Download - Oracle Part3

7/29/2019 Oracle Part3

1/106

Chapter 4:

Introduction to PL/SQL

Guide to Oracle 10g


2/106

Guide to Oracle 10g 2

Lesson A Objectives

After completing this lesson, you should be able to:

Describe the fundamentals of the PL/SQL

programming language Write and execute PL/SQL programs in SQL*Plus

Execute PL/SQL data type conversion functions

Display output through PL/SQL programs

Manipulate character strings in PL/SQL programs

Debug PL/SQL programs


3/106


Introduction

DB users use GUI applications to enter data intoinput boxes, and then click buttons to executeprograms that perform DB functions.

We will use the following utilities:Form builder

Report builder

To use these utilities effectively an understanding ofPL/SQL is needed.

PL/SQL is a procedural programming language thatOracle utilities use to manipulate DB data.

Programming or scripting language conceptsunderstanding is necessary.


4/106


Fundamentals of PL/SQL

PL/SQL program combines SQL queries withprocedural commands (if/then, loops).

It is a full-featured programming language Executed using Oracle 10gutilities such as:SQL*Plus or

Forms Builder It is an interpreted language

Its commands are not a case sensitive. Inside single

quotes are case sensitive.. M != m Semicolon ends each command

It contains reserved words and built in functions.


5/106


PL/SQL Command Capitalization

Styles


6/106


PL/SQL Variables and Data Types

Variable names must follow the Oracle namingstandard

Strongly typed language:Explicitly declare each variable including data type

before using variable

Variable declaration syntax: var i abl e_name dat a_t ype_decl ar at i on;

Default value always NULL PL/SQL supports: scalar, composite, reference and

LOB.


7/106


Scalar Variables

Reference single value Data types correspond to Oracle 10gdatabase data

types

VARCHAR2CHAR

DATE

NUMBER


8/106


General Scalar Data Types


9/106


Composite Variables

Data Structure is a data object made up of multiple

individual data elements Composite variable is a data structure contains

multiple scalar variables

Types:

RECORD

TABLEVARRAY


10/106


Reference Variables

Reference variables directly reference specificdatabase column or row

Reference variables assume data type of associatedcolumn or row

%TYPE data declaration syntax: var i abl e_namet abl ename. f i el dname%TYPE;

%ROWTYPE data declaration syntax: var i abl e_name t abl ename%ROWTYPE;


11/106


Reference Variables

The (%TYPE) reference data type specifies avariable that references a single DB field.

current_f_last FACULTY.F_LAST%TYPE;

The current_f_last variable assumes a data type ofVARCHAR2(30), because this is the data type ofthe f_last column in the FACULTY table.


12/106


Reference Variables

The (%ROWTYPE) reference data type createscomposite variables that reference the entire data

record. Facul t y_r ow FACULTY%ROWTYPE;

The variable faculty_row references all of thecolumns in the FACULTY table, and each column

has the same data type as its associated DB column.


13/106


LOB Data Types

It declares variables that referencebinary orcharacterdata objects up to 4 GB.

LOB values in PL/SQL programs must be

manipulated using special package calledDBMS_LOB.


14/106


PL/SQL Program Blocks

Declaration section (DECLARE) Optional (variable declaration)

Execution section (BEGIN) Required (Program statements)

Exception section (EXCEPTION) Optional (Error handling statements)

Comment statements

Enclosed within /* and */ Two hyphens (--) for a single line comment

END;


15/106


PL/SQL Arithmetic Operators in

Describing Order of Precedence


16/106


Assignment Statements

Assigns value to variable Operator

:=

Syntax var i abl e_name : = val ue;

String literal must be enclosed in a singlequotation marks.


17/106


Assignment Statements You can use the := in the declaration section to

give initial values. s_id NUMBER := 100;

You can use DEFAULT instead: s_id NUMBER DEFAULT 100;

IfNULL value is used in an assignment statementwith arithmetic operation, the result is always

NULL.


18/106


Executing a PL/SQL Program in

SQL*Plus

Create and execute PL/SQL program blockswithin variety of Oracle 10gdevelopmentenvironments:

SQL *PlusForm Builder

Report Builder


19/106


Displaying PL/SQL Program Output in SQL*Plus

PL/SQL outputbuffer

Memory area on database server

Stores programs output values before they aredisplayed to user

Buffer size should be increased before output is

displayed: SET SERVEROUTPUT ON SI ZE buf f er _si ze

Default buffer size

2000 bytes

If the output is more than buffer_size, error apperas.


20/106


Displaying PL/SQL Program Output in SQL*Plus

(continued)

Display program output

DBMS_OUTPUT. PUT_LI NE( ' di spl ay_t ext' ) ;

Display maximum of 255 characters of text data.

More than 255 characters gives an error.

Example: s_name := Ahmed Mahmoud;

DBMS_OUTPUT. PUT_LI NE( s_name) ;


21/106


Writing a PL/SQL Program

Write PL/SQL program in Notepad or another texteditor

Copy and paste program commands into SQL*Plus

Press Enter after last program command Type front slash ( / ) (to instruct the interpreter to execute the code)

Then press Enter again


22/106


Writing a PL/SQL Program

(continued)

Good programming practicePlace DECLARE, BEGIN, and END commandsflush with left edge of text editor window

Indent commands within each section


23/106


PL/SQL Program Commands

Output: Todays data is

18-OCT-07PL/SQL procedure successfully completed

If no output appears use: SQL> set serveroutput on


24/106


PL/SQL Data Conversion Functions

Implicit data conversions Interpreter automatically converts value from one data

type to another(the command DBMS_OUTPUT.PUT_LINE(todays_date), theinterpreter automatically converted the todas_date DATE variable to a text string foroutput

). If PL/SQL interpreter unable to implicitly convert value

error occurs

Explicit data conversionsConvert variables to different data typesUsing built in data conversion functions


25/106


PL/SQL Data Conversion Functions of

PL/SQL


26/106


Manipulating Character Strings with

PL/SQL String

Character data value

Consists of one or more characters

Concatenating

Joining two separate strings Parse

Separate single string consisting of two data items

separated by commas or spaces Built in functions to concatenate and parse strings

in SQL*Plus are as follows:


27/106


Concatenating Character Strings

Double bar Operator

||

Syntax:

new_st r i ng : = st r i ng1 | | st r i ng2;

Exampl e: Ful l _name : = Qusai | | | | Abuei n

You must conver t NUMBER and DATE dat at ype var i abl es ( TO_CHAR) i nt o a st r i ngbef or e concat enat i ng i t .


28/106


Removing Blank Leading and Trailing

Spaces from Strings

LTRIM functionRemove blank leading spaces

st r i ng : = LTRI M( st r i ng_var i abl e_name) ;

RTRIM functionRemove blank trailing spaces

st r i ng : = RTRI M( st r i ng_var i abl e_name) ;


29/106


Finding the Length of Character

Strings

LENGTH function syntax

st r i ng_l engt h : = LENGTH( st r i ng_var i abl e_name) ;

Space i s i ncl uded wher ever i t appear s. DBMS_OUTPUT. PUT_LI NE( LENGTH( ' Today' ' s dat e i s ' ) ) ;

Gi ves 20.


30/106


Character String Case Functions

Modify case of character strings Functions and syntax:

st r i ng : =

UPPER( st r i ng_var i abl e_name) ; st r i ng : =

LOWER( st r i ng_var i abl e_name) ;

st r i ng : =I NI TCAP( st r i ng_var i abl e_name) ;


31/106


Parsing Character Strings

INSTRfunction

Searches string for specific substring

If substring is found, the starting position of itwithin the original_string is returned as integervalue. Otherwise, zero is returned.

Syntax: st ar t _posi t i on : =

I NSTR( or i gi nal _st r i ng, subst r i ng) ;

st ar t _posi t i oi n shoul d be i nt egerdat a t ype.


32/106


Parsing Character Strings

SUBSTRfunctionExtracts specific number of characters from characterstring starting at given point

Syntax: ext r act ed_st r i ng : = SUBSTR( st r i ng_var i abl e,

st ar t i ng_poi nt, number _of _char act er s) ;

To parse a string, usually INSTR is used to find delimiter and then use theSUBSTR.

Example page 206, 207.


33/106


Debugging PL/SQL Programs

Syntax errorOccurs when command does not follow guidelinesof programming language

Generate compiler or interpreter error messages Logic error

Does not stop program from running

Results in incorrect result


34/106


Program with a Syntax Error


35/106


Program with a Logic Error


36/106


Finding Syntax Errors

Often involve:Misspelling reserved word

Omitting required character in command

Using built-in function improperly Interpreter

Flags line number and character location of syntaxerrors

May actually be on preceding line

Displays error code and message


37/106


Finding Syntax Errors (continued)

Comment out program linesTo find error

Cascading errors are called so when

One syntax error can generate many more errorsFix the first error andrerun the program. Do not fix

next errors before rerunning the program.


38/106


Finding Logic Errors

Caused by:

Not using proper order of operations in arithmeticfunctions

Passing incorrect parameter values to built-infunctions

Creating loops that do not terminate properly

Using data values that are out of range or not ofright data type


39/106


Finding Logic Errors (continued)

Debugger

Program that enables software developers to pauseprogram execution and examine current variable

valuesBest way to find logic errors

SQL*Plus environment does not provide PL/SQL

debuggerUse DBMS_OUTPUT to print variable values


40/106


Lesson A Summary

PL/SQL data types:

ScalarComposite

Reference

LOB

Program block

DeclarationExecution

Exception


41/106


Lesson B Objectives

After completing this lesson, you should be able to:

Create PL/SQL decision control structures Use SQL queries in PL/SQL programs

Create loops in PL/SQL programs Create PL/SQL tables and tables of records

Use cursors to retrieve database data into PL/SQLprograms


42/106


Lesson B Objectives (continued)

Use the exception section to handle run-time errors inPL/SQL programs


43/106


PL/SQL Decision Control Structures

Sequentialprocessing

Processes statements one after another

Decision control structures

Alter order in which statements executeBased on values of certain variables


44/106


IF/THEN

Syntax:I F condi t i onTHEN

commands t hat execut e i f condi t i oni s TRUE;

END I F;

Condition

Expression evaluates to TRUE or FALSE If TRUE commands execute

Example


45/106

Example



46/106


PL/SQL Comparison Operators

Do not get confused with the equal (=) sign and the assignment (:=) operator


47/106


IF/THEN/ELSE

Syntax:

I F condi t i onTHEN

commands t hat execut e i f condi t i oni s TRUE;

ELSE

commands t hat execut e i f condi t i oni s FALSE;

END I F; Evaluates ELSE command if condition FALSE


48/106

Example



49/106


Nested IF/THEN/ELSE

Placing one or more IF/THEN/ELSE statementswithin program statements that execute after IF orELSE command

Important to properly indent program lines

IF/ELSIF Syntax:


50/106

50

Syntax:I F condi t i on1 THEN

commands t hat execut e i f condi t i on1 i s TRUE;

ELSI F condi t i on2THEN

commands t hat execut e i f condi t i on2 i s TRUE;

. . .

ELSE

commands t hat execut e i f no condi t i ons TRUE;

END I F;

Not e t her e i s no E f i r e I F i n ELSI F, and no spaces.


51/106


Logical Operators AND, OR, and

NOT

Create complex expressions for decision controlstructure condition

AND

Expressions on both sides of operator must be truefor combined expression to be TRUE

OR

Expressions on either side of operator must be truefor combined expression to be TRUE


52/106


Logical Operators AND, OR, and

NOT (continued)

Order of evaluation:

NOT (evaluated first)

AND

OR


53/106


Evaluating AND and OR in an

Expression


54/106


Using SQL Queries in PL/SQL Programs

Use SQL action query

Put query or command in PL/SQL program

Use same syntax as execute query or command inSQL*Plus

Can use variables instead of literal values to specify

data values

Ex: curr_f_name := Ahmad

We need to insert Ahmad into a column S_FIRST

INSERT INTO student (s_first)

VALUES (curr_f_name);

Ex: WHERE s_first = curr_f_name;

PL/SQL Command


55/106


Using SQL Commands in PL/SQL

Programs


56/106


Loops

Systematically executes program statements

Periodically evaluates exit condition to determine

if loop should repeat or exit Pretest loop

Evaluates exit condition before any programcommands execute

Posttest loop

Executes program commands before loop evaluatesexit condition for first time

In PL/SQL there are FIVE loop structures.


57/106


The LOOP...EXIT Loop This kind of loop can be either Pretest or posttest

Syntax:LOOP[ pr ogr am st at ement s]

I F condi t i onTHENEXI T;

END I F;

[ addi t i onal pr ogr am st at ement s]END LOOP;

The LOOP EXIT Loop


58/106

Guide to Oracle 10 58

The LOOP...EXIT Loop LOOP keyword signals the beginning of the loop.

If the IF/THEN decision structure is the first code in theloop, it is a pretest loop.

If the IF/THEN decision structure is the last code in the

loop, it is a posttest loop.

The LOOP...EXIT WHEN Loop


59/106


p Can be either a Pretest or posttest

Syntax:

LOOP

pr ogr am st at ement s

EXI T WHEN condi t i on;END LOOP;

The WHILE...LOOP


60/106


It is a Pretest loop that test the condition before it executeany program statements.

Syntax:

WHI LE condi t i on LOOP

pr ogr am st at ement sEND LOOP;


61/106


The Numeric FOR Loop

In the previous LOOP structures we had toincrement the counter variable.

The numeric FOR does not require explicitcounter increment

Automatically increments counter

Syntax:FOR count er _var i abl e I N st ar t _val ue . . . end_val ue

LOOP

pr ogr am st at ement s

END LOOP;

Start_value and end_value must be integer. The increment is always done

by 1.


62/106

Example


C FOR L


63/106


Cursors FOR Loop

Pointer to memory location on database server thatthe DBMS uses to process a SQL query

A cursor is a mechanism by which you can assign

a name to a "select statement" and manipulate theinformation within that SQL statement

Use to:

Retrieve and manipulate database data from thetables into PL/SQL programs

Types: Implicit

Explicit

Implicit Cursors


64/106


For every SQL statement execution certain area inmemory is allocated. PL/SQL allow you to name thisarea. This private SQL area is calledcontext area or

cursor. A cursor acts as a handle orpointerinto thecontext area. A PL/SQL program controls the contextarea using the cursor.

Context area Is a memory location on the DB server contains

information about query, such as the number of rowsthe query processed.

Created by INSERT, UPDATE, DELETE, or SELECT


65/106

Implicit Cursors

Active setSet of data rows that query retrieves using

SELECT

Implicit cursor Is a Pointer to the context area.

Called so because you do not need explicitcommand to create it.


Cont


66/106

For SQL queries returning single row PL/SQL

declares implicit cursors. Implicit cursors are simple SELECT statements

and are written in the BEGIN block (executablesection) of the PL/SQL.

Implicit cursors are easy to code, and they retrieve

exactly one row. PL/SQL implicitly declares cursors for all DML

statements.

The most commonly raised exceptions here areNO_DATA_FOUND orTOO_MANY_ROWS.

Guide to Oracle 10g 66To be discussed later

Implicit Cursors (continued)


67/106


Can be used to assign output of SELECT query toPL/SQL program variables:

When query will return only one record

If the query returns more than one record or zerorecords, an error occurs.

Syntax:

SELECT ename, sal INTO ena, esa FROM EMP WHERE EMPNO = 7844;


68/106

Cont



69/106



Syntax:SELECT f i el d1, f i el d2, . . .

I NTO var i abl e1, var i abl e2, . . .

FROM t abl e1, t abl e2, . . .WHERE j oi n_condi t i ons

AND sear ch_condi t i on_t o_r et r i eve_1_r ecor d;

The val ue of f i el d1 i s assi gned t o var i abl e1, f i el d2t o var i abl e2 and so on.

Dat a t ypes of f i el d1, f i el d2 and var i abl e1, var i abl e2must be t he same.

After we discuss join



70/106



Useful to use %TYPE reference data type To declare variables used with implicit cursors

Syntax: var_name tablename.fieldname%TYPE

Ex: curr_f_last faculty.f_last%TYPE Error ORA-01422: exact fetch returns more than requested

number of rows .

Implicit cursor query tried to retrieve multiple records Error ORA-014032: no data found.

If one of these errors occur, you shouldmodify the query to

retrieve only one record, oruse Explicit cursor.


71/106


72/106

Example



73/106


Explicit Cursors

Retrieve and display data in PL/SQL programs for

query that might:Retrieve multiple records

Return no records at all

Steps for creating and using explicit cursorDeclare cursor

Open cursorFetch data rows

Close cursor


74/106

Explicit cursor functions



75/106

Controlling Explicit Cursors



76/106

Cont


Explicit Cursors (continued)


77/106



Declare explicit cursor syntax: by declaring an explicit cursor, you create amemory location on the DB server that processes a query and stores the retrieved records .

Syntax:CURSOR cur sor _name I S sel ect _quer y; cur sor _name: Any val i d PL/ SQL name.

sel ect _quer y: The quer y t hat r et r i eves t he dat a val ues.

Open explicit cursor syntax: by doing so the interpreter checks for syntaxerror, upon correct it translate the query into machine-language format. The system thencreates the memory location that stores the active set (contains the retrieved data using

SELECT) .Syntax

OPEN cur sor _name;


78/106




79/106


p ( )

Fetchcommand is used to retrieve the data from the DB into the active set, one row

at a time.

Because a query might return several rows, you execute theFETCH command within a (LOOPEXIT WHEN) loop:

Syntax:

LOOP

FETCH cur sor _name I NTO

var i abl e_name( s) ;

EXI T WHEN cur sor _name%NOTFOUND;

Cursor_name is name of the cursor defined previously.

Variable_name(s) represents either a single variable or a list of variables that receivesdata from the cursors SELECT query. (every retrieved column from the DB table isassociated with a program variable, %TYPE or %ROWTYPE)


80/106




81/106


p ( )

Active set pointeris created when processing explicitcursor that:

Indicates memory location of the next record retrievedfrom database

After the last record of the query is executed, that

pointer points to an empty record.The condition (EXI T WHEN cur sor _name%NOTFOUND)

guarantees the exit when the pointer points to an empty

record.



82/106



Close cursor syntax:

CLOSE cur sor _name;

You shoul d cl ose a cur sor af t erpr ocessi ng al l r ecor ds so as t he memor y

and ot her r esour ces ar e avai l abl e t ot he syst em f or ot her t asks.

The syst em aut omat i cal l y cl oses t hecur sor when t he pr ogr am ends, i f youf or get t o do so.

Processing Explicit Cursors Using a


83/106

LOOP...EXIT WHEN Loop

Often used to process explicit cursors that retrieveand display database records

Use %TYPE variable to display explicit cursorvalues (See example Figure 4-29 page 243)

Use %ROWTYPE variable to display explicit

cursor values


84/106



Cursor FOR Loop

Make it easier than LOOP...EXIT WHEN to processexplicit cursors.

For Loop automatically:

Opens cursorFetches records

Closes cursor


C FOR L ( i d)


85/106

85

Cursor FOR Loop (continued)

Syntax:FOR var i abl e_name( s) I N cur sor _name

LOOP

pr ocessi ng commands

END LOOP;

Explicit Cursors Using a Cursor FOR Loop


86/106


Explicit Cursors Using a Cursor FOR Loop,

LOOP...EXIT WHEN

The cur sor var i abl e can be r ef er encedout si de t he Loop Exi t When. Whi l e cannot be r ef er enced out si de t he For Loop.


87/106


Handling Runtime Errors in PL/SQL

Programs

Pl/SQL supports Exception handling, that is:

Programmers place commands for displaying error messagesand

Give users options for fixing errors in programs exception

section (Optional section in a PL/SQL program block. Page 195) Runtime errors

Cause program to fail during execution

Exception section is used to handle runtime error. Unwanted event.


88/106

Handling Runtime Errors in PL/SQL Programs (cont )


89/106


Handling Runtime Errors in PL/SQL Programs (cont.)

An exception orunwantedevent is raisedwhen aruntime error occurs.

The control is transferred to the exception section,where the exception handlerhas the following options:Correct error without notifying user of problem

Inform user of error without taking corrective actionCorrect and inform

Inform the user and allow him to decide what action to

take. After exception handler executesProgram ends

Handling Runtime Errors in PL/SQL Programs (cont )


90/106


Handling Runtime Errors in PL/SQL Programs (cont.)

There are three kinds of exceptions:

Predefined

UndefinedUser defined

Predefined Exceptions


91/106


p

Most common errors that occur in programs PL/SQL language:

Assigns exception name

Provides built-in exception handlerfor each predefinedexception

System automatically displays error message informinguser of the nature of the problem

One can create exception handlers to display alternate

error messages for predefined exception

C PL/SQL P d fi d


92/106


Common PL/SQL Predefined

Exceptions

Exception Handler Syntax


93/106



The syntax to write an alternate error messages for predefined handler.



94/106



exception_name1, exception_name2, refer to thename of the predefined exception (table 4-10).

Exception handler commandsrepresent commandsthat display the error message.

WHEN OTHERS THEN clause is a handler

presents a general message to describe errors that donot have specific exception handlers.


95/106

95



96/106


When using alternate error message, such as in the example 4-

35, which handles only the NO_DATA_FOUND. Anunanticipated error might occur. That way no handler exists tohandle such unanticipated errors.

For that reason it is helpful to use the WHEN OTHERSexception handler too, so as to display messages for thatsituation.

To do this, use SQLERRMbuilt-in function (SQL ERRorMessage).

That function returns a string that contains the most recent error

code and its description. First define a varchar2 variable of length 512 to assign it the

error message: error_message := SQLERRM;


97/106


Undefined Exceptions


98/106


p

Less common errors

Do not have predefined names

One must explicitly declare exception in programsdeclaration section and

Associate the new exception with a specific Oracle

error code then

Create an exception handler in the exception section

Using the same syntax as for predefined exceptions

Syntax of Undefined Exceptions


99/106


DECLARE

e-exception_name EXCEPTION;

PRAGMA EXCEPTION INIT(e-exception_name, -Oracle_error_code);

e-exception_name represents name you assign to theexception.

PRAGMA EXCEPTION INIT: a command associates your

exception name with a specific Oracle error code. The hyphen (-) must precede the Oracle_error_code. See for

example Figure 4-37 next slide, the code is2291.

See example Figure 4-38 on undefined exception, next slide.


100/106

100

User-defined Exceptions


101/106


User-defined exceptions do not raise Oracle runtimeerror. BUT

Require exception handling to:

Enforce business rules (special rules of a company,example page 253. Rule: one can only delete grades withvaluesNULL. Other grades can notbe deleted. If a user

tries to delete a grade that its value is not NULL, theprogram should raise an exception and display a message)or

Ensure the integrity of database

General Syntax for Declaring, Raising, and

H dli U d fi d E i


102/106


Handling a User-defined Exception

General Syntax for Declaring, Raising, and Handling a User-

defined Exception


103/106


de ed cept o

To raise the exception, you must use the IF/THEN,which evaluates the exception condition (grade IS NOT

NULL).

If the condition is TRUE, the RAISE command isexecuted and raises the exception (transfer execution to

the exception section). The exception handler executes anddisplays the error-

handling messages.


104/106



105/106


Lesson B Summary

Decision control structures:

IF/THEN

IF/THEN/ELSE

IF/ELSIF Loop

Repeats action multiple times until it reaches exit

condition

Five types of loops


106/106

Lesson B Summary (continued)

Cursor

Pointer to memory location that DBMS uses toprocess SQL query

Types: Implicit

Explicit

Exception handlingThree exception types