app.py

from threading import Thread
import gradio as gr
from gradio_client import Client as GrClient
import inspect
from gradio import routes
from typing import List, Type

import requests, os, re, asyncio, queue
import math
import time
import datetime

loop = asyncio.get_event_loop()
gradio_client = GrClient(os.environ.get('GrClient_url'))
# Monkey patch
def get_types(cls_set: List[Type], component: str):
    docset = []
    types = []
    if component == "input":
        for cls in cls_set:
            doc = inspect.getdoc(cls)
            doc_lines = doc.split("\n")
            docset.append(doc_lines[1].split(":")[-1])
            types.append(doc_lines[1].split(")")[0].split("(")[-1])
    else:
        for cls in cls_set:
            doc = inspect.getdoc(cls)
            doc_lines = doc.split("\n")
            docset.append(doc_lines[-1].split(":")[-1])
            types.append(doc_lines[-1].split(")")[0].split("(")[-1])
    return docset, types
routes.get_types = get_types


q = queue.Queue()

def chat():
    global q
    if !q.empty():
        arr = q.get()
        start = time.time()
        result = gradio_client.predict(
            arr[0],
            # str representing input in 'User input' Textbox component
            arr[1],
            arr[2],
    		fn_index=0
        )
        result = str(result)
        
        end = time.time()
    
    
        sec = (end - start)
        result_list = str(datetime.timedelta(seconds=sec)).split(".")
        print("응답 시간 : " + result_list[0] +"\nx:"+ arr[0] +"\nid:"+ arr[1] +"\ncdata:" + arr[2] +"\nresult:"+ result + "\ncollback_url : " + arr[3])
        return result

th_a = Thread(target = chat)
th_a.daemon = True
th_a.start()

# App code
def res(x, id, cdata, url):    
    global q

    arr = [x, id, cdata, url]
    q.put(arr)

    print("\n_Done\n\n")
    return "Done"

with gr.Blocks() as demo:
    count = 0
    aa = gr.Interface(
      fn=res,
      inputs=["text","text", "text", "text"],
      outputs="text",
      description="chat",
    )

    demo.queue(max_size=32).launch(enable_queue=True)