Language specific SQL Server collation

I have a production SQL Server 2008 R2 server, there are couple databases, one for each client, each in different country.

I've been using the same default collation for all of them but then Bulgarian client has been added, and they use Cyrillic for adding their customers.

The first problem is that I wasn't开发者_如何学Go able to establish for sure what collation should I use. I think Cyrillic_General_CI_AS is the right one. In other words I think that Bulgarian is case insensitive and accent sensitive. Sometimes they use English but i think it's not an issue.

The other question I have is will it work properly with this collation if server uses Polish_CI_AS and therefore system databases like tempdb use it as well, maybe some less specific collation would be better for such mixed environment ?

Is it safe to change collation on working database ?

---

I can only try and answer part of your questions. According to the all-knowing Wikipedia, yes, Bulgarian uses the Cyrillic alphabet (that probably is not your question). But, your statement that

Bulgarian is case insensitive and accent sensitive

does not make sense to me. Languages, like English are neither case (in)sensitive nor accent (in)sensitive. That is simply a design choice on how you want SQL to do string compares internally by default (can be overridden if needed by casting the collation).

In short though, I would agree that Cyrillic_General_CI_AS is what you want if you are using "_CI_AS" for all of your other databases.

As far as your other questions, perhaps someone else can help.