Summary

In this chapter, we've looked at manipulating base relations using various relational operators and seen some examples implemented in the SQL language. We've also had yet another look at the issue of Nulls and three-valued logic.

Of the standard relational operators, the restriction and projection operators select subsets of a single recordset. The join, union, intersection, difference, and Cartesian product operators control the way two recordsets are combined. All of these operators, with the exception of difference, can be implemented using a SQL SELECT statement. Difference can sometimes be implemented using SELECT and sometimes requires other techniques that are outside the scope of this book.

We've also looked at a few special operators. The summarize and extend operators perform calculations on the data. The rename operator controls the column headings that are displayed for the view. TRANSFORM, ROLLUP, and CUBE are special extensions to the SQL language implemented by Microsoft, each of which also provides a special way of summarizing and viewing data.

With this overview of relational algebra, we've completed Part I. Relational database theory is complex, and of course, there are issues and subtleties that I haven't been able to cover in the scope of an introductory book. But you've now seen all the major components of the theory. In the next section, we'll turn to the other type of modeling employed with relational databases: dimensional modeling.