说明:部分示例在练习中使用了两种实现
1. 基于排序机制的wordcount
/**
* 基于排序机制的 wordcount
* 需求:
* 1. 对文本文件内出现的每个单词统计出其出现的次数
* 2. 按照每个单词出现次数降序排序
* @author Z -Jay
*
*/
public class SortedWordCount {
public static void main(String[] args) {
SparkConf conf = new SparkConf().setAppName("SortedWordCount" ).setMaster("local");
JavaSparkContext sc = new JavaSparkContext(conf );
JavaRDD<String> linesRDD = sc.textFile("C:\\Users\\Z-Jay\\Desktop\\spark.txt" );
JavaRDD<String> words = linesRDD .flatMap(new FlatMapFunction<String, String>() {
private static final long serialVersionUID = 1L;
@Override
public Iterator<String> call(String line ) throws Exception {
String[] split = line .split (" " );
return Arrays.asList( split).iterator();
}
});
JavaPairRDD<String, Integer> pairs = words .mapToPair(new PairFunction<String, String, Integer>() {
private static final long serialVersionUID = 1L;
@Override
public Tuple2<String, Integer> call(String word) throws Exception {
return new Tuple2<>(word,1);
}
});
JavaPairRDD<String, Integer> wordCounts = pairs.reduceByKey(new Function2<Integer, Integer, Integer>() {
private static final long serialVersionUID = 1L;
@Override
public Integer call(Integer v1 , Integer v2) throws Exception {
return v1 +v2 ;
}
});
//至此,已经得到每个单词出现的次数,wordCountsRDD中元素形式("hello",2)
//开始新需求,按照每个单词出现的次数降序排序,则必须重构为元素形式为(2,"hello")的新RDD才能进行排序
//KV反转以排序
JavaPairRDD<Integer, String> countWords = wordCounts.mapToPair(new PairFunction<Tuple2<String,Integer>, Integer, String>() {
private static final long serialVersionUID = 1L;
@Override
public Tuple2<Integer, String> call(Tuple2<String, Integer> tuple) throws Exception {
return new Tuple2<>(tuple._2,tuple._1);
}
});
//按照Key排序
JavaPairRDD<Integer, String> sortedPairs = countWords.sortByKey(false);
//KV再反转以显示
JavaPairRDD<String, Integer> resPairs = sortedPairs.mapToPair(new PairFunction<Tuple2<Integer,String>, String ,Integer>() {
private static final long serialVersionUID = 1L;
@Override
public Tuple2<String, Integer> call(Tuple2<Integer, String> tuple) throws Exception {
return new Tuple2<>(tuple._2,tuple._1);
}
});
resPairs.foreach(new VoidFunction<Tuple2<String,Integer>>() {
private static final long serialVersionUID = 1L;
@Override
public void call(Tuple2<String, Integer> tuple) throws Exception {
System. out.println(tuple ._1 +":" +tuple ._2 );
}
});
sc.close();
}
}
def main(args:Array[String]){
val conf = new SparkConf().setAppName ("SortedWordsCount" ).setMaster ("local" )
val sc = new SparkContext(conf )
val lines = sc.textFile("C:\\ Users\\Z-Jay \\Desktop\\ spark.txt",1 )
val words = lines.flatMap{ line => line.split( " ") }
val pairs = words. map { word => (word, 1) }
val wordCounts = pairs. reduceByKey(pairs)
val countWords = wordCounts.map { wordCount => (wordCount._2,wordCount._1) }
val sortedCountWords = countWords.sortByKey (false)
val sortedWordCounts = sortedCountWords.map { countWord => (countWord._2,countWord._1) }
sortedWordCounts. foreach{ wordCount => println(wordCount._1+ " : "+wordCount._2) }
}
2. 二次排序(ORDER BY col1,col2)
/**
* 二次排序
* 需求:
* 1. 按照文件中的第一列排序
* 2. 如果第一列相同,则按照第二列排序
*
* 规范:
* 1. 必须自定义二次排序Key,实现Ordered <T>接口及序列化
* 2. 在自定义实现中,定义参与排序的列
* 3. 提供排序列的getter、setter、hashCode、equal
* @author Z -Jay
*
*/
public class DoubleSortKey implements Ordered<DoubleSortKey>, Serializable {
private static final long serialVersionUID = 1L;
//定义参与排序的列
private Integer firstCol ;
private Integer secondCol ;
// this > other 的定义
@Override
public boolean $greater(DoubleSortKey other) {
if(this .firstCol > other.firstCol){
return true ;
} else if (this.firstCol == other.firstCol && this. secondCol> other.secondCol ){
return true ;
}
return false ;
}
// this >= other 的定义
@Override
public boolean $greater$eq(DoubleSortKey other) {
if(this .$greater(other)){
return true ;
} else if (this.firstCol == other.firstCol && this. secondCol == other.secondCol ){
return true ;
}
return false ;
}
// this < other 的定义
@Override
public boolean $less(DoubleSortKey other) {
if(this .firstCol < other.firstCol){
return true ;
} else if (this.firstCol == other.firstCol && this. secondCol < other.secondCol ){
return true ;
}
return false ;
}
@Override
public boolean $less$eq(DoubleSortKey other) {
if(this .$less(other)){
return true ;
} else if (this.firstCol == other.firstCol && this. secondCol == other.secondCol ){
return true ;
}
return false ;
}
@Override
public int compare(DoubleSortKey other) {
if(this .firstCol - other.getFirstCol() !=0){
return this.firstCol - other.getFirstCol();
} else{
return this.secondCol - other.getSecondCol();
}
}
//compareTo 与 compare的定义一致
@Override
public int compareTo(DoubleSortKey other) {
if(this .firstCol - other.getFirstCol() !=0){
return this .firstCol - other.getFirstCol();
} else{
return this .secondCol - other.getSecondCol();
}
}
//为参与排序的多个列提供getter,setter,hashCode,equals
public Integer getFirstCol() {
return firstCol ;
}
public void setFirstCol(Integer firstCol) {
this.firstCol = firstCol;
}
public Integer getSecondCol() {
return secondCol ;
}
public void setSecondCol(Integer secondCol) {
this.secondCol = secondCol;
}
@Override
public int hashCode() {
final int prime = 31;
int result = 1;
result = prime * result + (( firstCol == null ) ? 0 : firstCol.hashCode());
result = prime * result + (( secondCol == null ) ? 0 : secondCol.hashCode());
return result ;
}
@Override
public boolean equals(Object obj) {
if (this == obj)
return true ;
if (obj == null)
return false ;
if (getClass() != obj .getClass())
return false ;
DoubleSortKey other = (DoubleSortKey) obj ;
if (firstCol == null) {
if (other .firstCol != null)
return false ;
} else if (!firstCol.equals( other.firstCol ))
return false ;
if (secondCol == null) {
if (other .secondCol != null)
return false ;
} else if (!secondCol.equals( other.secondCol ))
return false ;
return true ;
}
}
public class DoubleSort {
public static void main(String[] args) {
SparkConf conf = new SparkConf().setAppName("DoubleSort" ).setMaster("local");
JavaSparkContext sc = new JavaSparkContext(conf );
JavaRDD<String> lines = sc.textFile("C:\\Users\\Z-Jay\\Desktop\\sort.txt" );
//引入自定义数据结构,中间封装形式 (DoubleSortKey,"{line}")
JavaPairRDD<DoubleSortKey, String> pairs = lines.mapToPair(new PairFunction<String, DoubleSortKey, String>() {
private static final long serialVersionUID = 1L;
@Override
public Tuple2<DoubleSortKey, String> call(String line) throws Exception {
String[] cols = line .split(" ");
DoubleSortKey doubleSortKey = new DoubleSortKey(Integer.parseInt(cols[0]),Integer. parseInt(cols[1]));
return new Tuple2<>(doubleSortKey, line);
}
});
//将按照自定义规则进行排序
JavaPairRDD<DoubleSortKey, String> sortedLines = pairs.sortByKey();
//还原回原始文本,消除自定义数据结构,也就是说自定义Key是中间排序的工具
JavaRDD<String> res = sortedLines .map(new Function<Tuple2<DoubleSortKey,String>, String>() {
private static final long serialVersionUID = 1L;
@Override
public String call(Tuple2<DoubleSortKey, String> tuple) throws Exception {
return tuple ._2 ;
}
});
res.foreach(new VoidFunction<String>() {
private static final long serialVersionUID = 1L;
@Override
public void call(String line) throws Exception {
System. out.println(line );
}
});
sc.close();
}
}
class DoubleSortKey(val first:Int, val second:Int) extends Ordered [DoubleSortKey] with Serializable {
def compare(that: DoubleSortKey) : Int = {
if(this. first == that. first != 0){
this. first-that .first
}else{
this. second-that .second
}
}
}
object DoubleSort {
def main(args : Array[ String]):Unit = {
val conf = new SparkConf().setAppName ("DoubleSort" ).setMaster ("local" )
val sc = new SparkContext(conf )
val lines = sc.textFile("C:\\ Users\\Z-Jay \\Desktop\\ sort.txt",1 )
val pairs = lines.map{ line => (
new DoubleSortKey(line.split(" " ){0 }.toInt, line.split(" "){ 1}.toInt),line
)}
val sortedPairs = pairs. sortByKey()
val sortedLines = sortedPairs.map {sortedPair => sortedPair._2}
sortedLines. foreach{sortedLine => println(sortedLine)}
}
}
3.TOPN
(1) 简单不分组
/**
* 简单Top N
* 对文本文件内的数字,取最大的前3个
*
* @author Z -Jay
*
*/
public class TopN1 {
public static void main(String[] args) {
SparkConf conf = new SparkConf().setAppName("TopN1" ).setMaster("local");
JavaSparkContext sc = new JavaSparkContext(conf );
JavaRDD<String> lines = sc.textFile("C:\\Users\\Z-Jay\\Desktop\\topn1.txt" );
//将RDD中全部元素String转换为Integer
JavaPairRDD<Integer, String> numbers = lines .mapToPair(new PairFunction<String, Integer, String>() {
private static final long serialVersionUID = 1L;
@Override
public Tuple2<Integer, String> call(String numStr) throws Exception {
Tuple2<Integer,String> temp = new Tuple2<>(Integer.parseInt(numStr),"");
return temp ;
}
});
JavaPairRDD<Integer, String> sortedNums = numbers.sortByKey(false);
JavaRDD<Integer> sortedNumsOnly = sortedNums .map(new Function<Tuple2<Integer,String>, Integer>() {
private static final long serialVersionUID = 1L;
@Override
public Integer call(Tuple2<Integer, String> tuple) throws Exception {
return tuple ._1;
}
});
List<Integer> res = sortedNumsOnly .take(3);
System. out.println(res );
sc.close();
}
}
object TopN {
def main(args: Array[ String]): Unit = {
val conf = new SparkConf().setAppName ("TopN" ).setMaster ("local" )
val sc = new SparkContext(conf )
val numStrs = sc.textFile("C:\\ Users\\Z-Jay \\Desktop\\ topn1.txt")
val pairs = numStrs.map{ numStr => (line.toInt, line) }
val sortedPairs = pairs. sortByKey(false )
val sortedNums = sortedPairs.map { sortedPair => sortedPair._1 }
val res = sortedNums.take(3)
for ( num <- res ){
println(num)
}
}
}
object TopN {
def main(args: Array[ String]): Unit = {
val conf = new SparkConf().setAppName ("TopN" ).setMaster ("local" )
val sc = new SparkContext(conf )
val numStrs = sc.textFile("C:\\ Users\\Z-Jay \\Desktop\\ topn1.txt")
val pairs = numStrs.map{ numStr => (line.toInt, line) }
val sortedPairs = pairs. sortByKey(false )
val sortedNums = sortedPairs.map { sortedPair => sortedPair._1 }
val res = sortedNums.take(3)
for ( num <- res ){
println(num)
}
}
}
(2) 分组
/**
* 分组TOPN
* 显示每个班的前三名
* @author Z -Jay
*
*/
public class TopN2 {
public static void main(String[] args) {
SparkConf conf = new SparkConf().setAppName( "TopN2").setMaster("local" );
JavaSparkContext sc = new JavaSparkContext(conf );
JavaRDD<String> lines = sc.textFile("C:\\Users\\Z-Jay\\Desktop\\top2.txt" );
JavaPairRDD<String, Integer> classScores = lines.mapToPair(new PairFunction<String, String, Integer>() {
private static final long serialVersionUID = 1L;
@Override
public Tuple2<String, Integer> call(String line) throws Exception {
String[] split = line .split(" ");
return new Tuple2<>(split[0],Integer.parseInt( split[1]));
}
});
JavaPairRDD<String, Iterable<Integer>> groupClassScores = classScores.groupByKey();
JavaPairRDD<String, Iterable<Integer>> resScores = groupClassScores.mapToPair(new PairFunction<Tuple2<String,Iterable<Integer>>, String, Iterable<Integer>>() {
private static final long serialVersionUID = 1L;
@Override
public Tuple2<String, Iterable<Integer>> call(Tuple2<String, Iterable<Integer>> tuple ) throws Exception {
Integer[] tempVal = new Integer[3];
Iterator<Integer> tempScores = tuple._2 .iterator();
while(tempScores .hasNext()){
Integer score = tempScores .next();
for(Integer i = 0;i < 3; i++){
if(tempVal [i ] == null){
tempVal[i ] = score;
break;
} else if (score >tempVal[i]){
for(Integer j =2; j>i; j--){
tempVal[j ] = tempVal[j -1];
}
tempVal[i ] = score;
break;
}
}
}
return new Tuple2<>(tuple._1 ,Arrays.asList( tempVal));
}
});
resScores.foreach(new VoidFunction<Tuple2<String,Iterable<Integer>>>() {
private static final long serialVersionUID = 1L;
@Override
public void call(Tuple2<String, Iterable<Integer>> t) throws Exception {
System. out.println(t ._1 +":" +t ._2 );
}
});
sc.close();
}
}