Slovak_CI_AS collation not recognizing case insensitivity in SQL Server

Question

Hi All,

Greetings! Hope you all are doing well..

We are facing an issue with the “Slovak_CI_AS” collation and need your inputs. When specified with lowered-case column names, scripts are not recognizing a subset of columns while they are working for others.

Below are quick steps to reproduce the issue. In this case, the SELECT script is failing for a lowered-case “chkey3” column but working fine for “cname”.

1. Create a new database "CollationTest" with collation "Slovak_CI_AS"
2. Create a new Table "TestTable" as:

CREATE TABLE [dbo].[TestTable](
[cHKey3] nvarchar NULL,
[cName] nvarchar NULL
) ON [PRIMARY]

3. Run the below SELECT scripts:

SELECT * FROM TestTable; // runs fine
SELECT cHKey3 FROM TestTable; // runs fine
SELECT cName FROM TestTable; // runs fine
SELECT chkey3 FROM TestTable; // fails with error message "Invalid column name 'chkey3'"
SELECT cname FROM TestTable; // runs fine

4. Close any open sessions on the database "CollationTest" and alter its collation from "Slovak_CI_AS" to the default "SQL_Latin1_General_CP1_CI_AS"

ALTER DATABASE CollationTest COLLATE SQL_Latin1_General_CP1_CI_AS

5. Run the scripts in step 3. After altering the collation, all scripts run as expected.

It seems that Slovak_CI_AS is not working correctly for all characters. Appreciate your feedback/suggestion at the earliest.

Thanks,
Sanjoy Jana

Answer 1

I could be wrong but I believe this is because Ch and ch are the upper and lower case variants of the same single character according to this Slovak orthography. However, cH are 2 separate characters, c and H.