AW: Western European database characterset

First, some of our database applications don’t support Unicode. That’s why we have to use a single byte characterset. My point has been that I might run into conversion errors using the combination client WE8MSWIN1252 / database WE8ISO8859P1[5]. That’s why I would prefer the combination client WE8MSWIN1252 / database WE8MSWIN1252 for non-unicode applications.

Markus

Von: Martin Berger [mailto:martin.a.berger_at_gmail.com] Gesendet: Donnerstag, 6. November 2014 15:14 An: Zwettler Markus (OIZ)
Cc: oracle-l_at_freelists.org
Betreff: Re: Western European database characterset

Markus,

In general I'd user AL32UTF8 for every database. (my opinion) If a client does not use it, still it will work and should be capable for future upgrades.

NLS_LANG on client side must set according to the client application (NOT identical to the DB characterset).

By these settings, writing to the DB should be possible in any case, and clients can display every data, unless it's not mappable in their codepage.

Martin

On Thu, Nov 6, 2014 at 10:32 AM, Zwettler Markus (OIZ) <Markus.Zwettler_at_zuerich.ch<mailto:Markus.Zwettler_at_zuerich.ch>> wrote: Hi,

We are using AL32UTF8 as our standard database characterset mostly having Windows clients using WE8MSWIN1252 as client characterset. Applications not able to use unicode are using the database characterset WE8ISO8859P15 today and WE8ISO8859P1 before (legacy).

I don't feel comfortable with both ISO database charactersets as there might be conversion problems with windows clients on some codepages. I would also use WE8MSWIN1252 as database characterset in that case although we have some interfaces and middleware running on unix. Everyone agrees? Am I missing something?

Cheers,
Markus

--
http://www.freelists.org/webpage/oracle-l