Question

To quote from Guidelines and rules for GetHashCode by Eric Lippert:

Rule: Consumers of GetHashCode cannot rely upon it being stable over time or across appdomains

Suppose you have a Customer object
that has a bunch of fields like Name,
Address, and so on. If you make two
such objects with exactly the same
data in two different processes, they
do not have to return the same hash
code. If you make such an object on
Tuesday in one process, shut it down,
and run the program again on
Wednesday, the hash codes can be
different.

This has bitten people in the past.
The documentation for
System.String.GetHashCode notes
specifically that two identical
strings can have different hash codes
in different versions of the CLR, and
in fact they do. Don’t store string hashes in databases and expect them to be the same forever, because they won’t be.

So what is the correct way to create a HashCode of a string that I can store in a database?

(Please tell me I am not the first person to have left this bug in software I have written!)

Answer 1

It depends what properties you want that hash to have. For example, you could just write something like this:

public int HashString(string text)
{
    // TODO: Determine nullity policy.

    unchecked
    {
        int hash = 23;
        foreach (char c in text)
        {
            hash = hash * 31 + c;
        }
        return hash;
    }
}

So long as you document that that is how the hash is computed, that’s valid. It’s in no way cryptographically secure or anything like that, but you can persist it with no problems. Two strings which are absolutely equal in the ordinal sense (i.e. with no cultural equality etc applied, exactly character-by-character the same) will produce the same hash with this code.

The problems come when you rely on undocumented hashing – i.e. something which obeys GetHashCode() but is in no way guaranteed to remain the same from version to version… like string.GetHashCode().

Writing and documenting your own hash like this is a bit like saying, “This sensitive information is hashed with MD5 (or whatever)”. So long as it’s a well-defined hash, that’s fine.

EDIT: Other answers have suggested using cryptographic hashes such as SHA-1 or MD5. I would say that until we know there’s a requirement for cryptographic security rather than just stability, there’s no point in going through the rigmarole of converting the string to a byte array and hashing that. Of course if the hash is meant to be used for anything security-related, an industry-standard hash is exactly what you should be reaching for. But that wasn’t mentioned anywhere in the question.