The Language of SQL- P39 doc

The previous example would need to be broken down into two statements, such as: INSERT INTO Customers FirstName, LastName, State VALUES 'Virginia', 'Jones', 'OH'; INSERT INTO Customers F

D A T A B A S E D I F F E R E N C E S : O r a c l e

Oracle doesn’t permit multiple rows to be specified after the VALUES keyword The previous example would need to be broken down into two statements, such as:

INSERT INTO Customers

(FirstName, LastName, State)

VALUES ('Virginia', 'Jones', 'OH');

INSERT INTO Customers

(FirstName, LastName, State)

VALUES ('Clark', 'Woodland', 'CA');

Also note that the order of values after theVALUESkeyword corresponds to the

order of columns listed in the columnlist after the INSERT TO The order in which the columns are listed does not have to be the same as it is in the database

In other words, the above insert just as easily could have been accomplished with this statement:

INSERT INTO Customers

(State, LastName, FirstName)

VALUES

('OH', 'Jones', 'Virginia'),

('CA', 'Woodland', 'Clark')

In the aboveINSERT, we listed the State column first instead of last Again, the order in which columns are listed doesn’t matter

To sum up, the general format for theINSERT INTOstatement is:

INSERT INTO table

(columnlist)

VALUES

(RowValues1),

(RowValues2)

(repeat any number of times)

The columns in columnlist need to correspond to the columns in RowValues Also, if all the columns in columnlist are listed in the same order as they

physi-cally exist in the database, and if there are no auto-increment columns in the Chapter 17 ■ Modifying Data

176

table, then theINSERT INTOstatement can be executed without specifying the

columnlist However, this practice is strongly discouraged since it is prone to

error

What happens when not all columns are specified in anINSERT? Simple Those

columns that are not specified are given NULL values For example, let’s say we

want to insert one additional row into the Customers table, for a customer

named Tom Monroe However, we don’t know Tom’s state Here’s theINSERT:

INSERT INTO Customers

(FirstName, LastName)

VALUES

('Tom', 'Monroe')

Afterwards, his row in the table appears as:

CustomerID FirstName LastName State

Since we didn’t specify a value for the State column for this new row, it is given a

NULL value

There are two variations of the INSERT INTO statement The second format

applies to situations where you insert data that is obtained from a SELECT

statement, which means that instead of listing values after a VALUESkeyword,

you substitute aSELECTstatement that obtains similar values

Let’s say that we have another table named CustomerTransactions, which holds

data that we would like to insert into the Customers table The

Customer-Transactions table might look like:

CustomerID State Name1 Name2

If we wanted to add all customers in the state of Rhode Island from the CustomerTransactions table to the Customers table, the following would accomplish that objective:

INSERT INTO Customers

(FirstName, LastName, State)

SELECT

Name1,

Name2,

State

FROM CustomerTransactions

WHERE State = 'RI'

After thisINSERT, the Customers table contains:

CustomerID FirstName LastName State

The aboveINSERT simply substituted a SELECT statement for the previously seenVALUES clause As would be expected, Michael Blake didn’t get added to the Customers table, since he is not in Rhode Island Also notice that the column names in the Customers and CustomerTransactions tables are not identical The column names don’t matter as long as the columns are listed in the correct cor-responding order

Deleting Data

Deleting data is much simpler than adding it TheDELETEstatement is used to handle deletes When aDELETEis executed, it deletes entire rows, not individual columns in a row The general format is:

DELETE

FROM table

WHERE condition

Chapter 17 ■ Modifying Data

178

Here’s a simple example Let’s say we want to delete rows from the previously

mentioned Customers table if the customer is in Rhode Island The statement to

accomplish this is:

DELETE

FROM Customers

WHERE State = 'RI'

That’s all there is to it If you wanted to test the results of the above DELETE

before executing it, you would simply substitute a SELECT for DELETE, as

follows:

SELECT

COUNT (*)

FROM Customers

WHERE State = 'RI'

This would provide a count of the rows to be deleted, which supplies some level

of validation for the delete

There is one other option for deleting data that is worth mentioning If you want

to delete all the data in a single table, you can employ a TRUNCATE TABLE

statement to delete everything The advantage of theTRUNCATE TABLEover the

DELETEstatement is that it is much faster Unlike theDELETE, theTRUNCATE

TABLEdoesn’t log the results of the transaction We haven’t talked about

data-base log processes, but this is a function that most datadata-bases provide that allows

database administrators to recover databases in the event of system crashes and

other similar problems

If you want to delete all rows in the Customers table, you can issue this

statement:

TRUNCATE TABLE Customers

This has the same result as this statement:

DELETE

FROM Customers

One other slight difference between DELETE and TRUNCATE TABLE is that

TRUNCATE TABLE resets the current values used for auto-increment columns

TheDELETEdoesn’t affect those values

Updating Data

The procedure for updating data involves specifying which columns are to be updated, as well as logic for selecting rows The general format for anUPDATE

statement is:

UPDATE table

SET Column1 = Expression1,

Column2 = Expression2

(repeat any number of times)

WHERE condition

This statement is similar to the basic SELECT, except that theSET keyword is used to assign new values to specified columns TheWHERE condition specifies which rows are to be updated, but theUPDATE statement can update multiple columns at the same time If more than one column is being updated, theSET

keyword is only listed once, but a comma must separate all update expressions Starting with a simple example, let’s say we want to change the name of customer William Smith to Bill Smythe His row in the Customers table currently looks like:

CustomerID FirstName LastName State

TheUPDATEstatement to accomplish the change is:

UPDATE Customers

SET FirstName = 'Bill',

LastName = 'Smythe'

WHERE CustomerID = 1

After executing this statement, this row in the Customers table appears as:

CustomerID FirstName LastName State

Notice that the value of the State column is unchanged since that column was not included in the UPDATE statement Also note that the WHERE clause is Chapter 17 ■ Modifying Data

180