On Tue, Dec 13, 2016 at 6:06 PM, <sdic...@watchtower-security.com> wrote: > > So, I was talking with a friend about channels and mutexs, and the use of > mutexs instead of channels. While I'm not sure why you'd want to, I was > given this example and not an answer. > > In the example, he declares two channels. Input and Output. Then, he has a > function running a for select loop in a gorutine. The input object accepts a > text of body(string) and an ID (int), to which it stores it in a map. Then, > the output channel accepts a request id (int), and a chan string. When > something is put on the output channel, he gets the body of text out of the > map and writes it to the output object's return channel and then closes it. > > My question, is this a good practice? Why would one want to use channels in > place of mutexs? Is this efficient?
I expect that a mutex will be more efficient, and I would probably write this code with a Mutex or a RWMutex. But the advantage of this code is that it is clearly free of concurrency problems and race conditions. Especially for more complex cases, that is not a negligible feature. Ian -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
