phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-05 12:21:52 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

[api_updates_3] fix CLI for routing_table, bug fixes for memory & safety (#90)

Browse source 
* fix llama stack build

* fix configure

* fix configure for simple case

* configure w/ routing

* move examples config

* fix memory router naming

* issue w/ safety

* fix config w/ safety

* update memory endpoints

* allow providers in api_providers

* configure script works

* all endpoints w/ build->configure->run simple local works

* new example run.yaml

* run openapi generator
This commit is contained in:
Xi Yan 2024-09-23 08:46:33 -07:00 • committed by GitHub
parent 8cf634e615
commit ddebf9b6e7
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
18 changed files with 725 additions and 605 deletions
Download patch file Download diff file Expand all files Collapse all files

20
llama_stack/distribution/server/server.py
View file

@ -35,9 +35,6 @@ from fastapi import Body, FastAPI, HTTPException, Request, Response
from fastapi.exceptions import RequestValidationError
from fastapi.responses import JSONResponse, StreamingResponse
from fastapi.routing import APIRoute
from pydantic import BaseModel, ValidationError
from termcolor import cprint
from typing_extensions import Annotated
from llama_stack.providers.utils.telemetry.tracing import (
end_trace,
@ -45,6 +42,9 @@ from llama_stack.providers.utils.telemetry.tracing import (
SpanStatus,
start_trace,
)
from pydantic import BaseModel, ValidationError
from termcolor import cprint
from typing_extensions import Annotated
from llama_stack.distribution.datatypes import * # noqa: F403
from llama_stack.distribution.distribution import (
@ -307,6 +307,10 @@ async def resolve_impls_with_routing(run_config: StackRunConfig) -> Dict[Api, An
# TODO: check that these APIs are not in the routing table part of the config
providers = all_providers[api]
# skip checks for API whose provider config is specified in routing_table
if isinstance(config, PlaceholderProviderConfig):
continue
if config.provider_id not in providers:
raise ValueError(
f"Unknown provider `{config.provider_id}` is not available for API `{api}`"
@ -315,9 +319,8 @@ async def resolve_impls_with_routing(run_config: StackRunConfig) -> Dict[Api, An
configs[api] = config
apis_to_serve = run_config.apis_to_serve or set(
list(specs.keys()) + list(run_config.routing_tables.keys())
list(specs.keys()) + list(run_config.routing_table.keys())
)
print("apis_to_serve", apis_to_serve)
for info in builtin_automatically_routed_apis():
source_api = info.routing_table_api
@ -331,15 +334,16 @@ async def resolve_impls_with_routing(run_config: StackRunConfig) -> Dict[Api, An
if info.router_api.value not in apis_to_serve:
continue
if source_api.value not in run_config.routing_tables:
print("router_api", info.router_api)
if info.router_api.value not in run_config.routing_table:
raise ValueError(f"Routing table for `{source_api.value}` is not provided?")
routing_table = run_config.routing_tables[source_api.value]
routing_table = run_config.routing_table[info.router_api.value]
providers = all_providers[info.router_api]
inner_specs = []
for rt_entry in routing_table.entries:
for rt_entry in routing_table:
if rt_entry.provider_id not in providers:
raise ValueError(
f"Unknown provider `{rt_entry.provider_id}` is not available for API `{api}`"