Mysql your visual blueprint for creating open source databases- P7 doc

SELECT columns or expressions FROM table or tables WHERE conditions GROUP BY columns ORDER BY columns LIMIT number; BASIC SELECT SYNTAX The LIMITclause allows you to limit the number o

ˇ Type UPDATE mail SET

address2=address, and press

Enter

■You are prompted for the

next line

Á Type city2=city, state2=state, postal2=postal;

and press Enter

■This completes the UPDATE query All of the

‡ Type SELECT name, address, address2 FROM mail;

and press Enter

■The values of the columns you specified are displayed for all rows Verify that the address values were copied

modified by any UPDATEquery If you want to update a row and preserve the current value of the timestamp column, you must explicitly set the column to its original value.

For example, if you were to add a second address to the address table and perform a similar update, you may want to avoid updating the timestamp

in the updatetime column The following query accomplishes this:

Example:

UPDATE address SET address2=address, city2=city, state2=state,

updatetime=updatetime;

While setting a column to its own value normally has no effect, in a timestamp column this prevents the MySQL server from automatically updating the field You can also set the timestamp column explicitly to a different value For example, the following UPDATEquery sets all rows to

a specified updatetime value:

Example:

UPDATE address SET address2=address, city2=city, state2=state,

updatetime="20030101120000";

Note: This example uses the testdb

database and creates a new table

⁄ From the MySQL monitor,

type USE testdb; and press

Enter

■The database is now

selected

¤ Type CREATE TABLE exams (name VARCHAR(80), and press Enter

‹ Type numtests INT, totalscore INT, avgscore TINYINT); and press Enter

■The table is created.

› Type INSERT INTO exams (name, numtests, totalscore)

and press Enter

■You are prompted for the next line

ˇ Type VALUES ("Sam", 5, 350), and press Enter

Á Type ("Ted",3, 220),("Sue",6, 510);

and press Enter

■This completes the INSERT query and adds three sample rows to the table

Often, you will find it useful to update a column's

value based on its existing value The simplest

example of this is to increment a numeric column's

value This is easy to do in an UPDATEquery by referring to

the column's current value.

For example, suppose you created a table to store exam

statistics for students The following CREATE TABLEquery

creates this simple table:

CREATE TABLE exams (name VARCHAR(80),

numtests INT, totalscore INT, avgscore

TINYINT);

This creates a table called exams with four columns: name

for the student name, numtests for the number of tests the

student has taken, totalscore for the total of all test scores,

and avgscore for an average.

When a new test is administered to students, you may want

to increment the numtests column for all of the rows in the database You can use a simple UPDATEquery to

accomplish this:

UPDATE exams SET numtests = numtests + 1;

This query adds one to the current value of the numtests column for each row and stores the resulting value in that row's numtests column, replacing the original value The net effect is to increment the numtests column for every student.

As with other UPDATEqueries, you could optionally add a

WHEREclause Specifying a WHEREclause may be useful to increment the number of tests for only a single student or group of students You can use a wide variety of arithmetic operations on MySQL column values; these are listed in Chapter 7.

INCREMENT A COLUMN VALUE

108

INCREMENT A COLUMN VALUE

‡ Type UPDATE exams and

press Enter

■You are prompted for the

next line

° Type SET numtests = numtests + 1; and press Enter

■This completes the UPDATE query The column is incremented in all rows

· Type SELECT * FROM exams; and press Enter

■The contents of the table are displayed The number of tests has been incremented for each row

columns For example, you could use an UPDATEquery to automatically set the avgscore column for each student to be an average calculated by dividing totalscore by numtests.

Example:

UPDATE exams SET avgscore = totalscore / numtests;

This example uses the /(division) operator to calculate the average Because there

is no WHEREclause, this operation will be performed on all rows of the table.

Because UPDATEcan modify multiple columns at once, you could combine this example with the previous example to increment the number of tests and calculate the average at the same time.

Example:

UPDATE exams SET numtests = numtests + 1, avgscore = totalscore / numtests;

Because MySQL processes the UPDATEquery from left to right, the numtests column will be incremented for each row first, after which the new value will be used in the calculation of the average.

SELECTis one of the most powerful MySQL query

commands, and one you will use frequently A SELECT

query returns one or more rows from one or more

tables You can use SELECTfrom the MySQL monitor to

display data or from an application to retrieve data.

USING SELECT QUERIES

110

The SELECTquery has a specific

syntax The various clauses and

keywords are optional, but must be

specified in this order Many of the

clauses require that you specify a

table with FROM.

SELECT columns or expressions

FROM table or tables WHERE

conditions

GROUP BY columns ORDER BY

columns

LIMIT number;

BASIC SELECT SYNTAX

The LIMITclause allows you to limit the

number of rows the SELECTquery can return.

If you specify a single number with LIMIT,

only that number of rows will be returned This

clause can be combined with a WHEREclause

to display a limited number of rows that match

the condition.

If you specify two numbers in the LIMIT

clause, the first is the number of the first result

row to be returned Rows are numbered

starting with zero The second number is the

limit You can use this to display pages of data

from a query.

Example:

SELECT * FROM address LIMIT 10;

THE LIMIT CLAUSE

If you specify more than one table name in the FROM

clause, the SELECTquery will return data from multiple tables This is known as a JOINquery, and requires a

WHEREclause to match a column from each table with corresponding items in other tables.

You can also use the JOINkeyword to combine tables When you use INNER JOINor simply specify multiple tables, only rows that match between tables are displayed When you use LEFT JOIN, all of the rows in the left table are displayed If no corresponding values exist for the right table,NULLvalues are returned.

Example:

SELECT * FROM address, mail WHERE address.name = mail.name;

USING JOIN TO COMBINE TABLES

Specify Columns

You can specify one or more column names in the SELECTquery The columns you specify will be displayed for each row returned by the query You can use commas to separate multiple columns or use *to return all columns.

Example:

SELECT name, city FROM address;

Specify Tables

You use the FROMkeyword in SELECTto specify one or more tables from which to retrieve rows In most cases, a single table name is used.

The ORDER BYclause specifies one or more columns by which to sort the results of the SELECT

query You can specify a single column name or multiple columns separated by commas You can optionally specify the keyword ASC(ascending) or

DESC(descending) for each column An ascending sort is the default.

Example:

SELECT * FROM address ORDER BY name;

THE ORDER BY CLAUSE

The GROUP BYclause is similar to ORDER

BY, but all of the rows for each value of the

group column are combined into a single

row You can use functions such as COUNT

to perform calculations on the combined

data.

Example:

SELECT state, COUNT(*) FROM address

GROUP BY state;

THE GROUP BY CLAUSE

Compare Numeric Values

MySQL includes a number of comparison operators you

can use with numeric values:

OPERATOR DESCRIPTION

>= Is greater than or equal to

<= Is less than or equal to

<>or != Is not equal to

Work with NULL Values

The following comparison operators work with NULL

values in columns:

OPERATOR DESCRIPTION

IS NOT NULL Is not the NULL value

<=> Is equal to (allows NULL values)

Compare Text Strings

You can compare text values using the standard equal, greater-than, and less-than operators Additionally, you can use LIKEor NOT LIKEto compare text strings These operators allow the wildcard values %for any characters or no characters, and _for one character.

Combine Conditions

You can use the ANDkeyword to combine two conditions in a WHEREclause Only the rows that match both conditions will be returned The ORkeyword also combines conditions In this case, any row that matches one condition or the other is returned The NOT

keyword negates one or more conditions.

You can use AND,OR, and NOTto combine any number

of conditions for a WHEREclause You can use parentheses to indicate the conditions that should be evaluated first.

Example:

SELECT * FROM address WHERE (state="CA" OR state="AZ") AND name LIKE "%Smith%";

THE WHERE CLAUSE

You can use the WHEREclause to select only the rows

that match a condition You can use any of MySQL's

available functions and comparison operators to form a

WHEREcondition.

Example:

SELECT * FROM address WHERE name LIKE "%Smith%";

Note: This example uses the testdb

database and the mail table See

Chapter 5 or the CD-ROM for

instructions to create this table

⁄ From the MySQL monitor, type USE testdb; and press Enter

■You are prompted for the next line

¤ Type SELECT * FROM mail;

and press Enter

■All columns and rows of the table are displayed

The SELECTquery is one of the most powerful options

available in MySQL Using SELECT, you can display

data from a table or retrieve data into an application.

A basic SELECTquery specifies the fields to display

followed by the FROMkeyword and the table to draw them

from:

SELECT name, address FROM mail;

This example displays all of the rows of the mail table For

each row, the name and address columns are displayed The

columns for each row are displayed in the order you

specified in the SELECTstatement You can also use a

wildcard character (*) to select all of the table's columns:

SELECT * FROM mail;

This example displays all of the rows of the table Each

row includes the value of all columns The columns are

displayed in the order they were defined when the table

was created.

You can also use SELECTwithout the FROMkeyword to test MySQL functions and expressions When you do not specify a table to select data from, MySQL will evaluate the expression in the SELECTstatement and display the result For example, this query displays the sum of several numbers:

SELECT 3 + 12 + 33;

When you include the FROMkeyword to specify a table, you can also combine the table's fields into functions and expressions For example, this SELECTquery displays the name and address of each row in the mail table, converting all text to uppercase:

SELECT UPPER(name), UPPER(address) FROM mail;

DISPLAY DATA WITH SELECT

112

DISPLAY DATA WITH SELECT

‹ Type SELECT name FROM

mail; and press Enter

■Only the name column is displayed

› Type SELECT city, address, name FROM mail; and press Enter

■The city, address, and name columns are displayed

in order

Sometimes, when you perform a function or calculation on one

or more fields of a table in a SELECTquery, you want to view the result as well as the original fields You can do this by assigning a

name, or alias, to the calculated value For example, the following

query displays the name field followed by its uppercase equivalent:

Example:

SELECT name, UPPER(name) AS name2 FROM mail;

The ASkeyword is used to indicate an alias name for the calculated value This name is displayed in the results when you use a query from the MySQL monitor When you are using MySQL with a language such as PHP or Perl, the alias is available to your program

as well as the columns you name directly in the SELECTquery.

As another example, the CONCATfunction in MySQL combines multiple text values into a single string This example returns the name and an addr alias that combines the address, city, and state columns.

Example:

SELECT name, CONCAT(address, "/", city, "/", state) AS addr FROM mail;

Note: This example uses the mail

table in the testdb database See

Chapter 5 or the CD-ROM if you

need to create this table

⁄ From the MySQL monitor,

type USE testdb; and press

Enter

■The database is now selected

¤ Type SELECT * FROM mail;

and press Enter

■All rows of the table are displayed

‹ Type SELECT * FROM mail

and press Enter

■You are prompted for the next line of the query

› Type WHERE name = "John Smith"; and press Enter

■The row that matches the WHERE clause is displayed

By default, when you use a SELECTquery, all of the

rows of the table are returned You can add a WHERE

clause to your SELECTqueries to select one or more

specific rows You can select rows by performing

comparisons with one or more of the table's fields The

following is a simple example of a SELECTquery with a

WHEREclause:

SELECT * FROM mail

WHERE name = "John Smith";

This example looks for a specific value within the name

field of the mail table and displays the rows that match.

Because the name field is the primary key in this case, only

one row will be displayed.

MySQL uses case-insensitive matching with text fields.

Aside from differences in case, the =operator will only

match if the name is exactly as specified You can also use

other operators, such as less than (<) and greater than (>).

Numbers are compared numerically, and text values are

compared alphabetically If the values you are comparing are different types, MySQL converts them to a compatible type if possible.

With text fields, you can use the LIKEkeyword to find partial matches.LIKEallows you to use wildcard characters The first wildcard,_, matches a single character The following example finds "John Smith", "John Smitt", or a name with any other final character:

SELECT * FROM mail WHERE name = "John Smit_";

The other wildcard,%, can represent any number of characters or no characters You can use this at the beginning and ending of a word to find the word anywhere

in the column The example below will find the name "John Smith" as well as any other name containing "Smith":

SELECT * FROM mail WHERE name LIKE "%Smith%";

USING THE WHERE CLAUSE

114

USING THE WHERE CLAUSE

ˇ Type SELECT * FROM mail

and press Enter

■You are prompted for the

next line

Á Type WHERE name LIKE

"%Smith%"; and press Enter

■The rows that match the WHERE clause are displayed

‡ Type SELECT name, address FROM mail and press Enter

■You are prompted for the next line

° Type WHERE postal >

10000; and press Enter

■The rows that match the WHERE clause are displayed Note: The fields you use in the

WHERE clause do not have to be

returned by the SELECT query.

MySQL supports a variety of operators for comparing numeric or text values The table below lists each operator and its purpose.

OPERATOR DESCRIPTION

>= Is greater than or equal to

<>or != Is not equal to

IS NOT NULL Is not the NULL value

<=> Is equal to (allows NULL values)

NOT LIKE Text strings do not match

Note: This example uses the mail

table in the testdb database See

Chapter 5 or the CD-ROM if you

need to create this table

⁄ From the MySQL monitor, type USE testdb; and press Enter

■The database is now selected

¤ Type SELECT * FROM mail WHERE and press Enter

■You are prompted for the next line

‹ Type name LIKE

"%Smith%" OR name LIKE

"%Jones%"; and press Enter

■This completes the query Rows that match either condition are displayed

Often, in a large database, specifying a single

condition would still return a huge number of

rows You can combine several conditions in a

WHEREclause to make your search more specific The

logical operators AND,OR, and NOTare used to combine

conditions.

The ORoperator allows you to make a search more general.

The following example displays records from the mail table

that contain the names Smith or West:

SELECT * FROM mail WHERE

name LIKE "%Smith%" OR name LIKE "%West%";

The ANDoperator allows you to make a search more

specific The following example displays only the records

with a name column containing "Smith" and a value of

"CA" in the state column:

SELECT * FROM mail WHERE

name LIKE "%Smith%" AND state = "CA";

If you have not used ANDand ORwith computer languages before, they may be confusing Remember that using OR

will allow more rows to match the query, and using ANDwill allow less rows to match.

Finally, the NOToperator inverts a condition If you use NOT LIKEor !=(not equal), rows that do not match the condition are returned You can use NOTto make any existing condition into its opposite.

You can combine any number of conditions with AND,OR, and NOTto create complex conditions When you use AND

and ORtogether, often the meaning is ambiguous You can enclose conditions in parentheses to ensure that they are considered first, before combining the result with the other conditions.

SPECIFY MULTIPLE WHERE CONDITIONS

116

SPECIFY MULTIPLE WHERE CONDITIONS