Spark笔记：复杂RDD的API的理解（下）
本篇接着谈谈那些稍微复杂的API。
1) flatMapValues：针对Pair RDD中的每个值应用一个返回迭代器的函数，然后对返回的每个元素都生成一个对应原键的键值对记录
这个方法我最开始接触时候，总是感觉很诧异，不是太理解，现在回想起来主要原因是我接触的第一个flatMapValues的例子是这样的，代码如下：
val rddPair: RDD[(String, Int)] = sc.parallelize(List(("x01", 2), ("x02", 5), ("x03", 8), ("x04", 3), ("x01", 12), ("x03", 9)), 1)<br/>
val rddFlatMapVals1:RDD[(String,Int)] = rddPair.flatMapValues { x => x to (6) }<br/>
/* 结果：(x01,2),(x01,3),(x01,4),(x01,5),(x01,6),(x02,5),(x02,6),(x04,3),(x04,4),(x04,5),(x04,6) */<br/>
println("====flatMapValues 1====:" + rddFlatMapVals1.collect().mkString(","))
这个实例使用的是scala里Range这种数据类型，Range数据类型是一个数字的范围，细细讲它的理论也没啥意思，我们看下面的例子吧，代码如下：
val list:List[Int] = List(1,2,3,4,5,6)<br/>
val len:Int = list.size - 1<br/>
val r:Range = 0 to len
for (ind <- r){<br/>
print(list(ind) + ";")// 1;2;3;4;5;6;<br/>
}<br/>
println("")<br/>
for (ind <- 0 to len){<br/>
print(list(ind) + ";")// 1;2;3;4;5;6;<br/>
}<br/>
println("")
由以上代码我们可以看到0 to 3就是指代0，1，2，3这三个数字，所以我们可以在for循环里指代这个范围。
其实flatMapValues接受的外部方法的返回类型是一个Seq类型，Seq类型在scala里就是一个序列，一个有序的序列，我们可以把他理解成数组，我们来看看下面的例子：
def flatMapRange(par:Int):Range = {<br/>
par to 6<br/>
}
def flatMapList(par:Int):List[Int] = {<br/>
List(par + 1000)<br/>
}
def flatMapSeq(par:Int):Seq[Int] = {<br/>
Seq(par + 6000)<br/>
}<br/>
val rddFlatMapVals2:RDD[(String,Int)] = rddPair.flatMapValues { x => flatMapRange(x) }<br/>
/* 结果：(x01,2),(x01,3),(x01,4),(x01,5),(x01,6),(x02,5),(x02,6),(x04,3),(x04,4),(x04,5),(x04,6) */<br/>
println("====flatMapValues 2====:" + rddFlatMapVals2.collect().mkString(","))<br/>
val rddFlatMapVals3:RDD[(String,Int)] = rddPair.flatMapValues { x => flatMapList(x) }<br/>
/* 结果：(x01,1002),(x02,1005),(x03,1008),(x04,1003),(x01,1012),(x03,1009) */<br/>
println("====flatMapValues 3====:" + rddFlatMapVals3.collect().mkString(","))<br/>
val rddFlatMapVals4:RDD[(String,Int)] = rddPair.flatMapValues { x => flatMapSeq(x) }<br/>
/* 结果：(x01,6002),(x02,6005),(x03,6008),(x04,6003),(x01,6012),(x03,6009) */<br/>
println("====flatMapValues 4====:" + rddFlatMapVals4.collect().mkString(","))
谈到flatMapValues这个方法，让我不得不回忆起另外一个与之类似的方法flatMap，我们来看看这个方法的实例吧，代码如下：
val rddFlatMap1:RDD[(String,Int)] = rddPair.flatMap(x => List((x._1,x._2 + 3000)))<br/>
// 结果：(x01,3002),(x02,3005),(x03,3008),(x04,3003),(x01,3012),(x03,3009)<br/>
println("=====flatMap 1======:" + rddFlatMap1.collect().mkString(","))<br/>
val rddFlatMap2:RDD[Int] = rddPair.flatMap(x => List(x._2 + 8000))<br/>
// 结果:8002,8005,8008,8003,8012,8009<br/>
println("=====flatMap 2======:" + rddFlatMap2.collect().mkString(","))<br/>
val rddFlatMap3:RDD[String] = rddPair.flatMap(x => List(x._1 + "@!@" + x._2))<br/>
// 结果：x01@!@2,x02@!@5,x03@!@8,x04@!@3,x01@!@12,x03@!@9<br/>
println("=====flatMap 3======:" + rddFlatMap3.collect().mkString(","))
由此可见flatMap方法里的参数也是一个Seq，而且他们之间可以相互替代使用，只不过flatMapValues是让二元组里的第一个元素保持不变的情况下进行计算的（及key值不发生变化）。不过spark不会无缘无故的定义一个flatMapValues,它其实和spark里的分区紧密相关，关于spark的分区知识我会在后面文章里谈谈的。
2) rightOuterJoin，leftOuterJoin，rddCogroup及右连接，左连接和分组函数
我们首先看看他们的使用吧，代码如下：
val rdd:RDD[(String,Int)] = sc.makeRDD(List(("x01",2),("x02",5),("x03",9),("x03",21),("x04",76)))<br/>
val other:RDD[(String,Int)] = sc.makeRDD(List(("x01",4),("x02",6),("x03",11)))
val rddRight:RDD[(String,(Option[Int],Int))] = rdd.rightOuterJoin(other)<br/>
/* 结果：(x02,(Some(5),6)),(x03,(Some(9),11)),(x03,(Some(21),11)),(x01,(Some(2),4)) */<br/>
println("====rightOuterJoin====:" + rddRight.collect().mkString(","))
val rddLeft:RDD[(String,(Int,Option[Int]))] = rdd.leftOuterJoin(other)<br/>
/* 结果： (x02,(5,Some(6))),(x04,(76,None)),(x03,(9,Some(11))),(x03,(21,Some(11))),(x01,(2,Some(4))) */<br/>
println("====leftOuterJoin====:" + rddLeft.collect().mkString(","))<br/>
val rddSome = rddLeft.filter(x => x._2._2.isEmpty == false)// 过滤掉None的记录<br/>
/* 结果: (x02,(5,Some(6))),(x03,(9,Some(11))),(x03,(21,Some(11))),(x01,(2,Some(4)))*/<br/>
println("====rddSome===:" + rddSome.collect().mkString(","))
val rddCogroup: RDD[(String, (Iterable[Int], Iterable[Int]))] = rdd.cogroup(other)<br/>
/* 结果: (x02,(CompactBuffer(5),CompactBuffer(6))),(x04,(CompactBuffer(76),CompactBuffer())),(x03,(CompactBuffer(9, 21),CompactBuffer(11))),(x01,(CompactBuffer(2),CompactBuffer(4)))*/<br/>
println("===cogroup====:" + rddCogroup.collect().mkString(","))
这三个方法很好理解，就和关系数据库里的左右连接，分组一样，不过它们的返回值在我刚学习spark时候很是疑惑了半天，这里就好好说下它们的返回值，这其实就是学习下scala的数据类型了。
首先是Some数据类型，Some并不是一个直接操作的数据类型，它属于Option这个数据结构的，其实None也是Option里的数据结构，Some里面只能放一个元素，例如Some(1),Some((1,2)),为什么scala里还要这么繁琐的定义一个Option，并在其中还定义一个Some和一个None的结构呢？我们首先看看下面代码：
def optionSome():Unit = {<br/>
/*<br/>
* =======option for 1=========<br/>
0:3<br/>
2:8<br/>
3:11<br/>
=======option for 1=========<br/>
=======option for 2=========<br/>
0:3<br/>
1:None<br/>
2:8<br/>
3:11<br/>
=======option for 2=========<br/>
*/<br/>
val list:List[Option[Int]] = List(Some(3),None,Some(8),Some(11))<br/>
println("=======option for 1=========")<br/>
for (i <- 0 until list.size){<br/>
if (!list(i).isEmpty){<br/>
println(i + ":" + list(i).get)<br/>
}<br/>
}<br/>
println("=======option for 1=========")<br/>
println("=======option for 2=========")<br/>
for (j <- 0 until list.size){<br/>
val res = list(j) match {<br/>
case None => println(j + ":None")<br/>
case _ => println(j + ":" + list(j).get)<br/>
}<br/>
}<br/>
println("=======option for 2=========")<br/>
}
Option数据结构其实想要表达的是一个数据集合，这个数据集合里要么有值，要么没值，这点在左右连接查询里就非常有用，其实左右连接最后的结果就是要么关联上了要么没有关联上。
分组cogroup返回的结构是CompactBuffer，CompactBuffer并不是scala里定义的数据结构，而是spark里的数据结构，它继承自一个迭代器和序列，所以它的返回值是一个很容易进行循环遍历的集合，这点很符合cogroup的返回值类型。
好了，这篇内容就写完了，下一篇文章我将要简单聊聊spark分区，后面应该暂时会停停spark的学习，要搞搞前端的一些技术，这都是因为工作需要了。
最后我将完整示例代码给大家分享下，代码如下：
package cn.com.sparktest
import org.apache.spark.SparkConf<br/>
import org.apache.spark.SparkContext<br/>
import org.apache.spark.SparkContext._<br/>
import org.apache.spark.rdd.RDD<br/>
import scala.collection.immutable.List<br/>
import org.apache.spark.util.collection.CompactBuffer
object ScalaTest {<br/>
val conf: SparkConf = new SparkConf().setAppName("spark scala").setMaster("local[2]")<br/>
val sc: SparkContext = new SparkContext(conf)
def aggrFtnOne(par: ((Int, Int), Int)): (Int, Int) = {<br/>
/*<br/>
*aggregate的初始值为(0,0):<br/>
====aggrFtnOne Param===:((0,0),1)<br/>
====aggrFtnOne Param===:((1,1),2)<br/>
====aggrFtnOne Param===:((3,2),3)<br/>
====aggrFtnOne Param===:((6,3),4)<br/>
====aggrFtnOne Param===:((10,4),5)*/<br/>
/*<br/>
*aggregate的初始值为(1,1):<br/>
====aggrFtnOne Param===:((1,1),1)<br/>
====aggrFtnOne Param===:((2,2),2)<br/>
====aggrFtnOne Param===:((4,3),3)<br/>
====aggrFtnOne Param===:((7,4),4)<br/>
====aggrFtnOne Param===:((11,5),5)<br/>
* */<br/>
println("====aggrFtnOne Param===:" + par.toString())<br/>
val ret: (Int, Int) = (par._1._1 + par._2, par._1._2 + 1)<br/>
ret<br/>
}
def aggrFtnTwo(par: ((Int, Int), (Int, Int))): (Int, Int) = {<br/>
/*aggregate的初始值为(0,0):::::((0,0),(15,5))*/<br/>
/*aggregate的初始值为(1,1):::::((1,1),(16,6))*/<br/>
println("====aggrFtnTwo Param===:" + par.toString())<br/>
val ret: (Int, Int) = (par._1._1 + par._2._1, par._1._2 + par._2._2)<br/>
ret<br/>
}
def foldFtn(par: (Int, Int)): Int = {<br/>
/*fold初始值为0：<br/>
=====foldFtn Param====:(0,1)<br/>
=====foldFtn Param====:(1,2)<br/>
=====foldFtn Param====:(3,3)<br/>
=====foldFtn Param====:(6,4)<br/>
=====foldFtn Param====:(10,5)<br/>
=====foldFtn Param====:(0,15)<br/>
* */<br/>
/*<br/>
* fold初始值为1:<br/>
=====foldFtn Param====:(1,1)<br/>
=====foldFtn Param====:(2,2)<br/>
=====foldFtn Param====:(4,3)<br/>
=====foldFtn Param====:(7,4)<br/>
=====foldFtn Param====:(11,5)<br/>
=====foldFtn Param====:(1,16)<br/>
* */<br/>
println("=====foldFtn Param====:" + par.toString())<br/>
val ret: Int = par._1 + par._2<br/>
ret<br/>
}
def reduceFtn(par:(Int,Int)):Int = {<br/>
/*<br/>
* ======reduceFtn Param=====:1:2<br/>
======reduceFtn Param=====:3:3<br/>
======reduceFtn Param=====:6:4<br/>
======reduceFtn Param=====:10:5<br/>
*/<br/>
println("======reduceFtn Param=====:" + par._1 + ":" + par._2)<br/>
par._1 + par._2<br/>
}
def sparkRDDHandle(): Unit = {<br/>
val rddInt: RDD[Int] = sc.parallelize(List(1, 2, 3, 4, 5), 1)
val rddAggr1: (Int, Int) = rddInt.aggregate((0, 0))((x, y) => (x._1 + y, x._2 + 1), (x, y) => (x._1 + y._1, x._2 + y._2))<br/>
println("====aggregate 1====:" + rddAggr1.toString()) // (15,5)
val rddAggr2: (Int, Int) = rddInt.aggregate((0, 0))((x, y) => aggrFtnOne(x, y), (x, y) => aggrFtnTwo(x, y)) // 参数可以省略元组的括号<br/>
println("====aggregate 2====:" + rddAggr2.toString()) // (15,5)
val rddAggr3: (Int, Int) = rddInt.aggregate((1, 1))((x, y) => aggrFtnOne((x, y)), (x, y) => aggrFtnTwo((x, y))) // 参数使用元组的括号<br/>
println("====aggregate 3====:" + rddAggr3.toString()) // (17,7)
val rddAggr4: (Int, Int) = rddInt.aggregate((1, 0))((x, y) => (x._1 * y, x._2 + 1), (x, y) => (x._1 * y._1, x._2 + y._2))<br/>
println("====aggregate 4====:" + rddAggr4.toString()) // (120,5)
val rddFold1: Int = rddInt.fold(0)((x, y) => x + y)<br/>
println("====fold 1====:" + rddFold1) // 15
val rddFold2: Int = rddInt.fold(0)((x, y) => foldFtn(x, y)) // 参数可以省略元组的括号<br/>
println("====fold 2=====:" + rddFold2) // 15
val rddFold3: Int = rddInt.fold(1)((x, y) => foldFtn((x, y))) // 参数使用元组的括号<br/>
println("====fold 3====:" + rddFold3) // 17
val rddReduce1:Int = rddInt.reduce((x,y) => x + y)<br/>
println("====rddReduce 1====:" + rddReduce1)// 15
val rddReduce2:Int = rddInt.reduce((x,y) => reduceFtn(x,y))<br/>
println("====rddReduce 2====:" + rddReduce2)// 15
}
def combineFtnOne(par:Int):(Int,Int) = {<br/>
/*<br/>
* ====combineFtnOne Param====:2<br/>
====combineFtnOne Param====:5<br/>
====combineFtnOne Param====:8<br/>
====combineFtnOne Param====:3<br/>
*/<br/>
println("====combineFtnOne Param====:" + par)<br/>
val ret:(Int,Int) = (par,1)<br/>
ret<br/>
}
def combineFtnTwo(par:((Int,Int),Int)):(Int,Int) = {<br/>
/*<br/>
====combineFtnTwo Param====:((2,1),12)<br/>
====combineFtnTwo Param====:((8,1),9)<br/>
* */<br/>
println("====combineFtnTwo Param====:" + par.toString())<br/>
val ret:(Int,Int) = (par._1._1 + par._2,par._1._2 + 1)<br/>
ret<br/>
}
def combineFtnThree(par:((Int,Int),(Int,Int))):(Int,Int) = {<br/>
/*<br/>
* 无结果打印<br/>
*/<br/>
println("@@@@@@@@@@@@@@@@@@")<br/>
println("====combineFtnThree Param===:" + par.toString())<br/>
val ret:(Int,Int) = (par._1._1 + par._2._1,par._1._2 + par._2._2)<br/>
ret<br/>
}
def flatMapRange(par:Int):Range = {<br/>
par to 6<br/>
}
def flatMapList(par:Int):List[Int] = {<br/>
List(par + 1000)<br/>
}
def flatMapSeq(par:Int):Seq[Int] = {<br/>
Seq(par + 6000)<br/>
}
def sparkPairRDD(): Unit = {<br/>
val rddPair: RDD[(String, Int)] = sc.parallelize(List(("x01", 2), ("x02", 5), ("x03", 8), ("x04", 3), ("x01", 12), ("x03", 9)), 1)
/* def combineByKey[C](createCombiner: Int => C, mergeValue: (C, Int) => C, mergeCombiners: (C, C) => C): RDD[(String, C)] */<br/>
val rddCombine1:RDD[(String,(Int,Int))] = rddPair.combineByKey(x => (x, 1), (com: (Int, Int), x) => (com._1 + x, com._2 + 1), (com1: (Int, Int), com2: (Int, Int)) => (com1._1 + com2._1, com1._2 + com2._2))<br/>
println("====combineByKey 1====:" + rddCombine1.collect().mkString(",")) // (x02,(5,1)),(x03,(17,2)),(x01,(14,2)),(x04,(3,1))
val rddCombine2:RDD[(String,(Int,Int))] = rddPair.combineByKey(x => combineFtnOne(x), (com: (Int, Int), x) => combineFtnTwo(com,x), (com1: (Int, Int), com2: (Int, Int)) => combineFtnThree(com1,com2))<br/>
println("=====combineByKey 2====:" + rddCombine2.collect().mkString(",")) // (x02,(5,1)),(x03,(17,2)),(x01,(14,2)),(x04,(3,1))
val rddKeys:RDD[String] = rddPair.keys<br/>
/*结果:x01,x02,x03,x04,x01,x03 注意调用keys方法时候不能加上括号，否则会报错*/<br/>
println("====keys====:" + rddKeys.collect().mkString(","))
val rddVals:RDD[Int] = rddPair.values<br/>
/*结果：2,5,8,3,12,9 注意调用values方法时候不能加上括号，否则会报错*/<br/>
println("=====values=====:" + rddVals.collect().mkString(","))
val rddFlatMapVals1:RDD[(String,Int)] = rddPair.flatMapValues { x => x to (6) }<br/>
/* 结果：(x01,2),(x01,3),(x01,4),(x01,5),(x01,6),(x02,5),(x02,6),(x04,3),(x04,4),(x04,5),(x04,6) */<br/>
println("====flatMapValues 1====:" + rddFlatMapVals1.collect().mkString(","))<br/>
val rddFlatMapVals2:RDD[(String,Int)] = rddPair.flatMapValues { x => flatMapRange(x) }<br/>
/* 结果：(x01,2),(x01,3),(x01,4),(x01,5),(x01,6),(x02,5),(x02,6),(x04,3),(x04,4),(x04,5),(x04,6) */<br/>
println("====flatMapValues 2====:" + rddFlatMapVals2.collect().mkString(","))<br/>
val rddFlatMapVals3:RDD[(String,Int)] = rddPair.flatMapValues { x => flatMapList(x) }<br/>
/* 结果：(x01,1002),(x02,1005),(x03,1008),(x04,1003),(x01,1012),(x03,1009) */<br/>
println("====flatMapValues 3====:" + rddFlatMapVals3.collect().mkString(","))<br/>
val rddFlatMapVals4:RDD[(String,Int)] = rddPair.flatMapValues { x => flatMapSeq(x) }<br/>
/* 结果：(x01,6002),(x02,6005),(x03,6008),(x04,6003),(x01,6012),(x03,6009) */<br/>
println("====flatMapValues 4====:" + rddFlatMapVals4.collect().mkString(","))
val rddFlatMap1:RDD[(String,Int)] = rddPair.flatMap(x => List((x._1,x._2 + 3000)))<br/>
// 结果：(x01,3002),(x02,3005),(x03,3008),(x04,3003),(x01,3012),(x03,3009)<br/>
println("=====flatMap 1======:" + rddFlatMap1.collect().mkString(","))<br/>
val rddFlatMap2:RDD[Int] = rddPair.flatMap(x => List(x._2 + 8000))<br/>
// 结果:8002,8005,8008,8003,8012,8009<br/>
println("=====flatMap 2======:" + rddFlatMap2.collect().mkString(","))<br/>
val rddFlatMap3:RDD[String] = rddPair.flatMap(x => List(x._1 + "@!@" + x._2))<br/>
// 结果：x01@!@2,x02@!@5,x03@!@8,x04@!@3,x01@!@12,x03@!@9<br/>
println("=====flatMap 3======:" + rddFlatMap3.collect().mkString(","))<br/>
}
def optionSome():Unit = {<br/>
/*<br/>
* =======option for 1=========<br/>
0:3<br/>
2:8<br/>
3:11<br/>
=======option for 1=========<br/>
=======option for 2=========<br/>
0:3<br/>
1:None<br/>
2:8<br/>
3:11<br/>
=======option for 2=========<br/>
*/<br/>
val list:List[Option[Int]] = List(Some(3),None,Some(8),Some(11))<br/>
println("=======option for 1=========")<br/>
for (i <- 0 until list.size){<br/>
if (!list(i).isEmpty){<br/>
println(i + ":" + list(i).get)<br/>
}<br/>
}<br/>
println("=======option for 1=========")<br/>
println("=======option for 2=========")<br/>
for (j <- 0 until list.size){<br/>
val res = list(j) match {<br/>
case None => println(j + ":None")<br/>
case _ => println(j + ":" + list(j).get)<br/>
}<br/>
}<br/>
println("=======option for 2=========")<br/>
}
def pairRDDJoinGroup():Unit = {<br/>
val rdd:RDD[(String,Int)] = sc.makeRDD(List(("x01",2),("x02",5),("x03",9),("x03",21),("x04",76)))<br/>
val other:RDD[(String,Int)] = sc.makeRDD(List(("x01",4),("x02",6),("x03",11)))
val rddRight:RDD[(String,(Option[Int],Int))] = rdd.rightOuterJoin(other)<br/>
/* 结果：(x02,(Some(5),6)),(x03,(Some(9),11)),(x03,(Some(21),11)),(x01,(Some(2),4)) */<br/>
println("====rightOuterJoin====:" + rddRight.collect().mkString(","))
val rddLeft:RDD[(String,(Int,Option[Int]))] = rdd.leftOuterJoin(other)<br/>
/* 结果： (x02,(5,Some(6))),(x04,(76,None)),(x03,(9,Some(11))),(x03,(21,Some(11))),(x01,(2,Some(4))) */<br/>
println("====leftOuterJoin====:" + rddLeft.collect().mkString(","))<br/>
val rddSome = rddLeft.filter(x => x._2._2.isEmpty == false)// 过滤掉None的记录<br/>
/* 结果: (x02,(5,Some(6))),(x03,(9,Some(11))),(x03,(21,Some(11))),(x01,(2,Some(4)))*/<br/>
println("====rddSome===:" + rddSome.collect().mkString(","))
val rddCogroup: RDD[(String, (Iterable[Int], Iterable[Int]))] = rdd.cogroup(other)<br/>
/* 结果: (x02,(CompactBuffer(5),CompactBuffer(6))),(x04,(CompactBuffer(76),CompactBuffer())),(x03,(CompactBuffer(9, 21),CompactBuffer(11))),(x01,(CompactBuffer(2),CompactBuffer(4)))*/<br/>
println("===cogroup====:" + rddCogroup.collect().mkString(","))<br/>
}
def scalaBasic(){<br/>
val its:Iterable[Int] = Iterable(1,2,3,4,5)<br/>
its.foreach { x => print(x + ",") }// 1,2,3,4,5,
val tuple2Param1:Tuple2[String,Int] = Tuple2("x01",12)// 标准定义二元组<br/>
val tuple2Param2:(String,Int) = ("x02",29)// 字面量定义二元组
/* 结果: x01:12*/<br/>
println("====tuple2Param1====:" + tuple2Param1._1 + ":" + tuple2Param1._2)<br/>
/* 结果: x02:29 */<br/>
println("====tuple2Param2====:" + tuple2Param2._1 + ":" + tuple2Param2._2)
val tuple6Param1:Tuple6[String,Int,Int,Int,Int,String] = Tuple6("xx01",1,2,3,4,"x1x")// 标准定义6元组<br/>
val tuple6Param2:(String,Int,Int,Int,Int,String) = ("xx02",1,2,3,4,"x2x")// 字面量定义6元组
/* 结果: xx01:1:2:3:4:x1x */<br/>
println("====tuple6Param1====:" + tuple6Param1._1 + ":" + tuple6Param1._2 + ":" + tuple6Param1._3 + ":" + tuple6Param1._4 + ":" + tuple6Param1._5 + ":" + tuple6Param1._6)<br/>
/* 结果: xx02:1:2:3:4:x2x */<br/>
println("====tuple6Param2====:" + tuple6Param2._1 + ":" + tuple6Param2._2 + ":" + tuple6Param2._3 + ":" + tuple6Param2._4 + ":" + tuple6Param2._5 + ":" + tuple6Param2._6)
val list:List[Int] = List(1,2,3,4,5,6)<br/>
val len:Int = list.size - 1<br/>
val r:Range = 0 to len
for (ind <- r){<br/>
print(list(ind) + ";")// 1;2;3;4;5;6;<br/>
}<br/>
println("")<br/>
for (ind <- 0 to len){<br/>
print(list(ind) + ";")// 1;2;3;4;5;6;<br/>
}<br/>
println("")<br/>
}
def main(args: Array[String]): Unit = {<br/>
scalaBasic()<br/>
optionSome()
sparkRDDHandle()<br/>
sparkPairRDD()<br/>
pairRDDJoinGroup()
}<br/>
}
转发申明：
本文转自互联网，由小站整理并发布，在于分享相关技术和知识。版权归原作者所有，如有侵权，请联系本站 top8488@163.com，将在24小时内删除。谢谢
