Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Java LinkedHashMap get first or last entry

People also ask

Is LinkedHashMap ordered?

Yes. See: LinkedHashMap: This linked list defines the iteration ordering, which is normally the order in which keys were inserted into the map (insertion-order).

Does LinkedHashMap follow insertion order?

Important Features of a LinkedHashMap are listed as follows: It contains only unique elements. It may have one null key and multiple null values. It is non-synchronized. It is the same as HashMap with an additional feature that it maintains insertion order.

Does LinkedHashMap maintain insertion order in Java?

LinkedHashMap extends HashMap. It maintains a linked list of the entries in the map, in the order in which they were inserted. This allows insertion-order iteration over the map. That is,when iterating through a collection-view of a LinkedHashMap, the elements will be returned in the order in which they were inserted.

Is LinkedHashMap values ordered?

The whole purpose of the LinkedHashMap class is to keep the ordering of the elements while iterating the map and this behavior is well specified.


The semantics of LinkedHashMap are still those of a Map, rather than that of a LinkedList. It retains insertion order, yes, but that's an implementation detail, rather than an aspect of its interface.

The quickest way to get the "first" entry is still entrySet().iterator().next(). Getting the "last" entry is possible, but will entail iterating over the whole entry set by calling .next() until you reach the last. while (iterator.hasNext()) { lastElement = iterator.next() }

edit: However, if you're willing to go beyond the JavaSE API, Apache Commons Collections has its own LinkedMap implementation, which has methods like firstKey and lastKey, which do what you're looking for. The interface is considerably richer.


Can you try doing something like (to get the last entry):

linkedHashMap.entrySet().toArray()[linkedHashMap.size() -1];

I know that I came too late but I would like to offer some alternatives, not something extraordinary but some cases that none mentioned here. In case that someone doesn't care so much for efficiency but he wants something with more simplicity(perhaps find the last entry value with one line of code), all this will get quite simplified with the arrival of Java 8 . I provide some useful scenarios.

For the sake of the completeness, I compare these alternatives with the solution of arrays that already mentioned in this post by others users. I sum up all the cases and i think they would be useful(when performance does matter or no) especially for new developers, always depends on the matter of each problem

Possible Alternatives

Usage of Array Method

I took it from the previous answer to to make the follow comparisons. This solution belongs @feresr.

  public static String FindLasstEntryWithArrayMethod() {
        return String.valueOf(linkedmap.entrySet().toArray()[linkedmap.size() - 1]);
    }

Usage of ArrayList Method

Similar to the first solution with a little bit different performance

public static String FindLasstEntryWithArrayListMethod() {
        List<Entry<Integer, String>> entryList = new ArrayList<Map.Entry<Integer, String>>(linkedmap.entrySet());
        return entryList.get(entryList.size() - 1).getValue();
    }

Reduce Method

This method will reduce the set of elements until getting the last element of stream. In addition, it will return only deterministic results

public static String FindLasstEntryWithReduceMethod() {
        return linkedmap.entrySet().stream().reduce((first, second) -> second).orElse(null).getValue();
    }

SkipFunction Method

This method will get the last element of the stream by simply skipping all the elements before it

public static String FindLasstEntryWithSkipFunctionMethod() {
        final long count = linkedmap.entrySet().stream().count();
        return linkedmap.entrySet().stream().skip(count - 1).findFirst().get().getValue();
    }

Iterable Alternative

Iterables.getLast from Google Guava. It has some optimization for Lists and SortedSets too

public static String FindLasstEntryWithGuavaIterable() {
        return Iterables.getLast(linkedmap.entrySet()).getValue();
    }

Here is the full source code

import com.google.common.collect.Iterables;
import java.math.BigDecimal;
import java.math.RoundingMode;
import java.util.ArrayList;
import java.util.LinkedHashMap;
import java.util.List;
import java.util.Map;
import java.util.Map.Entry;

public class PerformanceTest {

    private static long startTime;
    private static long endTime;
    private static LinkedHashMap<Integer, String> linkedmap;

    public static void main(String[] args) {
        linkedmap = new LinkedHashMap<Integer, String>();

        linkedmap.put(12, "Chaitanya");
        linkedmap.put(2, "Rahul");
        linkedmap.put(7, "Singh");
        linkedmap.put(49, "Ajeet");
        linkedmap.put(76, "Anuj");

        //call a useless action  so that the caching occurs before the jobs starts.
        linkedmap.entrySet().forEach(x -> {});



        startTime = System.nanoTime();
        FindLasstEntryWithArrayListMethod();
        endTime = System.nanoTime();
        System.out.println("FindLasstEntryWithArrayListMethod : " + "took " + new BigDecimal((endTime - startTime) / 1000000.000).setScale(3, RoundingMode.CEILING) + " milliseconds");


         startTime = System.nanoTime();
        FindLasstEntryWithArrayMethod();
        endTime = System.nanoTime();
        System.out.println("FindLasstEntryWithArrayMethod : " + "took " + new BigDecimal((endTime - startTime) / 1000000.000).setScale(3, RoundingMode.CEILING) + " milliseconds");

        startTime = System.nanoTime();
        FindLasstEntryWithReduceMethod();
        endTime = System.nanoTime();

        System.out.println("FindLasstEntryWithReduceMethod : " + "took " + new BigDecimal((endTime - startTime) / 1000000.000).setScale(3, RoundingMode.CEILING) + " milliseconds");

        startTime = System.nanoTime();
        FindLasstEntryWithSkipFunctionMethod();
        endTime = System.nanoTime();

        System.out.println("FindLasstEntryWithSkipFunctionMethod : " + "took " + new BigDecimal((endTime - startTime) / 1000000.000).setScale(3, RoundingMode.CEILING) + " milliseconds");

        startTime = System.currentTimeMillis();
        FindLasstEntryWithGuavaIterable();
        endTime = System.currentTimeMillis();
        System.out.println("FindLasstEntryWithGuavaIterable : " + "took " + (endTime - startTime) + " milliseconds");


    }

    public static String FindLasstEntryWithReduceMethod() {
        return linkedmap.entrySet().stream().reduce((first, second) -> second).orElse(null).getValue();
    }

    public static String FindLasstEntryWithSkipFunctionMethod() {
        final long count = linkedmap.entrySet().stream().count();
        return linkedmap.entrySet().stream().skip(count - 1).findFirst().get().getValue();
    }

    public static String FindLasstEntryWithGuavaIterable() {
        return Iterables.getLast(linkedmap.entrySet()).getValue();
    }

    public static String FindLasstEntryWithArrayListMethod() {
        List<Entry<Integer, String>> entryList = new ArrayList<Map.Entry<Integer, String>>(linkedmap.entrySet());
        return entryList.get(entryList.size() - 1).getValue();
    }

    public static String FindLasstEntryWithArrayMethod() {
        return String.valueOf(linkedmap.entrySet().toArray()[linkedmap.size() - 1]);
    }
}

Here is the output with performance of each method

FindLasstEntryWithArrayListMethod : took 0.162 milliseconds
FindLasstEntryWithArrayMethod : took 0.025 milliseconds
FindLasstEntryWithReduceMethod : took 2.776 milliseconds
FindLasstEntryWithSkipFunctionMethod : took 3.396 milliseconds
FindLasstEntryWithGuavaIterable : took 11 milliseconds

LinkedHashMap current implementation (Java 8) keeps track of its tail. If performance is a concern and/or the map is large in size, you could access that field via reflection.

Because the implementation may change it is probably a good idea to have a fallback strategy too. You may want to log something if an exception is thrown so you know that the implementation has changed.

It could look like:

public static <K, V> Entry<K, V> getFirst(Map<K, V> map) {
  if (map.isEmpty()) return null;
  return map.entrySet().iterator().next();
}

public static <K, V> Entry<K, V> getLast(Map<K, V> map) {
  try {
    if (map instanceof LinkedHashMap) return getLastViaReflection(map);
  } catch (Exception ignore) { }
  return getLastByIterating(map);
}

private static <K, V> Entry<K, V> getLastByIterating(Map<K, V> map) {
  Entry<K, V> last = null;
  for (Entry<K, V> e : map.entrySet()) last = e;
  return last;
}

private static <K, V> Entry<K, V> getLastViaReflection(Map<K, V> map) throws NoSuchFieldException, IllegalAccessException {
  Field tail = map.getClass().getDeclaredField("tail");
  tail.setAccessible(true);
  return (Entry<K, V>) tail.get(map);
}

One more way to get first and last entry of a LinkedHashMap is to use toArray() method of Set interface.

But I think iterating over the entries in the entry set and getting the first and last entry is a better approach.

The usage of array methods leads to warning of the form " ...needs unchecked conversion to conform to ..." which cannot be fixed [but can be only be suppressed by using the annotation @SuppressWarnings("unchecked")].

Here is a small example to demonstrate the usage of toArray() method:

    public static void main(final String[] args) {
        final Map<Integer,String> orderMap = new LinkedHashMap<Integer,String>();
        orderMap.put(6, "Six");
        orderMap.put(7, "Seven");
        orderMap.put(3, "Three");
        orderMap.put(100, "Hundered");
        orderMap.put(10, "Ten");

        final Set<Entry<Integer, String>> mapValues = orderMap.entrySet();
        final int maplength = mapValues.size();
        final Entry<Integer,String>[] test = new Entry[maplength];
        mapValues.toArray(test);

        System.out.print("First Key:"+test[0].getKey());
        System.out.println(" First Value:"+test[0].getValue());

        System.out.print("Last Key:"+test[maplength-1].getKey());
        System.out.println(" Last Value:"+test[maplength-1].getValue());
    }

    // the output geneated is :
    First Key:6 First Value:Six
    Last Key:10 Last Value:Ten

It's a bit dirty, but you can override the removeEldestEntry method of LinkedHashMap, which it might suit you to do as a private anonymous member:

private Splat eldest = null;
private LinkedHashMap<Integer, Splat> pastFutures = new LinkedHashMap<Integer, Splat>() {

    @Override
    protected boolean removeEldestEntry(Map.Entry<Integer, Splat> eldest) {

        eldest = eldest.getValue();
        return false;
    }
};

So you will always be able to get the first entry at your eldest member. It will be updated every time you perform a put.

It should also be easy to override put and set youngest ...

    @Override
    public Splat put(Integer key, Splat value) {

        youngest = value;
        return super.put(key, value);
    }

It all breaks down when you start removing entries though; haven't figured out a way to kludge that.

It's very annoying that you can't otherwise get access to head or tail in a sensible way ...