Unicode support (multi-language)

[JBaileys]

Trying to understand the unicode data options...

If we don't support unicode character data, does that mean there will be limitations on how data is entered into our system in languages such as Chinese, Russian, Greek and Arabic? Or is it possible to use the SQL Collation option? What other limitations exist?

 

[NathanGriffiths]

Hi,

If you don't support Unicode data then you will not be able to store Unicode character sets like Chinese or Greek at all.

For non-Unicode data types like char and varchar, SQL Server uses the code page of the local machine (this is defined by the Windows locale setting) to determine the character set and the collation settings to determine the sort order of the data.

For Unicode data, the collation defines both the character set and the sort order.

SQL Server 2000 supports multiple, column specific collation settings.

If your system only has to support one Latin-based language e.g. US English, then storing Unicode data is inefficient, however if you have to support multiple languages or non-Latin based languages then you will have to use Unicode data types.

See SQL Server Collation Fundamentals in BOL for more details.

Nathan