Обсуждение: Copy command and duplicate items (Support Replace?)

Hi,

Writing a script to pull data from SQL server into a flat-file (or just
piped in directly to PG using Perl DBI)

Just wondering if the copy command is able to do a replace if there are
existing data in the Db already. (This is usually in the case of updates
to specific rows and there be a timestamp indicating it has been changed
etc.)

In MySQL, the mysqlimport util has the --replace function which will
replace the data if there is any event of a duplicate.

Does PG support this?

On Tue, Aug 14, 2007 at 10:50:33AM +0800, Ow Mun Heng wrote:
> Hi,
>
> Writing a script to pull data from SQL server into a flat-file (or just
> piped in directly to PG using Perl DBI)
>
> Just wondering if the copy command is able to do a replace if there are
> existing data in the Db already. (This is usually in the case of updates
> to specific rows and there be a timestamp indicating it has been changed
> etc.)
>
> In MySQL, the mysqlimport util has the --replace function which will
> replace the data if there is any event of a duplicate.
>
> Does PG support this?

No; you'll need to COPY into a temporary or staging table and then
proceed from there. Alternatively, you could use
http://pgfoundry.org/projects/pgloader/.
--
Decibel!, aka Jim Nasby                        decibel@decibel.org
EnterpriseDB      http://enterprisedb.com      512.569.9461 (cell)