Using Unicode Data Types in the Eval Database

The Unicode data types, introduced in SQL Server 7.0, make it much easier to create a database that can function in an international context. Before the Unicode standard was developed, no single character set (code page) could accommodate the three languages the PT application supports—U.S. English, German, and Japanese. All columns in the Eval database that contain character data have a Unicode data type. The columns have nchar, nvarchar, and ntext data types instead of char, varchar, and text data types.

The following columns in the Eval database have Unicode data types:

• The Description column in the Activity table
• The Name and Description columns in the ActivityType table
• The Name column in the Grade table
• The Name and Description columns in the GradeScale table
• The Name and Description columns in the Group table
• The Address, Phone, Email, and Comment columns in the Location table
• The Password and UserName columns in the Logon table.
• The Name, Address, and Contact columns in the Person table
• The Evaluator column in the PersonActivity table
• The PersonType column in the PersonGroup table

Using Unicode data types also allows data to be moved between SQL Server installations that have different code pages. Because each Unicode character uses 2 bytes, the Eval database uses more space when table columns have Unicode data types. To conserve as much space as possible, the database designers use the nvarchar data type instead of the nchar data type for columns in which the length of the entry varies. Because a Unicode character uses 2 bytes, the maximum size of nchar and nvarchar columns is 4,000 characters—not 8,000 characters as in char and varchar columns. The database designers expect the entries in the Eval database tables to be much smaller than 4,000 characters, and this limit has no effect on the PT application.