In a previous post I was exploring the use of the EitherT to make it easier to work Either nested in a Future. I’m currently reading the book Advanced Scala with Cats and decided to rewrite some of my code using the Cats library instead.
There’s also a page on Herding Cats where Eugene Yokota covers the same ground. I wanted to expand my examples from last post so that they actually execute in a Future so I can map that to my own error handling code in real programs. For example in the Herding Cats blog the demonstration code returns values like this:
What I wanted to figure out was how this looks in real code where you may have a function that works with a Future[Either]. I went back to my code from last post and modified the dummy functions so that:
- The code executes in a Future
- The function returns Cats EitherT type response
This makes things easier at the call site because instead of converting the response from Future[Either[String, A]] as I did then, you can simply use the EitherT directly. So instead of:
you can use the results directly
If you check the example below the only thing needed to make your function return an EitherT[Future] is to use the EitherT constructor on the final value
The other thing you need to know about EitherT in Cats is that you need to use ‘value’ instead of ‘run’ to get into the results at the end.
I found this post useful for more ways to create a Future[Either] stack.
Final thoughts; whilst the the syntax is slightly different when working with EitherT and Cats, Scalaz and the Hamsters library, the concept is the same and it comes down to finding a way to use them that makes them easier to work with at the calling site. I think I can make things even cleaner with an implicit conversion from Future[Either] to EitherT[Future, String, A] but that will be possibly a later post.
Again for reference the libraries used when writing this post are as follow: