如何使用groupingBy返回列表<String, String>

Question

I have a List of GroupUser instances with String getGroup() and String getName() methods. I want to return a Map<String, List<String>> where the key is the group and the value is the list of names for that group.

If I use

Map<String, List<GroupUser>> groupUserList = users.stream()
           .collect(Collectors.groupingBy(GroupUser::getGroup));

that returns Map<String, List<GroupUser>>

How can I get a Map<String, List<String>> directly

Answer 1

You can get a Map<String, List<String>> where the key is the group and the value is a list of names with:

    Map<String, List<String>> groupUserList = users.stream()
            .collect(
                    Collectors.groupingBy(GroupUser::getGroup, 
                            Collectors.mapping(GroupUser::getName, 
                                    Collectors.toList())));

Edit

Explaination:

Looking at groupingBy javadocs we see that 1 of the overloads has the signature:

public static <T,K,A,D> Collector<T,?,Map<K,D>> groupingBy(Function<? super T,? extends K> classifier, Collector<? super T,A,D> downstream)

You can think of this downstream as a collector that tells the stream what to do with the elements that it separated by key (in our case the group is the key).

Notice that it is another collector . Next we want to transform a GroupUser into a String which is it's name. We know we transform elements of a stream with map . However this takes a Collector .

Look further into the Collectors API we see that we already have a collector that maps elements and takes another Collector

public static <T,U,A,R> Collector<T,?,R> mapping(Function<? super T,? extends U> mapper, Collector<? super U,A,R> downstream)

Now from the docs: downstream - a collector which will accept mapped values . Since we mapped GroupedUser to it's name the last thing left to do is to collect them in a list and that is one of the most commonly used collectors. Collectors.toList()

Bottom line: Simple usage of collectors tells you that this is where you finished the processing of your stream, however we can compose collectors in more complex scenarios like this one.