Mysql your visual blueprint for creating open source databases- P4 docx

■ The MySQL monitor prompts for another line. › Type url VARCHAR(255), and press Enter. ˇ Type description TEXT ); and press Enter. Note: If you want to add additional fields, include them before the closing parenthesis. ■ The table is now created. Á To verify that the command worked, type SHOW COLUMNS FROM links; and press Enter. ■ The columns you created for this table are displayed. If you are not sure what length to use for a CHAR or VARCHAR column, specify a value large enough to store a typical value. You can always change the length later. You can also change VARCHAR columns to TEXT columns without losing data. This is explained in Chapter 3. The MySQL server may automatically change the type of some of your columns if there is a conflict between them. In particular, if you use fixed-length CHAR columns and at least one variable-length VARCHAR or TEXT column, the entire row must be variable length, so your CHAR columns are converted to VARCHAR. In addition, the server automatically changes VARCHAR columns with a length of one to three characters to CHAR columns, because it is not practical to use variable length when the column is this small. While it is important to choose the correct column types when you create a table, after you have created it, you can work with VARCHAR and CHAR columns in exactly the same ways. TEXT columns also work in most of the same ways, except that you cannot use the entire body of a TEXT field as an index. 47 MANAGE DATABASES AND TABLES 2 516922 Ch02.F 9/26/02 11:32 AM Page 47 ⁄ From the MySQL monitor, type USE testdb; and press Enter. ■ The database is now selected. Note: If you did not create the testdb database in Chapter 1, you need to create it first. ¤ Type CREATE TABLE music ( and press Enter. ■ This starts the table definition. ■ The MySQL monitor prompts for the next line. ‹ Type title VARCHAR(100), and press Enter. S ets and enumerations are different from normal text columns in that they are limited to a specific list of values. You can use these columns whenever you need to assign categories to the items in a table. For example, suppose you were using a database table to catalog music recordings. You could use a SET column to store the category of music that each recording contains: category SET("rock", "pop", "blues", "country", "dance") The one potential disadvantage of a SET column is that each item can have only one value. Anything that fits more than one category could only be listed in one. This is where ENUM columns are useful. With an ENUM column, each item can be assigned any number of values from the list you specify when you create the table. You can look at an ENUM column as a list of flags, each of which can be assigned or not assigned to an item. In the music catalog example, an ENUM column would be ideal to store the media the music is available on: media ENUM("CD","DVD","LP","Cassette") Although you could use separate columns to achieve the same effect, ENUM columns are convenient because they can be assigned values as a single unit. They are also stored on the server in a highly efficient way, using only one bit per item, and are thus especially useful when you are storing many rows of data in a table. Because ENUM and SET columns can have a large number of possible values, you may need to split the column specification into two or more lines when you create the table. USING SETS AND ENUMERATIONS MySQL 48 USING SETS AND ENUMERATIONS 516922 Ch02.F 9/26/02 11:32 AM Page 48 › Type category SET("rock", "pop", and press Enter. ˇ Type "blues", "country", "dance"), and press Enter. ■ These two lines define a SET column. Á Type media ENUM("CD","DVD", and press Enter. ‡ Type "LP","Cassette") ); and press Enter. ■ This completes the definition for the ENUM column and the CREATE TABLE command. The table is now created. The values for an ENUM column are actually stored as integers. A value of 1 represents the first possible value, 2 represents the second possible value, and so on. ENUM columns can have a maximum of 65,535 total values available. ENUM values use one byte of storage if there are less than 256 possible values and two bytes if there are 256 or more. The values for a SET column are stored using individual bits. This means that one byte, or eight bits, of storage is required for every eight members of the set. A SET column can have a maximum of 64 members, which requires eight bytes of storage. You can change the definition for a SET or ENUM column using the ALTER TABLE query in MySQL, described in Chapter 3. However, changing the definition does not change the values stored for existing data. If you add a value at the beginning of the list, the numeric values for the entire list will change. You can safely add values at the end of the list, but the best strategy is to determine the possible values in advance and include them when you first create the table. 49 MANAGE DATABASES AND TABLES 2 516922 Ch02.F 9/26/02 11:32 AM Page 49 ⁄ From the MySQL monitor, type USE testdb; and press Enter. ■ This selects the testdb database. ¤ Type CREATE TABLE MailList ( and press Enter. ■ The MySQL monitor prompts for the next line. ‹ Type name VARCHAR(80) PRIMARY KEY, and press Enter. ■ This defines the name field as the primary key. A long with the various columns the table will contain, you can specify a primary key and one or more indexed columns. In general, you will want to have at least a primary key for any table. If there is no primary key, there may be no way to uniquely identify a single record. This makes it impossible to edit or delete a single row without affecting other rows in the table. For the primary key, you should choose a column that will have a unique value for each row of the table. For example, for a table that stores a list of names and mailing addresses, the name field is usually a good primary key, assuming that the list is small enough that duplicate names is not a concern. You can define the primary key by including the keywords PRIMARY KEY with one of the column definitions when you create the table, or with a separate PRIMARY KEY definition that specifies a column for the key in parentheses. In this case, you can also use multiple fields as a primary key. The following CREATE TABLE query creates a MailList table with columns for the name, address, city, state, and postal code. The name column is defined as the primary key. CREATE TABLE MailList ( name VARCHAR(80) PRIMARY KEY, address VARCHAR(120), city VARCHAR(50), state VARCHAR(2), postal VARCHAR(5) ); CREATE AN INDEXED TABLE MySQL 50 CREATE AN INDEXED TABLE 516922 Ch02.F 9/26/02 11:32 AM Page 50 › Type address VARCHAR(120), and press Enter. ˇ Type city VARCHAR(50), and press Enter. Á Type state VARCHAR(2), and press Enter. ‡ Type postal VARCHAR(5) ); and press Enter. ■ This completes the query, and the table is created. ° Type SHOW COLUMNS FROM MailList; and press Enter. ■ The columns for the table you created are displayed. Along with the primary key, one or more additional indexes are often useful. You should only define an index on an additional column if you frequently need to search for values in that column. You can add an index with the INDEX keyword, and optionally specify the UNIQUE keyword to require a unique value for each row. The following example shows an expanded mailing list table with an index on the postal code field. In this case, the UNIQUE keyword is not used because multiple records can have the same code. Example: CREATE TABLE MailList2 ( name VARCHAR(80) PRIMARY KEY, address VARCHAR(120), city VARCHAR(50), state VARCHAR(2), postal VARCHAR(5), INDEX (postal) ); 51 MANAGE DATABASES AND TABLES 2 516922 Ch02.F 9/26/02 11:32 AM Page 51 ⁄ From the MySQL monitor, type USE testdb; and press Enter. ■ This selects the database. Note: This task uses the prices table created earlier in this chapter. ¤ Type DROP TABLE prices; and press Enter. ■ The table is deleted. ‹ Type SHOW TABLES; and press Enter. ■ Verify that the deleted table is not listed. I f you no longer have any use for a table, you can delete it using the DROP TABLE command. This command immediately deletes the table, including all data. You can specify more than one table to drop with DROP TABLE, separated by commas. The following command deletes the prices table: DROP TABLE prices; When you attempt to drop a table that does not exist, MySQL returns an error. You can optionally specify the IF EXISTS keywords to prevent the error. The following example deletes the test1 table only if it exists: DROP TABLE IF EXISTS test1; If you want to delete an entire database, you can use the DROP DATABASE command. This command deletes the entire database including all tables and data. As with DROP TABLE, you can use the IF EXISTS keyword to prevent an error if the database does not exist. The following command deletes a database called newdb: DROP DATABASE newdb; The DROP command does not prompt you for confirmation, whether you are deleting a table or an entire database. After you have issued the DROP command, there is no way to recover the data unless you have a backup copy. Be sure that you have a backup before you use the DROP command. In order to drop a database or table, the MySQL username you are using must have the correct privileges. Only the root user is allowed to delete databases by default. See Chapter 11 for more information about MySQL security. In this example you delete the prices table and the newdb database, which you created in the sections "Create a Database" and "Create a Simple Table," earlier in this chapter. Be sure you specify the names correctly to avoid deleting the wrong data. DELETE TABLES AND DATABASES MySQL 52 DELETE A TABLE 516922 Ch02.F 9/26/02 11:32 AM Page 52 The DROP TABLE command deletes the disk files that MySQL uses to store a table's data. There are typically three separate files for each table. The DROP DATABASE command deletes the files for each table in the database and additionally deletes the directory for the database. If non-MySQL files are also in this directory, it may prevent the directory from being entirely deleted. Because the DROP command is drastic, be sure you have a backup of all data before deleting a table or database. You will learn more about backing up MySQL databases in Chapter 8. If you are unsure whether to delete a table, you can use the SHOW COLUMNS command to display the columns of the table. Example: SHOW COLUMNS FROM tablename; You can also use a SELECT query to display the data in the table before deleting it. The following command displays all of the rows of a table. Example: SELECT * FROM tablename; 53 DELETE A DATABASE ⁄ From the MySQL monitor, type DROP DATABASE newdb; and press Enter. ■ The database is deleted. ¤ Type SHOW DATABASES; and press Enter. ■ The list of databases is displayed. Verify that the database has been deleted. MANAGE DATABASES AND TABLES 2 516922 Ch02.F 9/26/02 11:32 AM Page 53 W hen you create a table with the CREATE TABLE command in MySQL, you specify the column definitions and other options. If you later decide to change any aspect of the table's definition, you can do so using the ALTER TABLE command. This command allows you to change column names, column types, and other aspects of a table's definition. USING ALTER TABLE QUERIES MySQL 54 Basic ALTER TABLE Syntax You can make any number of changes with a single ALTER TABLE command. You can use a variety of commands within ALTER TABLE to add columns, remove columns, and make other changes. If you use multiple keywords to make changes, separate them with commas. Example: ALTER TABLE address ADD COLUMN lastvisit DATE, DROP COLUMN postal; ADD COLUMN Use the ADD COLUMN command to add a column to the table. Specify the new column name, the column type, and any attributes. You can use the same syntax as you use when creating a table, as described in Chapter 2. You can optionally specify the keyword FIRST after ADD COLUMN to add the column at the beginning of the table, or the AFTER keyword and a column name to add it after an existing column. If you do not specify either of these, the column is added at the end of the table. DROP COLUMN The DROP COLUMN command enables you to delete an existing table column. Use this command with caution, because it deletes all of the data stored in that column in existing table rows without asking for confirmation. To use DROP COLUMN, simply specify the column name. CHANGE The CHANGE command changes the definition of an existing column. To use this command, specify the old column name followed by the new name, the column type, and any options such as DEFAULT or NULL. Specify the old name twice if you are not renaming the column. The MySQL server attempts to convert any existing data in the column to the new column type. However, in some cases, such as when you change the length of a VARCHAR column to a smaller amount, the data in the column will be truncated and cannot be restored. Example: ALTER TABLE address CHANGE name name CHAR(120) NOT NULL; MODIFY The MODIFY command changes the definition for a column without changing its name. To use this command, specify the column name, the new column type, and any options. As with CHANGE, the data is converted wherever possible to the new format. Example: ALTER TABLE address MODIFY name CHAR(120) NOT NULL; 516922 Ch03.F 9/26/02 11:32 AM Page 54 MODIFY TABLES 3 55 ADD INDEX Use the ADD INDEX command to add an index to the table for an existing column. To use this command, specify an optional index name followed by the column or columns to index in parentheses. If you use ADD UNIQUE instead of ADD INDEX, a unique index is created. Before adding a unique index, be sure the existing rows of the table have unique values for the column or columns you plan to index. Example: ALTER TABLE address ADD INDEX postindex (postal); DROP INDEX The DROP INDEX command deletes an existing index. To use this command, specify the index name. The index name is the column name by default, or the name you specified when creating the index. Example: ALTER TABLE address DROP INDEX postindex; ADD PRIMARY KEY The ADD PRIMARY KEY command adds a primary key. This can only be used if the table does not have an existing primary key. To use this command, specify one or more columns to act as the primary key. Each existing row of the table must have a unique value for the column or columns specified. The column you specify must also have the NOT NULL attribute. You can alter the column to add this attribute if necessary, as described earlier in this section. Example: ALTER TABLE address ADD PRIMARY KEY (name, address); DROP PRIMARY KEY The DROP PRIMARY KEY command removes an existing primary key. This only removes the indexing information, not the column or columns that act as the key. This command does not require any parameters. Example: ALTER TABLE address DROP PRIMARY KEY; RENAME The RENAME command renames an existing table. To use this command, specify the existing table name and the new name. You can use RENAME TO as a synonym for RENAME. Example: ALTER TABLE temp RENAME TO temp2; Table Options You can change table options using an ALTER TABLE command. These options include TYPE, the table type, COMMENT, the optional comment field, and other options. The complete list of options is in Chapter 2. To change options, specify each option followed by an equal sign and the new value. Example: ALTER TABLE temp TYPE=Heap, COMMENT="I changed the table type."; ORDER BY The ORDER BY command sorts the existing data in a table by the value of a specified column. While you can retrieve the rows of the table in any order, this command is useful to make the order permanent when the table is rarely changed and usually retrieved in a particular order. Example: ALTER TABLE address ORDER BY name; 516922 Ch03.F 9/26/02 11:32 AM Page 55 Note: This example uses the testdb database and the address table. If you have not created these, follow the instructions in Chapter 1 or on the CD-ROM. ⁄ From the MySQL monitor, type USE testdb; and press Enter. ■ The database is now selected. ¤ Type SHOW COLUMNS FROM address; and press Enter. ■ The current list of columns is displayed. I f you have created a few tables in MySQL, you learned that there are many decisions to be made — the columns to include, their types, and other aspects of the table. Fortunately, you can change most of these after the table is created with the ALTER TABLE command. You can use ALTER TABLE ADD COLUMN to add a column to an existing table. The basic query syntax specifies the table, the name of the new column, and its column type: ALTER TABLE address ADD COLUMN country VARCHAR(10); You can optionally specify where the new column should be added. Either specify the keyword FIRST to insert the new column as the first column in the table's definition, or specify the AFTER keyword followed by the name of an existing column. The following example adds a country column to the address table after the existing state column: ALTER TABLE address ADD COLUMN country VARCHAR(10) AFTER state; When you add a column, you can specify the same attributes that you use when creating a table, as described in Chapter 2. You can specify the column's type and display width, select a default value for the column, and specify whether null values are allowed. You cannot specify values for the data in the table's rows for the new column. The default value of the column will be used for all rows of the table until you change the data. Be sure the new column's name does not conflict with other columns in the table. The column must also be compatible with the existing columns: In particular, if there are existing variable-length text columns, such as VARCHAR or TEXT, you cannot add a fixed-length CHAR column. ADD A COLUMN TO A TABLE MySQL 56 ADD A COLUMN TO A TABLE 516922 Ch03.F 9/26/02 11:32 AM Page 56 [...]... Ch03.F 9/26/02 11:32 AM Page 58 MySQL ADD AN INDEX A s you work with a table in MySQL, you may find that an index on an additional field would improve performance You can use the ALTER TABLE command to add an index to an existing table To add an index, you specify the ADD INDEX keywords, an optional name for the index, and the existing column or columns to be indexed For example, the following ALTER... INDEX (custnum); Keep in mind that adding an index does not always improve performance In fact, extra indexes on fields that are not frequently used for searching can slow down access to the table See Chapter 10 for information on determining whether an added index will be beneficial You can optionally specify more than one column for the index, separated by commas In this case, the index will be ADD AN... unique for each row In this case, you can use an auto-increment column as a key When you specify the AUTO_INCREMENT attribute for a column, each row you add is automatically assigned a new unique value for that column ALTER TABLE links ADD COLUMN num INTEGER UNSIGNED NOT NULL AUTO_INCREMENT PRIMARY KEY; Auto-increment columns are ideal for use as primary keys or unique indexes, and in fact MySQL requires... them with a new name The rules for the new column name are the If you specify a different column type or attributes with the CHANGE keyword, the column's definition will be modified This is explained in the next section RENAME A COLUMN Note: This example uses the testdb database and the links table See Chapter 1 or the CD-ROM for information on creating them ⁄ From the MySQL monitor, ¤ Type DESCRIBE... data in the table when adding a unique index or primary key Because these indexes require a unique value for each row, MySQL will return an error if there are duplicate values for the column you are indexing To successfully add the index, you will need to ensure that the rows have unique values for the column that you are making into a key If you are adding a column that will act as the new primary... table, because existing rows need unique values for the primary key ADD A PRIMARY KEY Note: This example uses the testdb database and the address table To create them, see Chapter 1 or the CD-ROM s The database is selected ⁄ From the MySQL monitor, s The MySQL monitor type USE testdb; and press Enter 60 ¤ Type ALTER TABLE address and press Enter prompts for the next line ‹ Type ADD PRIMARY KEY (name);... using the ALTER TABLE command in MySQL For example, the following command adds a timestamp column called updatetime with a display width of 14 characters to the address table: ALTER TABLE address ADD COLUMN updatetime TIMESTAMP(14); When you add a new timestamp column to an existing table, it is filled with the default value for each existing row Because the default value for the first timestamp column... rows will be non-unique You must assign unique values before adding a unique index ADD A TIMESTAMP COLUMN Note: This example uses the testdb database and the address table See Chapter 1 or the CD-ROM to create them 62 ⁄ From the MySQL monitor, ¤ Type ALTER TABLE address s The MySQL monitor type USE testdb; and press Enter and press Enter prompts for the next line s The database is now selected 516922... use any of MySQL' s integer types for an autoincrement column However, be sure to use a type that can store a wide enough range of values to account for the maximum amount of rows you will be adding to the table You can change the column type later if you need a larger range, as explained in the section "Change a Column Type," later in this chapter hile some tables have an obvious choice for a primary... uses the testdb database and the address table The instructions for creating these are in Chapter 1 and on the CD-ROM ⁄ Type USE testdb; and press Enter 58 selected ‹ Type stateindex (state); and press Enter ¤ Type ALTER TABLE address s The index is now created s The database is now ADD INDEX and press Enter s The MySQL monitor prompts for the next line › Type SHOW COLUMNS FROM address; and press Enter . row without affecting other rows in the table. For the primary key, you should choose a column that will have a unique value for each row of the table. For example, for a table that stores a list of names. three separate files for each table. The DROP DATABASE command deletes the files for each table in the database and additionally deletes the directory for the database. If non -MySQL files are also. does not always improve performance. In fact, extra indexes on fields that are not frequently used for searching can slow down access to the table. See Chapter 10 for information on determining whether