Question

I need to sort an array based on the positions held in another array.

What I have works, but it is kinda slow, is there a faster/better way to implement this?

2 Parts:

Part1

int i = mArrayName.size(); 
int temp = 0;
for(int j=0;j<i;j++){
            temp = mArrayPosition.get(j);
            mArrayName.set(temp, mArrayNameOriginal.get(j));
        }

In this part, mArrayPosition is the position I would like the mArrayName to be in.

Ex.
input:
mArrayName= (one, two, three)
mArrayPosition = (2,0,1)

output:
mArrayName= (three, one two)

Part 2

int k=0;
int j=0;
do{
    if(mArrayName.get(k)!=mArrayNameOriginal.get(j)){
        j++;
    }else{

        mArrayIdNewOrder.set(k, mArrayId.get(j)); 
            k++;
            j=0;
        }
    }while(k < mArrayName.size());
}

In this part, mArrayName is the reordered name array, mArrayNameOriginal is the original name array.

Ex.
mArrayName = (three, one, two)
mArrayNameOriginal = (one, two, three)

Now I want to compare these two arrays, find which entries are equal and relate that to a new array that has their rowId number in it.

Ex.
input:
mArrayId = (001,002,003)

output:
mArrayIdNewOrder = (003,001,002)

So then I will have mArrayIdNewOrder id’s matching up with the correct names in mArrayName.

Like I said these methods work, but is there a faster/better way to do it? I tried looking at Arrays.sort and comparators but they only seem to sort alphabetically or numerically. I saw something like I can create my own rules inside the comparator but it would probably end up being similar to what I already have.

Sorry for the confusing question. I’ll try to clear up any ambiguities if needed.

Answer 1

The best performance read I’ve found is Android’s Designing For Performance doc. You are violating a couple of the “Android way” style of doing things that will help you.

You are using multiple internal getters inside each loop for what looks like a simple value. Redo this by accessing the fields directly.

For extra credit, post your performance comparison results! I’d love to see em!