对数据分组处理的常用处理方式

数据准备

List<Result> list = new ArrayList<>();
        list.add(new Result("张三", "语文", 93, 1));
        list.add(new Result("李四", "语文", 88, 0));
        list.add(new Result("王五", "语文", 75, 1));
        list.add(new Result("韩六", "语文", 66, 0));
        list.add(new Result("张三", "数学", 63, 1));
        list.add(new Result("李四", "数学", 74, 0));
        list.add(new Result("王五", "数学", 85, 1));
        list.add(new Result("韩六", "数学", 96, 0));
        list.add(new Result("张三", "英语", 88, 1));
        list.add(new Result("李四", "英语", 88, 0));
        list.add(new Result("王五", "英语", 88, 1));
        list.add(new Result("韩六", "英语", 88, 0));

需求描述：

按照性别分组后找到每科成绩最高的数据

代码实现

使用Collectors.groupingBy处理

Map<Integer, Map<String, Optional<Result>>> collect = list.stream().collect(
                Collectors.groupingBy(Result::getGender
                        , Collectors.groupingBy(Result::getSubject
                                , Collectors.maxBy(Comparator.comparing(Result::getScore)))));

结果

针对上面写法的修改

使用Collectors.collectingAndThen处理Optional问题

Map<Integer, Map<String, Result>> collect1 = list.stream().collect(
                Collectors.groupingBy(Result::getGender
                        , Collectors.groupingBy(Result::getSubject
                                , Collectors.collectingAndThen(Collectors.maxBy(
                                		Comparator.comparing(Result::getScore)), Optional::get))));

结果

替换写法

使用toMap替换第二个groupingby

Map<Integer, Map<String, Result>> map1 = list.stream().collect(
                Collectors.groupingBy(Result::getGender
                        , Collectors.toMap(Result::getSubject
                        , Function.identity()
                        , BinaryOperator.maxBy(Comparator.comparing(Result::getScore)))));

要注意toMap第三个参数的类型是BinaryOperator mergeFunction，故使用的是BinaryOperator.maxBy()而非Collectors.maxBy()

如果想只要结果中一个字段如：名字

Map<Integer, Map<String, String>> collect2 = list.stream().collect(
                Collectors.groupingBy(Result::getGender
                        , Collectors.groupingBy(Result::getSubject
                                , Collectors.collectingAndThen(
                                		Collectors.maxBy(
                                			Comparator.comparing(Result::getScore))
                                					, a -> a.get().getName()
                                					)
                                			)
                                	)
                             );

注意

使用Lambda时，请及时处理数据为null的情况。