Author ygale
Recipients
Date 2004-05-02.02:50:20
SpamBayes Score
Marked as misclassified
Message-id
In-reply-to
Content
Logged In: YES 
user_id=1033539

I suggest leaving the function as is, but replacing the
constant with a value that varies as we step through the
tuple. Take the values from a fixed table. When we reach the
end of the table, start again from the beginning and mung
the values slightly (e.g., increment them).

True, there will always be the possibilities of collisions,
but this will make it very unlikely except in very weird
cases. Using a table of constants is a standard technique
for hashes.
History
Date User Action Args
2007-08-23 14:21:13adminlinkissue942952 messages
2007-08-23 14:21:13admincreate