, AsyncIterator[AgentTurnResponseStreamChunk]]: ...
@webmethod(

31
llama_stack/providers/inline/agents/meta_reference/agent_instance.py
View file

@ -244,22 +244,13 @@ class ChatAgent(ShieldRunnerMixin):
)
await self.storage.add_turn_to_session(request.session_id, turn)
if output_message.tool_calls and request.allow_turn_resume:
chunk = AgentTurnResponseStreamChunk(
event=AgentTurnResponseEvent(
payload=AgentTurnResponseTurnAwaitingInputPayload(
turn=turn,
)
)
)
else:
chunk = AgentTurnResponseStreamChunk(
event=AgentTurnResponseEvent(
payload=AgentTurnResponseTurnCompletePayload(
turn=turn,
)
chunk = AgentTurnResponseStreamChunk(
event=AgentTurnResponseEvent(
payload=AgentTurnResponseTurnCompletePayload(
turn=turn,
)
)
)
yield chunk
@ -686,10 +677,16 @@ class ChatAgent(ShieldRunnerMixin):
message.content = [message.content] + output_attachments
yield message
else:
logcat.debug("agents", f"completion message with EOM (iter: {n_iter}): {str(message)}")
logcat.debug(
"agents",
f"completion message with EOM (iter: {n_iter}): {str(message)}",
)
input_messages = input_messages + [message]
else:
logcat.debug("agents", f"completion message (iter: {n_iter}) from the model: {str(message)}")
logcat.debug(
"agents",
f"completion message (iter: {n_iter}) from the model: {str(message)}",
)
# 1. Start the tool execution step and progress
step_id = str(uuid.uuid4())
yield AgentTurnResponseStreamChunk(
@ -810,7 +807,7 @@ class ChatAgent(ShieldRunnerMixin):
) -> Tuple[List[ToolDefinition], Dict[str, str]]:
# Determine which tools to include
agent_config_toolgroups = {
toolgroup.name if isinstance(toolgroup, AgentToolGroupWithArgs) else toolgroup
(toolgroup.name if isinstance(toolgroup, AgentToolGroupWithArgs) else toolgroup)
for toolgroup in self.agent_config.toolgroups
}
toolgroups_for_turn_set = (

2
llama_stack/providers/inline/agents/meta_reference/agents.py
View file

@ -140,7 +140,6 @@ class MetaReferenceAgentsImpl(Agents):
documents: Optional[List[Document]] = None,
stream: Optional[bool] = False,
tool_config: Optional[ToolConfig] = None,
allow_turn_resume: Optional[bool] = False,
) -> AsyncGenerator:
request = AgentTurnCreateRequest(
agent_id=agent_id,
@ -150,7 +149,6 @@ class MetaReferenceAgentsImpl(Agents):
toolgroups=toolgroups,
documents=documents,
tool_config=tool_config,
allow_turn_resume=allow_turn_resume,
)
if stream:
return self._create_agent_turn_streaming(request)