Java 8 stream group by min and max

问题

Suppose you run an SQL query against an employees table:

SELECT department, team, MIN(salary), MAX(salary)
  FROM employees
 GROUP BY department, team

And in the java client you map the result set to a list of Aggregate instances by making a DAO call like below:

List<Aggregate> deptTeamAggregates = employeeDao.getMinMaxSalariesByDeptAndTeam()

And 'Aggregate' has getter methods for department, team, minSalary, maxSalary and there is a Pair<T, T> tuple

What would be the clearest and possible the most optimal way to map the result set into the two maps below:

Map<String, Pair<Integer, Integer>> byDepartmentMinMax = ...
Map<Pair<String, String>, Pair<Integer, Integer>> byDepartmentAndTeamMinMax  = ...

I know I could map my result set in a different way and/or make two trips to the database and achieve the same thing in an easier way but I am more about understanding the java 8 capabilities.

Thank you in advance for your inputs.

回答1:

    class Pair<T, U> {
        public final T x;
        public final U y;

        public Pair(T x, U y) {
            this.x = x;
            this.y = y;
        }
    }

    Collector<Aggregate, ?, Pair<Integer, Integer>> aggregateSalary = 
        mapping(a -> new Pair<>(a.getMinSalary(), a.getMaxSalary()),
            reducing(new Pair<>(Integer.MAX_VALUE, Integer.MIN_VALUE),
                (a, b) -> new Pair<>(Math.min(a.x, b.x), Math.max(a.y, b.y))));

    Map<String, Pair<Integer, Integer>> byDepartmentMinMax =
        deptTeamAggregates.stream()
            .collect(groupingBy(a -> a.getDepartment(), aggregateSalary));

    Map<Pair<String, String>, Pair<Integer, Integer>> byDepartmentAndTeamMinMax =
        deptTeamAggregates.stream()
            .collect(toMap(a -> new Pair<>(a.getDepartment(), a.getTeam()), a -> new Pair<>(a.getMinSalary(), a.getMaxSalary())));

来源：https://stackoverflow.com/questions/30958516/java-8-stream-group-by-min-and-max