| You're the second person to suggest a hash table, so maybe I
should look into it. Could I ask a couple questions? As I
understand it, a hash table is a table of key/value pairs. I
give it a key, it returns a value. In this case, I will
need multiple values: the primary keys of the records that have
the non-unique indexed field. Of course I could parse a
comma delimited string, but there's no way to know how long that
string might get, and IIRC most hash table programs use
fixed-length value data files.
So what am I missing? How would you use hash tables to store the
locations of multiple records?
-Miko |
