# drift_actor

[![Package Version](https://img.shields.io/hexpm/v/drift_actor)](https://hex.pm/packages/drift_actor)
[![Hex Docs](https://img.shields.io/badge/hex-docs-ffaff3)](https://hexdocs.pm/drift_actor/)

`drift_actor` provides an actor wrapper for running `drift` steppers on Erlang.

```sh
gleam add gleam_erlang@1
gleam add drift@1
gleam add drift_actor@1
```
```gleam
import drift
import drift/actor
import gleam/erlang/process
import gleam/function
import gleam/io
import gleam/option.{None, Some}
import gleam/string

pub fn main() {
  // Start a stepper that adds all the numbers sent to it,
  // until None is encountered
  let assert Ok(subject) =
    actor.using_io(
      // No inputs in this example
      fn() { process.new_selector() },
      function.identity,
      fn(ctx, output) {
        io.println(string.inspect(output))
        Ok(ctx)
      },
    )
    |> actor.start(100, 0, fn(ctx, state, input) {
      case input {
        Some(input) -> drift.continue(ctx, state + input)
        None ->
          ctx
          |> drift.output(state)
          |> drift.stop(state)
      }
    })

  process.send(subject, Some(40))
  process.send(subject, Some(2))

  // This will terminate the actor, and print 42
  process.send(subject, None)

  // But we need to wait for the message to be handled...
  process.sleep(100)
}
```