In java whats the most efficient way to remove duplicate characters from string?

Question

I have a helper function which finds the indexes of duplicate characters in the String. Now whats the best way to remove these duplicates? Thanks!

Answer 1

This is the best way I know to do it. It takes a string, separates it into characters, put it into a hashset (nonrepeating, ordered) and then prints (or could return the string.

This is the best way out of the ones listed

String example = "thiscode";
char[] chars = example.toCharArray();
Set<Character> str = new LinkedHashSet<Character>();
for (char c : chars) {
    str.add(c);
}

StringBuilder sb = new StringBuilder();
for (Character character : str) {
    sb.append(character);
}
System.out.println(sb.toString());

Alternatively:

public static String convert(String example){
    char[] chars = example.toCharArray();
    Set<Character> str = new LinkedHashSet<Character>();
    for (char c : chars) {
        str.add(c);
    }

    StringBuilder sb = new StringBuilder();
    for (Character character : str) {
        sb.append(character);
    }
    return sb.toString();
}

Another way to do it:

    String example = "thiscode";
    StringBuilder sb = new StringBuilder(example);
    for (int i=0; i<example.length(); i++)                            //itterate throught the characters
        if (!sb.toString().contains(example.charAt(i) + ""))          //determine if its in the stringbuilder
            sb.append(example.charAt(i));                             //if not add it
    example = sb.toString();                                          //take result
    System.out.println(example);

Inefficient, but easy implementation

String example = "thiscode";
String empty = "";
boolean alphabet[] = new boolean[26];
for (char c : example.toCharArray())
    if (alphabet[(int) ((c + "").toLowerCase().charAt(0) - 'a')] == false)
        empty += c;
example = empty;
System.out.println(example);

Hope this helps.

Answer 2

You can create a set of the chracters used and utilize the add method as it returns false if the set already contains the value listed, no reason to loop over the elements more than once

    String input = "somesortoftestwords";
    Set<Character> charSet = new HashSet<Character>();
    StringBuilder sb = new StringBuilder();
    for (char c : input.toCharArray()) {
        if (charSet.add(c)){
            sb.append(c);
        }
    }
    System.out.println(sb.toString());

Answer 3

Another Possible Solution:

1. Convert the String into characters

   char[] charz = inputString.toCharArray();

2. Sort the characters
   

   Arrays.sort(charz);

3. Now use a loop and check for duplicates

Answer 4

In the comments of one of the answers here there was talk of removing chars from a StringBuilder without the changing indexes causing problems. So I wrote this. I'm not saying this is the best way to process a String. I would used the Linked Set solution or something similar. (So don't vote down - or up for that matter :)

Here we loop for StringBuilder length -1 and check all chars AFTER that for duplicates and remove them. we only go to -1 because the last char doesn't have anything after it to check. The len - 1 calculation is done every time in the for loop, so it can't run over when chars are removed.

public String removeDuplicates(String string) {
    StringBuilder stringBuilder = new StringBuilder(string);
    for (int x=0;x<stringBuilder.length()-1;x++) {
        String character = Character.toString(stringBuilder.charAt(x));
        int i;
        while ((i = stringBuilder.indexOf(character, x+1)) != -1) {
            stringBuilder.replace(i, i+1, "");
        }
    }
    return stringBuilder.toString();
}