GradLLM / app.py
johnbridges's picture
.
b2c2f23
raw
history blame
3.85 kB
import asyncio
import gradio as gr
from fastapi import FastAPI
from config import settings
from listener import RabbitListenerBase
from rabbit_repo import RabbitRepo
from service import LLMService
from runners.base import ILLMRunner
# --- Runner factory (stub) ---
class EchoRunner(ILLMRunner):
Type = "EchoRunner"
async def StartProcess(self, llmServiceObj: dict): pass
async def RemoveProcess(self, sessionId: str): pass
async def StopRequest(self, sessionId: str): pass
async def SendInputAndGetResponse(self, llmServiceObj: dict):
# Emits a message back (you can choose queue names per your topology)
pass
async def runner_factory(llmServiceObj: dict) -> ILLMRunner:
# Use llmServiceObj["LLMRunnerType"] to instantiate different runners
return EchoRunner()
# --- Publisher and Service ---
publisher = RabbitRepo(external_source="https://space.external") # put your ExternalUrl if you have one
service = LLMService(publisher, runner_factory)
# --- Handlers mapping .NET FuncName -> service method ---
async def h_start(data): await service.StartProcess(data or {})
async def h_user(data): await service.UserInput(data or {})
async def h_remove(data): await service.RemoveSession(data or {})
async def h_stop(data): await service.StopRequest(data or {})
async def h_qir(data): await service.QueryIndexResult(data or {})
async def h_getreg(data): await service.GetFunctionRegistry(False)
async def h_getreg_f(data): await service.GetFunctionRegistry(True)
handlers = {
"llmStartSession": h_start,
"llmUserInput": h_user,
"llmRemoveSession": h_remove,
"llmStopRequest": h_stop,
"queryIndexResult": h_qir,
"getFunctionRegistry": h_getreg,
"getFunctionRegistryFiltered": h_getreg_f,
}
listener = RabbitListenerBase(service_id=settings.SERVICE_ID, handlers=handlers)
# Declarations mirror your C# InitRabbitMQObjs()
DECLS = [
{"ExchangeName":"llmStartSession"+settings.SERVICE_ID, "FuncName":"llmStartSession", "MessageTimeout":600000, "RoutingKeys":[settings.RABBIT_ROUTING_KEY]},
{"ExchangeName":"llmUserInput"+settings.SERVICE_ID, "FuncName":"llmUserInput", "MessageTimeout":600000, "RoutingKeys":[settings.RABBIT_ROUTING_KEY]},
{"ExchangeName":"llmRemoveSession"+settings.SERVICE_ID, "FuncName":"llmRemoveSession", "MessageTimeout":60000, "RoutingKeys":[settings.RABBIT_ROUTING_KEY]},
{"ExchangeName":"llmStopRequest"+settings.SERVICE_ID, "FuncName":"llmStopRequest", "MessageTimeout":60000, "RoutingKeys":[settings.RABBIT_ROUTING_KEY]},
{"ExchangeName":"queryIndexResult"+settings.SERVICE_ID, "FuncName":"queryIndexResult", "MessageTimeout":60000, "RoutingKeys":[settings.RABBIT_ROUTING_KEY]},
{"ExchangeName":"getFunctionRegistry"+settings.SERVICE_ID, "FuncName":"getFunctionRegistry", "MessageTimeout":60000, "RoutingKeys":[settings.RABBIT_ROUTING_KEY]},
{"ExchangeName":"getFunctionRegistryFiltered"+settings.SERVICE_ID, "FuncName":"getFunctionRegistryFiltered", "MessageTimeout":60000, "RoutingKeys":[settings.RABBIT_ROUTING_KEY]},
]
# --- Gradio UI (for smoke test) ---
async def ping():
return "ok"
with gr.Blocks() as demo:
gr.Markdown("### LLM Runner (Python) listening on RabbitMQ")
btn = gr.Button("Ping")
out = gr.Textbox()
btn.click(ping, inputs=None, outputs=out)
# --- FastAPI mount + lifecycle ---
app = FastAPI()
app = gr.mount_gradio_app(app, demo, path="/")
@get("/health")
async def health():
return {"status":"ok"}
@on_event("startup")
async def on_start():
await publisher.connect()
await service.init()
await listener.start(DECLS)
@on_event("shutdown")
async def on_stop():
# aio-pika RobustConnection closes on GC; optionally add explicit closes if you add references
pass
if __name__ == "__main__":
import uvicorn
uvicorn.run(app, host="0.0.0.0", port=7860)