unicode, collation puzzle

Hi:

Production server SQL2000 sp4 ent version.
collation: SQL_Latin1_General_CP1_CI_AS

Question:

1. if we need to add a customer from Germany, what we should do?

2. is it possible one databases could handle different language in storing/retriving data with different language?

is that related to Unicode?

thanks in advance

David

is the following statement true?

Looks like we should be able to support German and French since all their characters already exist in the ASCII character set, so Unicode is not necessarily required.

and how could we translating the current default collation to other languages?

thanks
David

If they are in same code page, do nothing. Otherwise, you are on your own without unicode.

SELECT @@serviceName serviceName, @@version as version,
SERVERPROPERTY('productversion') productVersion,
SERVERPROPERTY ('productlevel') productLevel,
SERVERPROPERTY ('edition') edition,
SERVERPROPERTY ('collation') collation
serviceName version productVersion productLevel edition collation
MSSQLSERVER Microsoft SQL Server 2000 - 8.00.2039 (Intel X86)
May 3 2005 23:18:38
Copyright (c) 1988-2003 Microsoft Corporation
Enterprise Edition on Windows NT 5.2 (Build 3790: Service Pack 2)
8.00.2039 SP4 Enterprise Edition SQL_Latin1_General_CP1_CI_AS
(1 row(s) affected)

Hi Rmiao:

1. not quite sure about what you mentioned "if they are in the same code page"

2. our current collation is SQL_Latin1_General_CP1_CI_AS, we have not use nchar, nvarchar, nor ntext in 99% of our table columns datatype.

Does it mean, we need to "Replace all uses of the char, varchar, and text data types with nchar, nvarchar, and ntext. This eliminates the need to consider code page conversion issues.".

3. Have not dealed with another language before....

thanks for your help
David

1. A code page is an ordered set of characters, different code page has different set of characters.

2. Better to use unicode if need support multiple languages, and should put N in front of string to insert like N'xxxxx'.

3. Yes but relevant here.

Hi Rmiao:

create table tempNVarchar
(column1_varchar20 varchar(20),
column2_nvarchar20 nvarchar(20))

insert into tempNVarchar (column1_varchar20, column2_nvarchar20)
values ('ie, Ä ä Ö ö Ü ü ß', 'ie, Ä ä Ö ö Ü ü ß')

1. I expect the column1 will be error since it is not nvarchar.
2. I expect the column2 will be truncated since nvarchar(20) could only store 10 characters.
3. but the insert result is ok withour error. And the retriving is fine too. some puzzle...

again, thanks for your help.
David

1. sql will not generate error even you insert two-byte character in it, you may get undisable characters when query it.
2. not really, you can verify it with len and datalength functions.