- •Contents
- •Send Us Your Comments
- •Preface
- •What's New in PL/SQL?
- •1 Overview of PL/SQL
- •Advantages of PL/SQL
- •Tight Integration with SQL
- •Support for SQL
- •Better Performance
- •Higher Productivity
- •Full Portability
- •Tight Security
- •Support for Object-Oriented Programming
- •Understanding the Main Features of PL/SQL
- •Block Structure
- •Variables and Constants
- •Processing Queries with PL/SQL
- •Declaring PL/SQL Variables
- •Control Structures
- •Writing Reusable PL/SQL Code
- •Data Abstraction
- •Error Handling
- •PL/SQL Architecture
- •In the Oracle Database Server
- •In Oracle Tools
- •2 Fundamentals of the PL/SQL Language
- •Character Set
- •Lexical Units
- •Delimiters
- •Literals
- •Comments
- •Declarations
- •Using DEFAULT
- •Using NOT NULL
- •Using the %TYPE Attribute
- •Using the %ROWTYPE Attribute
- •Restrictions on Declarations
- •PL/SQL Naming Conventions
- •Scope and Visibility of PL/SQL Identifiers
- •Assigning Values to Variables
- •Assigning Boolean Values
- •Assigning a SQL Query Result to a PL/SQL Variable
- •PL/SQL Expressions and Comparisons
- •Logical Operators
- •Boolean Expressions
- •CASE Expressions
- •Handling Null Values in Comparisons and Conditional Statements
- •Summary of PL/SQL Built-In Functions
- •3 PL/SQL Datatypes
- •PL/SQL Number Types
- •PL/SQL Character and String Types
- •PL/SQL National Character Types
- •PL/SQL LOB Types
- •PL/SQL Boolean Types
- •PL/SQL Date, Time, and Interval Types
- •Datetime and Interval Arithmetic
- •Avoiding Truncation Problems Using Date and Time Subtypes
- •Overview of PL/SQL Subtypes
- •Using Subtypes
- •Converting PL/SQL Datatypes
- •Explicit Conversion
- •Implicit Conversion
- •Choosing Between Implicit and Explicit Conversion
- •DATE Values
- •RAW and LONG RAW Values
- •4 Using PL/SQL Control Structures
- •Overview of PL/SQL Control Structures
- •Testing Conditions: IF and CASE Statements
- •Using the IF-THEN Statement
- •Using the IF-THEN-ELSE Statement
- •Using the IF-THEN-ELSIF Statement
- •Using the CASE Statement
- •Guidelines for PL/SQL Conditional Statements
- •Controlling Loop Iterations: LOOP and EXIT Statements
- •Using the LOOP Statement
- •Using the EXIT Statement
- •Using the EXIT-WHEN Statement
- •Labeling a PL/SQL Loop
- •Using the WHILE-LOOP Statement
- •Using the FOR-LOOP Statement
- •Sequential Control: GOTO and NULL Statements
- •Using the GOTO Statement
- •Using the NULL Statement
- •5 Using PL/SQL Collections and Records
- •What Is a Collection?
- •Understanding Nested Tables
- •Understanding Varrays
- •Understanding Associative Arrays (Index-By Tables)
- •How Globalization Settings Affect VARCHAR2 Keys for Associative Arrays
- •Choosing Which PL/SQL Collection Types to Use
- •Choosing Between Nested Tables and Associative Arrays
- •Choosing Between Nested Tables and Varrays
- •Defining Collection Types
- •Declaring PL/SQL Collection Variables
- •Initializing and Referencing Collections
- •Referencing Collection Elements
- •Assigning Collections
- •Comparing Collections
- •Using PL/SQL Collections with SQL Statements
- •Using PL/SQL Varrays with INSERT, UPDATE, and SELECT Statements
- •Manipulating Individual Collection Elements with SQL
- •Using Multilevel Collections
- •Using Collection Methods
- •Checking If a Collection Element Exists (EXISTS Method)
- •Counting the Elements in a Collection (COUNT Method)
- •Checking the Maximum Size of a Collection (LIMIT Method)
- •Finding the First or Last Collection Element (FIRST and LAST Methods)
- •Looping Through Collection Elements (PRIOR and NEXT Methods)
- •Increasing the Size of a Collection (EXTEND Method)
- •Decreasing the Size of a Collection (TRIM Method)
- •Deleting Collection Elements (DELETE Method)
- •Applying Methods to Collection Parameters
- •Avoiding Collection Exceptions
- •What Is a PL/SQL Record?
- •Using Records as Procedure Parameters and Function Return Values
- •Assigning Values to Records
- •Comparing Records
- •Inserting PL/SQL Records into the Database
- •Updating the Database with PL/SQL Record Values
- •Restrictions on Record Inserts/Updates
- •Querying Data into Collections of Records
- •6 Performing SQL Operations from PL/SQL
- •Overview of SQL Support in PL/SQL
- •Data Manipulation
- •Transaction Control
- •SQL Functions
- •SQL Pseudocolumns
- •SQL Operators
- •Performing DML Operations from PL/SQL (INSERT, UPDATE, and DELETE)
- •Overview of Implicit Cursor Attributes
- •Using PL/SQL Records in SQL INSERT and UPDATE Statements
- •Issuing Queries from PL/SQL
- •Selecting At Most One Row: SELECT INTO Statement
- •Selecting Multiple Rows: BULK COLLECT Clause
- •Looping Through Multiple Rows: Cursor FOR Loop
- •Performing Complicated Query Processing: Explicit Cursors
- •Querying Data with PL/SQL
- •Querying Data with PL/SQL: Implicit Cursor FOR Loop
- •Querying Data with PL/SQL: Explicit Cursor FOR Loops
- •Overview of Explicit Cursors
- •Using Subqueries
- •Using Correlated Subqueries
- •Writing Maintainable PL/SQL Queries
- •Using Cursor Attributes
- •Overview of Explicit Cursor Attributes
- •Using Cursor Variables (REF CURSORs)
- •What Are Cursor Variables (REF CURSORs)?
- •Why Use Cursor Variables?
- •Declaring REF CURSOR Types and Cursor Variables
- •Controlling Cursor Variables: OPEN-FOR, FETCH, and CLOSE
- •Avoiding Errors with Cursor Variables
- •Restrictions on Cursor Variables
- •Using Cursor Expressions
- •Restrictions on Cursor Expressions
- •Example of Cursor Expressions
- •Constructing REF CURSORs with Cursor Subqueries
- •Overview of Transaction Processing in PL/SQL
- •Using COMMIT, SAVEPOINT, and ROLLBACK in PL/SQL
- •How Oracle Does Implicit Rollbacks
- •Ending Transactions
- •Setting Transaction Properties with SET TRANSACTION
- •Overriding Default Locking
- •Doing Independent Units of Work with Autonomous Transactions
- •Advantages of Autonomous Transactions
- •Controlling Autonomous Transactions
- •Using Autonomous Triggers
- •Calling Autonomous Functions from SQL
- •7 Performing SQL Operations with Native Dynamic SQL
- •What Is Dynamic SQL?
- •Why Use Dynamic SQL?
- •Using the EXECUTE IMMEDIATE Statement
- •Specifying Parameter Modes for Bind Variables in Dynamic SQL Strings
- •Building a Dynamic Query with Dynamic SQL
- •Examples of Dynamic SQL for Records, Objects, and Collections
- •Using Bulk Dynamic SQL
- •Using Dynamic SQL with Bulk SQL
- •Examples of Dynamic Bulk Binds
- •Guidelines for Dynamic SQL
- •When to Use or Omit the Semicolon with Dynamic SQL
- •Improving Performance of Dynamic SQL with Bind Variables
- •Passing Schema Object Names As Parameters
- •Using Duplicate Placeholders with Dynamic SQL
- •Using Cursor Attributes with Dynamic SQL
- •Passing Nulls to Dynamic SQL
- •Using Database Links with Dynamic SQL
- •Using Invoker Rights with Dynamic SQL
- •Using Pragma RESTRICT_REFERENCES with Dynamic SQL
- •Avoiding Deadlocks with Dynamic SQL
- •Backward Compatibility of the USING Clause
- •8 Using PL/SQL Subprograms
- •What Are Subprograms?
- •Advantages of PL/SQL Subprograms
- •Understanding PL/SQL Procedures
- •Understanding PL/SQL Functions
- •Using the RETURN Statement
- •Declaring Nested PL/SQL Subprograms
- •Passing Parameters to PL/SQL Subprograms
- •Actual Versus Formal Subprogram Parameters
- •Using Positional, Named, or Mixed Notation for Subprogram Parameters
- •Specifying Subprogram Parameter Modes
- •Using Default Values for Subprogram Parameters
- •Overloading Subprogram Names
- •Guidelines for Overloading with Numeric Types
- •Restrictions on Overloading
- •How Subprogram Calls Are Resolved
- •How Overloading Works with Inheritance
- •Using Invoker's Rights Versus Definer's Rights (AUTHID Clause)
- •Advantages of Invoker's Rights
- •Specifying the Privileges for a Subprogram with the AUTHID Clause
- •Who Is the Current User During Subprogram Execution?
- •How External References Are Resolved in Invoker's Rights Subprograms
- •Overriding Default Name Resolution in Invoker's Rights Subprograms
- •Granting Privileges on Invoker's Rights Subprograms
- •Using Roles with Invoker's Rights Subprograms
- •Using Views and Database Triggers with Invoker's Rights Subprograms
- •Using Database Links with Invoker's Rights Subprograms
- •Using Object Types with Invoker's Rights Subprograms
- •Using Recursion with PL/SQL
- •What Is a Recursive Subprogram?
- •Calling External Subprograms
- •Creating Dynamic Web Pages with PL/SQL Server Pages
- •Controlling Side Effects of PL/SQL Subprograms
- •Understanding Subprogram Parameter Aliasing
- •9 Using PL/SQL Packages
- •What Is a PL/SQL Package?
- •What Goes In a PL/SQL Package?
- •Example of a PL/SQL Package
- •Advantages of PL/SQL Packages
- •Understanding The Package Specification
- •Referencing Package Contents
- •Understanding The Package Body
- •Some Examples of Package Features
- •Private Versus Public Items in Packages
- •Overloading Packaged Subprograms
- •How Package STANDARD Defines the PL/SQL Environment
- •About the DBMS_ALERT Package
- •About the DBMS_OUTPUT Package
- •About the DBMS_PIPE Package
- •About the UTL_FILE Package
- •About the UTL_HTTP Package
- •Guidelines for Writing Packages
- •Separating Cursor Specs and Bodies with Packages
- •10 Handling PL/SQL Errors
- •Overview of PL/SQL Runtime Error Handling
- •Guidelines for Avoiding and Handling PL/SQL Errors and Exceptions
- •Advantages of PL/SQL Exceptions
- •Summary of Predefined PL/SQL Exceptions
- •Defining Your Own PL/SQL Exceptions
- •Declaring PL/SQL Exceptions
- •Scope Rules for PL/SQL Exceptions
- •Associating a PL/SQL Exception with a Number: Pragma EXCEPTION_INIT
- •How PL/SQL Exceptions Are Raised
- •Raising Exceptions with the RAISE Statement
- •How PL/SQL Exceptions Propagate
- •Reraising a PL/SQL Exception
- •Handling Raised PL/SQL Exceptions
- •Handling Exceptions Raised in Declarations
- •Handling Exceptions Raised in Handlers
- •Branching to or from an Exception Handler
- •Retrieving the Error Code and Error Message: SQLCODE and SQLERRM
- •Catching Unhandled Exceptions
- •Tips for Handling PL/SQL Errors
- •Continuing after an Exception Is Raised
- •Retrying a Transaction
- •Using Locator Variables to Identify Exception Locations
- •Overview of PL/SQL Compile-Time Warnings
- •PL/SQL Warning Categories
- •Controlling PL/SQL Warning Messages
- •Using the DBMS_WARNING Package
- •11 Tuning PL/SQL Applications for Performance
- •How PL/SQL Optimizes Your Programs
- •When to Tune PL/SQL Code
- •Guidelines for Avoiding PL/SQL Performance Problems
- •Avoiding CPU Overhead in PL/SQL Code
- •Avoiding Memory Overhead in PL/SQL Code
- •Profiling and Tracing PL/SQL Programs
- •Using The Trace API: Package DBMS_TRACE
- •Reducing Loop Overhead for DML Statements and Queries (FORALL, BULK COLLECT)
- •Using the FORALL Statement
- •Retrieving Query Results into Collections with the BULK COLLECT Clause
- •Writing Computation-Intensive Programs in PL/SQL
- •Tuning Dynamic SQL with EXECUTE IMMEDIATE and Cursor Variables
- •Tuning PL/SQL Procedure Calls with the NOCOPY Compiler Hint
- •Restrictions on NOCOPY
- •Compiling PL/SQL Code for Native Execution
- •Setting Up Transformation Pipelines with Table Functions
- •Overview of Table Functions
- •Using Pipelined Table Functions for Transformations
- •Writing a Pipelined Table Function
- •Returning Results from Table Functions
- •Pipelining Data Between PL/SQL Table Functions
- •Querying Table Functions
- •Optimizing Multiple Calls to Table Functions
- •Fetching from the Results of Table Functions
- •Passing Data with Cursor Variables
- •Performing DML Operations Inside Table Functions
- •Performing DML Operations on Table Functions
- •Handling Exceptions in Table Functions
- •12 Using PL/SQL Object Types
- •Overview of PL/SQL Object Types
- •What Is an Object Type?
- •Why Use Object Types?
- •Structure of an Object Type
- •Components of an Object Type
- •What Languages can I Use for Methods of Object Types?
- •How Object Types Handle the SELF Parameter
- •Overloading
- •Changing Attributes and Methods of an Existing Object Type (Type Evolution)
- •Defining Object Types
- •Overview of PL/SQL Type Inheritance
- •Declaring and Initializing Objects
- •Declaring Objects
- •Initializing Objects
- •How PL/SQL Treats Uninitialized Objects
- •Accessing Object Attributes
- •Defining Object Constructors
- •Calling Object Constructors
- •Calling Object Methods
- •Sharing Objects through the REF Modifier
- •Manipulating Objects through SQL
- •Selecting Objects
- •Inserting Objects
- •Updating Objects
- •Deleting Objects
- •13 PL/SQL Language Elements
- •Assignment Statement
- •AUTONOMOUS_TRANSACTION Pragma
- •Blocks
- •CASE Statement
- •CLOSE Statement
- •Collection Methods
- •Collections
- •Comments
- •COMMIT Statement
- •Constants and Variables
- •Cursor Attributes
- •Cursor Variables
- •Cursors
- •DELETE Statement
- •EXCEPTION_INIT Pragma
- •Exceptions
- •EXECUTE IMMEDIATE Statement
- •EXIT Statement
- •Expressions
- •FETCH Statement
- •FORALL Statement
- •Functions
- •GOTO Statement
- •IF Statement
- •INSERT Statement
- •Literals
- •LOCK TABLE Statement
- •LOOP Statements
- •MERGE Statement
- •NULL Statement
- •Object Types
- •OPEN Statement
- •OPEN-FOR Statement
- •OPEN-FOR-USING Statement
- •Packages
- •Procedures
- •RAISE Statement
- •Records
- •RESTRICT_REFERENCES Pragma
- •RETURN Statement
- •ROLLBACK Statement
- •%ROWTYPE Attribute
- •SAVEPOINT Statement
- •SCN_TO_TIMESTAMP Function
- •SELECT INTO Statement
- •SERIALLY_REUSABLE Pragma
- •SET TRANSACTION Statement
- •SQL Cursor
- •SQLCODE Function
- •SQLERRM Function
- •TIMESTAMP_TO_SCN Function
- •%TYPE Attribute
- •UPDATE Statement
- •Where to Find PL/SQL Sample Programs
- •Exercises for the Reader
- •Assigning Character Values
- •Comparing Character Values
- •Inserting Character Values
- •Selecting Character Values
- •Advantages of Wrapping PL/SQL Procedures
- •Running the PL/SQL Wrap Utility
- •Input and Output Files for the PL/SQL Wrap Utility
- •Limitations of the PL/SQL Wrap Utility
- •What Is Name Resolution?
- •Examples of Qualified Names and Dot Notation
- •Differences in Name Resolution Between SQL and PL/SQL
- •Understanding Capture
- •Inner Capture
- •Same-Scope Capture
- •Outer Capture
- •Avoiding Inner Capture in DML Statements
- •Qualifying References to Object Attributes and Methods
- •Calling Parameterless Subprograms and Methods
- •Name Resolution for SQL Versus PL/SQL
- •When Should I Use Bind Variables with PL/SQL?
- •When Do I Use or Omit the Semicolon with Dynamic SQL?
- •How Can I Use Regular Expressions with PL/SQL?
- •How Do I Continue After a PL/SQL Exception?
- •How Do I Pass a Result Set from PL/SQL to Java or Visual Basic (VB)?
- •How Do I Specify Different Kinds of Names with PL/SQL's Dot Notation?
- •What Can I Do with Objects and Object Types in PL/SQL?
- •How Do I Create a PL/SQL Procedure?
- •How Do I Input or Output Data with PL/SQL?
- •How Do I Perform a Case-Insensitive Query?
- •Index
- •Symbols
Assigning Values to Records
DECLARE
TYPE EmpRec IS RECORD (
emp_id |
NUMBER(4) |
last_name |
VARCHAR2(10), |
dept_num |
NUMBER(2), |
job_title |
VARCHAR2(9), |
salary |
NUMBER(7,2)); |
FUNCTION nth_highest_salary (n INTEGER) RETURN EmpRec IS ...
BEGIN NULL;
END;
/
Like scalar variables, user-defined records can be declared as the formal parameters of procedures and functions:
DECLARE
TYPE EmpRec IS RECORD ( emp_id emp.empno%TYPE, last_name VARCHAR2(10), job_title VARCHAR2(9), salary NUMBER(7,2));
...
PROCEDURE raise_salary (emp_info EmpRec); BEGIN
...
END;
/
Assigning Values to Records
To set all the fields in a record to default values, assign to it an uninitialized record of the same type:
DECLARE
TYPE RecordTyp IS RECORD (field1 NUMBER, field2 VARCHAR2(32) DEFAULT 'something');
rec1 RecordTyp; rec2 RecordTyp;
BEGIN
--At first, rec1 has the values we assign. rec1.field1 := 100; rec1.field2 := 'something else';
--Assigning an empty record to rec1 resets fields to their default values.
--Field1 is NULL and field2 is 'something' (because of the DEFAULT clause above). rec1 := rec2;
dbms_output.put_line('Field1 = ' || NVL(TO_CHAR(rec1.field1),'<NULL>') || ',
field2 = ' || rec1.field2); END;
/
You can assign a value to a field in a record using an assignment statement with dot notation:
emp_info.last_name := 'Fields';
Instead of assigning values separately to each field in a record, you can assign values to all fields at once.
You can assign one user-defined record to another if they have the same datatype. Having fields that match exactly is not enough. Consider the following example:
5-34 PL/SQL User's Guide and Reference
Assigning Values to Records
DECLARE
--Two identical type declarations.
TYPE DeptRec1 IS RECORD ( dept_num NUMBER(2), dept_name VARCHAR2(14)); TYPE DeptRec2 IS RECORD ( dept_num NUMBER(2), dept_name VARCHAR2(14)); dept1_info DeptRec1;
dept2_info DeptRec2; dept3_info DeptRec2;
BEGIN
--Not allowed; different datatypes, even though fields are the same.
--dept1_info := dept2_info;
--This assignment is OK because the records have the same type. dept2_info := dept3_info;
END;
/
You can assign a %ROWTYPE record to a user-defined record if their fields match in number and order, and corresponding fields have the same datatypes:
DECLARE
TYPE RecordTyp IS RECORD (last employees.last_name%TYPE, id employees.employee_ id%TYPE);
CURSOR c1 IS SELECT last_name, employee_id FROM employees;
--Rec1 and rec2 have different types. But because rec2 is based on a %ROWTYPE, we can
--assign is to rec1 as long as they have the right number of fields and the fields
--have the right datatypes. rec1 RecordTyp;
rec2 c1%ROWTYPE;
BEGIN
SELECT last_name, employee_id INTO rec2 FROM employees WHERE ROWNUM < 2; rec1 := rec2;
dbms_output.put_line('Employee #' || rec1.id || ' = ' || rec1.last); END;
/
You can also use the SELECT or FETCH statement to fetch column values into a record. The columns in the select-list must appear in the same order as the fields in your record.
DECLARE
TYPE RecordTyp IS RECORD (last employees.last_name%TYPE, id employees.employee_ id%TYPE);
rec1 RecordTyp; BEGIN
SELECT last_name, employee_id INTO rec1 FROM employees WHERE ROWNUM < 2; dbms_output.put_line('Employee #' || rec1.id || ' = ' || rec1.last);
END;
/
You cannot assign a list of values to a record using an assignment statement. There is no constructor-like notation for records.
Comparing Records
Records cannot be tested for nullity, or compared for equality, or inequality.
Using PL/SQL Collections and Records 5-35
Assigning Values to Records
If you want to make such comparisons, write your own function that accepts two records as parameters and does the appropriate checks or comparisons on the corresponding fields.
Inserting PL/SQL Records into the Database
A PL/SQL-only extension of the INSERT statement lets you insert records into database rows, using a single variable of type RECORD or %ROWTYPE in the VALUES clause instead of a list of fields. That makes your code more readable and maintainable.
If you issue the INSERT through the FORALL statement, you can insert values from an entire collection of records.
The number of fields in the record must equal the number of columns listed in the INTO clause, and corresponding fields and columns must have compatible datatypes. To make sure the record is compatible with the table, you might find it most convenient to declare the variable as the type table_name%ROWTYPE.
Example 5–34 Inserting a PL/SQL Record Using %ROWTYPE
This example declares a record variable using a %ROWTYPE qualifier. You can insert this variable without specifying a column list. The %ROWTYPE declaration ensures that the record attributes have exactly the same names and types as the table columns.
DECLARE
dept_info dept%ROWTYPE; BEGIN
--deptno, dname, and loc are the table columns.
--The record picks up these names from the %ROWTYPE. dept_info.deptno := 70;
dept_info.dname := 'PERSONNEL'; dept_info.loc := 'DALLAS';
--Using the %ROWTYPE means we can leave out the column list
--(deptno, dname, loc) from the INSERT statement. INSERT INTO dept VALUES dept_info;
END;
/
Updating the Database with PL/SQL Record Values
A PL/SQL-only extension of the UPDATE statement lets you update database rows using a single variable of type RECORD or %ROWTYPE on the right side of the SET clause, instead of a list of fields.
If you issue the UPDATE through the FORALL statement, you can update a set of rows using values from an entire collection of records.
Also with an UPDATE statement, you can specify a record in the RETURNING clause to retrieve new values into a record. If you issue the UPDATE through the FORALL statement, you can retrieve new values from a set of updated rows into a collection of records.
The number of fields in the record must equal the number of columns listed in the SET clause, and corresponding fields and columns must have compatible datatypes.
Example 5–35 Updating a Row Using a Record
You can use the keyword ROW to represent an entire row:
5-36 PL/SQL User's Guide and Reference
Assigning Values to Records
DECLARE
dept_info dept%ROWTYPE; BEGIN
dept_info.deptno := 30; dept_info.dname := 'MARKETING'; dept_info.loc := 'ATLANTA';
--The row will have values for the filled-in columns, and null
--for any other columns.
UPDATE dept SET ROW = dept_info WHERE deptno = 30;
END;
/
The keyword ROW is allowed only on the left side of a SET clause.
The argument to SET ROW must be a real PL/SQL record, not a subquery that returns a single row.
The record can also contain collections or objects.
Example 5–36 Using the RETURNING Clause with a Record
The INSERT, UPDATE, and DELETE statements can include a RETURNING clause, which returns column values from the affected row into a PL/SQL record variable. This eliminates the need to SELECT the row after an insert or update, or before a delete.
By default, you can use this clause only when operating on exactly one row. When you use bulk SQL, you can use the form RETURNING BULK COLLECT INTO to store the results in one or more collections.
The following example updates the salary of an employee and retrieves the employee's name, job title, and new salary into a record variable:
DECLARE
TYPE EmpRec IS RECORD (last_name employees.last_name%TYPE, salary employees.salary%TYPE);
emp_info EmpRec; emp_id NUMBER := 100;
BEGIN
UPDATE employees SET salary = salary * 1.1 WHERE employee_id = emp_id RETURNING last_name, salary INTO emp_info;
dbms_output.put_line('Just gave a raise to ' || emp_info.last_name || ', who now makes ' || emp_info.salary);
ROLLBACK;
END;
/
Restrictions on Record Inserts/Updates
Currently, the following restrictions apply to record inserts/updates:
■Record variables are allowed only in the following places:
■On the right side of the SET clause in an UPDATE statement
■In the VALUES clause of an INSERT statement
■In the INTO subclause of a RETURNING clause
Record variables are not allowed in a SELECT list, WHERE clause, GROUP BY clause, or ORDER BY clause.
Using PL/SQL Collections and Records 5-37