Learn about Java hashCode() and equals() methods, their default implementation and how to correctly override them. Also learn to implement these methods using Apache Commons package’s utility classes HashCodeBuilder and EqualsBuilder.
hashCode()andequals()methods have been defined inObjectclass which is parent class for java objects. For this reason, all java objects inherit a default implementation of these methods.
Table of Contents: 1) Usage of hashCode() and equals() Methods 2) Override the default behavior 3) EqualsBuilder and HashCodeBuilder 4) Generate hashCode() and equals() using Eclipse 5) Important things to remember 6) Special Attention When Using in ORM
1. Usage of hashCode() and equals() Methods
equals(Object otherObject)– As method name suggests, is used to simply verify the equality of two objects. It’s default implementation simply check the object references of two objects to verify their equality. By default, two objects are equal if and only if they are stored in the same memory address.hashcode()
This hash code is used for determining the bucket location, when this object needs to be stored in some HashTable like data structure.
1.1. Contract between hashCode() and equals()
It is generally necessary to override the hashCode() method whenever equals() method is overridden, so as to maintain the general contract for the hashCode() method, which states that equal objects must have equal hash codes.
- Whenever it is invoked on the same object more than once during an execution of a Java application, the
hashCodemethod must consistently return the same integer, provided no information used inequalscomparisons on the object is modified.
This integer need not remain consistent from one execution of an application to another execution of the same application. - If two objects are equal according to the
equals(Object)method, then calling thehashCodemethod on each of the two objects must produce the same integer result. - It is not required that if two objects are unequal according to the
equals(java.lang.Object)method, then calling thehashCodemethod on each of the two objects must produce distinct integer results.
However, the programmer should be aware that producing distinct integer results for unequal objects may improve the performance of hash tables.
2. Override the default behavior of hashCode() and equals()
Everything works fine until you do not override any of these methods in your classes. But, sometimes the application needs to change the default behavior of some objects. Lets understand why we need to override equals and hashcode methods.
2.1. Default Behavior
Lets take an example where your application has Employee object. Lets create a minimal possible structure of Employee class:
public class Employee
{
private Integer id;
private String firstname;
private String lastName;
private String department;
//Setters and Getters
}Above Employee class has some very basic attributes and their accessor methods. Now consider a simple situation where you need to compare two employee objects.
public class EqualsTest {
public static void main(String[] args) {
Employee e1 = new Employee();
Employee e2 = new Employee();
e1.setId(100);
e2.setId(100);
System.out.println(e1.equals(e2)); //false
}
}
No prize for guessing. Above method will print "false". But, is it really correct after knowing that both objects represent the same employee. In a real-time application, this should return true.
2.2. Should we override only equals() method?
To achieve correct application behavior, we need to override equals() method as below:
public boolean equals(Object o) {
if(o == null)
{
return false;
}
if (o == this)
{
return true;
}
if (getClass() != o.getClass())
{
return false;
}
Employee e = (Employee) o;
return (this.getId() == e.getId());
}
Add this method to your Employee class, and EqualsTest will start returning "true".
So are we done? Not yet. Lets test again above modified Employee class in different way.
import java.util.HashSet;
import java.util.Set;
public class EqualsTest
{
public static void main(String[] args)
{
Employee e1 = new Employee();
Employee e2 = new Employee();
e1.setId(100);
e2.setId(100);
//Prints 'true'
System.out.println(e1.equals(e2));
Set<Employee> employees = new HashSet<Employee>();
employees.add(e1);
employees.add(e2);
System.out.println(employees); //Prints two objects
}
}Above class prints two objects in the second print statement. If both employee objects have been equal, in a Set which stores only unique objects, there must be only one instance inside HashSet, after all both objects refer to the same employee. What is it we are missing??
2.3. Override hashCode() method also
We are missing the second important method hashCode(). As java docs say, if you override equals() method then you must override hashCode() method. So lets add another method in our Employee class.
@Override
public int hashCode()
{
final int PRIME = 31;
int result = 1;
result = PRIME * result + getId();
return result;
}Once above method is added in Employee class, the second statement start printing only single object in second statement, and thus validating the true equality of e1 and e2.
3. EqualsBuilder and HashCodeBuilder utility classes
Apache commons provide two excellent utility classes HashCodeBuilder and EqualsBuilder for generating hash code and equals methods. Below is its usage:
import org.apache.commons.lang3.builder.EqualsBuilder;
import org.apache.commons.lang3.builder.HashCodeBuilder;
public class Employee
{
private Integer id;
private String firstname;
private String lastName;
private String department;
//Setters and Getters
@Override
public int hashCode()
{
final int PRIME = 31;
return new HashCodeBuilder(getId()%2==0?getId()+1:getId(), PRIME).toHashCode();
}
@Override
public boolean equals(Object o) {
if (o == null)
return false;
if (o == this)
return true;
if (o.getClass() != getClass())
return false;
Employee e = (Employee) o;
return new EqualsBuilder().
append(getId(), e.getId()).
isEquals();
}
}4. Generate hashCode() and equals() using Eclipse
If you are using any code editor, then most editors also are capable of generating some good structure for you. For example, Eclipse IDE has option to generate a very good implementation of hashCode() and equals() for you.
Right click on java file -> Source -> Generate hashCode() and equals() …
5. Java hashCode() and equals() best practices
- Always use same attributes of an object to generate
hashCode()andequals()both. As in our case, we have used employeeid. equals()must be consistent (if the objects are not modified, then it must keep returning the same value).- Whenever a.equals(b), then a.hashCode() must be same as b.hashCode().
- If you override one, then you should override the other.
6. Special Attention When Using in ORM
If you’re dealing with an ORM, make sure to always use getters, and never field references in hashCode() and equals(). This is for reason, in ORM, occasionally fields are lazy loaded and not available until called their getter methods.
For example, In our Employee class if we use e1.id == e2.id. It is very much possible that id field is lazy loaded. So in this case, one might be zero or null, and thus resulting in incorrect behavior.
But if uses e1.getId() == e2.getId(), we can be sure even if field is lazy loaded; calling getter will populate the field first.
This is all i know about hashCode() and equals() methods. I hope, it will help someone somewhere.
If you feel, I am missing something or wrong somewhere, please leave a comment. I will update this post again to help others.
Happy Learning !!
sachin rokade
why PRIME value is 31
sindhu
Hi, your blog really helps a lot for all, I appreciate for sharing your knowledge with us.
I am having a small problem, i would be glad if you help me out.
While i am working with hashmap<String,List>, my requirement is to read data from file and set to the bean class, then i need to get the data from bean class and based on conditions i need to add to the list then add to the map. In each if- block i am adding to the map–>map.put(stringvalue,listobject), in each if-block in map entry value is getting replaced with new value and again new key value is getting returned.
like–>when i added first record in first if block
1) v->first record
when again i add to map the second record in another second if block then
2) [v—>{first record,second record},M–>{second record}]
But my requirement is like
v->{first record} M–>{second record}
and when finally i retrieve from map all the values are getting replaced with the latest values.
please suggest me with a solution.
@anonymous
Very nice article. I understood clearly about the contract between equals and hashcode methods.
tushar
Thanks for the elaborative article
Lakshman
Another Important Note you can add is:
– If two objects are not equal according to
then it doesn’t necessarily mean that their hashcode must be different.
Raj kiran
good one.somehow i understand
Marcos
when in doubt, use constant value for hashCode
vaibhav
Where did you find that hashCode method return memory address of an object?
Lokesh Gupta
It’s not memory address; it’s integer representation of memory address.
Hint: http://docs.oracle.com/javase/7/docs/api/java/lang/Object.html#hashCode%28%29
Please let me know if you think otherwise.
Palash Kanti Kundu
Nice Article. It helped me refreshing my knowledge on Hashcode and Equals. Please keep up the good work. i will surely check your blog time to time…
đ
Sharda Singh
very nice and well explained. Keep posting..
Ananth
I guess it alsow returns false because both are Wrapper class.
this.getId() == e.getId()
Correct me if i am wrong!
rakesh
how two different object can have same hashcode,if they have it means they try to address two different value on same position in memory??????
Lokesh
Usually this happens when somebody override the hashCode() method.
Bhagwati Prasad
Two objects may be located in separate location on heap, but they can still be equal. But the Equals-Hashcode contract says, if two objects are equal then they must have the same hashcode.
That’s why your hashcode cannot be integer representation of memory address anymore. you need to override it.
Ivan
@Override
public int hashCode()
{
final int PRIME = 31;
int result = 1;
result = PRIME * result + getId();
return result;
}
What’s the purpose of result variable?
Lokesh
I have used only “id” variable that’s why it looks like un-necessary. A complete method will look like this:
@Override public int hashCode() { final int prime = 31; int result = 1; result = prime * result + ((department == null) ? 0 : department.hashCode()); result = prime * result + ((firstname == null) ? 0 : firstname.hashCode()); result = prime * result + ((id == null) ? 0 : id.hashCode()); result = prime * result + ((lastName == null) ? 0 : lastName.hashCode()); return result; }Morten
I find it tedious to implement equals() and hashCode() by hand. I also see too many mistakes in code in this area. Better to automate this, so I have just released a free open source tool VALJOGen (valjogen.41concepts.com) which can generate your value classes with setters/getters (*), Object.hashCode, Object.equals, Object.toString, Comparable.compareTo and more from plain java interfaces. The generated output can be automatically updated when you change your interfaces – no need to maintain the generated code.
The generated output can be automatically updated when you change your interfaces so no need to maintain the generated code. Compared to alternatives it is much more customisable so no problem if you want to add your own value based custom methods, have the generated code subclass a base class with a non-default constructor, change hashCode to cache results etc. Everything can be customised.
Let me know what you think?
Anshuman Dwivedi
I’ve absorbed every line of article except “If you override one, then you should override the other”.
Why implementations of both are required ?
What does it convey ?
What if I override only hashCode() ?
If possible then kindly explain by any scenario.
Lokesh
You must override hashCode() in every class that overrides equals(). Failure to do so will result in a violation of the general contract for Object.hashCode(), which will prevent your class from functioning properly in conjunction with all hash-based collections, including HashMap, HashSet, and Hashtable. – from Effective Java, by Joshua Bloch
Remember that Java puts a rule that “If two objects are equal using Object class equals method, then the hashcode method should give the same value for these two objects.”
In fact, there is very are chance that you will be first modifying your hashcode() method. In most of the cases, you will need to compare two instances logically rather than default compare which compare the object references in memory. e.g. Two user objects created with new keyword will be stored in different memory locations so not equal according to default comparison using “==” operator. But for application, they actually represent same user entity and they should be treated as same instance. If one session in application says, user with id 1 is logged in, then in another part of application some code should be able to say same thing about user (1) using different instance.
To do this, you must override equals() method first of all (before even overriding hashcode()). Now, above rule says that “If two objects are equal using Object class equals method, then the hashcode method should give the same value for these two objects.” If you are able to satisfy this rule then there is no need to override any more function. If rule is broken, you must override to make rule happy. You see, java has not made it mandatory. No compilation errors.
For more discussion, please read comments below. Some good questions already have been answered there.
rajesh
I have a doubt, what happens if we override only hashcode() method.
Nagesh
“Above class prints two objects in second print statement. If both employee objects have been equal, in a Set which stores only unique objects, there must be only one instance inside HashSet, after all both objects refer to same employee. What is it we are missing??”
————————–
At this point, you could probably explain why it is printing two objects… (probably some discussion on what happens internally when you add e1 object to HasSet, e2 object to HashSet and how do you end up with 2 objects in the HashSet).
This would help add more detail about the working of the Hashing mechanism in Java.
————————–
“We are missing the second important method hashCode(). “
Lokesh
HI Nagesh, Thanks for asking a good question. I would suggest anybody who wants to know why it printed two objects : Please drill down the sourcecode of HashSet.add() method.
In short, HashSet uses a hashmap to store unique values as “keys” in map. Now how hashmap works, is another topic you may be interested in.
Serge
Hello! Why do you use “31” number and multiplication on 1 (!) in hash code generating?
final int PRIME = 31;
int result = 1;
result = PRIME * result + getId();
If employee id is int, and hashcode is integer, you can just return and employee’s id. Am I wrong?
Lokesh
A good discussion happened here: https://stackoverflow.com/questions/3613102/why-use-a-prime-number-in-hashcode
But as far as above example is concerned, you seems right. There is not much difference.
Subhankar
Excellent Explanation. Finally i understood why we override hashcode() and eqals() methods.
Thank you Sir.
swekha
in which case two objects have the same hashcode??
Lokesh
If class has overridden the hashCode() method then it is possible. e.g. Integer class. You create 10 instances of ‘new Integer(10)’, all will have same hashcode i.e. 10.
vinod
Then why you have written this line? “If you override one, then you should override the other.”
Pallabi Kar
Very nice article.
abhi
Nicely explained !! .. Thanks
anonymous
this explanation is the best after a long search for a good neat example with explanation.Thankyou đ
Santhosh Varma
Thanks for explaining the ORM related issue
Jin
Is it O.K to use getId() which returns an Integer and which can be null? And is there any reason that you compared the id(Integer) with a double-equal-sign?
arti
very well explained..
Hemant Sharma
Hello Sir, Tutorials are very good, only the red mark on the image is misplace, i think it should on hashCode() and equals(), but its at generate getters and setters…
You may rectify this…
Lokesh
Are you sure that you posted the comment in correct page? There is no image in this post.
Hemant Sharma
where you told about generating the hashCode and equals method using eclipse
Lokesh
Ohh.. I am sorry. I am behind some firewall which is preventing image loading from wordpress.com domain. Please ignore my comment.
HIMANSU NAYAK
Hi Lokesh,
hashcode() of the object is only used during collections with hashing principle but if i don’t have to use the object in any of the collections then is it fine to override only the equal() method.. is there any scenario where my object will behave strangely because of not overriding the hashcode().
Lokesh
No. it will not. Everything works fine until you try to use methods where hashCode() is needed.
Sagar
A. Suppose I have a class Employee and I am trying to add the same instance of Employee, thrice to a HashSet like below:
Employee e1 = new Employee(123); //Consider 123 as EmpId.
and now:
Set sEmployee = new HashSet();
sEmployee.add(e1);
sEmployee.add(e1);
sEmployee.add(e1);
Also, I have written my equals in such that it always return true and hashcode() always returns a diff number.
1.) Will the above set contain 3 entries of the duplicate obj?
2.) What is the order in which equals() and hashcode() are called?Because ,if hashcode is diff then there is no need to even look at equals method
3.) How many times equals and hashcode will be called in above case? Please elaborate a little on this
B. How can we sort a HashMap(empid(key),empDeptName(value)) first based on empid and then based on DeptName.
Collections.sort() takes only List as parameter, so cant even send a custom comparator here.
Lokesh
Have you tried above anything yourself. I will appreciate if you try first. I am happy to discuss if something results into un-expected behavior.
Jinesh
I believe, since you are overriding hashcode which returns different numbers for the same object it still depends on which bucket these object go to. If
1. Lets say, if hashcodes are 11, 12 and 13 but the hashset api applies another hashing function on the hashcode which determines the bucket. so if these all still fall under same bucket, then it will start checking equals which it always returns true and it will keep replacing the new object on exsiting object. The result is you will have only one bucket and that will have the last object entered. So Size = 1.
2. On the other hand, if the hashset api puts these into different buckets then you will have same objects saved thrice in different buckets and hence SIZE = 3.
3. With the same argument, SIZE can even be 2.
Guys, reply if you concur or think otherwise.
allen
thanks very much
sara
excellent!!!!!! simply superb… very clear and easy to understand..
shweta ramteke
best article after searching for so long time,i found this article the best n helpful.
Mandar
it is awesome .. after lot of googling I find this article is the very best ..
it gives clear idea .. very good example thanks for posting this ..
sanjay
When we are comparing string obj like String str=”ABC”;
str.equals(“XYZ”);
In this case “XYZ” is a object or not?
How it is working in default implementation of equals method.
Lokesh Gupta
“XYZ” is also an object at lower level in JVM, So JVM interpret str.equals(âXYZâ); and str.equals(new String(âXYZâ)); same.
Nawazish
Hi Lokesh,
For some reason I was in a situation where I needed to customize equals() and hashCode () methods. The situation was that based on the primary key, say the employee ID of an employee, I needed to insert elements into an HashSet. However to my discomfort, I am not getting the desired behavior out of the HashSet collection. I am quoting the code below with my comments:
class Student
{
int id;
Student (int id)
{
this.id=id;
}
public boolean equals (Student st)
{
if (this.id==st.id)
return true;
else
return false;
}
public int hashCode ()
{
return 31*id;
}
}
class StudentSet
{
public static void main (String [] args)
{
Set set=new HashSet();
Student s1=new Student (1001);
Student s2=new Student (1001);
System.out.println (“hashCode: “+s1.hashCode());
System.out.println (“hashCode: “+s2.hashCode());
System.out.println (“is s1 Added: “+set.add (s1)); // true
System.out.println (“is s2 Added: “+set.add (s2)); // true ???
System.out.println (“equals: “+s1.equals(s2));
System.out.println (“size: “+set.size()); //size: 2 ???
}
}
As I re-implemented equals() for employee “id” based equality and hashCode() as well, I expected the size of the collection to be 1. However it gives “2”. Could you explain the behavior please?
Lokesh Gupta
It is because you are not overriding the equals() method correctly. Correct way is to pass Object as method argument. Best way to detect is add an @override annotation. It will tell you that method is overridden or local.
@Overridepublic boolean equals(Object st) {
if (this.id == ((Student)st).id)
return true;
else
return false;
}
H Singh
Why wait(), notify() and notifyAll() are delared final in Object class ?
mani
because lock taken place on object
DH
Thanks for the post, very useful reference.
“Whenever a.equals(b), then a.hashCode() must be same as b.hashCode()” – is that “is and only if”? i.e. are there cases where hashCode() are the same but equals() should return false?
Also, I’ve been searching for any mentions of whether or not there’s a pitfall with the following code:
public boolean equals( Object o )
{
// assume that hashCode() has been implemented correctly like the examples in this article
return ( o == this || ( o != null && o.getClass() == getClass() && o.hashCode() == hashCode() ) );
}
Basically, this enforces the contract that equals must return true if the hashCode() of two objects are the same… does it not?
DH
Nevermind, I think I answered my own question… that is, objects that are not logically equal can end up with the same hash value due to imperfect hash function – can’t believe I totally forgot that đ
Rahul Godara
very well explained… thank you đ
Prabu
Very Useful one…thank u….
Krepil23
Can u explain situation like master detail when id is unknown for more than one child objects (ids will be generated on persist). When initial id is 0 (if primitive) for more then one object, but I want put it in HashSet like childs, and then persist the Master with childs.
Creating hashCode from just childId is not good in that case because I’ll get just one object in that childs HashSet?
For situation like this, I create hash and eqals from more than one property. Am I doing wrong?
P.S. Sry on my english :/
Lokesh Gupta
In this specific situation, you sounds right to me.
idledevi
Thank you for helpful information.
For whatever reason, is it ok to override hashcode() and do not override equals?
All blogs talk only about if you override equals(), then you have to override hashcode(). what about reverse?
Lokesh Gupta
It all depends on the usecase where you will use your class.
e.g. if class instances will used as keys in HashMap then you must carefully design your class and check the behavior of both methods because both participate in put and get method in hashmap.
If you might want to implement logical comparison instead of default comparison mechanism in java, you can live happy with only overriding equals() method only.
If there is none of above scenario, then there is no need to override any of above. But if still you want to override, do at your will. It will not make much difference.
Amrit
Thanks Lokesh for clearing out my doubts …I went through many blogs but this one was the best, giving a clear picture .
shriman
Thanks for explaining the ORM related issue.
Tarun Snoi
thank it clear all my doubt about hashcode() or equals () method
sujit
Excellent !!
Jake
I have read that the default hashcode is computed by using the class name, method names and member variables. source : hashcode internals. Is it right or not?
Lokesh Gupta
It depends on JVM implementations of hashCode() native method.
Most implementations are like:
public int hashCode() {
return VMMemoryManager.getIdentityHashCode(this);
}
This getIdentityHashCode() method is native method. Please note that identity hashcode takes no account of the content of the object, just where it is located.
System.identityHashCode(java.lang.Object) is another method similar to VMMemoryManager.getIdentityHashCode().
I guess, I gave enough pointers to you look for more information, if still not convinced.
neeraj Patel
Very Helpful.Keep posting..
Lakhan Singh
thanks it is really very helpful …
swapna
good explanation
B
Excellent explanation,I appreciate you very much,
Javed
i have a question that,
while calculating hashcode is it possible that we got hashcode value out of int range ?
Lokesh
No. It’s not possible.
rajeshwar nagampet (@rajeshwarn433)
Well Explained… Thanks for the article…