Re: Problems with NLS_LANG settings

On Wed, 31 Oct 2001 15:53:28 +0100, "Ralf Pickart" <ralf.pickart.NOSPAM_at_gmx.net> wrote:

>Hi,
>
>until now we used the WE8ISO8859P1 characterset for our databases. To
>support other languages/charsets I have to set up a database with multibyte
>unicode characterset. Ok, so far so good... I choose UTF8 as database
>characterset and set NLS_LANG on client side to this value too (environment
>variable and windows registry).
>
> Now I can insert data using SQL*Plus for Windows and get the same data
>back... but if I use the character-based SQL*Plus for DOS I get different
>(special) characters AND if I insert values using SQL*Plus for DOS I get
>these values only with SQL*Plus for DOS correctly. If I use SQL*Plus for
>Windows to retrieve these values I get wrong characters back.
>
>Now my question:
>Have I made a mistake in my NLS_LANG settings or is this the normal
>behaviour of national language support?
>
>Regards
>Ralf Pickart
>

If the strange feeling this has to do with your MS-DOS codepage [Quoted] settings. Remember the MODE command?
[Quoted] AFAIK you still need to use it to support codepage 850 which is equivalent to latin-1 which is equivalent to we8iso8859p1. UTF8 may be a similar case, but then again the (cut-down) DOS shell might not support 2-byte charactersets at all.

Hth

>Oracle DBA - empolis content management GmbH
>
>
>

Sybrand Bakker, Senior Oracle DBA

To reply remove -verwijderdit from my e-mail address Received on Wed Oct 31 2001 - 16:48:52 CET