On 26.11.2011 00:42, Adrian Klaver wrote:
> On Friday, November 25, 2011 1:19:29 am Condor wrote:
>> Hello,
>> early postgresql versions (< 9.1) did not show this error message:
>> FATAL conversion between LATIN1 and WIN1251 is not supported
>> and connect to db. I access server over the network and every time
>> when
>> I try to login because I work on latin1 I should export
>> LANG="ru_RU.CP1251"
>> for example. With few terminals on few different databases that is
>> pretty disgustingly.
>> Any way how I can avoid it ?
>
> Did the 9.1 database get created with a different character set then
> on previous
> versions?
>
> See here for automatic conversions:
>
> http://www.postgresql.org/docs/9.1/interactive/multibyte.html#AEN32070
>
> Going back to 8.3 at least I do not see that it has changed.
>
> Going to release notes:
> http://www.postgresql.org/docs/9.1/interactive/release-9-1.html
>
> I see:
> "
> Have psql set the client encoding from the operating system locale by
> default
> (Heikki Linnakangas)
>
> This only happens if the PGCLIENTENCODING environment variable is not
> set.
> "
> This led me to:
> http://www.postgresql.org/docs/9.1/interactive/runtime-config-
> client.html#RUNTIME-CONFIG-CLIENT-FORMAT
> "
> client_encoding (string)
>
> Sets the client-side encoding (character set). The default is to
> use the
> database encoding. The character sets supported by the PostgreSQL
> server are
> described in Section 22.3.1.
> "
>
>
> --
> Adrian Klaver
> adrian.klaver@gmail.com
No, charset of databases is the same. I use the same ENV when I upgrade
sql servers
and recreate psql database directory.
About client encoding, I never ever has before a configured postgresql
on my work station
where I connect to servers. Even postgres user and config file did not
exists and this
worked fine in psql versions below 9.1
--
Regards,
Condor
