BigInt in SQL Server

Hi! All: I am using CT Library comes with freeTDS to get some data from a 
Microsoft SQL Server. I noticed if the field is defined as BigInt(which is a 64 
bit int) in SQL Server, the CT Library somehow translates it into a 
CS_FLOAT_TYPE and ended up losing some precision. I wonder if this is a known 
bug?

e.g.
In SQL Server, do this:
create table test1(AA  bigint, BB  int,  CC  smallint,  DD  tinyint, EE  float) 
;
insert into test1 values(1234567891234567899, 12345678,123,12, 13.245678) ;

When I retrieve that record using CT Library, it becomes:
1234567891234567936,12345678,123,12,13.245678

You can see the first coumn is different from database. but If I do the select 
command directly in SQL Server console, I got the original number, which 
indicates the server does not lose the precision.

Thanks!