Question

Putting Hashmaps by reference and putting hashmaps by copy. How do i do the latter? The other issue is the number of String[] types is not really pre-known, so creating multiple instances of Multiset<String> textAndCount = TreeMultiset.create(); isn't very helpful. I've the following code, but my output for both types are the same.

import java.util.HashMap;
import java.util.Iterator;
import java.util.Map;
import java.util.Map.Entry;

import com.google.common.collect.Multiset;
import com.google.common.collect.TreeMultiset;


public class TestIterator {

private static String[] foobarness  = {"foo", "bar", "ness", "foo", "bar", "foo", "ness", "bar", "foo", "ness", "foo", "bar", "foo", "ness", "bar", "ness", "foo", "bar", "foo", "ness"};

private static String[] foobarness2  = {"bar", "ness", "foo", "bar", "foo", "ness", "bar", "foo", "ness", "foo", "foo", "bar", "foo", "ness", "bar", "foo", "ness", "bar", "foo", "ness", "ness", "bar", "foo", "ness"};

private static String[] types = {"type::1", "type::2"};


public static void main(String[] args) {
    Map<String, Multiset<String>> typeTextCount = 
        new HashMap<String, Multiset<String>>();

    Multiset<String> textAndCount = TreeMultiset.create();
    for (int i = 0; i < types.length; i++) {
        if ("type::1".equals(types[i])) {
            for (String text : foobarness)
                textAndCount.add(text, 1);
        }
        if ("type::2".equals(types[i])) {
            for (String text : foobarness2) 
                textAndCount.add(text, 1);
        }
        typeTextCount.put(types[i], textAndCount);
    }

    Iterator<Entry<String, Multiset<String>>> itTTC = 
        typeTextCount.entrySet().iterator();

    while (itTTC.hasNext()) {
        Map.Entry textCt = (Map.Entry) itTTC.next();
        System.out.println(textCt.getKey() + " :\t" + textCt.getValue());
        itTTC.remove();
    }
}

My outputs are from the above code:

type::2 :   [bar x 13, foo x 17, ness x 14]
type::1 :   [bar x 13, foo x 17, ness x 14]

The correct outputs should be:

type::1 :   [bar x 6, foo x 8, ness x 6]
type::2 :   [bar x 7, foo x 9, ness x 8]
Was it helpful?

Solution

Move Multiset<String> textAndCount = TreeMultiset.create() inside your for-loop. This same multiset is being shared by both "types", so your counts are doubled.

Your for-loop might then look like this:

    for (int i = 0; i < types.length; i++) {
        Multiset<String> textAndCount = TreeMultiset.create();
        if ("type::1".equals(types[i])) {
            for (String text : foobarness)
                textAndCount.add(text, 1);
        }
        if ("type::2".equals(types[i])) {
            for (String text : foobarness2)
                textAndCount.add(text, 1);
        }
        typeTextCount.put(types[i], textAndCount);
    }

While you're at it, you could improve your iteration of the map as well by using a for-each style loop. If you're keen on removing each entry as you iterate over it, you could wrap your entrySet in a consumingIterable for the same functionality.

    for (Entry<String, Multiset<String>> textCt : Iterables.consumingIterable(typeTextCount
            .entrySet())) {
        System.out.println(textCt.getKey() + " :\t" + textCt.getValue());
    }

This yields the output:

type::2 :   [bar x 7, foo x 9, ness x 8]
type::1 :   [bar x 6, foo x 8, ness x 6]

If you don't like that order, I'd suggest using an Ordering to get a sorted list of your entries.

OTHER TIPS

Multiset<String> textAndCount = TreeMultiset.create(); should be inside the loop. If you put a copy of the set the output would be

type::1 :   [bar x 6, foo x 8, ness x 6]
type::2 :   [bar x 13, foo x 17, ness x 14]
Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top