More efficient way to remove elements from an array list

Question

I have developed a array list something like this

ArrayList<String> list = new ArrayList<String>();
list.add("1");
list.add("8");
list.add("8");
list.add("3");
list.add("4");

Now my question is: if I want to remove the "8"s from the list, which way is better?

first way:

for(int i = 0; i < list.size(); i++) {
    if(list.get(i).equals("8")) {
        list.remove(i);
        i--;
    }
}

second way:

Iterator<String> iterator = list.iterator();
    while(iterator.hasNext())
        if(iterator.next().equals("8"))
            iterator.remove();

Now please advise which one of them is more efficient and faster from performance point of view and also is there any other way that is something like built in function by using it we can remove duplicate without iterating that much.

add to set duplicate are removed

Balaswamy Vaddeman
– Balaswamy Vaddeman

2012-04-29 07:35:31 +00:00
Commented Apr 29, 2012 at 7:35 — Balaswamy Vaddeman
– Balaswamy Vaddeman, Commented Apr 29, 2012 at 7:35

Marko Topolnik · Accepted Answer · 2012-04-29 07:54:41Z

5

If you just need a set of numbers, then use HashSet and not a List. If you need to preserve the order by which you are putting the numbers, use LinkedHashSet. As for removal, always prefer the version with the iterator, even though in your particular case the performance may be comparable. The idiom with iterator is more widely applicable than indexing, for example if you used a LinkedList, indexing would result in disastrous performance.

answered Apr 29, 2012 at 7:54

Marko Topolnik

201k31 gold badges336 silver badges455 bronze badges

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

Comments

assylias · Accepted Answer · 2012-04-29 08:20:33Z

3

Performance wise they should be similar. Have you tested? If you want to use the built in methods, you can do this with a similar performance.(to be confirmed by testing):

list.removeAll(Arrays.asList("8"));

Finally if you want a list without duplicates, use a Set as others have mentioned.

edited Apr 29, 2012 at 8:20

answered Apr 29, 2012 at 7:55

assylias

330k84 gold badges680 silver badges806 bronze badges

2 Comments

Marko Topolnik Over a year ago

This will be less performant than OP's code. list.remove will start from the beginning of the list every time -- O(n^2) instead of OP's O(n).

dghtr Over a year ago

list.removeAll(Collections.singleton("4")); this is also good I think

aviad · Accepted Answer · 2012-04-29 08:10:26Z

0

If you need to store the duplicates I suggest you to use Map<Integer,Integer> where the 1st Integer is the key and the 2nd is the number of key occurences . So when you add the key that already exists you just increment the corresponding counter. (In remove operation you doing vice - versa). When you need all the distinct values, you just use Map.keySet().

edited Apr 29, 2012 at 8:10

answered Apr 29, 2012 at 7:41

aviad

8,2789 gold badges55 silver badges100 bronze badges

Collectives™ on Stack Overflow

More efficient way to remove elements from an array list

3 Answers 3

Comments

2 Comments

Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

2 Comments

Comments

Linked

Related