> On Mar 27, 2018, at 3:47 PM, armand pirvu <armand.pi...@gmail.com> wrote:
>
> Folks
>
>
> From another system (non postgres) as data gets changed we collect in the
> form like below into a named pipe which is tailed into psql -U csidba -d
> repdb -hrephost
>
> insert into foo (mydata) values ('{"mail_stop": "" , "city_id": "Garching bei
> M?nchen"}');
>
> And got a failure
> ERROR: invalid byte sequence for encoding "UTF8": 0xfc
>
>
> insert into foo (mydata) values ('{"mail_stop": "" , "city_id": "Garching bei
> M?nchen"}');
>
> Problem is said system to convert to UTF8 is nicely put a huge amount of
> effort and we don't have the resources anyway at this point
>
>
> This is not a copy, just a tailed named pipe into psql
>
> Is there any trick/way I can shove this data into Postgres ? I really don't
> care whether it read "Garching bei M?nchen" or "Garching bei Mfoonchen"
>
>
>
> Thanks
> — Armand
>
>
Sorry for double posting but I found that if I change the client_encoding to
WIN1252 I am able to move forward
"city_id": "Garching bei München”
