Given the following example seed:
List<EmailGroupingDataDTO> emailDataToGroup = Arrays.asList(
buildEmailGroupingDto(1L, "member_number_1", "username_1", "group_code_1"),
buildEmailGroupingDto(2L, "member_number_1", "username_1", "group_code_1"),
buildEmailGroupingDto(3L, "member_number_1", "username_2", "group_code_1"),
buildEmailGroupingDto(4L, "member_number_1", "username_2", "group_code_1"),
buildEmailGroupingDto(5L, "member_number_1", "username_2", "group_code_3"),
buildEmailGroupingDto(6L, "member_number_2", "username_1", "group_code_1"),
buildEmailGroupingDto(7L, "member_number_2", "username_1", "group_code_2"),
buildEmailGroupingDto(8L, "member_number_2", "username_1", "group_code_2"),
buildEmailGroupingDto(9L, "member_number_3", "username_1", "group_code_1"),
buildEmailGroupingDto(10L, "member_number_3", "username_1", "group_code_1"),
buildEmailGroupingDto(11L, "member_number_3", "username_1", "group_code_2"),
buildEmailGroupingDto(12L, "member_number_3", "username_1", "group_code_2"),
buildEmailGroupingDto(13L, "member_number_3", "username_1", "group_code_3"),
buildEmailGroupingDto(14L, "member_number_4", "username_1", "group_code_3"),
buildEmailGroupingDto(15L, "member_number_4", "username_1", "group_code_3")
);
I want to create a List of Lists (batches) based on whether either one of the following properties changes when iterating through the list:
private boolean isEmailDataEqual(EmailGroupingDataDTO currDto, EmailGroupingDataDTO nextDto) {
return currDto.getMemberNumber().equals(nextDto.getMemberNumber())
&& currDto.getUsername().equals(nextDto.getGroupCode())
&& currDto.getGroupCode().equals(nextDto.getGroupCode());
}
Through regular iteration I can manage this myself probably.
However the question I have is whether there is a stream api that can manage this declaratively.
CodePudding user response:
StreamEx will do this for you:
List<List<EmailGroupingDataDTO>> groups = StreamEx.of(emailDataToGroup)
.groupRuns(this::isEmailDataEqual)
.toList();
CodePudding user response:
You could use AtomicInteger & AtomicReference
for the classifier of Collectors.groupingBy
and build a map (and get the grouped batches using Map.values
). Integer as group key and reference as a representative element of a group and to enable an easy way of comparison for your attributes.
import java.util.concurrent.atomic.AtomicInteger;
import java.util.concurrent.atomic.AtomicReference;
....
List<EmailGroupingDataDTO> emailDataToGroup = //your list
AtomicInteger ai = new AtomicInteger();
AtomicReference<EmailGroupingDataDTO> myRef = new AtomicReference<>(emailDataToGroup.get(0));
Collection<List<EmailGroupingDataDTO>> result =
emailDataToGroup.stream().collect(Collectors.groupingBy(myDto -> {
boolean equal = isEmailDataEqual(myDto, myRef.get());
if (!equal) myRef.getAndSet(myDto);
return equal ? ai.get() : ai.incrementAndGet();
})).values();
NOTE: as @shmosel pointed out correctly, this approach will only work if you stream over your list sequentally.