Changing a column collation - problems?

Question

Hello,
I have a VFP database that was migrated to SQL Server 2012. One of the VFP table's PK's utilized uppercase and lowercase values to distinguish rows, which of course does not fly by default in SQL Server.
Rather than having to redesign this part of the application and go through a data conversion, I'm thinking of changing the collation of this PK column and related FK fields to be case sensitive. On the surface it sounds like this would solve my
issue and the app would function as it always has.
Is there anything that would make me think twice about doing this?
Thanks.
Bill

Answer 1

Hi,
Be aware that changing collation at individual fields may break your queries.
Look at the following sample :
USE YourTestDB
SET NOCOUNT ON
SET ANSI_NULLS ON
SET QUOTED_IDENTIFIER ON
SET ANSI_PADDING ON
CREATE TABLE tblX(X CHAR(10) COLLATE SQL_Latin1_General_CP1_CI_AS NOT NULL, CONSTRAINT PK_tblX PRIMARY KEY CLUSTERED (X ASC))
CREATE TABLE tblY(Y CHAR(10) COLLATE Latin1_General_100_BIN NOT NULL, CONSTRAINT PK_tblY PRIMARY KEY CLUSTERED (Y ASC))
INSERT INTO tblX VALUES ('A');
INSERT INTO tblX VALUES ('B');
INSERT INTO tblX VALUES ('C');
INSERT INTO tblX VALUES ('D');
INSERT INTO tblX VALUES ('E');
INSERT INTO tblY VALUES ('A');
INSERT INTO tblY VALUES ('C');
INSERT INTO tblY VALUES ('E');
INSERT INTO tblY VALUES ('G');
INSERT INTO tblY VALUES ('I');
SELECT tblX.X, tblY.Y FROM tblX INNER JOIN tblY ON tblX.X = tblY.Y
-- Gives the following error
-- Msg 468, Level 16, State 9, Line 20
-- Cannot resolve the collation conflict between "Latin1_General_100_BIN" and "SQL_Latin1_General_CP1_CI_AS" in the equal to operation.
-- You'll have to rewrite your query as follows
SELECT tblX.X, tblY.Y FROM tblX INNER JOIN tblY ON tblX.X COLLATE Latin1_General_100_BIN = tblY.Y
-- Or
SELECT tblX.X, tblY.Y FROM tblX INNER JOIN tblY ON tblX.X = tblY.Y COLLATE SQL_Latin1_General_CP1_CI_AS
Sebastian Sajaroff Senior DBA Pharmacies Jean Coutu