papapep wrote:
> Anyone can give me some guide to how the function should do the control
> of duplicated rows?
First tell us what the function should do.
If you just want to ignore the 'duplicate' rows, that's easy:
define 1 column as unique and just insert (without transaction) the data
row by row.
Every 'duplicate' row will throw an error, but since we don't use a
transaction here the row is 'ignored', leaving only rows that are
distinct for that unique column.
I write 'duplicate' in quotes, because you didn't define what you mean
by duplicate.
HTH
--
e-Trolley Sayegh & John, Nabil Sayegh
Tel.: 0700 etrolley /// 0700 38765539
Fax.: +49 69 8299381-8
PGP : http://www.e-trolley.de
