On Fri, 25 Nov 2011, Scott Mead wrote:
> Why don't you first load the data into a table (no primary key), then use
> SQL to find your dups?
>
> once loaded:
> SELECT <primary_key_column>, count(1) from <table> group by 1 having
> count(1) > 1;
>
> At least then, you'll really know what you're in for. You can either
> script a DELETE or... whatever you want to do, once clean, you can add the
> PK.
Scott,
I'll do this. First I need to find why 2 rows (of 143260) are not in the
table after reading the source file. I have those 2 rows somewhat isolated
as one's in the third quarter of the file and one in the fourth quarter.
Thanks,
Rich