On 06/05/2012 04:49 PM, Rich Shepard wrote:
> On Tue, 5 Jun 2012, David Johnston wrote:
>
>> As for the duplicate key I would first check to see if you are inserting
>> into a non-empty table and that one or more of your newly inserted
>> records
>> conflicts with existing records on the waterchem table. The easiest way
>> would be to insert into a staging table and then join the staging and
>> live
>> tables together on the PK and see if any records are returned.
>
> David,
>
> Since the two separate tables have no duplicates I find it difficult to
> understand how combining the two generates a duplicate row.
>
> In any case, I always drop the table before running the script after
> correcting errors so there should not be a previous row with the same
> values. Puzzling,
Information that would be helpful:
The schema of the waterchem table
In particular, what is unique constraint "waterchem_pkey" a constraint on?
>
> Thanks,
>
> Rich
>
>
--
Adrian Klaver
adrian.klaver@gmail.com