Collation types

[marina_sirota]

Hello, dear Friends!
My situation is a following:
I have some not big databases with only English data and some conteining data in Hebrew.
My question is: May I manage two collation types- meaning the database collation type will be for example some kind of Latin and some databases' collation will be in Hebrew.
If you have any experience of managing DB's in two absolutely different languages (Japanesse, Russian, Chinesse, Arabic with English) please tell me about it.
Best regards.
Marina.

[MAK]

In sql 2000 you can have different collation for different columns. example

CREATE TABLE [TEST21] (
[MATURAL_DATE] [char] (8) COLLATE Korean_Wansung_Unicode_CS_AS_KS NOT NULL ,
[RISK_RATING] [char] (2) COLLATE SQL_Latin1_General_CP1_CI_AS NOT NULL ,
) ON [PRIMARY]



CREATE TABLE [TEST22] (
[colu] char(10) COLLATE Albanian_CI_AI_KS_WS NULL,
[MATURAL_DATE] [char] (8) COLLATE Korean_Wansung_Unicode_CS_AS_KS NOT NULL ,
[RISK_RATING] [char] (2) COLLATE SQL_Latin1_General_CP1_CI_AS NOT NULL ,
) ON [PRIMARY]

[MAK]

ofcourse you can have different collation on different databases too.

[rmiao]

Right, you can specify collation at server, db, table and column level in sql2k.

[marina_sirota]

Hello, dear Friends again!
Thank you for all of you for your kind responces!
My situation is: SQL server has Latin type of collation name and my databases have Hebrew collation type.
After I perform any sort operation (data pass temp_db) I receive gibberish.
This is my problem.
After I changed all my databases to Latin supporting Hebrew (non-sensitive)
it seems to be ok...
but my original problem was very strange.
Thank you.
Marina.

[skhanal]

Tempdb collation is same as model database collation. If your db collation is different that that, it causes all sorts of problems.

If you have to maintain different colallations use unicode datatypes like nchar, nvarchar and ntext.