esarbe

-----BEGIN PRIVATE KEY-----
MIIEvQIBADANBgkqhkiG9w0BAQEFAASCBKcwggSjAgEAAoIBAQC10dxEGINZbF0nIoMtM8705Nqm6ZWdb72DqTdFJ+UzQIRIUS59lQkYLvdQp71767vz0dVlPTikHmiv
dYHRc7Fo6JsmSUsGR3th+fU6d1Wt6cwpMTUXj/qODmubDK/ioVDW7wz9OFlSsCBvylOYp9v2+u/VXwACnBXNxCDezjx4RKcqMFT31WTxqU9OM9J86ChMOW4bFA41aLAJ
ozB+02xis7OV175XdQ5vkVXM9ys6ZoRF/K6NXeHiwcZFtMKyphXAxqU7uGY2a16bC3TEG5/km6Jru3Wxy4nKlDyUjWISwH4llWjdSi99r2c1fSCXlMCrW0CHoznn+22l
YCKtYe8JAgMBAAECggEAGOPDJvFCHd43PFG9qlTyylR/2CSWzigLRfhGsClfd24oDaxLVHav+YcIZRqpVkr1flGlyEeittjQ1OAdptoTGbzp7EpRQmlLqyRoHRpT+MxO
Hf91+KVFk+fGdEG+3CPgKKQt34Y0uByTPCpy2i10b7F3Xnq0Sicq1vG33DhYT9A/DRIjYr8Y0AVovq0VDjWqA1FW5OO9p7vky6e+PDMjSHucQ+uaLzVZSc7vWOh0tH5M
0GVk17YpBiB/iTpw4zBUIcaneQX3eaIfSCDHK0SCD6IRF7kl+uORzvWqiWlGzpdG2B96uyP4hd3WoPcZntM79PKm4dAotdgmalbueFJfpwKBgQDUy0EyA9Fq0aPF4LID
HqDPduIm4hEAZf6sQLd8Fe6ywM4p9KOEVx7YPaFxQHFSgIiWXswildPJl8Cg5cM2EyMU1tdn5xaR4VIDk8e2JEDfhPtaWskpJp2rU2wHvAXOeAES7UFMrkhKVqqVOdbo
IhlLdcYp5KxiJ3mwINSSO94ShwKBgQDavJvF+c8AINfCaMocUX0knXz+xCwdP430GoPQCHa1rUj5bZ3qn3XMwSWa57J4x

esarbe

Monocle, like many other Scala FP libraries, was inspired by Haskell. In our case, it is the Lens library by Edward Kmett and al.

In Monocle, we experimented with various optics encoding: pair of functions, Van Laarhoven, profunctor (see LensImpl). The JVM and Haskell runtime are hugely different, and an encoding that works well in Haskell can be inefficient in Scala. For example, Haskell relies on zero cost wrapper (newtype) to effectively select typeclass instances, but we don't have an equivalent in Scala/JVM yet (opaque types may help). You can find some of the benchmarks we made for Lenses in 2015 here.

However, something we didn't do very well was to adapt the API to the specificity of Scala. If you look at Monocle 1.x or 2.x, it has the same interface as Haskell's Lens but in much more clunky way. The example

esarbe

Scala indentation syntax

An interesting choice, probably too late. Right now Scala should focus on reducing paper cuts and puzzlers. This said; if indentation syntax is comming, I want it to be an improvement rather than

What is proposed

extend optionality of braces

we already do it! This is already legal for single-line method definitions

esarbe

Generalized Algebraic Data Types in Scala

Basic GADTs

Here's an ADT which is not a GADT, in Haskell:

data Expr = IntExpr Int | BoolExpr Bool

esarbe

object Main {

  trait Language { self: Literal =>
    import self._

    def add(l: F[Int], r: F[Int]): F[Int]
  }

  trait Literal {

esarbe

object Main {

  trait Language {
    type F[X]

    def num(i: Int): F[Int]
    def add(l: F[Int], r: F[Int]): F[Int]
  }

esarbe

/// Original C++ implementation found at http://www.wilmott.com/messageview.cfm?catid=10&threadid=38771
/// C# implementation found at http://weblogs.asp.net/esanchez/archive/2010/07/29/a-quick-and-dirty-implementation-of-excel-norminv-function-in-c.aspx
/*
 *     Compute the quantile function for the normal distribution.
 *
 *     For small to moderate probabilities, algorithm referenced
 *     below is used to obtain an initial approximation which is
 *     polished with a final Newton step.
 *
 *     For very large arguments, an algorithm of Wichura is used.

esarbe

object FunctorComposition extends App {

  trait Functor[F[_]] {
    def map[A, B](fa: F[A])(f: A => B): F[B]
  }

  sealed trait Maybe[+T]
  case class Some[T](value: T) extends Maybe[T]
  case object None extends Maybe[Nothing]
  object Maybe {