I have a table like so: id, title, yield, directions and would like to prevent duplicate records from being added (i.e. according to the title, yield, and directions fields). I won't normally be querying on the yield or directions fields, so I just have indexes for id and title. What's the best way to prevent duplicates from being added? - Before inserting, do a 'select id from stuff where title=? and yield=? and directions=?'. This would want the title and directions fields indexed (which seems like a waste of space since they won't be used except for rare inserts). - Create a unique index across the title, yield, and directions fields. - Create a 'hash' field by md5'ing the title, yield, and directions fields, and create a unique index on it. Then when inserting new records, first create a hash and check if it already exists, or have the database automatically handle this (trigger to compute hash field at insert time - unique index will raise an exception). Thanks for any help, insights, suggestions, etc. CSN __________________________________ Yahoo! Mail - PC Magazine Editors' Choice 2005 http://mail.yahoo.com ---------------------------(end of broadcast)--------------------------- TIP 5: don't forget to increase your free space map settings
