Service Adapters

Note

Be sure you have followed the instructions in Setup for importing dependencies before running examples.

A Stitch Service Adapter provides the detailed querying implementation that represents a service in a Stitch query. Each service has an associated adaptor which maps Stitch atomic queries onto the underlying service, and provides convenience methods and batching.

Writing a Service Adapter

Before writing a Service Adapter, you should be familiar with Calls, Groups, and Runners.

For most services that use Stitch, the call to a backend is hidden inside of their adapter’s code, so you never actually see it. Instead, adapters often have a method that, when given arguments returns a Stitch of the results. In this example, adapter takes in an argument and returns a Stitch of the results. For simplicity, let’s say that the service just adds 1 to the arg.

// in the adapter
private val g = new SeqGroup[Int, Int]{
def run(keys: Seq[Int]): Future[Seq[Try[Int]]] =
  Future.value(keys.map(i => Return(i + 1)))
}
def adapter(arg: Int): Stitch[Int] = Stitch.call(arg, g)

// adapter's user’s code
val s = Stitch.value(0).flatMap(arg => adapter(arg))
Await.result(Stitch.run(s)) // result: 1

For some APIs, all Stitch calls can go into the same batch. For others, the service-level call includes additional contextual information along with the key. That context is included in the Group, so calls with the same contexts may be batched together.

When writing a Service Adapter its important to encapsulate the Group and .call, see Groups.

Next Stitch Concepts