ZeroFusion (5/25/2010)
Suppose you have a Czech_CI_AS collation set up as your DB / table collation. Then you do this:
create table #t (string varchar (5))
insert into #t (string) values ('ACH')
insert into #t (string) values ('ACY')
insert into #t (string) values ('AHHC')
select string from #t
where string COLLATE Czech_BIN like '%C%' -- will miss ACH
drop table #t
You are right, it works.
Are you really doing it? I must say, sometimes I change collation from case-insensitive to case-sensitive, but I have never used BIN type collation yet.