| From: | Daniele Varrazzo <daniele(dot)varrazzo(at)gmail(dot)com> |
|---|---|
| To: | Adrian Klaver <adrian(dot)klaver(at)gmail(dot)com> |
| Cc: | Denis Papathanasiou <denis(dot)papathanasiou(at)gmail(dot)com>, "psycopg(at)postgresql(dot)org" <psycopg(at)postgresql(dot)org> |
| Subject: | Re: Best strategy for bulk inserts where some violate unique constraint? |
| Date: | 2013-11-06 15:02:32 |
| Message-ID: | CA+mi_8Z4y4L=xNC_xnmiELPnJ4tdp4QXUbEDogr71whyVL0XsQ@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | psycopg |
On Wed, Nov 6, 2013 at 2:54 PM, Adrian Klaver <adrian(dot)klaver(at)gmail(dot)com> wrote:
> On 11/06/2013 06:50 AM, Denis Papathanasiou wrote:
>>
>> On Tue, Nov 5, 2013 at 6:31 PM, Daniele Varrazzo
>>
>> I saw Adrian's reply about how this still won't do what I need in terms
>> of ignoring the constraint violators and keeping the valid entries; will
>> switching autocommit to True first have the desired effect?
>
> No, a particular COPY is treated as a single transaction.
If I'm not mistaken, copying into a temporary table and then inserting
only the wanted record into the target table should work as expected
and leave no bloat. Staying into the psycopg realm it should be the
most efficient way to load data; outside psycopg world there could be
more efficient stand-alone solutions.
-- Daniele
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Adrian Klaver | 2013-11-06 15:07:04 | Re: Best strategy for bulk inserts where some violate unique constraint? |
| Previous Message | Adrian Klaver | 2013-11-06 14:54:15 | Re: Best strategy for bulk inserts where some violate unique constraint? |