ImplementingObjects•Chapter14 647 Working with Tables, Indexes, and Constraints Use the ˛ CREATE TABLE statement to define a table by listing columns in the table along with corresponding data types. Indexes are useful for fast searching and sorting data. One clustered index ˛ is allowed per table, and the underlying table data is stored in the order of the clustered index. Nonclustered indexes are separate lookup structures that point to the table heap or the clustered index. Full-text indexes are used for specialized querying using functions like ˛ FREETEXT. Only one full-text index is allowed per table. Indexes and constraints can be defined separately and are bound to an ˛ existing table. Viewing and Modifying Data A view is a ˛ SELECT statement saved with a name. A view can be updated if it is based on a single table, or if it has INSTEAD OF triggers defined on it. Indexes can be created on a view as well as on a table. A stored procedure is any ˛ Transact-SQL statement saved with a name. Stored procedures can update data by using DML statements. A function is a ˛ Transact-SQL statement that usually performs a calculation. Functions must return a value. Functions that return a single value are known as scalar functions, whereas functions that return a table-valued expression are known as table-valued functions. A trigger is a statement that runs automatically, when data in a particular ˛ table or view is modified. Triggers can cancel transactions by using the ROLLBACK TRANSACTION statement. Triggers can be specified as FOR, AFTER, or INSTEAD OF. You can access special INSERTED and DELETED tables within the trigger to find out the old and new values of rows that are being updated. 648 Chapter14•ImplementingObjects Exam Objectives Frequently Asked Questions Q: When should I use the TRUNCATE TABLE statement rather than the DELETE statement? A: You should use the TRUNCATE TABLE statement if you want to quickly and indiscriminately empty the table. A performance advantage is achieved with the TRUNCATE TABLE statement when the database recovery model is set to Bulk-Logged. The DELETE statement allows you to restrict the data you will be deleting using a WHERE clause. Q: Does the TRUNCATE TABLE statement remove the table from the database? A: No, the TRUNCATE TABLE statement removes all data from the table, but the table structure and its definition remain intact. Q: What is the best indexing strategy? A: Indexing strategies vary depending on your data access pattern and parameters such as the size of your table. As a rule of thumb, it is recommended that you create a clustered index on the PRIMARY KEY column and multiple nonclus- tered indexes for other frequently searched-on columns. Q: When should I use FILLFACTOR and PAD_INDEX options? A: Use these options when creating or rebuilding an index. Bear in mind that FILLFACTOR and PAD_INDEX optimize write performance, but slightly decrease read performance because more data pages have to be accessed. Do not bother padding indexes based on identity columns. In these cases, new values will never go in the middle of a page, always at the end. Q: Why would I use the FREETEXT search function instead of multiple LIKE ‘%value%’ comparisons? A: Using LIKE comparisons is a highly time- and resource-consuming operation that always requires a table scan. The FREETEXT utilizes the full-text index structure and delivers much better performance than the LIKE comparison. ImplementingObjects•Chapter14 649 Q: What is the advantage of using an updateable view over updating the table directly, given that the updateable view will by definition always be based on a single table? A: A view is more flexible. For example, you may wish to restructure the under- lying tables without the need to change your client applications. You will only need to update the view in this case, not everything that referenced the tables being restructured. Q: Why should I use stored procedures in preference to functions? A: Stored procedures are usually used to perform an action, like update, insert, or delete, whereas functions are usually used to perform calculations. Functions cannot be used to execute DML statements. Q: If I define multiple triggers on a single table, what is the order of the triggers firing? A: The triggers will execute in random order; you cannot rely on the order of triggers firing. 650 Chapter14•ImplementingObjects Self Test 1. You are creating a view named WeeklySales. This view is used to create a sales report that is presented to management at the beginning of each week. You want to ensure that the underlying tables on which this view is based are not accidentally modified causing the report to break. What is the easiest way to implement this? A. Use a CREATE VIEW WITH CHECK constraint to create a view. B. Use a CREATE VIEW WITH SCHEMABINDING statement to create the view. C. Do nothing. When a view is based on a table, the underlying table cannot be modified until the view is dropped. D. Use a DDL trigger to roll back any statement that attempts to modify the table that the view depends on. 2. You have a view named YearlySales that lists all sales for the year. The reporting application your organization uses allows you to query the YearlySales view by CustomerName or by OrderDate. You receive unfavorable feedback from users that report generation is painfully slow. What is the best way to optimize report performance? A. Create indexes on CustomerName and OrderDate columns. B. Create a UNIQUE constraint on CustomerName and OrderDate columns. C. Create a DEFAULT constraint on CustomerName and OrderDate columns. D. Create a full-text index on CustomerName and OrderDate columns. 3. You have a very large table containing documents stored as a column of varbi- nary data type. The table is named Documents and is not referenced by FOREIGN KEY constraints. What is the most efficient way of removing all records from this table, while leaving the table ready for inserting new records (select all that apply)? A. TRUNCATE TABLE Documents B. DELETE Documents C. DELETE FROM Documents D. DROP TABLE Documents ImplementingObjects•Chapter14 651 E. Set the Database Recovery Model to Full F. Set the Database Recovery Model to Bulk-Logged 4. You have a table named Products, which contains the ProductID, ProductName, Model, and Color columns. The ProductID is marked as IDENTITY. You wish to ensure that there are never two products with the same combination of name, model, and color. What is the easiest way to achieve this? A. Create a PRIMARY KEY constraint on the ProductName, Model, and Color columns. B. Create a DEFAULT constraint on the ProductName, Model, and Color columns. C. Create a UNIQUE constraint on the ProductName, Model, and Color columns. D. Create a trigger FOR INSERT that checks that there is not already a combination of name, model, and color in the table. 5. You have two tables: PrizeDraw and Employees. Both tables have a PersonName column. You must ensure that employees cannot enter the prize draw. A record with the name that already exists in the Employees table cannot be inserted into the PrizeDraw table. What is the best way to achieve this? A. Create a CHECK constraint on the PrizeDraw table. B. Create a CHECK constraint on the Employees table. C. Create a trigger on the Employees table. D. Create a trigger on the PrizeDraw table. 6. You are tasked with creating a Reseller table, with the Commission column containing the commission percent. When a new reseller is added, the default commission level is 30%. What is the easiest way to implement this rule? A. Create a FOR INSERT trigger on the Reseller table. B. Create an INSTEAD OF INSERT trigger on the Reseller table. C. Create a DEFAULT constraint on the Commission column. D. Create a CHECK constraint on the Commission column. 7. You have a table named EmployeePhoto. This table is not referenced by any FOREIGN KEY constraints. What is the most efficient way of deleting the EmployeePhoto table entirely, including data and structure? . using a WHERE clause. Q: Does the TRUNCATE TABLE statement remove the table from the database? A: No, the TRUNCATE TABLE statement removes all data from the table, but the table structure and its. Chapter14•ImplementingObjects Exam Objectives Frequently Asked Questions Q: When should I use the TRUNCATE TABLE statement rather than the DELETE statement? A: You should use the TRUNCATE TABLE statement. empty the table. A performance advantage is achieved with the TRUNCATE TABLE statement when the database recovery model is set to Bulk-Logged. The DELETE statement allows you to restrict the