Question

I tried to use HashSet to remove the duplications from an ArrayList<StringBuilder>.

E.g. Here is an ArrayList, each line is a StringBuilder object.

"u12e5 u13a1 u1423"
"u145d"
"u12e5 u13a1 u1423"
"u3ab4 u1489"

I want to get the following:

"u12e5 u13a1 u1423"
"u145d"
"u3ab4 u1489"

My current implementation is:

static void removeDuplication(ArrayList<StringBuilder> directCallList) {
    HashSet<StringBuilder> set = new HashSet<StringBuilder>();
    for(int i=0; i<directCallList.size()-1; i++) {
        if(set.contains(directCallList.get(i)) == false)
            set.add(directCallList.get(i));
    }   
    StringBuilder lastString = directCallList.get(directCallList.size()-1);
    directCallList.clear();
    directCallList.addAll(set);
    directCallList.add(lastString);
} 

But the performance becomes worse and worse as the ArrayList size grows. Is there any problem with this implementation? Or do you have any better ones in terms of performance?

Answer 1

So you had some other options, but I like my solutions short, simple, and to the point. I’ve changed your method to no longer manipulate the parameter, but rather return a new List. I used a Set<String> to see if the contents of each StringBuilder was already included and returned the unique Strings. I also used a for each loop instead of accessing by index.

static List<StringBuilder> removeDuplication(List<StringBuilder> directCallList) {
    HashSet<String> set = new HashSet<String>();
    List<StringBuilder> returnList = new ArrayList<StringBuilder>();
    for(StringBuilder builder : directCallList) {
        if(set.add(builder.toString())
            returnList.add(builder);
    }   
    return returnList;
}