主题:收集器
一、需求
- 需要更加强大的reduce操作(收集器的实质是redcue操作的包装体)
- Stream的reduce方法,也可以将数据reduce到加复杂的结果上Lamda的深入认知(七),但我们更加需要库一级的reduce操作!!!
二、API布局
- 由Collector接口的实现类,完成reduce的算法行为。
- 由Collectors扮演Collector的工厂,其工厂方法中会传入lamda(通常会这样),从而产生由用户定制的Collector。
- 将Collector传递给Stream的collect方法
- collect方法为终端操作,产生结果。
- Demo:
//将数据收集到Set中
Arrays.asList(2,-3,1).stream()
.collect(Collectors.toSet());
// 将数据收集到定制的set
Arrays.asList(2,-3,1).stream()
.collect(Collectors.toCollection(LinkedHashSet::new));
三、常见的收集器
1. 计数统计:
Collectors.counting()
2. 简单算数统计:
- 原型:
Collectors.summarizingInt(ToIntFunction);
Collectors.summarizingLong(ToLongFunction);
Collectors.summarizingDouble(ToDoubleFunction);
Collectors.averagingInt(ToIntFunction);
Collectors.averagingLong(ToLongFunction);
Collectors.averagingDouble(ToDoubleFunction);
- demo: 统计数组中元素的个数
int len=Arrays.asList(new int[]{1,2,8,1},new int[]{4,8,1})
.stream()
.collect(Collectors.summingInt(s->s.length));
3. 综合统计:
- 原型:
Collectors.summarizingInt(ToIntFunction)
Collectors.summarizingLong(ToLongFunction)
Collectors.summarizingDouble(ToDoubleFunction)
- demo:
IntSummaryStatistics mm = Arrays.asList(new int[]{1, 2, 8, 1}, new int[]{4, 8, 1}).stream()
.collect(Collectors.summarizingInt(ary->ary.length));
//结果: IntSummaryStatistics{count=2, sum=7, min=3, average=3.500000, max=4}
4. 收集到集合对象中
- 原型:
public static Collector> toList()
public static Collector> toSet()
public static > Collector toCollection(Supplier collectionFactory)
- demo:
5. 找出最大及最小值
- 原型:
public static Collector> maxBy(Comparator super T> comparator)
- demo:
//两种不同的方式
String r1=Arrays.asList("abc","x","cdef","dd").stream()
.collect(Collectors.maxBy((s1,s2)->s1.length()-s2.length()))
.get();
String r2=Arrays.asList("abc","x","cdef","dd").stream()
.collect(Collectors.maxBy(Comparator.comparing(String::length)))
.get();
//结果:cdef
6. 连接操作
- 原型:
public static Collector joining(CharSequence delimiter)
- demo:
String r3=Arrays.asList("abc","x","cdef","dd").stream()
.collect(Collectors.joining("-"));
//结果:abc-x-cdef-dd
7. 广义reduce: 功能最强大的Collector(除自定制Collector外)
- 原型:
public static Collector reducing(U identity,
Function super T, ? extends U> mapper, BinaryOperator op)
- demo:统计字符的总个数
int totalLen1=Arrays.asList("abc","x","cdef","dd").stream()
.collect(Collectors.reducing(0,s->s.length(),(a,b)->a+b));
int totalLen2=Arrays.asList("abc","x","cdef","dd").stream()
.collect(Collectors.reducing(0,String::length,Integer::sum));
//result:10