How to Convert a List Into Map of Map Map<String, Map<String,Object>>

Question

I have a List<Employee> e which I want to convert into Map<String, Map<String,Emp>> where the outer string should be "Name" and the inner string should be "Domain".

       Name Id Domain
e(0) - Emp1, 1, Insurance
e(1) - Emp1, 2, Sales
e(2) - Emp2, 3, Sales
e(3) - Emp4, 4, Marketing

I am tried the following-

e.stream().collect(Collectors.groupingBy(
                                   Employee::getName,
                                   toMap(Employee::getDomain,Emp)));

So the expected output map should look like

<Emp1>
     <Insurance, e(0)>
     <Sales, e(1)>
<Emp2>
     <Sales, e(2)>
<Emp4>
     <Marketing, e(3)>

But I get only unique values, actual output-

<Emp1>
     <Insurance, e(0)>
<Emp2>
     <Sales, e(2)>
<Emp4>
     <Marketing, e(3)>

Can someone tell the best way to do it?

Naman · Accepted Answer · 2019-10-17 07:38:53Z

1

What you are mostly looking for is nested grouping such as :

Map<String, Map<String, List<Employee>>> groupedMap = employees.stream()
        .collect(Collectors.groupingBy(Employee::getName,
                Collectors.groupingBy(Employee::getDomain, Collectors.toList())));

Note - The values are List<Employee> which are the employees grouped by name and then by domain. (Both same clubbed into a List.)

If you were to stricly adhere to getting a single employee corresponding to the specified grouping, the code pretty much works for me with a small modification:

Map<String, Map<String, Employee>> groupedReducedMap = employees.stream()
        .collect(Collectors.groupingBy(Employee::getName,
                Collectors.toMap(Employee::getDomain,
                        Function.identity(), // value as the employee instance
                        (a, b) -> a))); // choose first instance for similar 'domain'

edited Oct 17, 2019 at 7:38

answered Oct 17, 2019 at 7:11

Naman

32.7k32 gold badges240 silver badges385 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

ps_403 Over a year ago

Thanks, for the required data structure the second one works

Eran · Accepted Answer · 2019-10-17 07:31:37Z

0

Since the output should be Map<String, Map<String,Employee>> and not Map<String, Map<String,List<Employee>>> (i.e. based on your requested output, there cannot be two Employees having the same name and the same domain), you can chain two groupingBy and then use reducing to ensure that each inner group will have a single Employee instead of a List<Employee>:

Map<String, Map<String,Optional<Employee>>> output =
    e.stream()
     .collect(Collectors.groupingBy(Employee::getName,
                                    Collectors.groupingBy(Employee::getDomain,
                                                          Collectors.reducing((x1,x2)->x2))));

The problem with this version of reducing is that it returns an Optional<Employee> instead of Employee, even though we know the Optional will never be empty.

We get can around that with:

  Map<String, Map<String,Employee>> output =
      e.stream()
       .collect(Collectors.groupingBy(Employee::getName,
                                      Collectors.groupingBy(Employee::getDomain,
                                                            Collectors.reducing(e.get(0),
                                                                                (x1,x2)->x2))));

Now, we use a variant of reducing having an identity value, to which we pass an arbitrary Employee instance (it doesn't matter which, since it will always be replaced by the correct instance).

edited Oct 17, 2019 at 7:31

answered Oct 17, 2019 at 7:16

Eran

395k57 gold badges726 silver badges793 bronze badges

1 Comment

Naman Over a year ago

IIRC, sometime back Holger pointed this out, that toMap is preferable to groupingBy + reducing. Here - in this Q&A

Collectives™ on Stack Overflow

How to Convert a List Into Map of Map Map<String, Map<String,Object>>

2 Answers 2

1 Comment

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related