News for hackers. Stuff that matters.
Re:A Go “clone” of the great and famous Requests library by evmar on Friday June 26, 2015 @10:10AM (#9784449)
I immediately noticed this as well. This API has flags for "asynchronous" requests, but Go style is to provide a blocking API, and let the caller decide to use a goroutine if they need concurrency.
Re:A Go “clone” of the great and famous Requests library by XorNot on Friday June 26, 2015 @10:43AM (#9784697)
Constructing a channel (as it does though) is just pragmatic though and definitely not any less go-like.

You get a channel - you can read from it until it's closed which is always safe. Go-routines are cheap so you don't care about the background details. For simplifying HTTP, this is great.

Re:A Go “clone” of the great and famous Requests library by levigross on Friday June 26, 2015 @10:56AM (#9784843)
Originally the API was completely asynchronous (every "request" function returned a channel).

I got a lot of feedback that I shouldn't do that, so I moved the asynchronous requests to specific functions e.g GetAsync and made the standard API synchronous.

Re:A Go “clone” of the great and famous Requests library by sagichmal on Friday June 26, 2015 @01:35PM (#9786381)
You should go further, removing the async methods altogether. In Go, making sync APIs async is trivial and well-understood.