3.5. Creating and Destroying Tables
section discusses the operations SQL
can perform to manipulate data stored as rows within tables in the
database. However, there is a separate set of statements that covers
the manipulation of the tables (and other objects) within the
database themselves. These statements are known as
Data Definition Language commands, or
The operations that can be performed on tables are fairly basic,
since they are quite far-reaching in their consequences. The two
simplest operations available are:
Creating a new table
This is done via the
TABLE command, the
syntax of which varies depending on the database platform being used.
However, this statement generally specifies the name of the table to
be created and the definition of all the columns of the table (both
names and datatypes).
For example, the SQL we used to create the
megaliths table within our database was:
CREATE TABLE megaliths (
id INTEGER NOT NULL,
CREATE TABLE will create a brand-new table with the
given definition, which will be completely empty until you insert
rows into it.
Deleting, or dropping, an existing table
This action is as drastic as data modification can get. The actual
table structure within the database is completely removed, as are any
rows of data currently stored within that table. This operation
cannot usually be rolled back from. Once the fatal statement is
typed, the specified table has gone forever (unless you have made a
The syntax for dropping tables is fairly standard across databases
and is extremely straightforward. To
completely get rid of our megaliths table, we can
issue the SQL statement of:
Something so deadly
should have a far more complicated syntax!
DROP TABLE megaliths
There are other ways in which table definitions can be manipulated,
and also other database structures that can be created (such as views
and indexes). But these are beyond the scope of this book. You should
consult your database documentation for more information.
Copyright © 2001 O'Reilly & Associates. All rights reserved.