delete duplicate rows

This procedure was passed on to me by another netter, Ian Timms. I used it and found it to be very painless. Unload the table to a file, delete all rows from the table, create a unique index on the fields necessary to ensure that the duplicates can not recurr, then reload the table using dbload with the -e option to ignore errors from the attempt to load duplicate rows. The only problem will be if you are not happy with the idea of only getting the first of the duplicates loaded (ie. say you wanted the second one instead!), unless you can order the unloaded to suit (using order by clause on the select) then your almost stuck (almost because you could always use an editor to delete the dupes before loading<g>). Good Luck, Reggie