Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
kenyeoh
Starting Member
2 Posts |
 Posted - 2003-05-13 : 22:42:05
|
| It is a long but interesting journey ... here it goes ...We have implemented a Data Warehouse that consolidates multi-source data from various Asia Pac and worldwide data, using DTS. Currently, we have English, Chinese (simplified and traditional), and Korean. We are planning to add a couple more. To correctly handle changing dimensions, we use the standard lookups to check for record existence. In the first unicode-enabled iteration, we got the DTS lookup to work by turning on most of the server's native languages in W2000 Regional Setting. How do we know? From within DTS, when we do a test for a Data Pump transformation, we can see the unicode language characters correctly displayed and matched for the lookups. Also, changing dimensions seemed to be handled at the database level, i.e. new records being inserted and existing ones updated. Then we upgraded to SQL Server Patch 3. From that point, the lookups for Korean data are not able to match any data, resulting in inserts on changing dimension when they should be updating. We are using already processed data files to test, so we should only get updates. The current server has the following languages set up:- Japanese- Korean- Simplified Chinese (Default)- Thai- Traditional Chinese- Vietnamese- Western Europe and United StatesOn a test server, we switched the default language to Korean and it worked for Korean changing dimensions but not for the others. Isn't that strange?The assumption was SQL Server is unicode enabled at the DB level and DTS should be able to handle the different code pages. Or is there something we do not yet know or would have missed out? Can anyone help explain this and how we can resolve the issue? If you need some test data, i can provide you. |
|
|
kenyeoh
Starting Member
2 Posts |
Posted - 2003-05-14 : 03:25:58
|
| Guys, I have done some researchand surmised that it boils down to the type of SQL Collation that I select for SQL Server.Has anyone toyed with the collation settings for use of Chinese, Korean and Taiwan together? |
 |
|
|
|
|
|
|
|
DTS and Unicode