Implement Tornado Asynchronous Execution for GET and POST Request – Tornado Tutorial

In this tutorial, we will introduce how to implement tornado asynchronous execution for GET and POST request. It is very useful for us to increase the performance of the tornado web server.

We usually build a web server using tornado as follows:

import tornado
from tornado.web import RequestHandler

class ChexianVoiceprintHandler_V2(RequestHandler):
    def get(self, *args, **kwargs):
        pass

    def post(self, *args, **kwargs):
        # do something, for example: create voiceprint
		result = do()
        self.write(result)
		
def make_app():
    setting = dict(

    )
    return tornado.web.Application(
        [(r'/ChexianVoiceprintRequest_V2', ChexianVoiceprintHandler_V2)]
    )


if __name__ == '__main__':
    app = make_app()
    app.listen(9003)
    tornado.ioloop.IOLoop.current().start()

However, it is not asynchronous.

In order to implement tornado asynchronous execution, we should use @gen.coroutine. Here is an example:

import tornado
from tornado import gen
from tornado.web import RequestHandler

class ChexianVoiceprintHandler_V2(RequestHandler):
    def get(self, *args, **kwargs):
        pass
		
    @gen.coroutine
    def post(self):
        future = tornado.concurrent.Future()
        future.add_done_callback(self.do_simple())
        yield future

    def do_simple(self, *args, **kwargs):
        # do something, for example: create voiceprint
	result = do()
        self.write(result)
        self.finish()
		
def make_app():
    setting = dict(

    )
    return tornado.web.Application(
        [(r'/ChexianVoiceprintRequest_V2', ChexianVoiceprintHandler_V2)]
    )


if __name__ == '__main__':
    app = make_app()
    app.listen(9003)
    tornado.ioloop.IOLoop.current().start()

As to this example, we should return a future object and finish it manually.