Scala红皮书练习10.9探讨

今天学到第10章Monoid，在做练习10.9时有一些和作者不一样的解法，分享一下。先看题目：

使用foldMap来判定给定的IndexedSeq[Int]是否有序。需要想一个创造性的Monoid。

我们先看一下作者给出的答案吧：

// This implementation detects only ascending order,
	// but you can write a monoid that detects both ascending and descending
	// order if you like.
	def ordered(ints: IndexedSeq[Int]): Boolean = {
	// Our monoid tracks the minimum and maximum element seen so far
	// as well as whether the elements are so far ordered.
	val mon = new Monoid[Option[(Int, Int, Boolean)]] {
	def op(o1: Option[(Int, Int, Boolean)], o2: Option[(Int, Int, Boolean)]) =
	(o1, o2) match {
	// The ranges should not overlap if the sequence is ordered.
	case (Some((x1, y1, p)), Some((x2, y2, q))) =>
	Some((x1 min x2, y1 max y2, p && q && y1 <= x2))
	case (x, None) => x
	case (None, x) => x
	}
	val zero = None
	}
	// The empty sequence is ordered, and each element by itself is ordered.
	foldMapV(ints, mon)(i => Some((i, i, true))).map(_._3).getOrElse(true)
	}

作者的算法只能判断是否是升序，降序还要再写一个函数，而且比较难读懂（估计大牛的脑结构和我们普通程序渣不一样）。重新写了一下，支持同时判断升 降序：

val orderMonoid=new Monoid[(Int,Int,Boolean)] {
  def op(a1:(Int,Int,Boolean),a2:(Int,Int,Boolean))={
    val b1=if(a1._1>=a2._1) true else false
    val b2=if(a1._2<=a2._2) true else false
    (a1._1.min(a2._1),a1._2.max(a2._2),a1._3&&a2._3&&(b1||b2))
  }
  def zero=(Int.MaxValue,Int.MinValue,true)
}
val seq=IndexedSeq[Int](10,8,6,5,4,3)
val isOrdered=foldMap(seq,orderMonoid)(x=>(x,x,true))._3