I will be appearing for Google onsite interview in a week time.I understand that hash tables,hash maps,hash functions are very useful and come handy in many interview questions like dictionary,bucket sort,to check duplicacy of one whole document, duplicacy of a URL etc , be it on strings or be it on integers. I am wondering what are some of the popular hash functions both on integers and strings.
One I can think of is h(n)=n for integers, where say we want to rank students depending on their marks i.e. very limited possible values range.
Please help with more popular choices esp for strings,documents.
Thanks,
For strings, one may use the string’s cryptographic hash as the key for a hash table. This will usually lead to a uniform distribution of the hash keys, which is a good hash table property.
If you want to narrow the size of the key (for example only 32 bit), you can still select a cryptographic hash function such as SHA-256 and use the lower 32 bits.
One can also represent a number as a string or as binary data and calculate its cryptographic hash to ensure a uniform key distribution.
Once your keys are uniformally distributed, you don’t need to use a complex hash function- you can just map the key range into equally-sized bins.
To prepare yourself better for the interview, you may want to read this as well.