ryanrain2016

> > > > 我以为我说的挺清楚了，你设置nocwnd=1也可以解决这个问题。说白了就是窗口满了在队列里没有发出去 多谢指点，我不太了解其中的配置细节，我加了下面的代码，demo运行符合期望了 ``` c := conn.(*kcp.UDPSession) c.SetWindowSize(1, 1) ```

> 建议改成： @defnngj > > `elem = Browser.driver.find_elements(By.NAME, value)[self.index] if Browser.driver.find_elements(By.ID, value) else None` 你这是要执行两边find_elements? 那句应该改成elem = Browser.driver.find_elements(By.NAME, value)，然后在最后加一句 elem = elem and elem[0] or None

> @tiangolo when do you plan to add a fully-functional async support to this library? > > We are using the the SQL Alchemy at the moment and recently rewrote...

I met the same problem with redis, even with `rpc=False`. It acts like a sync program and takes no advantages of `async`.

I think there should be a parameters to decide how many task can run parallelly

> > I think there should be a parameters to decide how many task can run parallelly > > You can use other methods to implement RPC calls. I don't...

1

> Thank you for suggesting this feature. If you're interested in contributing, we'd warmly welcome a Pull Request with the implementation 看了下源码，目前代码没有引入任何第三方库，使用的是python自带的http库。没有python自带的http库支持async，加入该特性会引入第三方的库。 另外新的async/await的语法在python 2.7上会报错，如果加入，就不可能支持python 2.7。 感觉原库的设计太多跟线程相关，使用线程池来实现较简单，可以在每个方法上加个转异步的装饰器就可以。但是感觉没有必要，用户自己转就可以，但毕竟不是原生的异步，感觉不太优雅

> Anyway, you can consume messages in batches - it should speed up your services working. Do you mean code like this? ```python @rb.subscriber(list=ListSub("users", batch=True)) async def my_listener(user: List[User]): await...

messages in redis like this