Re: BUG #15192: Implement option to use columns order defined at CSV

From: KES <kes-kes(at)yandex(dot)ru>
To: Francisco Olarte <folarte(at)peoplecall(dot)com>, "pgsql-bugs(at)lists(dot)postgresql(dot)org" <pgsql-bugs(at)lists(dot)postgresql(dot)org>
Subject: Re: BUG #15192: Implement option to use columns order defined at CSV
Date: 2018-05-10 18:05:15
Message-ID: 1266741525975515@web55j.yandex.ru
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

Yes. This is feature request.
May you please provide link to better places. I will post this there.

10.05.2018, 17:59, "Francisco Olarte" <folarte(at)peoplecall(dot)com>:
> On Thu, May 10, 2018 at 1:18 PM, PG Bug reporting form
> <noreply(at)postgresql(dot)org> wrote:
>>  The following bug has been logged on the website:
>>
>>  Bug reference: 15192
>>  Logged by: Eugen Konkov
>>  Email address: kes-kes(at)yandex(dot)ru
>>  PostgreSQL version: 9.6.1
>>  Operating system: SMP Debian 4.9.82-1+deb9u3
>>  Description:
>>
>>  While dumping/restore data I faced into problem that column order are
>>  different on different servers.
>
> As many have been faced before..
>
>>  Restoring the data
>>
>>      COPY tariff_details FROM STDIN WITH( FORMAT CSV, HEADER )
>>
>>  cause error:
>>
>>      ERROR: null value in column "periodic" violates not-null constraint
>>      DETAIL: Failing row contains (17, 1, Setup fee, 5.000000000000000000,
>>  null, f, null, 0).
>>      CONTEXT: COPY tariff_details, line 2: "17,1,Setup
>>  fee,5.000000000000000000,,f,,0"
>
> That's not a bug, that's pilot error, explained by yourself.
>
>>  To work around this problem I use next makefile:
>>
>>          dbrestoretable: export PGPASSWORD = ${DB_PASS}
>>          dbrestoretable:
>>                  line=$$(head -n 1 ${APP_ROOT}/db/${TABLE}.dump.csv)
>>                  @cat ${APP_ROOT}/db/${TABLE}.dump.csv | \
>>                          psql -h ${DB_HOST} -p ${DB_PORT} -U ${DB_USER} ${DB_NAME} -c \
>>                                  "BEGIN;COPY ${TABLE}($$line) FROM STDIN WITH( FORMAT CSV, HEADER
>>  );COMMIT;" ||:
>
> I normally use explicit column list on both sides, but that is good (
> as long as first line is fully double quoted to avoid capitalisation
> errors ), so
> error has already been solved.
>
>>  Implement option to force postgres to use columns order from CSV
>
> That is a feature request, not a bug. I think there are better places
> to post those.
>
> Francisco Olarte.

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Jonathan Allen 2018-05-11 00:37:47 RE: BUG #15080: ecpg on windows doesn't define HAVE_LONG_LONG_INT
Previous Message Francisco Olarte 2018-05-10 14:58:53 Re: BUG #15192: Implement option to use columns order defined at CSV