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

On Thu, May 10, 2018 at 1:18 PM, PG Bug reporting form
<noreply@postgresql.org> wrote:
> The following bug has been logged on the website:
>
> Bug reference:      15192
> Logged by:          Eugen Konkov
> Email address:      kes-kes@yandex.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.