flink通过keyby()+reduce做行转列数据格式

//测试使用
//    val row=text.flatMap(line=>(line.split("\\n")))
//      .filter(f=>{
//        var rs=true
//        if(f.contains("autoktvroom") && f.contains("wsmessage")){
//          rs=true
//        }else{
//          rs=false
//        }
//         rs
//      }).print()

初始化部分省略了，下面是转换逻辑作为备忘
val ttstream= datastream
  .map(m=>{
    val actype=m._1
    var joinroom_uv=0L
    var waitformic_uv=0L
    var controlmic_uv=0L
    var exitroom_uv=0L
    if(actype.equals("joinroom"))joinroom_uv=m._3
    if(actype.equals("waitformic"))waitformic_uv=m._3
    if(actype.equals("controlmic"))controlmic_uv=m._3
    if(actype.equals("exitroom"))exitroom_uv=m._3

    (m._2,joinroom_uv,waitformic_uv,controlmic_uv,exitroom_uv)

})
  .keyBy(0)
  .window(TumblingEventTimeWindows.of(Time.minutes(2)))
  .reduce(new ReduceFunction[Tuple5[String,Long,Long,Long,Long]]() {
    def reduce(value1: Tuple5[String,Long,Long,Long,Long], value2: Tuple5[String,Long,Long,Long,Long]): Tuple5[String,Long,Long,Long,Long] = {

      return new Tuple5[String,Long,Long,Long,Long](value1._1, value1._2 + value2._2,value1._3+value2._3,value1._4+value2._4,value1._5+value2._5)
    }
  })
  .map(m1=>{
      val time: Date = new Date()
      val dateFormat: SimpleDateFormat = new SimpleDateFormat("yyyyMMdd")
      val date = dateFormat.format(time)

      var jsonStr = "ktvroomid:"+m1._1+"_"+date+"&" // json格式开始
      jsonStr += "{"+
        "\"room_id\":\""+m1._1+
        "\",\"joinroom_uv\":\""+m1._2+
        "\",\"waitformic_uv\":\""+m1._3+
        "\",\"controlmic_uv\":\""+m1._4+
        "\",\"exitroom_uv\":\""+m1._5+
        "\"}"
      jsonStr
})