HLL.scala

import com.twitter.algebird.{
  HLL,
  HyperLogLog,
  HyperLogLogAggregator,
  HyperLogLogMonoid
}

object HLL extends App {
  val hllMonoid = new HyperLogLogMonoid(bits = 4)

  val agg = HyperLogLogAggregator.withErrorGeneric[String](0.01)

  val data = List("hello", "world", "wowww")

  val combinedHll: HLL = agg(data)

  println(s"old size: ${combinedHll.estimatedSize}")

  // HLL to bytes
  val hbytes = HyperLogLog.toBytes(combinedHll)

  // Bytes to HLL
  val newHll = HyperLogLog.fromBytes(hbytes)

  // combine HLLS
  val sum = hllMonoid.sum(Seq(combinedHll, newHll))

  println(s"new size: ${newHll.estimatedSize}")
  println(s"combined size: ${sum.estimatedSize}")
}