Re: Select UNICODE (UTF8) data as octets from database Options

From: Laurenz Albe <invite_at_spam.to.invalid>
Date: 07 May 2007 08:32:34 GMT
Message-ID: <1178526751.982915@proxy.dienste.wien.at>

OracleServer <helge.hauan_at_applica.no> wrote:
> My database (10g) has been created with database character set to
> "Unicode standard UTF-8 AL32UTF8".
>
> The data has been stored correctly. Used select dump(name, 1016)
> from...
> Query returned:
> Len = 3 CharacterSet=AL32UTF8: e4, b8, 84
> UTF-8 encoding of Unicode character (Traditional Chinese): ?
>
>
> Can someone tell me how to select this col into a ProC VARCHAR2
> variable? The query in my ProC code returns this a value, but the
> lenght is equal to 1. How can I get all the 3 octets into my VARCHAR2
> variable?
> Does it depends on any compilation option, or the PC locale?

It should depend only on the NLS_LANG environment variable.

You must set it to something that supports this 'shang' character, like 'SIMPLIFIED CHINESE_CHINA.ZHS16GBK' or AMERICAN_AMERICA.AL32UTF8

There is a precompiler option NLS_CHAR with which you can specify which host variables will be treated as multibyte variables.

Yours,
Laurenz Albe Received on Mon May 07 2007 - 03:32:34 CDT