How to filter the age while grouping in map with l

Question

Similar to my previous question here, the User objects I have are these

new User("ayush","admin",23)
new User("ashish","guest",19) 
new User("ashish","admin",20) 
new User("garima","guest",29)
new User("garima","super",45)
new User("garima","guest",19)

Now I am trying to get the name to varying ages trend for these users. But I need to filter them above a threshold age. I could get the trend using

Map<String, List<Integer>> userNameAndAgeTrend = users.stream().collect(Collectors.groupingBy(user-> user.getName(), Collectors.mapping(u-> u.getAge(), toList())));

this gives me {ashish=[19, 20], garima=[29, 45, 19], ayush=[23]}. But I am unable to filter the List properly using threshold for example 21 years in my situation using such grouping. Can someone please help?

Also, using .filter(user -> user.getAge() > 21) gives no mapping for ashish, which is what I want to store too. I can use Java10 installed on my machine and trying the suggested solutions.

Answer 1

You need to add the .filter() operation before the .collect() one

Map<String, List<Integer>> userNameAndAgeTrend = 
                      users.stream()
                           .filter(user -> user.getAge() > 21)
                           .collect(Collectors.groupingBy(user-> user.getName(), Collectors.mapping(u-> u.getAge(), toList())));

Answer 2

Stream.filter

You could use filter as

Map<String, List<Integer>> userNameAndAgeTrend = users.stream()
        .filter(a -> a.getAge() > 21) // only above 21
        .collect(Collectors.groupingBy(User::getName, Collectors.mapping(User::getAge, Collectors.toList())));

As confirmed by you in comments, this would give you as output

{garima=[29, 45], ayush=[23]}

---

Collectors.filtering

If you're looking for all the names, you could also use Collectors.filtering since java-9 which explicitly calls out a similar behavior (formatting mine) :

Using a filtering collector as shown above would result in a mapping from that department to an empty Set.

If a stream filter() operation were done instead, there would be no mapping for that department at all.

Its usage should look something like:

Map<String, List<Integer>> userNameAndAgeTrend = users.stream()
        .collect(Collectors.groupingBy(User::getName, Collectors.mapping(User::getAge, 
                        Collectors.filtering(age -> age > 21, Collectors.toList()))));

and the output now would be

{ashish=[], garima=[29, 45], ayush=[23]}

Answer 3

if you want to filter before grouping:

users.stream()
     .filter(u -> u.getAge() > 21) //<--- apply the filter operation
     ...

filter is an intermediate operation which enables one to "keep the elements that satisfy the provided predicate" and exclude others that don't.

So, after the filter operation you have a new stream consisting of only the elements that pass the provided predicate. in this case only users where their age is older than 21.

---

if you want to filter after grouping (not to be confused with filter in a stream, this is a little different)

the filtering collector as of JDK9:

users.stream()
     .collect(groupingBy(User::getName, 
            filtering(u -> u.getAge() > 21, 
                   mapping(User::getAge, toList()))));

see, the accepted answer here for a JDK8 implementation.

With stream’s filter above, the values are filtered first and then it’s grouped. in other words, after filtering we have "no trace" of them.

However, with the filtering collector as of JDK9 we can maintain a trace.