Scalaz(32)- Free :lift - Monad生产线

  1. 云栖社区>
  2. 博客>
  3. 正文

Scalaz(32)- Free :lift - Monad生产线

雪川大虫 2016-04-15 17:39:28 浏览1103
展开阅读全文

    在前面的讨论里我们提到自由数据结构就是产生某种类型的最简化结构,比如:free monoid, free monad, free category等等。我们也证明了List[A]是个free monoid。我们再看看free monad结构Free的定义:scalaz/Free.scala


/** A free operational monad for some functor `S`. Binding is done using the heap instead of the stack,
  * allowing tail-call elimination. */
sealed abstract class Free[S[_], A] {
...
  /** Return from the computation with the given value. */
  private[scalaz] case class Return[S[_], A](a: A) extends Free[S, A]

  /** Suspend the computation with the given suspension. */
  private[scalaz] case class Suspend[S[_], A](a: S[Free[S, A]]) extends Free[S, A]
...


我们在上一篇里证明过Free就是free monad,因为Free是个Monad而且它的结构是最简单的了:

1、Free[S[_],A]可以代表一个运算

2、case class Return是一个数据结构。Return(a:A)代表运算结束,运算结果a存放在结构中。另一个意义就是Monad.point(a:A),把一个任意A值a升格成Free

3、case class Suspend也是另一个数据结构。Suspend(a: S[Free[S,A]])代表把下一个运算存放在结构中。如果用Monad.join(a: F[F[A]])表示,那么里面的F[A]应该是个Free[S,A],这样我们才可能把运算结束结构Return[S,A](a:A)存放到Suspend中来代表下一步结束运算。

注意,Suspend(a: S[Free[S,A]])是个递归类型,S必须是个Functor,但不是任何Functor,而是map over Free[S,A]的Functor,也就是运算另一个Free[S,A]值。如果这个Free是Return则返回运算结果,如果是Suspend则继续递归运算。

简单来说Free是一个把Functor S[_]升格成Monad的产生器。我们可以用Free.liftF函数来把任何一个Functor升格成Monad。看看Free.liftF的函数款式就知道了:


  /** Suspends a value within a functor in a single step. Monadic unit for a higher-order monad. */
  def liftF[S[_], A](value: => S[A])(implicit S: Functor[S]): Free[S, A] =
    Suspend(S.map(value)(Return[S, A]))

liftF可以把一个S[A]升格成Free[S,A]。我们用个例子来证明:


1 package Exercises
 2 import scalaz._
 3 import Scalaz._
 4 import scala.language.higherKinds
 5 import scala.language.implicitConversions
 6 object freelift {
 7 trait Config[+A] {
 8   def get: A
 9 }
10 object Config {
11   def apply[A](a: A): Config[A] = new Config[A] { def get = a}
12   implicit val configFunctor = new Functor[Config] {
13      def map[A,B](ca: Config[A])(f: A => B) = Config[B](f(ca.get))
14   }
15 }
16 
17 val freeConfig = Free.liftF(Config("hi config"))  //> freeConfig  : scalaz.Free[Exercises.freelift.Config,String] = Suspend(Exercises.freelift$Config$$anon$2@d70c109)

在上面的例子里Config是个运算A值的Functor。我们可以用Free.liftF把Config(String)升格成Free[Config,String]。实际上我们可以把这个必须是Functor的门槛取消,因为用Coyoneda就可以把任何F[A]拆解成Coyoneda[F,A],而Coyoneda天生是个Functor。我们先看个无法实现map函数的F[A]:


1 trait Interact[+A]  //Console交互
2 //println(prompt: String) then readLine 返回String
3 case class Ask(prompt: String) extends Interact[String]
4 //println(msg: String) 不反回任何值
5 case class Tell(msg: String) extends Interact[Unit]

由于Ask和Tell都不会返回泛类值,所以无需或者无法实现map函数,Interact[A]是个不是Functor的高阶类。我们必须把它转成Coyoneda提供给Free产生Monad:


1 Free.liftFC(Tell("hello"))                        //> res0: scalaz.Free.FreeC[Exercises.freelift.Interact,Unit] = Suspend(scalaz.Coyoneda$$anon$22@71423665)
2 Free.liftFC(Ask("how are you"))                   //> res1: scalaz.Free.FreeC[Exercises.freelift.Interact,String] = Suspend(scalaz.Coyoneda$$anon$22@20398b7c)

我们看看liftFC函数定义:


/** A version of `liftF` that infers the nested type constructor. */
  def liftFU[MA](value: => MA)(implicit MA: Unapply[Functor, MA]): Free[MA.M, MA.A] =
    liftF(MA(value))(MA.TC)

  /** A free monad over a free functor of `S`. */
  def liftFC[S[_], A](s: S[A]): FreeC[S, A] =
    liftFU(Coyoneda lift s)

Coyoneda lift s 返回结果Coyoneda[S,A], liftFU用Unapply可以把Coyoneda[S,A]转化成S[A]并提供给liftF。看看Unapply这段:


 /**Unpack a value of type `M0[A0, B0]` into types `[a]M0[a, B0]` and `A`, given an instance of `TC` */
  implicit def unapplyMAB1[TC[_[_]], M0[_, _], A0, B0](implicit TC0: TC[({type λ[α] = M0[α, B0]})#λ]): Unapply[TC, M0[A0, B0]] {
    type M[X] = M0[X, B0]
    type A = A0
  } = new Unapply[TC, M0[A0, B0]] {
    type M[X] = M0[X, B0]
    type A = A0
    def TC = TC0
    def leibniz = refl
  }

  /**Unpack a value of type `M0[A0, B0]` into types `[b]M0[A0, b]` and `B`, given an instance of `TC` */
  implicit def unapplyMAB2[TC[_[_]], M0[_, _], A0, B0](implicit TC0: TC[({type λ[α] = M0[A0, α]})#λ]): Unapply[TC, M0[A0, B0]] {
    type M[X] = M0[A0, X]
    type A = B0
  } = new Unapply[TC, M0[A0, B0]] {
    type M[X] = M0[A0, X]
    type A = B0
    def TC = TC0
    def leibniz = refl
  }

好了。但是又想了想,一个是Functor的Interact又是怎样的呢?那Ask必须返回一个值,而这个值应该是个Free,实际上是代表下一个运算:


 1 package Exercises
 2 import scalaz._
 3 import Scalaz._
 4 import scala.language.higherKinds
 5 object interact {
 6 trait Interact[+A] 
 7 case class Ask[Next](prompt: String, n: String => Next) extends Interact[Next]
 8 case class Tell[Next](msg: String, n: Next) extends Interact[Next]
 9 
10 object interFunctor extends Functor[Interact] {
11   def map[A,B](ia: Interact[A])(f: A => B): Interact[B] = ia match {
12       case Tell(m,n) => Tell(m, f(n))
13       case g: Ask[A] => Ask[B](g.prompt, g.n andThen f)
14   }
15 }

所以Ask返回Next类型值,应该是个Free,代表下一个运算。

 

网友评论

登录后评论
0/500
评论
雪川大虫
+ 关注