Re: Identifying a character in a character set

From: Howard J. Rogers <hjr_at_dizwell.com>
Date: Thu, 26 Aug 2004 05:32:20 +1000
Message-ID: <412ce941$0$25605$afc38c87@news.optusnet.com.au>

Harry_Boswell_at_deq.state.ms.us wrote:

> I've been searching for thsi information, but I must not be using the
> right search words. I need to find which Oracle character set
> supports the Greek character mu (ALT+0181). Anybody got a link?
>
> Thanks,
> Harry Boswell

Well, you could always peruse www.unicode.org. In particular, http://www.unicode.org/standard/where/ is always good reading on an extremely wet Wednesday afternoon.

But otherwise: AL32UTF8 will always work. As will its UTF16 cousin. Being Unicode character sets, they will be able to store pretty much anything you can throw at them. For the odd bit of Greek in an otherwise 'English', presumably actually American :-(, database, AL32UTF8 would be the better choice (it's variable width, so the majority English characters still only take one byte. Only the odd bit of foreign input would take double-byte storage).

You don't mention a version, but bear in mind that in 9i, the national (ie, foreign) character set of a database *must* be Unicode. Therefore, you may have to do nothing more than declare the relevant column in the table an NVARCHAR2 instead of just a VARCHAR2.

Regards
HJR Received on Wed Aug 25 2004 - 14:32:20 CDT