Joost Kraaijeveld schrieb:
> On Wed, 2005-08-31 at 12:40 +0200, Roman Neuhauser wrote:
>
>> check these man pages: pg_dump(1), pg_restore(1), alter_table(7)
>
>
> I am afraid that the problem is more complex. The original database (which is created with SQL_ASCII) contains
invalidbyte sequences in some columns (target database created with UNICODE):
>
> jkr@Panoramix:~/postgresql$ pg_dump -t artik munttest | psql muntfinal
>
>>output.txt
>
> ERROR: invalid byte sequence for encoding "UNICODE": 0xeb207a
> CONTEXT: COPY artik, line 11705, column omschrijving: "Anna v. Groot
> Brittannië zi prf 38.61 mm"
>
> So I cannot dump/restore/alter table. I was hoping that piping the text
> from stdout to psql that a valid conversion to unicode would take place
> but apparently that is not the case.
>
> Any other ideas?
If you know the implicit charset you used in your SQL_ASCII
db, try with: SET client_encoding TO 'the_charset';
In your import script if you use one.
