I am trying to maintain insertion order in ConcurrentSkipListSet. The item being added is a custom class type with value(String) and index (int) properties. It implements Comparable interface. The set behaves very inconsistently, sometimes adding duplicate items. Items are considered duplicate if they have same value.
// This is the Item class being added in the set.
final class Item implements Comparable<Item> {
private String value;
private int index;
Item(String val, int idx) {
this.value = val;
this.index = idx;
}
#Override
public int compareTo(Item o) {
// returns zero when values are equal indicating it's a duplicate item.
return this.value.equals(o.value) ? 0 : this.index - o.index;
}
#Override
public String toString() {
return this.value;
}
}
// Below is the main class.
public class Test {
ConcurrentSkipListSet<Item> set;
AtomicInteger index;
public Test() {
set = new ConcurrentSkipListSet<>();
index = new AtomicInteger(0);
}
public static void main(String[] args) {
for (int i = 1; i <= 10; i++) {
Test test = new Test();
test.addItems();
test.assertItems();
}
}
//trying to test it for 10 times. It always fails for once or twice.
private void assertItems() {
Iterator<Item> iterator = set.iterator();
String[] values = {"yyyy", "bbbb", "aaaa"};
for (String value : values) {
if (!value.equals(iterator.next().toString())) {
System.out.println("failed for :" + set);
return;
}
}
System.out.println("passed for :" + set);
}
//adding items with some duplicate values
private void addItems() {
set.add(new Item("yyyy", index.getAndIncrement()));
set.add(new Item("bbbb", index.getAndIncrement()));
set.add(new Item("yyyy", index.getAndIncrement()));
set.add(new Item("aaaa", index.getAndIncrement()));
}
Expected : passed for :[yyyy, bbbb, aaaa]
Actual : failed for :[yyyy, bbbb, yyyy, aaaa]
But as mentioned before, the result is very inconsistent. Most of the times, it passes.
Please let know what could be the reason for this behavior. Is the 'compareTo()' method wrong? If so, it should always fail.
Ideally we should override 'equals()' method also. But it doesn't matter from sorted set perspective.
Appreciate your help.
You have broken the contract of compareTo, which results in undefined behaviour.
Finally, the implementor must ensure that x.compareTo(y)==0 implies
that sgn(x.compareTo(z)) == sgn(y.compareTo(z)), for all z.
You can easily see that you fail this requirement by pulling your Items out into variables:
final Item x = new Item("yyyy", index.getAndIncrement());
final Item z = new Item("bbbb", index.getAndIncrement());
final Item y = new Item("yyyy", index.getAndIncrement());
System.out.println(x.compareTo(y));
System.out.println(x.compareTo(z));
System.out.println(y.compareTo(z));
Output:
0
-1
1
The signs are different, therefore the contract has been broken.
In your compareTo-implementation you are mixing two different properties in an illegal way. Thus you break the contract of the Comparable interface.
In your comparison, you look at the index only if the values are not equal. This way you do not define an overall natural order for your items. Depending on what comparison is done first, the result of sorting a list will be random.
#Override
public int compareTo(Item o) {
int vCompare = this.value.compareTo(o.value);
if (vCompare == 0) {
return this.index - o.index;
}
return vCompare;
}
This implementation will first compare by value and then by index. It adheres to the Comparable contract and actually defines a natural order for Items and works fine with the Set implementation.
Caution: This sample implementation will break the tests.
The tests are there to show the code behaves as intended. But in this case the intended behavior is the actual issue.
It is incompatible with the Comparable contract.
You cannot sort a list by numeric index and expect a lookup by alphabetical value to succeed. But that's exactly what is attempted here. Sort by index but find duplicate names. It does not work this way.
I don't know the implementation of ConcurrentSkipListSet in detail, but it looks like you need to override the equals method of your class to specify what qualifies two objects to be equal.
This is not an answer, rather a solution to achieve the objective based on root cause finding by #Michael and #Jochen. Modified the Item class comparator to below to have natural order of value String.
public int compareTo(Item o) {
return this.value.compareTo(o.value);
}
And then, added an index based comparator to achieve FIFO retrieval.
// This iterator would now be used in assertItems() method in main class.
private Iterator<Item> getFIFOIterator() {
ArrayList<Item> list = new ArrayList<>(set);
list.sort(Comparator.comparingInt(Item::getIndex));
return list.iterator();
}
#Michael and #Jochen : Appreciate you for taking your time and figuring out the root cause.
Related
I have list which contains a property class object, In the list i have 3 status
not_paid
paid
part_paid
I want to sort my list below mentioned order.
First - not_paid
second- part_paid
third -paid
How can I sort my list using Comparator class.?
public static Comparator<OrderHistoryItemData> COMPARE_BY_PAYMENT = new Comparator<OrderHistoryItemData>() {
public int compare(OrderHistoryItemData one, OrderHistoryItemData other) {
String p1 = one.getAttributes().getFieldPaymentStatus();
String p2 = other.getAttributes().getFieldPaymentStatus();
if (p1.equals(p2)) {
return 0;
}
if (p1.equals("not_paid") && (p2.equals("part_paid") || p2.equals("not_paid"))) {
return -1;
}
if (p1.equals("not_paid") && p2.equals("not_paid")) {
return -1;
}
return 1;
}
};
This is my Code. i am getting below order using this code.
paid-->not_paid-->part_paid
This is my Update Code. I got my result.
public static Comparator<OrderHistoryItemData> COMPARE_BY_PAYMENT = new Comparator<OrderHistoryItemData>() {
public int compare(OrderHistoryItemData one, OrderHistoryItemData other) {
String p1 = one.getAttributes().getFieldPaymentStatus();
String p2 = other.getAttributes().getFieldPaymentStatus();
if (p1.equals(p2)) {
return 0;
}
if (p1.equals("not_paid") && (p2.equals("part_paid") || p2.equals("paid"))) {
return -1;
}
if (p1.equals("part_paid") && p2.equals("paid")) {
return -1;
}
return 1;
}
};
To avoid complex comparator, I encourage you to export your statuses to an enum. (Plus this will work if you will add more statuses in the future, without the need to change logic in your comparator):
enum PaymentStatus { // Write them in order you want to be sorted
NOT_PAID,
PART_PAID,
PAID
}
Then sorting will be as simple as :
list.sort(Comparator.comparing(item ->item.getAttributes().getFieldPaymentStatus()));
What you can do is first mapping the strings to integers in the desired order, and then simply subtracting them from eachother.
private static Comparator<Payments> comparator = new Comparator<Payments>() {
// Use this mapping function to map the statuses to ints.
// The lowest number comes first
private int map(String str) {
switch (str) {
case "not_paid":
return 0;
case "part_paid":
return 1;
case "paid":
return 2;
default:
return 3;
}
}
// Alternatively, you can use the Map interface to define the sorting
// order.
#Override
public int compare(Payments o1, Payments o2) {
return map(o1.status) - map(o2.status);
}
};
I suggest – Schidu Luca already mentioned it in his answer – that you use enums to define a fixed set of known values, like payment statuses. This provides compile-time safety.
Note: I wouldn't, however, suggest to bind the enum declaration order to the sorting order.
I am looking to implement a sort feature for my address book application.
I want to sort an ArrayList<Contact> contactArray. Contact is a class which contains four fields: name, home number, mobile number and address. I want to sort on name.
How can I write a custom sort function to do this?
Here's a tutorial about ordering objects:
The Java Tutorials - Collections - Object Ordering
Although I will give some examples, I would recommend to read it anyway.
There are various way to sort an ArrayList. If you want to define a natural (default) ordering, then you need to let Contact implement Comparable. Assuming that you want to sort by default on name, then do (nullchecks omitted for simplicity):
public class Contact implements Comparable<Contact> {
private String name;
private String phone;
private Address address;
#Override
public int compareTo(Contact other) {
return name.compareTo(other.name);
}
// Add/generate getters/setters and other boilerplate.
}
so that you can just do
List<Contact> contacts = new ArrayList<Contact>();
// Fill it.
Collections.sort(contacts);
If you want to define an external controllable ordering (which overrides the natural ordering), then you need to create a Comparator:
List<Contact> contacts = new ArrayList<Contact>();
// Fill it.
// Now sort by address instead of name (default).
Collections.sort(contacts, new Comparator<Contact>() {
public int compare(Contact one, Contact other) {
return one.getAddress().compareTo(other.getAddress());
}
});
You can even define the Comparators in the Contact itself so that you can reuse them instead of recreating them everytime:
public class Contact {
private String name;
private String phone;
private Address address;
// ...
public static Comparator<Contact> COMPARE_BY_PHONE = new Comparator<Contact>() {
public int compare(Contact one, Contact other) {
return one.phone.compareTo(other.phone);
}
};
public static Comparator<Contact> COMPARE_BY_ADDRESS = new Comparator<Contact>() {
public int compare(Contact one, Contact other) {
return one.address.compareTo(other.address);
}
};
}
which can be used as follows:
List<Contact> contacts = new ArrayList<Contact>();
// Fill it.
// Sort by address.
Collections.sort(contacts, Contact.COMPARE_BY_ADDRESS);
// Sort later by phone.
Collections.sort(contacts, Contact.COMPARE_BY_PHONE);
And to cream the top off, you could consider to use a generic javabean comparator:
public class BeanComparator implements Comparator<Object> {
private String getter;
public BeanComparator(String field) {
this.getter = "get" + field.substring(0, 1).toUpperCase() + field.substring(1);
}
public int compare(Object o1, Object o2) {
try {
if (o1 != null && o2 != null) {
o1 = o1.getClass().getMethod(getter, new Class[0]).invoke(o1, new Object[0]);
o2 = o2.getClass().getMethod(getter, new Class[0]).invoke(o2, new Object[0]);
}
} catch (Exception e) {
// If this exception occurs, then it is usually a fault of the developer.
throw new RuntimeException("Cannot compare " + o1 + " with " + o2 + " on " + getter, e);
}
return (o1 == null) ? -1 : ((o2 == null) ? 1 : ((Comparable<Object>) o1).compareTo(o2));
}
}
which you can use as follows:
// Sort on "phone" field of the Contact bean.
Collections.sort(contacts, new BeanComparator("phone"));
(as you see in the code, possibly null fields are already covered to avoid NPE's during sort)
In addition to what was already posted by BalusC it may be worth pointing that since Java 8 we can shorten our code and write it like:
Collection.sort(yourList, Comparator.comparing(YourClass::getSomeComparableField));
or since List now have sort method also like
yourList.sort(Comparator.comparing(YourClass::getSomeComparableField));
Explanation:
Since Java 8, functional interfaces (interfaces with only one abstract method - they can have more default or static methods) can be easily implemented using:
lambdas arguments -> body
or method references source::method.
Since Comparator<T> has only one abstract method int compare(T o1, T o2) it is functional interface.
So instead of (example from #BalusC answer)
Collections.sort(contacts, new Comparator<Contact>() {
public int compare(Contact one, Contact other) {
return one.getAddress().compareTo(other.getAddress());
}
});
we can reduce this code to:
Collections.sort(contacts, (Contact one, Contact other) -> {
return one.getAddress().compareTo(other.getAddress());
});
We can simplify this (or any) lambda by skipping
argument types (Java will infer them based on method signature)
or {return ... }
So instead of
(Contact one, Contact other) -> {
return one.getAddress().compareTo(other.getAddress();
}
we can write
(one, other) -> one.getAddress().compareTo(other.getAddress())
Also now Comparator has static methods like comparing(FunctionToComparableValue) or comparing(FunctionToValue, ValueComparator) which we could use to easily create Comparators which should compare some specific values from objects.
In other words we can rewrite above code as
Collections.sort(contacts, Comparator.comparing(Contact::getAddress));
//assuming that Address implements Comparable (provides default order).
This page tells you all you need to know about sorting collections, such as ArrayList.
Basically you need to
make your Contact class implement the Comparable interface by
creating a method public int compareTo(Contact anotherContact) within it.
Once you do this, you can just call Collections.sort(myContactList);,
where myContactList is ArrayList<Contact> (or any other collection of Contact).
There's another way as well, involving creating a Comparator class, and you can read about that from the linked page as well.
Example:
public class Contact implements Comparable<Contact> {
....
//return -1 for less than, 0 for equals, and 1 for more than
public compareTo(Contact anotherContact) {
int result = 0;
result = getName().compareTo(anotherContact.getName());
if (result != 0)
{
return result;
}
result = getNunmber().compareTo(anotherContact.getNumber());
if (result != 0)
{
return result;
}
...
}
}
BalusC and bguiz have already given very complete answers on how to use Java's built-in Comparators.
I just want to add that google-collections has an Ordering class which is more "powerful" than the standard Comparators.
It might be worth checking out. You can do cool things such as compounding Orderings, reversing them, ordering depending on a function's result for your objects...
Here is a blog post that mentions some of its benefits.
You need make your Contact classes implement Comparable, and then implement the compareTo(Contact) method. That way, the Collections.sort will be able to sort them for you. Per the page I linked to, compareTo 'returns a negative integer, zero, or a positive integer as this object is less than, equal to, or greater than the specified object.'
For example, if you wanted to sort by name (A to Z), your class would look like this:
public class Contact implements Comparable<Contact> {
private String name;
// all the other attributes and methods
public compareTo(Contact other) {
return this.name.compareTo(other.name);
}
}
By using lambdaj you can sort a collection of your contacts (for example by their name) as it follows
sort(contacts, on(Contact.class).getName());
or by their address:
sort(contacts, on(Contacts.class).getAddress());
and so on. More in general, it offers a DSL to access and manipulate your collections in many ways, like filtering or grouping your contacts based on some conditions, aggregate some of their property values, etc.
Ok, I know this was answered a long time ago... but, here's some new info:
Say the Contact class in question already has a defined natural ordering via implementing Comparable, but you want to override that ordering, say by name. Here's the modern way to do it:
List<Contact> contacts = ...;
contacts.sort(Comparator.comparing(Contact::getName).reversed().thenComparing(Comparator.naturalOrder());
This way it will sort by name first (in reverse order), and then for name collisions it will fall back to the 'natural' ordering implemented by the Contact class itself.
The Collections.sort is a good sort implementation. If you don't have The comparable implemented for Contact, you will need to pass in a Comparator implementation
Of note:
The sorting algorithm is a modified mergesort (in which the merge is omitted if the highest element in the low sublist is less than the lowest element in the high sublist). This algorithm offers guaranteed n log(n) performance. The specified list must be modifiable, but need not be resizable. This implementation dumps the specified list into an array, sorts the array, and iterates over the list resetting each element from the corresponding position in the array. This avoids the n2 log(n) performance that would result from attempting to sort a linked list in place.
The merge sort is probably better than most search algorithm you can do.
I did it by the following way.
number and name are two arraylist. I have to sort name .If any change happen to name arralist order then the number arraylist also change its order.
public void sortval(){
String tempname="",tempnum="";
if (name.size()>1) // check if the number of orders is larger than 1
{
for (int x=0; x<name.size(); x++) // bubble sort outer loop
{
for (int i=0; i < name.size()-x-1; i++) {
if (name.get(i).compareTo(name.get(i+1)) > 0)
{
tempname = name.get(i);
tempnum=number.get(i);
name.set(i,name.get(i+1) );
name.set(i+1, tempname);
number.set(i,number.get(i+1) );
number.set(i+1, tempnum);
}
}
}
}
}
use this method:
private ArrayList<myClass> sortList(ArrayList<myClass> list) {
if (list != null && list.size() > 1) {
Collections.sort(list, new Comparator<myClass>() {
public int compare(myClass o1, myClass o2) {
if (o1.getsortnumber() == o2.getsortnumber()) return 0;
return o1.getsortnumber() < o2.getsortnumber() ? 1 : -1;
}
});
}
return list;
}
`
and use: mySortedlist = sortList(myList);
No need to implement comparator in your class.
If you want inverse order swap 1 and -1
With java 8 feature
List<Contact> contact = contactArray.stream().sorted((c1, c2) -> ((c1.getName().compareTo(c2.getName())))).collect(Collectors.toList());
You shoud use the Arrays.sort function. The containing classes should implement Comparable.
I'm trying to sort my List but this one doesn't work. Method collections.sort() does nothing.
public boolean schedule(){
List<Task> keys = new ArrayList<Task>(g.tasks.keySet());
for(int i = 0; i<keys.size();i++){
System.out.println(keys.get(i).getSize());
}
Collections.sort(keys);
for(int i = 0; i<keys.size();i++){
System.out.println(keys.get(i).getSize());
}
return true;
}
and this is my compareTo() method in Task class:
public int compareTo(Task t1) {
Integer csize = new Integer(t1.size);
int cmp = csize.compareTo(t1.size);
return cmp;
}
What is wrong in this method?
collections.sort doesn't work for custom objects
Sure it does, but it won't in your case because your compareTo method is broken. You're comparing t1's size to itself, not to the size of this
You've got:
public int compareTo(Task t1) {
Integer csize = new Integer(t1.size); // get t1's size
int cmp = csize.compareTo(t1.size); // ???? compare with t1's size ???
return cmp;
}
You need to change it to something like:
public int compareTo(Task t1) {
return Integer.compare(this.size, t1.size);
}
So now you're comparing the size of the parameter with the size of the current object.
You have an error in compareTo that has been pointed out in the accepted answer. I am supplying an additional answer only to provide a different idiom you might consider for defining natural order of a class:
class Task implements Comparable<Task> {
private static final Comparator<Task> ORDER = Comparator
.comparingInt(Task::getSize)
.reversed()
.thenComparing(Task::getPriority);
public int compareTo(Task other) {
return ORDER.compare(this, other);
}
}
The potential advantage of this delegation idiom is that on casual reading of a traditional compareTo implementation it's easy to miss things such as the order of arguments reversing the comparison. The declaration makes it very explicit. This also means that you have all the features of Comparator available (e.g. deciding if nulls are first or last).
Already answer was posted by #HoverCraft.
In addition to that
//For ASC
public int compareTo(Task t1) {
return (this.size - t1.size);
}
//For DESC
public int compareTo(Task t1) {
return (t1.size - this.size);
}
First of all sorry if my English bad, its not my first language..
I'm working on and android app project, that needed to sort ArrayList of an object..so I made this method to deal with that...
Lets say that I have an object of Restaurant that will contain this data:
private String name;
private float distance ;
And I sort it using the value of the variable distance from lowest to highest:
public void sort(RArrayList<RestaurantData> datas) {
RestaurantData tmp = new RestaurantData();
int swapped;
boolean b = true;
while (b) {
swapped = 0;
for (int i = 0; i < datas.size()-1; i++) {
if (datas.get(i).getDistance() > datas.get(i+1).getDistance()) {
tmp = datas.get(i);
datas.set(i, datas.get(i+1));
datas.set(i+1, tmp);
swapped = 1;
System.err.println("Swapped happening");
}
}
if (swapped == 0) {
System.err.println("Swapped end");
break;
}
}
But when i try the program..the result of an ArrayList is still random, is there any problem with my logic to sort the ArrayList of an object..
Please Help...Thankyou..
Why not use the Collections.sort method?
Here's how you could do it in your project:
public void sort(RArrayList<RestaurantData> datas) {
Collections.sort(datas, new Comparator<RestaurantData>() {
#Override
public int compare(RestaurantData lhs, RestaurantData rhs) {
return lhs.getDistance() - rhs.getDistance();
}
});
}
The above solution is a bit "destructive" in the sense that it changes the order of the elements in the original array - datas. If that's fine for you go ahead and use it. Personally I prefer things less destructive and if you have the memory to spare (meaning your array is small) you could consider this solution which copies the array before sorting. It also assumes your RArrayList is an implementation of ArrayList or backed up by it:
public List<RestaurantData> sort(RArrayList<RestaurantData> datas) {
// Create a list with enough capacity for all elements
List<RestaurantData> newList = new RArrayList<RestaurantData>(datas.size());
Collections.copy(newList, datas);
Collections.sort(newList, new Comparator<RestaurantData>() {
#Override
public int compare(RestaurantData lhs, RestaurantData rhs) {
return lhs.getDistance() - rhs.getDistance();
}
});
return newList;
}
Another thing to consider is also to create a single instance of the Comparator used in the method, since this implementation will create one instance per call. Not sure if it's worth it though, because it will also be destroyed quite soon since the scope is local.
Here's the documentation for the Collections api
One last thing, the comparator simply needs to return a value less than 0 if the elements are in the right order, bigger than 0 if they're in the wrong order or 0 if they're the same. Therefore it seems to be that it's enough to simply subtract the distances of each restaurant. However, if this isn't the case, please implement the comparator suiting your needs.
How do I get a PriorityQueue to sort on what I want it to sort on?
Also, is there a difference between the offer and add methods?
Use the constructor overload which takes a Comparator<? super E> comparator and pass in a comparator which compares in the appropriate way for your sort order. If you give an example of how you want to sort, we can provide some sample code to implement the comparator if you're not sure. (It's pretty straightforward though.)
As has been said elsewhere: offer and add are just different interface method implementations. In the JDK source I've got, add calls offer. Although add and offer have potentially different behaviour in general due to the ability for offer to indicate that the value can't be added due to size limitations, this difference is irrelevant in PriorityQueue which is unbounded.
Here's an example of a priority queue sorting by string length:
// Test.java
import java.util.Comparator;
import java.util.PriorityQueue;
public class Test {
public static void main(String[] args) {
Comparator<String> comparator = new StringLengthComparator();
PriorityQueue<String> queue = new PriorityQueue<String>(10, comparator);
queue.add("short");
queue.add("very long indeed");
queue.add("medium");
while (queue.size() != 0) {
System.out.println(queue.remove());
}
}
}
// StringLengthComparator.java
import java.util.Comparator;
public class StringLengthComparator implements Comparator<String> {
#Override
public int compare(String x, String y) {
// Assume neither string is null. Real code should
// probably be more robust
// You could also just return x.length() - y.length(),
// which would be more efficient.
if (x.length() < y.length()) {
return -1;
}
if (x.length() > y.length()) {
return 1;
}
return 0;
}
}
Here is the output:
short
medium
very long indeed
Java 8 solution
We can use lambda expression or method reference introduced in Java 8. In case we have some String values stored in the Priority Queue (having capacity 5) we can provide inline comparator (based on length of String) :
Using lambda expression
PriorityQueue<String> pq=
new PriorityQueue<String>(5,(a,b) -> a.length() - b.length());
Using Method reference
PriorityQueue<String> pq=
new PriorityQueue<String>(5, Comparator.comparing(String::length));
Then we can use any of them as:
public static void main(String[] args) {
PriorityQueue<String> pq=
new PriorityQueue<String>(5, (a,b) -> a.length() - b.length());
// or pq = new PriorityQueue<String>(5, Comparator.comparing(String::length));
pq.add("Apple");
pq.add("PineApple");
pq.add("Custard Apple");
while (pq.size() != 0)
{
System.out.println(pq.remove());
}
}
This will print:
Apple
PineApple
Custard Apple
To reverse the order (to change it to max-priority queue) simply change the order in inline comparator or use reversed as:
PriorityQueue<String> pq = new PriorityQueue<String>(5,
Comparator.comparing(String::length).reversed());
We can also use Collections.reverseOrder:
PriorityQueue<Integer> pqInt = new PriorityQueue<>(10, Collections.reverseOrder());
PriorityQueue<String> pq = new PriorityQueue<String>(5,
Collections.reverseOrder(Comparator.comparing(String::length))
So we can see that Collections.reverseOrder is overloaded to take comparator which can be useful for custom objects. The reversed actually uses Collections.reverseOrder:
default Comparator<T> reversed() {
return Collections.reverseOrder(this);
}
offer() vs add()
As per the doc
The offer method inserts an element if possible, otherwise returning
false. This differs from the Collection.add method, which can fail to
add an element only by throwing an unchecked exception. The offer
method is designed for use when failure is a normal, rather than
exceptional occurrence, for example, in fixed-capacity (or "bounded")
queues.
When using a capacity-restricted queue, offer() is generally preferable to add(), which can fail to insert an element only by throwing an exception. And PriorityQueue is an unbounded priority queue based on a priority heap.
Just pass appropriate Comparator to the constructor:
PriorityQueue(int initialCapacity, Comparator<? super E> comparator)
The only difference between offer and add is the interface they belong to. offer belongs to Queue<E>, whereas add is originally seen in Collection<E> interface. Apart from that both methods do exactly the same thing - insert the specified element into priority queue.
from Queue API:
The offer method inserts an element if possible, otherwise returning false. This differs from the Collection.add method, which can fail to add an element only by throwing an unchecked exception. The offer method is designed for use when failure is a normal, rather than exceptional occurrence, for example, in fixed-capacity (or "bounded") queues.
no different, as declare in javadoc:
public boolean add(E e) {
return offer(e);
}
Pass it a Comparator. Fill in your desired type in place of T
Using lambdas (Java 8+):
int initialCapacity = 10;
PriorityQueue<T> pq = new PriorityQueue<>(initialCapacity, (e1, e2) -> { return e1.compareTo(e2); });
Classic way, using anonymous class:
int initialCapacity = 10;
PriorityQueue<T> pq = new PriorityQueue<>(initialCapacity, new Comparator<T> () {
#Override
public int compare(T e1, T e2) {
return e1.compareTo(e2);
}
});
To sort in reverse order, simply swap e1, e2.
Just to answer the add() vs offer() question (since the other one is perfectly answered imo, and this might not be):
According to JavaDoc on interface Queue, "The offer method inserts an element if possible, otherwise returning false. This differs from the Collection.add method, which can fail to add an element only by throwing an unchecked exception. The offer method is designed for use when failure is a normal, rather than exceptional occurrence, for example, in fixed-capacity (or "bounded") queues."
That means if you can add the element (which should always be the case in a PriorityQueue), they work exactly the same. But if you can't add the element, offer() will give you a nice and pretty false return, while add() throws a nasty unchecked exception that you don't want in your code. If failure to add means code is working as intended and/or it is something you'll check normally, use offer(). If failure to add means something is broken, use add() and handle the resulting exception thrown according to the Collection interface's specifications.
They are both implemented this way to fullfill the contract on the Queue interface that specifies offer() fails by returning a false (method preferred in capacity-restricted queues) and also maintain the contract on the Collection interface that specifies add() always fails by throwing an exception.
Anyway, hope that clarifies at least that part of the question.
In here, We can define user defined comparator:
Below code :
import java.util.*;
import java.util.Collections;
import java.util.Comparator;
class Checker implements Comparator<String>
{
public int compare(String str1, String str2)
{
if (str1.length() < str2.length()) return -1;
else return 1;
}
}
class Main
{
public static void main(String args[])
{
PriorityQueue<String> queue=new PriorityQueue<String>(5, new Checker());
queue.add("india");
queue.add("bangladesh");
queue.add("pakistan");
while (queue.size() != 0)
{
System.out.printf("%s\n",queue.remove());
}
}
}
Output :
india
pakistan
bangladesh
Difference between the offer and add methods : link
As an alternative to using Comparator, you can also have the class you're using in your PriorityQueue implement Comparable (and correspondingly override the compareTo method).
Note that it's generally best to only use Comparable instead of Comparator if that ordering is the intuitive ordering of the object - if, for example, you have a use case to sort Person objects by age, it's probably best to just use Comparator instead.
import java.lang.Comparable;
import java.util.PriorityQueue;
class Test
{
public static void main(String[] args)
{
PriorityQueue<MyClass> queue = new PriorityQueue<MyClass>();
queue.add(new MyClass(2, "short"));
queue.add(new MyClass(2, "very long indeed"));
queue.add(new MyClass(1, "medium"));
queue.add(new MyClass(1, "very long indeed"));
queue.add(new MyClass(2, "medium"));
queue.add(new MyClass(1, "short"));
while (queue.size() != 0)
System.out.println(queue.remove());
}
}
class MyClass implements Comparable<MyClass>
{
int sortFirst;
String sortByLength;
public MyClass(int sortFirst, String sortByLength)
{
this.sortFirst = sortFirst;
this.sortByLength = sortByLength;
}
#Override
public int compareTo(MyClass other)
{
if (sortFirst != other.sortFirst)
return Integer.compare(sortFirst, other.sortFirst);
else
return Integer.compare(sortByLength.length(), other.sortByLength.length());
}
public String toString()
{
return sortFirst + ", " + sortByLength;
}
}
Output:
1, short
1, medium
1, very long indeed
2, short
2, medium
2, very long indeed
I was also wondering about print order. Consider this case, for example:
For a priority queue:
PriorityQueue<String> pq3 = new PriorityQueue<String>();
This code:
pq3.offer("a");
pq3.offer("A");
may print differently than:
String[] sa = {"a", "A"};
for(String s : sa)
pq3.offer(s);
I found the answer from a discussion on another forum, where a user said, "the offer()/add() methods only insert the element into the queue. If you want a predictable order you should use peek/poll which return the head of the queue."
Priority Queue has some priority assigned to each element, The element with Highest priority appears at the Top Of Queue. Now, It depends on you how you want priority assigned to each of the elements. If you don't, the Java will do it the default way. The element with the least value is assigned the highest priority and thus is removed from the queue first. If there are several elements with the same highest priority, the tie is broken arbitrarily. You can also specify an ordering using Comparator in the constructor PriorityQueue(initialCapacity, comparator)
Example Code:
PriorityQueue<String> queue1 = new PriorityQueue<>();
queue1.offer("Oklahoma");
queue1.offer("Indiana");
queue1.offer("Georgia");
queue1.offer("Texas");
System.out.println("Priority queue using Comparable:");
while (queue1.size() > 0) {
System.out.print(queue1.remove() + " ");
}
PriorityQueue<String> queue2 = new PriorityQueue(4, Collections.reverseOrder());
queue2.offer("Oklahoma");
queue2.offer("Indiana");
queue2.offer("Georgia");
queue2.offer("Texas");
System.out.println("\nPriority queue using Comparator:");
while (queue2.size() > 0) {
System.out.print(queue2.remove() + " ");
}
Output:
Priority queue using Comparable:
Georgia Indiana Oklahoma Texas
Priority queue using Comparator:
Texas Oklahoma Indiana Georgia
Else, You can also define Custom Comparator:
import java.util.Comparator;
public class StringLengthComparator implements Comparator<String>
{
#Override
public int compare(String x, String y)
{
//Your Own Logic
}
}
Here is the simple example which you can use for initial learning:
import java.util.Comparator;
import java.util.PriorityQueue;
import java.util.Queue;
import java.util.Random;
public class PQExample {
public static void main(String[] args) {
//PriorityQueue with Comparator
Queue<Customer> cpq = new PriorityQueue<>(7, idComp);
addToQueue(cpq);
pollFromQueue(cpq);
}
public static Comparator<Customer> idComp = new Comparator<Customer>(){
#Override
public int compare(Customer o1, Customer o2) {
return (int) (o1.getId() - o2.getId());
}
};
//utility method to add random data to Queue
private static void addToQueue(Queue<Customer> cq){
Random rand = new Random();
for(int i=0;i<7;i++){
int id = rand.nextInt(100);
cq.add(new Customer(id, "KV"+id));
}
}
private static void pollFromQueue(Queue<Customer> cq){
while(true){
Customer c = cq.poll();
if(c == null) break;
System.out.println("Customer Polled : "+c.getId() + " "+ c.getName());
}
}
}