This cookbook contains a wealth of solutions to problems that SQL programmers face all the time. Recipes inside range from how to perform simple tasks, like importing external data, to ways of handling issues that are more complicated, like set algebra. Each recipe includes a discussion that explains the logic and concepts underlying the solution. The book covers audit logging, hierarchies, importing data, sets, statistics, temporal data, and data structures.

SQL in general, and Transact-SQL in particular, is so deceptively simple that few people take the time to master it. Many of the SQL books currently on the market contribute to this state of affairs by not taking their readers very far beyond the simple SELECT column_name FROM table_name WHERE conditions pattern of writing SQL SELECT statements. In this book, we hope to take you far beyond the level of writing simple SQL statements and open your eyes to the full power of SQL and Transact-SQL.

For many years, SQL has been the domain of large corporations using high-end databases such as Oracle, Sybase, Informix, and DB2. Microsoft's entry into the market with Microsoft SQL Server brought Transact-SQL to the workgroup server and desktop level, thereby broadening the market for Transact-SQL programmers. More systems are being programmed in SQL and Transact-SQL, and if you're a Transact-SQL programmer, you'll want to leverage Transact-SQL to its fullest. Getting the most out of SQL and Transact-SQL requires a particular way of thinking. SQL is a set-oriented language in which you focus on end results, not on the procedures used to accomplish those results. If you're used to thinking procedurally, this shift to set-oriented thinking can be jarring. The SQL solutions to problems that are easily described procedurally are often not obvious. So why bother with the set-oriented SQL solutions? It's true that Transact-SQL provides you with procedural capabilities, but use of those capabilities is no excuse for not utilizing the core SQL language to its fullest. SQL solutions are often magnitudes more efficient than the corresponding procedural solutions.