Writing Lambdas with Scalambda

Scalambda automatically adds scalambda-core as a dependency to your project when you enable it. It is a library designed to help you get started coding lambda functions as quickly as possible.

The scalambda-core library currently comes with two traits for defining Lambda functions. Both use circe for encoding and decoding of your function’s input and output.

• io.carpe.scalambda.Scalambda gives you the most freedom and control for your functions.
• io.carpe.scalambda.effect.ScalambdaIO allows you to write functions using cats-effect’s powerful IO.

package io.carpe

import com.amazonaws.services.lambda.runtime.Context
import io.carpe.scalambda.Scalambda
 
class HelloWorld extends Scalambda[String, String] {
 
  override def handleRequest(input: String, context: Context): String = {
    "Hello, ${input}!"
  }
}

Inputs and Outputs

You can use any type you want for the input and output of your function, so long as you’ve defined a Encoder and Decoder for that given type.

You have a few options for how you’d like to define these encoders and decoders. At Carpe Data, we like to use Semi-Automatic Derivation.

By placing these encoders and decoders into companion objects within your Input and Output classes, you can guarantee that they will be in scope for your functions.

Example

Here is a custom case class we want to use as the output and/or input to our Lambda Function.

package io.carpe.views

case class Car(make: String, model: String)

object Car {
  import io.circe.generic.extras.semiauto._
  import io.circe.{Decoder, Encoder}

  implicit val decoder: Decoder[Car] = deriveConfiguredDecoder[Car]
  implicit val encoder: Encoder[Car] = deriveConfiguredEncoder[Car]
}

This function accepts a Car as an argument, then automatically turns that Car into a nice sports car. A dumb example, but hopefully it gives you the general idea.

Since it imports Car, the implicit decoder and encoder for a Car will be made visible to Scalambda. This allows Scalambda to use it to decode the input and encode the output for you,

package io.carpe

import com.amazonaws.services.lambda.runtime.Context
import io.carpe.scalambda.Scalambda
import io.carpe.views.Car
 
class UpgradeCar extends Scalambda[Car, Car] {
 
  override def handleRequest(input: Car, context: Context): Car = {
    // turn the input into a 911, then return it
    input.copy(make = "Porsche", model = "911 GT3")
  }
}