Question

• Can I use hash key as an id? Specify the reason.

• How do you handle the duplication of the hash code?

• How do you deal with a shared hash table being used on the web server?

Answer 1

1. Hash key can be used as id but is not recommended. But the hash function used should be collision resistant. A collision occurs when for two different values the computed hash is the same. The number of hash collisions for a given data set depends on the hash function used. Usually, hash functions with an extremely large hash key are collision-resistant as it is more difficult to obtain the same value for such a large key. For example, the most widely used cryptographic algorithm SHA-256 uses 256 bits hash key making it more collision resistant then SHA-1 and MD5.

Also, being collision resistant does not mean that the collisions cannot happen. It means that the collisions are hard to find. Many cryptographic algorithms which were initially described as collision resistant were broken later on.

2. If we specifically talk about duplication of hash codes in Java then we have two methods hashcode() and equals(). Basically, if two objects are equal then their hash codes should be equal. If the objects are not explicitly overridden then two objects are equal if and only if they are the same object which is created by a new call. It is called reference equality.

Now if we talk about maps in Java, each map key can have only one value. If a new entry is entered then the previous key-value pair is replaced by the new place. The map computes the hash code of the new key if the hash code does not already exist then it's value is simply entered but if the hashcode is already present then it looks for all entries with equal hashcodes and compares them to the new key using equals and it overrides the existing first key, if found.