Re: [BUG] - Invalid UNICODE character sequence found(0xc000)

Works fine on my sistem too, but I'm on a linux with UTF-8 encoding.
Anyway, if it gives you the same error from psql, then it's not a JDBC
bug, but rather a backend problem. I suggest you post the problem on the
pgsql-general list, giving all the details of your installation,
including the OS, the encoding of the terminal used, encoding of the DB,
version of the software... I suspect this is only reproducable under
your special circumstances, which I don't have here. On the general list
there are more chances somebody can reproduce it.

Cheers,
Csaba.

On Fri, 2004-01-09 at 15:05, Antonio Gallardo wrote:
> This is the ofending SELECT:
>
> SELECT A0.ROL_NAME,A0.ROL_ENABLE,A0.ROL_ID FROM AUTH_ROLE A0 WHERE
> (A0.ROL_NAME LIKE  'z%') AND A0.ROL_NAME <>  'admin'  ORDER BY 1
>
> I tried it using squirrelSQL - http://squirrel-sql.sourceforge.net/ and
> even in psql it returns:
>
> ERROR:  Invalid UNICODE character sequence found (0xc000)
>
> in psql the "SELECT version();" returns:
>
> PostgreSQL 7.3.4-RH on i386-redhat-linux-gnu, compiled by GCC
> i386-redhat-linux-gcc (GCC) 3.3.2 20031022 (Red Hat Linux 3.3.2-1)
>
> Lets focus on:
>
> (A0.ROL_NAME LIKE  'z%')
>
> If I query:
>
> (A0.ROL_NAME LIKE  'za%')   --- It is OK, but
>
> (A0.ROL_NAME LIKE  'az%')   --- throw the same exception
>
> Looks like PostgreSQL incorrect manage this sequence: "z%"
>
> Can you test it in a table while doing a similar cosntruction?
>
> Can you check this?
>
> Best Regards,
>
> Antonio Gallardo.
>