标签:def OLE 博文 BMI world 解决 gen 方案 阻塞
问题
如何在tornado的coroutine中调用同步阻塞的函数
解决方案
使用python内置标准库的concurrent.futures.ThreadPoolExecutor和tornado.concurrent.run_on_executor
解决示例
a.使用concurrent.futures.ThreadPoolExecutor
#-*-coding:utf-8-*- import time from tornado.gen import coroutine from tornado.ioloop import IOLoop from concurrent.futures import ThreadPoolExecutor def func(): time.sleep(2) print(10) def foo(): time.sleep(1) print(15) @coroutine def main(): pool = ThreadPoolExecutor(2) @coroutine def sync_func1(): yield pool.submit(func,) @coroutine def sync_func2(): yield pool.submit(foo,) t1 = time.time() yield [sync_func1(), sync_func2()] print(time.time() - t1) if __name__ == ‘__main__‘: IOLoop.current().run_sync(main)
b.使用run_on_executor
#-*-coding:utf-8-*- import os.path import time from tornado.gen import coroutine from tornado.ioloop import IOLoop from tornado.concurrent import run_on_executor from concurrent.futures import ThreadPoolExecutor class My(object): def __init__(self): self.executor = ThreadPoolExecutor(3) @run_on_executor def f(self): print(os.path.join(os.path.dirname(__file__), ‘python‘)) time.sleep(2) print(10) @run_on_executor def f1(self): time.sleep(1) print(15) @run_on_executor def f2(self): time.sleep(1.5) print(‘hello, world!‘) @coroutine def main(): m = My() t1 = time.time() yield [m.f1(), m.f2(), m.f()] print(time.time() - t1) if __name__ == ‘__main__‘: IOLoop.current().run_sync(main)
总结
我们直接运行上面的两个同步的函数,耗时需要3秒。但我们利用了ThreadPoolExecutor之后,总耗时只需要2秒左右。下面是运行结果:
---------------------
作者:Easy_to_python
来源:CSDN
原文:https://blog.csdn.net/hjhmpl123/article/details/53673108
版权声明:本文为博主原创文章,转载请附上博文链接!
标签:def OLE 博文 BMI world 解决 gen 方案 阻塞
原文地址:https://www.cnblogs.com/b02330224/p/10213899.html