| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | john snow <ofbizfanster(at)gmail(dot)com> |
| Cc: | pgsql-novice(at)postgresql(dot)org |
| Subject: | Re: copy from csv and postgresql 10's new identity column type |
| Date: | 2017-12-14 03:14:36 |
| Message-ID: | 29836.1513221276@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-novice |
john snow <ofbizfanster(at)gmail(dot)com> writes:
> we're porting old dbf's to postgresql 10. currently, we have a dbf that did
> not have a defined primary key. in the postgresql 10 table, we want to
> define a technical primary key using postgresql 10's identity column.
> when we export the dbf data to a csv, the csv does not have data for the
> new primary key column, so the COPY FROM command fails and we get the error
> message saying null is not a valid value for the primary key column.
This sounds like a bug that was fixed last week:
You could apply that patch locally, or avoid IDENTITY until 10.2 is out.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | john snow | 2017-12-14 20:14:50 | r there downsides to explicitly naming a pk column xxxx_pk |
| Previous Message | James Keener | 2017-12-14 02:44:18 | Re: copy from csv and postgresql 10's new identity column type |