Collectors是stream流处理最后一个环节collect的工具类，主要用于最后对汇集数据的处理，如将流处理结果存储为某种集合类型以及对结果进行分组、筛选、计算等操作.其主要功能如下：

本节内容还是以中奖记录对象来例：

/**
 * 中奖记录
 * @author shixinke
 * @date 2019-1-25
 */
class AwardRecord {
    /**
     * 记录ID
     */
    private Long recordId;
    /**
     * 中奖用户ID
     */
    private Long userId;
    /**
     * 数量
     */
    private Integer num;
    /**
     * 中奖物品ID
     */
    private Long itemId;
    /**
     * 中奖物品的单个价值
     */
    private Double worth;
    public Long getRecordId() {
        return recordId;
    }
    public void setRecordId(Long recordId) {
        this.recordId = recordId;
    }
    public Long getUserId() {
        return userId;
    }
    public void setUserId(Long userId) {
        this.userId = userId;
    }
    public Integer getNum() {
        return num;
    }
    public void setNum(Integer num) {
        this.num = num;
    }
    public Long getItemId() {
        return itemId;
    }
    public void setItemId(Long itemId) {
        this.itemId = itemId;
    }
    public Double getWorth() {
        return worth;
    }
    public void setWorth(Double worth) {
        this.worth = worth;
    }
    public AwardRecord(Long recordId, Long userId, Integer num, Long itemId, Double worth) {
        this.recordId = recordId;
        this.userId = userId;
        this.num = num;
        this.itemId = itemId;
        this.worth = worth;
    }
    @Override
    public String toString() {
        return "AwardRecord{" +
                "recordId=" + recordId +
                ", userId=" + userId +
                ", num=" + num +
                ", itemId=" + itemId +
                ", worth=" + worth +
                '}';
    }
}
public class CollectorDemo {
    public static void main(String[] args) {
        List<AwardRecord> awardRecordList = new ArrayList<AwardRecord>(5);
        awardRecordList.add(new AwardRecord(300001L, 5000001L, 1, 8754578L, 1.35));
        awardRecordList.add(new AwardRecord(300002L, 5200001L, 2, 8754579L, 2.35));
        awardRecordList.add(new AwardRecord(300003L, 5000001L, 1, 8754878L, 0.35));
        awardRecordList.add(new AwardRecord(300004L, 5300001L, 3, 8354578L, 0.21));
        awardRecordList.add(new AwardRecord(300005L, 5200001L, 1, 8154578L, 2.63));
    }
}

一、将流汇集成集合对象

• toList : 输出为List对象
• toMap : 输出为Map对象
• toSet : 输出为Set对象
• toCollection : 输出为集合对象
• toConcurrentMap : 输出为ConcurrentMap对象

List<AwardRecord> awardRecordList = new ArrayList<AwardRecord>(10);
//省略其他流操作
awardRecordList.stream().collect(Collectors.toList());

二、分组

• groupingBy : 按照某个分组函数进行分组
• groupingByConcurrent　：　按照某个分组函数进行分组(并发版，即线程安全版)

/**
* 1.分组(通过userId分组，获取按用户分组的用户的获奖列表)
 */
Map<Long, List<AwardRecord>> awardRecordMap = awardRecordList.stream().collect(Collectors.groupingBy(AwardRecord::getUserId));
System.out.println(awardRecordMap); //{5300001=[AwardRecord{recordId=300004, userId=5300001, num=3, itemId=8354578, worth=0.21}], 5000001=[AwardRecord{recordId=300001, userId=5000001, num=1, itemId=8754578, worth=1.35}, AwardRecord{recordId=300003, userId=5000001, num=1, itemId=8754878, worth=0.35}], 5200001=[AwardRecord{recordId=300002, userId=5200001, num=2, itemId=8754579, worth=2.35}, AwardRecord{recordId=300005, userId=5200001, num=1, itemId=8154578, worth=2.63}]}

三、筛选

• mapping
  • 第一个参数为组合的函数式接口
  • 第二个参数为Collector对象

/**
* 对结果进行组合并形成新的集合(中奖用户列表)
*/
List<Long> awardUserList = awardRecordList.stream().collect(Collectors.mapping(c->c.getUserId(), Collectors.toList()));
System.out.println(awardRecordList);

四、计算

1.最大值 maxBy

• 参数为一个Comparator对象

 /**
* 计算中奖价值最多的记录
*/
Optional<AwardRecord> maxRecord = awardRecordList.stream().collect(
                Collectors.maxBy(
                        (AwardRecord c1, AwardRecord c2)->Double.valueOf(c1.getNum()*c1.getWorth()).compareTo(
                                Double.valueOf(c2.getNum()*c2.getWorth()))));
System.out.println(maxRecord.get());  //AwardRecord{recordId=300002, userId=5200001, num=2, itemId=8754579, worth=2.35}

2.最小值 minBy

• 参数为一个Comparator对象

 /**
* 计算中奖价值最小的记录
*/
Optional<AwardRecord> minRecord = awardRecordList.stream().collect(
                Collectors.minBy(
                        (AwardRecord c1, AwardRecord c2)->Double.valueOf(c1.getNum()*c1.getWorth()).compareTo(
                                Double.valueOf(c2.getNum()*c2.getWorth()))));
System.out.println(minRecord.get()); //AwardRecord{recordId=300003, userId=5000001, num=1, itemId=8754878, worth=0.35}

3.总和

• summingInt : 计算元素为int型的总和
• summingLong　: 计算元素为Long型的总和
• summingDouble : 计算元素为Double型的总和

 /**
* 计算总数量
*/
System.out.println(awardRecordList.stream().collect(Collectors.summingInt(c -> c.getNum())));   //8
System.out.println(awardRecordList.stream().collect(Collectors.summingDouble(c -> c.getNum() * c.getWorth()))); //9.66

４.平均值

• averagingInt
• averagingLong
• averagingDouble

/**
* 平均中奖个数
*/
System.out.println(awardRecordList.stream().collect(Collectors.averagingInt(c -> c.getNum())));  //1.6
/**
* 平均中奖金额
*/
System.out.println(awardRecordList.stream().collect(Collectors.averagingDouble(c -> c.getWorth())));  //1.3780000000000001

5.个数　counting

/**
* 统计中奖人数
*/
System.out.println(awardRecordList.stream().map(AwardRecord::getUserId).distinct().collect(Collectors.counting())); //3

6.计算上面所有的

• summarizingInt
• summarizingLong
• summarizingDouble

 /**
* 所有统计指标汇总
*/
System.out.println(awardRecordList.stream().collect(Collectors.summarizingInt(c -> c.getNum()))); //IntSummaryStatistics{count=5, sum=8, min=1, average=1.600000, max=3}