We have a number of customers in which we supply a database that was built using the collation SQL_Latin1_General_CP1_C1_AS.
We now have to support Unicode to cater for other languages and thanks to the help of other at stackoverflow, I have now a script to change fields from varchar to nva开发者_开发百科rchar.
However, to cater for all languages what collation should I use and will I have to change the collation of the existing database (which we supply)? Do I have to alter my columns to use the collation and if so how would I manage that per installation? All I can see is a nightmare if I have to do this for each installation.
JD
Generally speaking, you should not have to change collation. The nvarchar will handle foreign-language strings, datetime is fully aware of time zones and apps take care of formatting numbers and dates for display.
精彩评论