为什么remove函数不适用于hashmaps？

Question

I am working on a simple project that obtains data from an input file, gets what it needs and prints it to a file. I am basically getting word frequency so each key is a string and the value is its frequency in the document. The problem however, is that I need to print out these values to a file in descending order of frequency. After making my hashmap, this is the part of my program that sorts it and writes it to a file.

//Hashmap I create
Map<String, Integer> map = new ConcurrentHashMap<String, Integer>();
int valueMax = -1;
//function to sort hashmap
while (map.isEmpty() == false){
            for (Entry<String, Integer> entry: map.entrySet()){
                if (entry.getValue() > valueMax){
                    max = entry.getKey();
                    System.out.println("max: " + max);
                    valueMax = entry.getValue();
                    System.out.println("value: " + valueMax);
                }
            }
            map.remove(max);
            out.write(max + "\t" + valueMax + "\n");
            System.out.println(max + "\t" + valueMax);  
        }   

When I run this i get:

t 9
t 9
t 9
t 9
t 9
....

so it appears the remove function is not working as it keeps getting the same value. I'm thinking i have an issue with a scope rule or I just don't understand hashmaps very well.

If anyone knows of a better way to sort a hashmap and print it, I would welcome a suggestion.

thanks

Answer 1

Your code doesn't work because on every subsequent iteration, entry.getValue() > valueMax is never true because you don't reset valueMax on re-entry into the while loop.

You don't need to muck around with double-looping over a concurrently accessible map though.

ConcurrentSkipListMap has a lastKey method that returns the greatest key and doesn't require iteration over the entire map.

Answer 2

From your code it looks like you aren't resetting valueMax at the end of your loop. This means the first time round the loop you'll find the maximum but you'll never find any subsequent values because you'll still be comparing to the overall maximum.

Answer 3

Hashmap: no order. You can use ArrayList, which implements List to have an order.

Take a look : http://docs.oracle.com/javase/7/docs/api/java/util/ArrayList.html

Answer 4

I guess it's because there no entries in map with key > initial value of valueMax, so condition if (entry.getValue() > valueMax) never true.

Additionally, there's TreeMap which holds it's contents sorted, so you can just iterate over it's entrySet() w/o any additional logic.

Answer 5

What about something like (not tested)

  final Map<String, Integer> map = new ConcurrentHashMap<String, Integer>();
  final Comparator <String, String> comparator = new Comparator ()
         {
             compare(String o1, String o2) 
             {
                  return map.get(o1) - map.get(o2);
             }
         };

  final TreeMap <String, Integer> sortedMap = new TreeMap (comparator);
  sortedMap.addAll(map);
  System.out.println(sortedMap);