The Book Functional and Reactive Domain Modeling (Debasish Ghosh) is one of the most interesting books I have read lately, in the realm of Scala. This is not another beginner Scala book, neither a functional programming primer. It is a much needed tour on domain modeling with functional patterns and concepts, reactive systems and modern software development in Scala. If you want to know how in the world one can use Akka streams, free monads and actors interacting with each other, this book is a great starting point :).
This is the second installment in a series of articles about Shapeless. The first post explained heterogeneous lists (HLists) and how to use them to do type-level recursion. Building on this ground, we will now talk about Generic derivation of case classes and sealed trait hierarchies. Along the way we shall cover other related subjects such as singleton-typed literals, products, coproducts (and labelled versions of those) and shapeless syntax goodies.
On April 9, Scala developers from Portugal, Spain and the UK gathered in Lisbon for LX Scala to discuss Scala’s present and future trends. These are my notes from the talks…
When using functional programming languages like Scala, developers spend a lot of their time composing functions and effects. One of the most common ways to express composability is to use monads. However, composing functions that return different monads can become quite messy and, without the right tools, quickly turn into a massive headache. That’s where Monad Transformers, which are the main focus of this post, come in handy!
Shapeless is a well known library for generic programming in scala. It is the basis for many widely used libraries and recent interesting projects in the Scala ecosystem, such as spray, Circe, Scodec and many others.
As part of an ongoing project at e.near, one of our Scala teams was recently tasked with a requirement to build an interpreter for executing workflows which are modelled with a textual DSL. These workflows had to be validated for errors, compiled to a simpler bytecode-like representation and stored, in order to be interpreted with as little overhead as possible.