Very Computer

by **Richard Infiel** » Wed, 14 Jan 1998 04:00:00

What is the story with SQL Server support for 2-byte character sets? Is it
due in V7.0? I have some Japanese customers who are keen to know...

Does anybody out there know if this is planned?
Richard.

by **Neil Pik** » Sat, 17 Jan 1998 04:00:00

Richard,

It should be in there.....

Neil Pike MVP/MCSE
Protech Computing Ltd

by **Dion Wiggin** » Sat, 17 Jan 1998 04:00:00

You can do this in 6.5
We are using it for Chinese and Japanese without a problem. The characterset
must be set to Multilingual when you install and your must set sorting to be
Case Sensitive. If you do this, then it can handle any language you give it.
There are a few limitations in searching as it does not know language you
are using, so if you give it half a double byte character as part of the
search using a LIKE query then it will give you lots of records you did not
want.
Hope this helps
--

Regards

Dion Wiggins
ActiveX Factory Inc.

>What is the story with SQL Server support for 2-byte character sets? Is it
>due in V7.0? I have some Japanese customers who are keen to know...

>Does anybody out there know if this is planned?
> Richard.

by **Hal Berenso** » Sun, 18 Jan 1998 04:00:00

There are two different kinds of 2-character support.

SQL Server 6.5 supports what is known as double-byte character sets (DBCS).
With DBCS you get the normal single byte character (SBCS. e.g. ISO Latin-1
a.k.a. Code Page 1252) set plus one additional set of 255 extended
characters. One byte is stolen from the SBCS code page as a flag byte
indicating this is an extended (2-byte) character. So if you use a
character out of your normal code page it takes one byte whereas if you use
an extended character it takes two. That allows you to run on a machine
with a DBCS variant of Windows NT installed and use that one extended
character set. What you can't do, for example, is mix Kanji and Chinese on
the same machine. This is a frequent requirement in the far east.

SQL Server 7 will support Unicode (as well as retaining SBCS/DBCS support).
In Unicode you can mix and match characters from any character set in the
same string thus meeting the full multilingual needs of the far east.
Unicode differs from DBCS in that characters are always 2-bytes with the
latin characters just a part . For details on Unicode itself see
http://www.unicode.org

--
All opinions expressed here are mine and mine alone

>What is the story with SQL Server support for 2-byte character sets? Is it
>due in V7.0? I have some Japanese customers who are keen to know...

>Does anybody out there know if this is planned?
> Richard.

by **Richard A. Schulm** » Sun, 18 Jan 1998 04:00:00

On Fri, 16 Jan 1998 08:34:43 GMT, Neil Pike

>Richard,

> It should be in there.....

> Neil Pike MVP/MCSE
> Protech Computing Ltd

Yes, but will this be DBCS still? Or will it at last be full Unicode
support, including support for queries that properly display Unicode
result sets, not just the Chinese "Big 5" character set or any of the
other more limited DBCS character sets hitherto supported?

In other words, will SQL Server finally catch up with the underlying
Unicode basis of Windows NT?

(posted and emailed)

Richard Schulman
---
To email me, remove the "XYZ"

by **Neil Pik** » Mon, 19 Jan 1998 04:00:00

Richard,

As far as I know it's full Unicode.

Neil Pike MVP/MCSE
Protech Computing Ltd