Oracle Database Application Developer’s Guide Fundamentals Developer 10g Release 2
Application%20Developer's%20Guide%20%E2%80%93%20Fundamentals
User Manual: Pdf
Open the PDF directly: View PDF .
Page Count: 440
Download | |
Open PDF In Browser | View PDF |
Oracle® Database Application Developer's Guide - Fundamentals 10g Release 2 (10.2) B14251-01 November 2005 Oracle Database Application Developer’s Guide - Fundamentals, 10g Release 2 (10.2) B14251-01 Copyright © 1996, 2005, Oracle. All rights reserved. Primary Author: Lance Ashdown Contributing Authors: D. Adams, M. Cowan, R. Moran, J. Melnick, E. Paapanen, J. Russell, R. Strohm Contributors: D. Alpern, G. Arora, C. Barclay, D. Bronnikov, T. Chang, M. Davidson, G. Doherty, D. Elson, A. Ganesh, M. Hartstein, J. Huang, N. Jain, R. Jenkins Jr., S. Kotsovolos, S. Kumar, C. Lei, D. Lorentz, V. Moore, J. Muller, R. Murthy, R. Pang, B. Sinha, S. Vemuri, W. Wang, D. Wong, A. Yalamanchi, Q. Yu The Programs (which include both the software and documentation) contain proprietary information; they are provided under a license agreement containing restrictions on use and disclosure and are also protected by copyright, patent, and other intellectual and industrial property laws. Reverse engineering, disassembly, or decompilation of the Programs, except to the extent required to obtain interoperability with other independently created software or as specified by law, is prohibited. The information contained in this document is subject to change without notice. If you find any problems in the documentation, please report them to us in writing. This document is not warranted to be error-free. Except as may be expressly permitted in your license agreement for these Programs, no part of these Programs may be reproduced or transmitted in any form or by any means, electronic or mechanical, for any purpose. If the Programs are delivered to the United States Government or anyone licensing or using the Programs on behalf of the United States Government, the following notice is applicable: U.S. GOVERNMENT RIGHTS Programs, software, databases, and related documentation and technical data delivered to U.S. Government customers are "commercial computer software" or "commercial technical data" pursuant to the applicable Federal Acquisition Regulation and agency-specific supplemental regulations. As such, use, duplication, disclosure, modification, and adaptation of the Programs, including documentation and technical data, shall be subject to the licensing restrictions set forth in the applicable Oracle license agreement, and, to the extent applicable, the additional rights set forth in FAR 52.227-19, Commercial Computer Software—Restricted Rights (June 1987). Oracle Corporation, 500 Oracle Parkway, Redwood City, CA 94065 The Programs are not intended for use in any nuclear, aviation, mass transit, medical, or other inherently dangerous applications. It shall be the licensee's responsibility to take all appropriate fail-safe, backup, redundancy and other measures to ensure the safe use of such applications if the Programs are used for such purposes, and we disclaim liability for any damages caused by such use of the Programs. Oracle, JD Edwards, PeopleSoft, and Retek are registered trademarks of Oracle Corporation and/or its affiliates. Other names may be trademarks of their respective owners. The Programs may provide links to Web sites and access to content, products, and services from third parties. Oracle is not responsible for the availability of, or any content provided on, third-party Web sites. You bear all risks associated with the use of such content. If you choose to purchase any products or services from a third party, the relationship is directly between you and the third party. Oracle is not responsible for: (a) the quality of third-party products or services; or (b) fulfilling any of the terms of the agreement with the third party, including delivery of products or services and warranty obligations related to purchased products or services. Oracle is not responsible for any loss or damage of any sort that you may incur from dealing with any third party. Contents Preface ............................................................................................................................................................... xix Audience..................................................................................................................................................... Documentation Accessibility ................................................................................................................... Related Documents ................................................................................................................................... Conventions ............................................................................................................................................... xix xix xx xxi What's New in Application Development? ................................................................................ xxiii Oracle Database 10g Release 2 (10.2) New Features........................................................................... Oracle Database 10g Release 1 (10.1) New Features........................................................................... 1 xxiii xxiv Orientation to Oracle Programmatic Environments Overview of Oracle Application Development ................................................................................. 1-1 Overview of PL/SQL................................................................................................................................ 1-2 What Is PL/SQL? ............................................................................................................................... 1-2 Advantages of PL/SQL..................................................................................................................... 1-3 Integration with Oracle Database............................................................................................. 1-4 High Performance....................................................................................................................... 1-4 High Productivity ....................................................................................................................... 1-4 Scalability ..................................................................................................................................... 1-4 Manageability .............................................................................................................................. 1-5 Object-Oriented Programming Support .................................................................................. 1-5 Portability..................................................................................................................................... 1-5 Security ......................................................................................................................................... 1-5 Built-In Packages......................................................................................................................... 1-6 PL/SQL Web Development Tools................................................................................................... 1-6 Overview of Java Support Built Into the Database ........................................................................... 1-6 Overview of Oracle JVM ................................................................................................................... 1-7 Overview of Oracle Extensions to JDBC......................................................................................... 1-8 JDBC Thin Driver........................................................................................................................ 1-8 JDBC OCI Driver......................................................................................................................... 1-8 JDBC Server-Side Internal Driver............................................................................................. 1-9 Oracle Database Extensions to JDBC Standards .................................................................... 1-9 Sample JDBC 2.0 Program ......................................................................................................... 1-9 Sample Pre-2.0 JDBC Program............................................................................................... 1-10 JDBC in SQLJ Applications..................................................................................................... 1-10 iii Overview of Oracle SQLJ ............................................................................................................... Benefits of SQLJ........................................................................................................................ Comparing SQLJ with JDBC .................................................................................................. SQLJ Stored Procedures in the Server................................................................................... Overview of Oracle JPublisher...................................................................................................... Overview of Java Stored Procedures............................................................................................ Overview of Oracle Database Web Services ............................................................................... Oracle Database as a Web Service Provider ........................................................................ Overview of Writing Procedures and Functions in Java........................................................... Overview of Writing Database Triggers in Java ................................................................. Why Use Java for Stored Procedures and Triggers?........................................................... Overview of Pro*C/C++ ....................................................................................................................... Implementing a Pro*C/C++ Application.................................................................................... Highlights of Pro*C/C++ Features............................................................................................... Overview of Pro*COBOL .................................................................................................................... How You Implement a Pro*COBOL Application ...................................................................... Highlights of Pro*COBOL Features ............................................................................................. Overview of OCI and OCCI................................................................................................................ Advantages of OCI.......................................................................................................................... Parts of the OCI ............................................................................................................................... Procedural and Non-Procedural Elements ................................................................................. Building an OCI Application......................................................................................................... Overview of Oracle Data Provider for .NET (ODP.NET) .............................................................. Using ODP.NET in a Simple Application ................................................................................... Overview of Oracle Objects for OLE (OO4O)................................................................................. OO4O Automation Server ............................................................................................................. OO4O Object Model ....................................................................................................................... OraSession................................................................................................................................. OraServer .................................................................................................................................. OraDatabase.............................................................................................................................. OraDynaset ............................................................................................................................... OraField..................................................................................................................................... OraMetaData and OraMDAttribute...................................................................................... OraParameters and OraParameter ........................................................................................ OraParamArray........................................................................................................................ OraSQLStmt.............................................................................................................................. OraAQ ....................................................................................................................................... OraAQMsg................................................................................................................................ OraAQAgent............................................................................................................................. Support for Oracle LOB and Object Datatypes........................................................................... OraBLOB and OraCLOB ......................................................................................................... OraBFILE................................................................................................................................... Oracle Data Control ........................................................................................................................ Oracle Objects for OLE C++ Class Library.................................................................................. Additional Sources of Information ............................................................................................... Choosing a Programming Environment ........................................................................................... Choosing Whether to Use OCI or a Precompiler ....................................................................... iv 1-11 1-12 1-12 1-13 1-13 1-13 1-14 1-14 1-15 1-15 1-15 1-15 1-16 1-17 1-18 1-18 1-19 1-19 1-20 1-20 1-21 1-21 1-22 1-22 1-23 1-24 1-25 1-25 1-25 1-26 1-26 1-26 1-27 1-27 1-27 1-27 1-27 1-28 1-28 1-28 1-29 1-29 1-29 1-29 1-29 1-30 1-30 Using Built-In Packages and Libraries......................................................................................... Java Compared to PL/SQL............................................................................................................ PL/SQL Is Optimized for Database Access ......................................................................... PL/SQL Is Integrated with the Database ............................................................................. Both Java and PL/SQL Have Object-Oriented Features .................................................... Java Is Used for Open Distributed Applications................................................................. Part I 2 1-30 1-31 1-31 1-31 1-31 1-32 SQL for Application Developers SQL Processing for Application Developers Grouping Operations into Transactions .............................................................................................. 2-1 Deciding How to Group Operations in Transactions................................................................... 2-1 Improving Transaction Performance ............................................................................................. 2-2 Committing Transactions.................................................................................................................. 2-2 Managing Commit Redo Behavior........................................................................................... 2-3 Rolling Back Transactions ................................................................................................................ 2-4 Defining Transaction Savepoints .................................................................................................... 2-5 An Example of COMMIT, SAVEPOINT, and ROLLBACK ................................................. 2-5 Ensuring Repeatable Reads with Read-Only Transactions ............................................................ 2-5 Using Cursors within Applications ...................................................................................................... 2-6 Declaring and Opening Cursors ..................................................................................................... 2-6 Using a Cursor to Execute Statements Again ................................................................................ 2-7 Closing Cursors ................................................................................................................................. 2-7 Cancelling Cursors ............................................................................................................................ 2-8 Locking Data Explicitly .......................................................................................................................... 2-8 Choosing a Locking Strategy ........................................................................................................... 2-9 When to Lock with ROW SHARE and ROW EXCLUSIVE Mode ....................................... 2-9 When to Lock with SHARE Mode............................................................................................ 2-9 When to Lock with SHARE ROW EXCLUSIVE Mode....................................................... 2-10 When to Lock in EXCLUSIVE Mode..................................................................................... 2-11 Privileges Required ................................................................................................................. 2-11 Letting Oracle Database Control Table Locking......................................................................... 2-11 Explicitly Acquiring Row Locks .................................................................................................. 2-12 About User Locks .................................................................................................................................. 2-13 When to Use User Locks ................................................................................................................ 2-13 Example of a User Lock.................................................................................................................. 2-13 Viewing and Monitoring Locks .................................................................................................... 2-14 Using Serializable Transactions for Concurrency Control ........................................................... 2-14 How Serializable Transactions Interact ....................................................................................... 2-15 Setting the Isolation Level of a Transaction ................................................................................ 2-16 The INITRANS Parameter ..................................................................................................... 2-16 Referential Integrity and Serializable Transactions .................................................................. 2-17 Using SELECT FOR UPDATE................................................................................................ 2-18 READ COMMITTED and SERIALIZABLE Isolation ............................................................... 2-18 Transaction Set Consistency .................................................................................................. 2-18 Comparison of READ COMMITTED and SERIALIZABLE Transactions....................... 2-19 v Choosing an Isolation Level for Transactions...................................................................... Application Tips for Transactions ................................................................................................ Autonomous Transactions ................................................................................................................... Examples of Autonomous Transactions ...................................................................................... Entering a Buy Order .............................................................................................................. Example: Making a Bank Withdrawal .................................................................................. Defining Autonomous Transactions ............................................................................................ Restrictions on Autonomous Transactions.................................................................................. Resuming Execution After a Storage Error Condition................................................................... What Operations Can Be Resumed After an Error Condition?................................................ Limitations on Resuming Operations After an Error Condition ............................................. Writing an Application to Handle Suspended Storage Allocation.......................................... Example of Resumable Storage Allocation ................................................................................. 3 2-20 2-20 2-20 2-22 2-23 2-23 2-26 2-27 2-27 2-27 2-27 2-28 2-28 Using SQL Datatypes in Application Development Representing Data with SQL Datatypes: Overview.......................................................................... 3-1 Representing Character Data ................................................................................................................. 3-2 Representing Character Data: Overview ........................................................................................ 3-2 Specifying Column Lengths as Bytes or Characters ..................................................................... 3-3 Choosing Between the CHAR and VARCHAR2 Datatypes........................................................ 3-3 Using Character Literals in SQL Statements .................................................................................. 3-4 Quoting Character Literals ........................................................................................................ 3-5 Representing Numeric Data................................................................................................................... 3-5 What Are the Numeric Datatypes? ................................................................................................. 3-5 Using Floating-Point Number Formats .......................................................................................... 3-6 Using a Floating-Point Binary Format ..................................................................................... 3-7 Representing Special Values with Native Floating-Point Formats ..................................... 3-8 Using Comparison Operators for Native Floating-Point Datatypes .......................................... 3-9 Performing Arithmetic Operations with Native Floating-Point Datatypes ........................... 3-10 Using Conversion Functions with Native Floating-Point Datatypes ...................................... 3-10 Client Interfaces for Native Floating-Point Datatypes .............................................................. 3-11 OCI Native Floating-Point Datatypes SQLT_BFLOAT and SQLT_BDOUBLE .............. 3-11 Native Floating-Point Datatypes Supported in Oracle OBJECT Types ........................... 3-12 Pro*C/C++ Support for Native Floating-Point Datatypes ................................................ 3-12 Representing Datetime Data............................................................................................................... 3-12 Representing Datetime Data: Overview ...................................................................................... 3-12 Using the DATE Datatype ...................................................................................................... 3-12 Using the TIMESTAMP Datatype ......................................................................................... 3-12 Using the TIMESTAMP WITH TIME ZONE Datatype...................................................... 3-12 Using the TIMESTAMP WITH LOCAL TIME ZONE Datatype....................................... 3-13 Representing the Difference Between Datetime Values..................................................... 3-13 Manipulating the Date Format ..................................................................................................... 3-13 Changing the Default Date Format ....................................................................................... 3-13 Displaying the Current Date and Time ................................................................................ 3-14 Manipulating the Time Format .................................................................................................... 3-14 Performing Date Arithmetic .......................................................................................................... 3-14 Converting Between Datetime Types .......................................................................................... 3-15 vi Importing and Exporting Datetime Types .................................................................................. Representing Specialized Data........................................................................................................... Representing Geographic Data ..................................................................................................... Representing Multimedia Data ..................................................................................................... Representing Large Amounts of Data.......................................................................................... Using RAW and LONG RAW Datatypes............................................................................. Representing Searchable Text ....................................................................................................... Representing XML .......................................................................................................................... Representing Dynamically Typed Data....................................................................................... Representing Data with ANSI/ISO, DB2, and SQL/DS Datatypes ....................................... Representing Conditional Expressions as Data .............................................................................. Identifying Rows by Address ............................................................................................................ Querying the ROWID Pseudocolumn ......................................................................................... Accessing the ROWID Datatype ................................................................................................... Restricted ROWID ................................................................................................................... Extended ROWID .................................................................................................................... External Binary ROWID.......................................................................................................... Accessing the UROWID Datatype................................................................................................ How Oracle Database Converts Datatypes ..................................................................................... Datatype Conversion During Assignments ................................................................................ Datatype Conversion During Expression Evaluation .............................................................. 4 3-15 3-16 3-16 3-16 3-16 3-17 3-18 3-18 3-19 3-21 3-22 3-23 3-24 3-24 3-24 3-24 3-25 3-25 3-25 3-25 3-26 Using Regular Expressions in Oracle Database Using Regular Expressions with Oracle Database: Overview ........................................................ 4-1 What Are Regular Expressions?....................................................................................................... 4-1 How Are Oracle Database Regular Expressions Useful?............................................................. 4-2 Oracle Database Implementation of Regular Expressions........................................................... 4-2 Oracle Database Support for the POSIX Regular Expression Standard..................................... 4-3 Regular Expression Metacharacters in Oracle Database .................................................................. 4-4 POSIX Metacharacters in Oracle Database Regular Expressions................................................ 4-4 Regular Expression Operator Multilingual Enhancements ......................................................... 4-6 Perl-Influenced Extensions in Oracle Regular Expressions ......................................................... 4-7 Using Regular Expressions in SQL Statements: Scenarios .............................................................. 4-9 Using an Integrity Constraint to Enforce a Phone Number Format........................................... 4-9 Using Back References to Reposition Characters ....................................................................... 4-10 5 Using Indexes in Application Development Guidelines for Application-Specific Indexes ..................................................................................... Create Indexes After Inserting Table Data .................................................................................... Switch Your Temporary Tablespace to Avoid Space Problems Creating Indexes................... Index the Correct Tables and Columns .......................................................................................... Limit the Number of Indexes for Each Table ................................................................................. Choose the Order of Columns in Composite Indexes .................................................................. Gather Statistics to Make Index Usage More Accurate ................................................................ Drop Indexes That Are No Longer Required ............................................................................... Privileges Required to Create an Index ......................................................................................... 5-1 5-2 5-2 5-2 5-3 5-3 5-4 5-4 5-5 vii Creating Indexes: Basic Examples ........................................................................................................ When to Use Domain Indexes ............................................................................................................... When to Use Function-Based Indexes .................................................................................................. Advantages of Function-Based Indexes.......................................................................................... Examples of Function-Based Indexes.............................................................................................. Example: Function-Based Index for Case-Insensitive Searches ........................................... Example: Precomputing Arithmetic Expressions with a Function-Based Index............... Example: Function-Based Index for Language-Dependent Sorting.................................... Restrictions for Function-Based Indexes......................................................................................... 6 5-5 5-6 5-6 5-6 5-7 5-8 5-8 5-8 5-8 Maintaining Data Integrity in Application Development Overview of Integrity Constraints ........................................................................................................ 6-1 When to Enforce Business Rules with Integrity Constraints ...................................................... 6-1 Example of an Integrity Constraint for a Business Rule ....................................................... 6-2 When to Enforce Business Rules in Applications ......................................................................... 6-2 Creating Indexes for Use with Constraints .................................................................................... 6-2 When to Use NOT NULL Integrity Constraints ........................................................................... 6-2 When to Use Default Column Values ............................................................................................ 6-3 Setting Default Column Values ....................................................................................................... 6-4 Choosing a Table's Primary Key ..................................................................................................... 6-4 When to Use UNIQUE Key Integrity Constraints ........................................................................ 6-5 Constraints On Views: for Performance, Not Data Integrity....................................................... 6-5 Enforcing Referential Integrity with Constraints .............................................................................. 6-6 About Nulls and Foreign Keys......................................................................................................... 6-7 Defining Relationships Between Parent and Child Tables .......................................................... 6-8 Rules for Multiple FOREIGN KEY Constraints ............................................................................ 6-9 Deferring Constraint Checks ............................................................................................................ 6-9 Guidelines for Deferring Constraint Checks .......................................................................... 6-9 Managing Constraints That Have Associated Indexes .................................................................. 6-10 Minimizing Space and Time Overhead for Indexes Associated with Constraints................ 6-10 Guidelines for Indexing Foreign Keys.............................................................................................. 6-11 About Referential Integrity in a Distributed Database................................................................. 6-11 When to Use CHECK Integrity Constraints..................................................................................... 6-11 Restrictions on CHECK Constraints ............................................................................................ 6-12 Designing CHECK Constraints .................................................................................................... 6-12 Rules for Multiple CHECK Constraints ...................................................................................... 6-12 Choosing Between CHECK and NOT NULL Integrity Constraints ...................................... 6-13 Examples of Defining Integrity Constraints ................................................................................... 6-13 Example: Defining Integrity Constraints with the CREATE TABLE Command .................. 6-13 Example: Defining Constraints with the ALTER TABLE Command...................................... 6-14 Privileges Required to Create Constraints ................................................................................. 6-14 Naming Integrity Constraints ...................................................................................................... 6-14 Enabling and Disabling Integrity Constraints ............................................................................... 6-14 Why Disable Constraints? ............................................................................................................. 6-15 About Exceptions to Integrity Constraints.................................................................................. 6-15 Enabling Constraints ..................................................................................................................... 6-15 Creating Disabled Constraints ..................................................................................................... 6-16 viii Enabling and Disabling Existing Integrity Constraints............................................................. Enabling Existing Constraints ............................................................................................... Disabling Existing Constraints .............................................................................................. Tip: Using the Data Dictionary to Find Constraints ........................................................... Guidelines for Enabling and Disabling Key Integrity Constraints.......................................... Fixing Constraint Exceptions ....................................................................................................... Altering Integrity Constraints ............................................................................................................ Renaming Integrity Constraints.................................................................................................... Dropping Integrity Constraints.......................................................................................................... Managing FOREIGN KEY Integrity Constraints ........................................................................... Datatypes and Names for Foreign Key Columns....................................................................... Limit on Columns in Composite Foreign Keys .......................................................................... Foreign Key References Primary Key by Default....................................................................... Privileges Required to Create FOREIGN KEY Integrity Constraints...................................... Choosing How Foreign Keys Enforce Referential Integrity .................................................... Viewing Definitions of Integrity Constraints ................................................................................ Examples of Defining Integrity Constraints................................................................................ Part II 7 6-16 6-16 6-16 6-17 6-17 6-17 6-17 6-18 6-19 6-19 6-19 6-19 6-20 6-20 6-20 6-21 6-21 PL/SQL for Application Developers Coding PL/SQL Procedures and Packages Overview of PL/SQL Program Units ................................................................................................... 7-1 Anonymous Blocks ........................................................................................................................... 7-2 Stored Program Units (Procedures, Functions, and Packages) .................................................. 7-3 Naming Procedures and Functions ......................................................................................... 7-4 Parameters for Procedures and Functions ............................................................................. 7-4 Creating Stored Procedures and Functions ........................................................................... 7-7 Altering Stored Procedures and Functions ............................................................................ 7-8 Dropping Procedures and Functions ...................................................................................... 7-8 External Procedures ................................................................................................................... 7-9 PL/SQL Packages ...................................................................................................................... 7-9 PL/SQL Object Size Limitation ............................................................................................ 7-11 Creating Packages ................................................................................................................... 7-11 Naming Packages and Package Objects .............................................................................. 7-12 Package Invalidations and Session State ............................................................................. 7-12 Packages Supplied With Oracle Database ........................................................................... 7-12 Overview of Bulk Binds .......................................................................................................... 7-12 When to Use Bulk Binds ......................................................................................................... 7-13 Triggers ..................................................................................................................................... 7-15 Compiling PL/SQL Procedures for Native Execution .................................................................... 7-15 Remote Dependencies ......................................................................................................................... 7-15 Timestamps ..................................................................................................................................... 7-16 Disadvantages of the Timestamp Model ............................................................................. 7-16 Signatures ........................................................................................................................................ 7-16 When Does a Signature Change?........................................................................................... 7-18 Examples of Changing Procedure Signatures .................................................................... 7-19 ix Controlling Remote Dependencies .............................................................................................. Dependency Resolution ......................................................................................................... Suggestions for Managing Dependencies ........................................................................... Cursor Variables ................................................................................................................................... Declaring and Opening Cursor Variables .................................................................................. Examples of Cursor Variables ...................................................................................................... Fetching Data ........................................................................................................................... Implementing Variant Records ............................................................................................. Handling PL/SQL Compile-Time Errors ......................................................................................... Handling Run-Time PL/SQL Errors .................................................................................................. Declaring Exceptions and Exception Handling Routines ........................................................ Unhandled Exceptions .................................................................................................................. Handling Errors in Distributed Queries ..................................................................................... Handling Errors in Remote Procedures ...................................................................................... Debugging Stored Procedures ............................................................................................................ Calling Stored Procedures ................................................................................................................... A Procedure or Trigger Calling Another Procedure.................................................................. Interactively Calling Procedures From Oracle Database Tools ............................................... Calling Procedures within 3GL Applications ............................................................................ Name Resolution When Calling Procedures .............................................................................. Privileges Required to Execute a Procedure .............................................................................. Specifying Values for Procedure Arguments ............................................................................. Calling Remote Procedures ................................................................................................................ Remote Procedure Calls and Parameter Values ......................................................................... Referencing Remote Objects .......................................................................................................... Synonyms for Procedures and Packages .................................................................................... Calling Stored Functions from SQL Expressions ........................................................................... Using PL/SQL Functions .............................................................................................................. Syntax for SQL Calling a PL/SQL Function ............................................................................... Naming Conventions ..................................................................................................................... Name Precedence .................................................................................................................... Arguments ............................................................................................................................... Using Default Values .............................................................................................................. Privileges .................................................................................................................................. Requirements for Calling PL/SQL Functions from SQL Expressions .................................... Controlling Side Effects ................................................................................................................. Restrictions................................................................................................................................ Declaring a Function ............................................................................................................... Parallel Query and Parallel DML ......................................................................................... PRAGMA RESTRICT_REFERENCES – for Backward Compatibility ............................. Serially Reusable PL/SQL Packages ........................................................................................... Package States .......................................................................................................................... Why Serially Reusable Packages? ......................................................................................... Syntax of Serially Reusable Packages ................................................................................... Semantics of Serially Reusable Packages.............................................................................. Examples of Serially Reusable Packages .............................................................................. Returning Large Amounts of Data from a Function....................................................................... x 7-20 7-21 7-22 7-22 7-22 7-23 7-23 7-24 7-24 7-26 7-27 7-28 7-28 7-28 7-29 7-31 7-32 7-32 7-33 7-33 7-33 7-34 7-34 7-35 7-35 7-36 7-36 7-37 7-37 7-37 7-38 7-39 7-39 7-39 7-39 7-40 7-40 7-41 7-42 7-43 7-46 7-47 7-47 7-47 7-47 7-48 7-51 Coding Your Own Aggregate Functions ........................................................................................... 7-52 8 Coding Dynamic SQL What Is Dynamic SQL? ........................................................................................................................... 8-1 Programming with Dynamic SQL ................................................................................................... 8-1 Why Use Dynamic SQL?......................................................................................................................... 8-2 Executing DDL and SCL Statements in PL/SQL .......................................................................... 8-3 Executing Dynamic Queries ............................................................................................................. 8-4 Referencing Database Objects that Do Not Exist at Compilation ............................................... 8-4 Optimizing Execution Dynamically ............................................................................................... 8-5 Executing Dynamic PL/SQL Blocks ............................................................................................... 8-5 Performing Dynamic Operations Using Invoker's Rights .......................................................... 8-6 Developing with Native Dynamic SQL: Scenario ............................................................................. 8-7 Sample DML Operation Using Native Dynamic SQL .................................................................. 8-7 Sample DDL Operation Using Native Dynamic SQL................................................................... 8-8 Sample Single-Row Query Using Native Dynamic SQL.............................................................. 8-8 Sample Multiple-Row Query with Native Dynamic SQL............................................................ 8-9 Choosing Between Native Dynamic SQL and the DBMS_SQL Package ..................................... 8-9 Advantages of Native Dynamic SQL ........................................................................................... 8-10 Native Dynamic SQL is Easy to Use ..................................................................................... 8-10 Native Dynamic SQL is Faster than DBMS_SQL ................................................................ 8-11 Native Dynamic SQL Supports User-Defined Types ......................................................... 8-12 Native Dynamic SQL Supports Fetching into Records ...................................................... 8-12 Advantages of the DBMS_SQL Package...................................................................................... 8-13 DBMS_SQL is Supported in Client-Side Programs ............................................................ 8-13 DBMS_SQL Supports Statements with Unknown Number of Inputs or Outputs ........ 8-13 DBMS_SQL Supports SQL Statements Larger than 32 KB ................................................ 8-13 DBMS_SQL Lets You Reuse SQL Statements ...................................................................... 8-13 Examples of DBMS_SQL Package Code and Native Dynamic SQL Code ............................. 8-14 Querying with Dynamic SQL: Example ............................................................................... 8-14 Performing DML with Dynamic SQL: Example ................................................................. 8-15 Performing DML with RETURNING Clause Using Dynamic SQL: Example ............... 8-16 Avoiding SQL Injection in PL/SQL ................................................................................................... 8-17 Overview of SQL Injection Techniques ....................................................................................... 8-17 Statement Modification ........................................................................................................... 8-18 Statement Injection .................................................................................................................. 8-19 Guarding Against SQL Injection................................................................................................... 8-21 Using Bind Variables to Guard Against SQL Injection ...................................................... 8-21 Using Validation Checks to Guard Against SQL Injection................................................ 8-22 9 Coding Triggers Designing Triggers................................................................................................................................... Creating Triggers ..................................................................................................................................... Types of Triggers ............................................................................................................................... Overview of System Events....................................................................................................... Getting the Attributes of System Events ................................................................................. 9-1 9-2 9-3 9-3 9-3 xi Naming Triggers ............................................................................................................................... 9-3 When Is the Trigger Fired? .............................................................................................................. 9-3 Do Import and SQL*Loader Fire Triggers?............................................................................. 9-4 How Column Lists Affect UPDATE Triggers ........................................................................ 9-4 Controlling When a Trigger Is Fired (BEFORE and AFTER Options) ...................................... 9-4 Ordering of Triggers ......................................................................................................................... 9-5 Modifying Complex Views (INSTEAD OF Triggers) ................................................................... 9-6 Views that Require INSTEAD OF Triggers............................................................................. 9-6 INSTEAD OF Trigger Example................................................................................................. 9-7 Object Views and INSTEAD OF Triggers .............................................................................. 9-8 Triggers on Nested Table View Columns ............................................................................... 9-8 Firing Triggers One or Many Times (FOR EACH ROW Option) .............................................. 9-9 Firing Triggers Based on Conditions (WHEN Clause) ............................................................. 9-10 Coding the Trigger Body ..................................................................................................................... 9-10 Accessing Column Values in Row Triggers ............................................................................... 9-12 Example: Modifying LOB Columns with a Trigger............................................................ 9-12 INSTEAD OF Triggers on Nested Table View Columns ................................................... 9-13 Avoiding Name Conflicts with Triggers (REFERENCING Option) ............................... 9-13 Detecting the DML Operation That Fired a Trigger ........................................................... 9-14 Error Conditions and Exceptions in the Trigger Body ...................................................... 9-14 Triggers on Object Tables............................................................................................................... 9-14 Triggers and Handling Remote Exceptions ............................................................................... 9-15 Restrictions on Creating Triggers ................................................................................................ 9-16 Who Is the Trigger User? ............................................................................................................... 9-19 Privileges Needed to Work with Triggers .................................................................................. 9-19 Compiling Triggers .............................................................................................................................. 9-20 Dependencies for Triggers ............................................................................................................ 9-20 Recompiling Triggers .................................................................................................................... 9-20 Modifying Triggers .............................................................................................................................. 9-21 Debugging Triggers ....................................................................................................................... 9-21 Enabling and Disabling Triggers ....................................................................................................... 9-21 Enabling Triggers ........................................................................................................................... 9-21 Disabling Triggers .......................................................................................................................... 9-21 Viewing Information About Triggers ............................................................................................... 9-22 Examples of Trigger Applications ..................................................................................................... 9-23 Responding to System Events through Triggers ............................................................................ 9-37 How Events Are Published Through Triggers ........................................................................... 9-37 Publication Context......................................................................................................................... 9-38 Error Handling ................................................................................................................................ 9-38 Execution Model.............................................................................................................................. 9-38 Event Attribute Functions.............................................................................................................. 9-38 List of Database Events .................................................................................................................. 9-41 System Events........................................................................................................................... 9-41 Client Events............................................................................................................................. 9-42 10 Developing Flashback Applications Overview of Flashback Features ....................................................................................................... 10-1 xii Application Development Features.............................................................................................. Database Administration Features ............................................................................................... Database Administration Tasks Before Using Flashback Features ............................................. Using Flashback Query (SELECT ... AS OF) ................................................................................... Examining Past Data: Example ..................................................................................................... Tips for Using Flashback Query ................................................................................................... Using the DBMS_FLASHBACK Package......................................................................................... Using ORA_ROWSCN......................................................................................................................... Using Flashback Version Query ........................................................................................................ Using Flashback Transaction Query................................................................................................ Flashback Transaction Query and Flashback Version Query: Example ............................... Flashback Tips ..................................................................................................................................... Flashback Tips – Performance..................................................................................................... Flashback Tips – General ............................................................................................................. 11 Developing Applications with the PL/SQL Web Toolkit Developing PL/SQL Web Applications: Overview ........................................................................ Invoking a PL/SQL Web Application.......................................................................................... Implementing a PL/SQL Web Application ................................................................................ PL/SQL Web Toolkit............................................................................................................... Using the mod_plsql Gateway ........................................................................................................... Generating HTML Output with PL/SQL.......................................................................................... Passing Parameters to a PL/SQL Web Application ......................................................................... Passing List and Dropdown List Parameters from an HTML Form ....................................... Passing Radio Button and Checkbox Parameters from an HTML Form ................................ Passing Entry Field Parameters from an HTML Form .............................................................. Passing Hidden Parameters from an HTML Form .................................................................... Uploading a File from an HTML Form........................................................................................ Submitting a Completed HTML Form......................................................................................... Handling Missing Input from an HTML Form .......................................................................... Maintaining State Information Between Web Pages ................................................................. Performing Network Operations within PL/SQL Stored Procedures....................................... Sending E-Mail from PL/SQL..................................................................................................... Getting a Host Name or Address from PL/SQL...................................................................... Working with TCP/IP Connections from PL/SQL ................................................................. Retrieving the Contents of an HTTP URL from PL/SQL ....................................................... Working with Tables, Image Maps, Cookies, and CGI Variables from PL/SQL ............... 12 10-2 10-2 10-3 10-4 10-5 10-5 10-6 10-7 10-8 10-10 10-10 10-12 10-12 10-13 11-1 11-1 11-2 11-2 11-3 11-4 11-5 11-5 11-6 11-6 11-8 11-8 11-8 11-9 11-9 11-10 11-10 11-10 11-11 11-11 11-13 Developing PL/SQL Server Pages PL/SQL Server Pages: Overview ........................................................................................................ What Are PL/SQL Server Pages and Why Use Them?............................................................. Prerequisites for Developing and Deploying PL/SQL Server Pages...................................... PSP and the HTP Package.............................................................................................................. PSP and Other Scripting Solutions ............................................................................................... Writing a PL/SQL Server Page ............................................................................................................ Specifying Basic Server Page Characteristics .............................................................................. 12-1 12-1 12-2 12-3 12-3 12-4 12-5 xiii Specifying the Scripting Language........................................................................................ Returning Data to the Client .................................................................................................. Handling Script Errors ............................................................................................................ Accepting User Input...................................................................................................................... Naming the PL/SQL Stored Procedure....................................................................................... Including the Contents of Other Files .......................................................................................... Declaring Global Variables in a PSP Script ................................................................................. Specifying Executable Statements in a PSP Script...................................................................... Substituting an Expression Result in a PSP Script ................................................................... Quoting and Escaping Strings in a PSP Script .......................................................................... Including Comments in a PSP Script ......................................................................................... Loading a PL/SQL Server Page into the Database ........................................................................ Querying PSP Source Code ......................................................................................................... Executing a PL/SQL Server Page Through a URL......................................................................... Examples of PL/SQL Server Pages ................................................................................................... Setup for PL/SQL Server Pages Examples................................................................................ Printing the Sample Table with a Loop ..................................................................................... Allowing a User Selection............................................................................................................ Using an HTML Form to Call a PL/SQL Server Page............................................................. Including JavaScript in a PSP File ....................................................................................... Debugging PL/SQL Server Page Problems .................................................................................... Putting PL/SQL Server Pages into Production .............................................................................. 13 Developing Applications with Database Change Notification What Is Database Change Notification? ........................................................................................... Using Database Change Notification in the Middle Tier ............................................................. Registering Queries for Database Change Notification ................................................................ Privileges .......................................................................................................................................... What Is a Database Change Registration?................................................................................... Supported Query Types ................................................................................................................. Registration Properties ................................................................................................................... Drop Table........................................................................................................................................ Interfaces for Database Change Registration .............................................................................. Creating a PL/SQL Stored Procedure as the Change Notification Recipient........................ Registering Queries for Change Notification Through PL/SQL ............................................. Creating a CHNF$_REG_INFO Object ................................................................................. Creating a Registration with DBMS_CHANGE_NOTIFICATION................................ Adding Objects to an Existing Registration ....................................................................... Querying Change Notification Registrations................................................................................ Interpreting a Database Change Notification................................................................................ Interpreting a CHNF$_DESC Object.......................................................................................... Interpreting a CHNF$_TDESC Object ................................................................................ Interpreting a CHNF$_RDESC Object ................................................................................ Configuring Database Change Notification: Scenario ................................................................ Creating a PL/SQL Callback Procedure.................................................................................... Registering the Query................................................................................................................... Best Practices ........................................................................................................................................ xiv 12-5 12-5 12-7 12-7 12-8 12-8 12-9 12-9 12-10 12-11 12-11 12-12 12-13 12-14 12-15 12-15 12-16 12-17 12-18 12-19 12-20 12-21 13-1 13-2 13-5 13-5 13-5 13-6 13-6 13-7 13-8 13-8 13-9 13-9 13-11 13-12 13-12 13-13 13-13 13-13 13-14 13-14 13-15 13-16 13-17 Troubleshooting................................................................................................................................... 13-18 Part III 14 Advanced Topics for Application Developers Calling External Procedures Overview of Multi-Language Programs ........................................................................................... What Is an External Procedure? .......................................................................................................... Overview of The Call Specification for External Procedures....................................................... Loading External Procedures .............................................................................................................. Loading Java Class Methods ......................................................................................................... Loading External C Procedures .................................................................................................... Publishing External Procedures ......................................................................................................... The AS LANGUAGE Clause for Java Class Methods ............................................................... The AS LANGUAGE Clause for External C Procedures .......................................................... LIBRARY .................................................................................................................................. NAME ..................................................................................................................................... LANGUAGE .......................................................................................................................... CALLING STANDARD ....................................................................................................... WITH CONTEXT .................................................................................................................. PARAMETERS ...................................................................................................................... AGENT IN .............................................................................................................................. Publishing Java Class Methods........................................................................................................ Publishing External C Procedures ................................................................................................... Locations of Call Specifications ....................................................................................................... Passing Parameters to External C Procedures with Call Specifications ................................... Specifying Datatypes .................................................................................................................... External Datatype Mappings....................................................................................................... BY VALUE/REFERENCE for IN and IN OUT Parameter Modes ....................................... The PARAMETERS Clause.......................................................................................................... Overriding Default Datatype Mapping ..................................................................................... Specifying Properties .................................................................................................................... INDICATOR ........................................................................................................................... LENGTH and MAXLEN....................................................................................................... CHARSETID and CHARSETFORM.................................................................................... Repositioning Parameters..................................................................................................... Using SELF.............................................................................................................................. Passing Parameters by Reference ........................................................................................ WITH CONTEXT ................................................................................................................... Inter-Language Parameter Mode Mappings...................................................................... Executing External Procedures with the CALL Statement .......................................................... Preconditions for External Procedures ...................................................................................... Privileges of External Procedures........................................................................................ Managing Permissions .......................................................................................................... Creating Synonyms for External Procedures..................................................................... CALL Statement Syntax ............................................................................................................... Calling Java Class Methods ......................................................................................................... 14-1 14-2 14-3 14-3 14-3 14-4 14-8 14-9 14-9 14-9 14-10 14-10 14-10 14-10 14-10 14-10 14-10 14-11 14-11 14-14 14-15 14-16 14-18 14-18 14-19 14-19 14-20 14-21 14-21 14-22 14-22 14-24 14-25 14-25 14-25 14-26 14-26 14-27 14-27 14-27 14-28 xv How the Database Server Calls External C Procedures .......................................................... Handling Errors and Exceptions in Multi-Language Programs................................................. Using Service Procedures with External C Procedures................................................................ Doing Callbacks with External C Procedures................................................................................ Object Support for OCI Callbacks............................................................................................... Restrictions on Callbacks ............................................................................................................. Debugging External Procedures ................................................................................................. Using Package DEBUG_EXTPROC .................................................................................... Demo Program .............................................................................................................................. Guidelines for External C Procedures........................................................................................ Restrictions on External C Procedures....................................................................................... 15 Developing Applications with Oracle XA X/Open Distributed Transaction Processing (DTP)........................................................................ DTP Terminology............................................................................................................................ Required Public Information ......................................................................................................... Oracle XA Library Interface Subroutines......................................................................................... XA Library Subroutines ................................................................................................................. Extensions to the XA Interface ...................................................................................................... Developing and Installing XA Applications ................................................................................... Responsibilities of the DBA or System Administrator .............................................................. Responsibilities of the Application Developer ........................................................................... Defining the xa_open() String ....................................................................................................... Syntax of the xa_open() String ............................................................................................... Required Fields for the xa_open() String.............................................................................. Optional Fields for the xa_open() String .............................................................................. Interfacing XA with Precompilers and OCI .............................................................................. Using Precompilers with the Oracle XA Library .............................................................. Using OCI with the Oracle XA Library .............................................................................. Managing Transaction Control with XA ................................................................................... Examples of Precompiler Applications .............................................................................. Migrating Precompiler or OCI Applications to TPM Applications....................................... Managing XA Library Thread Safety ......................................................................................... Specifying Threading in the Open String........................................................................... Restrictions on Threading in XA ......................................................................................... Troubleshooting XA Applications ................................................................................................... Accessing XA Trace Files ............................................................................................................. The xa_open() String DbgFl.................................................................................................. Trace File Locations ............................................................................................................... Managing In-Doubt or Pending Transactions .......................................................................... Using SYS Account Tables to Monitor XA Transactions......................................................... XA Issues and Restrictions ................................................................................................................ Using Database Links in XA Applications ................................................................................ Managing Transaction Branches in XA Applications.............................................................. Using XA with Oracle Real Application Clusters .................................................................... Managing Transaction Branches on Oracle Real Application Clusters (RAC)............. Managing Instance Recovery in Real Application Clusters ............................................ xvi 14-28 14-29 14-29 14-35 14-37 14-37 14-38 14-38 14-39 14-39 14-40 15-1 15-2 15-4 15-4 15-5 15-5 15-6 15-6 15-7 15-7 15-7 15-8 15-8 15-10 15-10 15-11 15-12 15-13 15-14 15-15 15-15 15-15 15-15 15-16 15-16 15-17 15-17 15-17 15-18 15-18 15-19 15-19 15-19 15-20 Global Uniqueness of XIDs in Real Application Clusters................................................ SQL-Based XA Restrictions.......................................................................................................... Rollbacks and Commits ........................................................................................................ DDL Statements ..................................................................................................................... Session State............................................................................................................................ EXEC SQL ............................................................................................................................... Miscellaneous Restrictions........................................................................................................... 16 15-21 15-21 15-21 15-22 15-22 15-22 15-22 Developing Applications on the Publish-Subscribe Model Introduction to Publish-Subscribe .................................................................................................... Publish-Subscribe Architecture ......................................................................................................... Publish-Subscribe Concepts ............................................................................................................... Examples of a Publish-Subscribe Mechanism ................................................................................ 16-1 16-2 16-3 16-5 Index xvii xviii Preface The Oracle Database Application Developer's Guide - Fundamentals describes basic application development features of Oracle Database 10g. Information in this guide applies to features that work the same on all supported platforms, and does not include system-specific information. Audience Oracle Database Application Developer's Guide - Fundamentals is intended for programmers developing new applications or converting existing applications to run in the Oracle Database environment. This book will also be valuable to systems analysts, project managers, and others interested in the development of database applications. To use this document, you need a working knowledge of application programming, and that you are acquainted with using the Structured Query Language (SQL) to access information in relational database systems. Some sections of this guide assume a familiar with object-oriented programming. Documentation Accessibility Our goal is to make Oracle products, services, and supporting documentation accessible, with good usability, to the disabled community. To that end, our documentation includes features that make information available to users of assistive technology. This documentation is available in HTML format, and contains markup to facilitate access by the disabled community. Accessibility standards will continue to evolve over time, and Oracle is actively engaged with other market-leading technology vendors to address technical obstacles so that our documentation can be accessible to all of our customers. For more information, visit the Oracle Accessibility Program Web site at http://www.oracle.com/accessibility/ Accessibility of Code Examples in Documentation Screen readers may not always correctly read the code examples in this document. The conventions for writing code require that closing braces should appear on an otherwise empty line; however, some screen readers may not always read a line of text that consists solely of a bracket or brace. xix Accessibility of Links to External Web Sites in Documentation This documentation may contain links to Web sites of other companies or organizations that Oracle does not own or control. Oracle neither evaluates nor makes any representations regarding the accessibility of these Web sites. TTY Access to Oracle Support Services Oracle provides dedicated Text Telephone (TTY) access to Oracle Support Services within the United States of America 24 hours a day, seven days a week. For TTY support, call 800.446.2398. Related Documents For more information, see these Oracle resources: ■ ■ ■ ■ ■ Oracle Database PL/SQL User's Guide and Reference to learn PL/SQL and to get a complete description of the PL/SQL high-level programming language, which is Oracle's procedural extension to SQL. Oracle Call Interface Programmer's Guide and Oracle C++ Call Interface Programmer's Guide to learn about the Oracle Call Interface (OCI). You can use the OCI to build third-generation language (3GL) applications that access the Oracle Database. Oracle Database Security Guide to learn about security features of the database that application developers and database administrators need to be aware of. The Oracle documentation for the Pro* series of precompilers, which allow you to embed SQL and PL/SQL in your programs. If you write 3GL application programs in C, C++, COBOL, or FORTRAN that incorporate embedded SQL, then refer to the corresponding precompiler manual. For example, if you program in C or C++, then refer to the Pro*C/C++ Programmer's Guide. Oracle JDeveloper 10g is an Integrated Development Environment (IDE) for building applications and Web services using the latest industry standards for Java, XML, and SQL. You can access the JDeveloper documentation at the following product page: http://www.oracle.com/technology/products/jdev ■ ■ ■ Oracle Database SQL Reference and Oracle Database Administrator's Guide for SQL information. Oracle Database Concepts for basic Oracle Database concepts Oracle XML Developer's Kit Programmer's Guide and Oracle XML DB Developer's Guide for developing applications that manipulate XML data. Many of the examples in this book use the sample schemas, which are installed by default when you select the Basic Installation option with an Oracle Database installation. Refer to Oracle Database Sample Schemas for information on how these schemas were created and how you can use them yourself. Printed documentation is available for sale in the Oracle Store at http://oraclestore.oracle.com/ To download free release notes, installation documentation, white papers, or other collateral, please visit the Oracle Technology Network (OTN). You must register online before using OTN; registration is free and can be done at http://www.oracle.com/technology/membership/ xx If you already have a username and password for OTN, then you can go directly to the documentation section of the OTN Web site at http://www.oracle.com/technology/documentation/ For additional information, see: ■ ■ ■ ■ ■ ■ Oracle JDeveloper 10g: Empowering J2EE Development by Harshad Oak. Apress, 2004. Oracle JDeveloper 10g Handbook by Avrom Faderman, Peter Koletzke, and Paul Dorsey. Oracle Press, 2004. Oracle PL/SQL Tips and Techniques by Joseph C. Trezzo. Oracle Press, 1999. Oracle PL/SQL Programming by Steven Feuerstein. 3rd Edition. O'Reilly & Associates, 2002. Oracle PL/SQL Developer's Workbook by Steven Feuerstein. O'Reilly & Associates, 2000. Oracle PL/SQL Best Practices by Steven Feuerstein. O'Reilly & Associates, 2001. Conventions The following text conventions are used in this document: Convention Meaning boldface Boldface type indicates graphical user interface elements associated with an action, or terms defined in text or the glossary. italic Italic type indicates book titles, emphasis, or placeholder variables for which you supply particular values. monospace Monospace type indicates commands within a paragraph, URLs, code in examples, text that appears on the screen, or text that you enter. xxi xxii What's New in Application Development? This section describes new features of the Oracle Database 10g Release 2 (10.2) and provides pointers to additional information. New features information from previous releases is also retained to help those users migrating to the current release. The following sections describe the new features in Oracle Oracle Database 10g: ■ Oracle Database 10g Release 2 (10.2) New Features ■ Oracle Database 10g Release 1 (10.1) New Features Oracle Database 10g Release 2 (10.2) New Features This section discusses new features introduced in Oracle Database 10g Release 1 (10.1). ■ Regular expression enhancements This release adds SQL support for common Perl-based extensions that are not included but do not conflict with the POSIX standard. See Also: Chapter 4, "Using Regular Expressions in Oracle Database" for more information ■ Unicode datatype literal enhancement You can avoid data loss if the database character set cannot represent all characters in the client character set. See Also: ■ "Using Character Literals in SQL Statements" on page 3-4 Database Change Notification Database Change Notification enables client applications to receive notifications when the result set of a registered query changes. For example, if the client registers a query of the hr.employees table, and if a user adds an employee, then the application can receive a database change notification when a new row is added to the table. A new query of hr.employees returns the changed result set. Database Change Notification supports both mid-tier caches and server-side stored procedures. For example, a stored procedure that performs DML on registered tables can automatically send change notifications to a mid-tier application, which can keep cached data up to date. See Also: Chapter 13, "Developing Applications with Database Change Notification" xxiii ■ Asynchronous Commit Oracle Database enables you to change the handling of commit redo depending on the needs of your application. You can change the default COMMIT options so that the application does not need to wait for Oracle Database to write data to the online redo logs. See Also: ■ "Managing Commit Redo Behavior" on page 2-3 Automatic Undo Retention Enhancement This feature provides maximum retention for the fixed-size undo tablespace, thereby improving flashback capability as well as other statistic collection queries in which the query length is unknown. See Also: "Database Administration Tasks Before Using Flashback Features" on page 10-3 ■ Failover Improvements for Real Application Clusters (RAC) Distributed Transactions Distributed Transactions in a RAC environment detect failures and start the failover and failback processes automatically. See Also: "Using XA with Oracle Real Application Clusters" on page 15-19 ■ XML DB Web Services Enables direct access to the Oracle database through a native Web Service. Developers can write and deploy web services that can query the database with SQL or XQuery or execute stored procedures. You can access web services through the Oracle XML DB listener. See Also: "Overview of Oracle Database Web Services" on page 1-14 Oracle Database 10g Release 1 (10.1) New Features This section discusses new features introduced in Oracle Database 10g Release 1 (10.1). ■ Regular Expression Support A set of SQL functions introduced in this release let you perform queries and manipulate string data by means of regular expressions. Refer to Chapter 4, "Using Regular Expressions in Oracle Database" for more information. ■ Oracle Expression Filter Oracle Expression Filter lets you store conditional expressions in a column that you can use in the WHERE clause of a database query. Refer to "Representing Conditional Expressions as Data" on page 3-22 for more information. See Also: ■ Oracle Database SQL Reference Native floating-point datatypes Column datatypes BINARY_FLOAT and BINARY_DOUBLE are introduced in this release. These datatypes provide an alternative to using the Oracle NUMBER datatype, with the following benefits: xxiv ■ More efficient use of storage resources ■ Faster arithmetic operations ■ Support for numerical algorithms specified in the IEEE 754 Standard Support for native floating-point datatypes in bind and fetch operations is provided for the following client interfaces: ■ SQL ■ PL/SQL ■ OCI ■ OCCI ■ Pro*C/C++ ■ JDBC See Also: ■ "Representing Numeric Data" on page 3-5 Terabyte-Size Large Object (LOB) support This release provides support for terabyte-size LOB values (from 8 to 128 terabytes) in the following programmatic environments: ■ Java (JDBC) ■ OCI ■ PL/SQL (package DBMS_LOB) You can store and manipulate LOB (BLOB, CLOB, and NCLOB) datatypes larger than 4GB. See Also: ■ For details on terabyte-size LOB support: ■ Oracle Database Application Developer's Guide - Large Objects ■ Oracle Call Interface Programmer's Guide Flashback This release has new and enhanced flashback features. You can now do the following: ■ ■ ■ Query the transaction history of a row. Obtain the SQL undo syntax for a row and perform row-level flashback operations. Perform remote queries of past data. See Also: ■ Chapter 10, "Developing Flashback Applications" Oracle Data Provider for .NET Oracle Data Provider for .NET (ODP.NET) is a new programmatic environment that implements a data provider for Oracle Database. It uses APIs native to Oracle Database to offer fast and reliable access from any .NET application to database features and data. ODP.NET also uses and inherits classes and interfaces available in the Microsoft .NET Framework Class Library. See Also: Oracle Data Provider for .NET Developer's Guide xxv xxvi 1 Orientation to Oracle Programmatic Environments This chapter contains these topics: ■ Overview of Oracle Application Development ■ Overview of PL/SQL ■ Overview of Java Support Built Into the Database ■ Overview of Pro*C/C++ ■ Overview of Pro*COBOL ■ Overview of OCI and OCCI ■ Overview of Oracle Data Provider for .NET (ODP.NET) ■ Overview of Oracle Objects for OLE (OO4O) ■ Choosing a Programming Environment Overview of Oracle Application Development As an application developer, you have many choices when writing a program to interact with an Oracle database. Client/Server Model In a traditional client/server program, the code of your application runs on a machine other than the database server. Database calls are transmitted from this client machine to the database server. Data is transmitted from the client to the server for insert and update operations and returned from the server to the client for query operations. The data is processed on the client machine. Client/server programs are typically written by using precompilers, whereas SQL statements are embedded within the code of another language such as C, C++, or COBOL. Server-Side Coding You can develop application logic that resides entirely inside the database by using triggers that are executed automatically when changes occur in the database or stored procedures that are called explicitly. Off-loading the work from your application lets you reuse code that performs verification and cleanup and control database operations from a variety of clients. For example, by making stored procedures callable through a Web server, you can construct a Web-based user interface that performs the same functions as a client/server application. Orientation to Oracle Programmatic Environments 1-1 Overview of PL/SQL Two-Tier Versus Three-Tier Models Client/server computing is often referred to as a two-tier model: your application communicates directly with the database server. In the three-tier model, a separate application server processes the requests. The application server might be a basic Web server, or might perform advanced functions like caching and load-balancing. Increasing the processing power of this middle tier lets you lessen the resources needed by client systems, resulting in a thin client configuration in which the client machine might need only a Web browser or other means of sending requests over the TCP/IP or HTTP protocols. User Interface The interface that your application displays to end users depends on the technology behind the application as well as the needs of the users themselves. Experienced users can enter SQL commands that are passed on to the database. Novice users can be shown a graphical user interface that uses the graphics libraries of the client system (such as Windows or X-Windows). Any of these traditional user interfaces can also be provided in a Web browser by means of HTML and Java. Stateful Versus Stateless User Interfaces In traditional client/server applications, the application can keep a record of user actions and use this information over the course of one or more sessions. For example, past choices can be presented in a menu so that they do not have to be entered again. When the application is able to save information in this way, the application is considered stateful. Web or thin-client applications that are stateless are easier to develop. Stateless applications gather all the required information, process it using the database, and then start over with the next user. This is a popular way to process single-screen requests such as customer registration. There are many ways to add stateful behavior to Web applications that are stateless by default. For example, an entry form on one Web page can pass information to subsequent Web pages, allowing you to construct a wizard-like interface that remembers the user's choices through several different steps. Cookies can be used to store small items of information on the client machine, and retrieve them when the user returns to a Web site. Servlets can be used to keep a database session open and store variables between requests from the same client. Overview of PL/SQL This section contains the following topics: ■ What Is PL/SQL? ■ Advantages of PL/SQL ■ PL/SQL Web Development Tools What Is PL/SQL? PL/SQL is Oracle's procedural extension to SQL, the standard database access language. It is an advanced 4GL (fourth-generation programming language), which means that it is an application-specific language. PL/SQL and SQL have built-in treatment of the relational database domain. In PL/SQL, you can manipulate data with SQL statements and control program flow with procedural constructs such as loops. You can also do the following: 1-2 Oracle Database Application Developer’s Guide - Fundamentals Overview of PL/SQL ■ Declare constants and variables ■ Define procedures and functions ■ Use collections and object types ■ Trap runtime errors Applications written in any of the Oracle programmatic interfaces can call PL/SQL stored procedures and send blocks of PL/SQL code to Oracle Database for execution. 3GL applications can access PL/SQL scalar and composite datatypes through host variables and implicit datatype conversion. A 3GL language is easier than assembler language for a human to understand and includes features such as named variables. Unlike 4GL, it is not specific to an application domain. Example 1–1 provides an example of a simple PL/SQL procedure. The procedure debit_account withdraws money from a bank account. It accepts an account number and an amount of money as parameters. It uses the account number to retrieve the account balance from the database, then computes the new balance. If this new balance is less than zero, then the procedure jumps to an error routine; otherwise, it updates the bank account. Example 1–1 Simple PL/SQL Example PROCEDURE debit_account (p_acct_id INTEGER, p_debit_amount REAL) IS v_old_balance REAL; v_new_balance REAL; e_overdrawn EXCEPTION; BEGIN SELECT bal INTO v_old_balance FROM accts WHERE acct_no = p_acct_id; v_new_balance := v_old_balance - p_debit_amount; IF v_new_balance < 0 THEN RAISE e_overdrawn; ELSE UPDATE accts SET bal = v_new_balance WHERE acct_no = p_acct_id; END IF; COMMIT; EXCEPTION WHEN e_overdrawn THEN -- handle the error END debit_account; See Also: ■ Oracle Database PL/SQL User's Guide and Reference ■ Oracle Database SQL Reference Advantages of PL/SQL PL/SQL is a portable, high-performance transaction processing language with the following advantages: ■ Integration with Oracle Database ■ High Performance ■ High Productivity Orientation to Oracle Programmatic Environments 1-3 Overview of PL/SQL ■ Scalability ■ Manageability ■ Object-Oriented Programming Support ■ Portability ■ Security ■ Built-In Packages Integration with Oracle Database PL/SQL enables you use all of the Oracle Database SQL data manipulation, cursor control, and transaction control statements. PL/SQL also supports the SQL functions, operators, and pseudocolumns. So, you can manipulate data in Oracle Database flexibly and safely. PL/SQL supports all SQL datatypes. Combined with the direct access that SQL provides, these shared datatypes integrate PL/SQL with the Oracle Database data dictionary. PL/SQL supports Dynamic SQL, which is a programming technique that enables you to build and process SQL statements "on the fly" at run time. It gives PL/SQL flexibility comparable to scripting languages such as Perl, Korn shell, and Tcl. The %TYPE and %ROWTYPE attributes enable your code to adapt as table definitions change. For example, the %TYPE attribute declares a variable based on the type of a database column. If the column datatype changes, then the variable uses the correct type at runtime. This provides data independence and reduces maintenance costs. High Performance If your application is database intensive, then you can use PL/SQL blocks to group SQL statements before sending them to Oracle Database for execution. This coding strategy can drastically reduce the communication overhead between your application and Oracle Database. PL/SQL stored procedures are compiled once and stored in executable form, so procedure calls are quick and efficient. A single call can start a compute-intensive stored procedure, reducing network traffic and improving round-trip response times. Executable code is automatically cached and shared among users, lowering memory requirements and invocation overhead. High Productivity PL/SQL adds procedural capabilities such as Oracle Forms and Oracle Reports. For example, you can use an entire PL/SQL block in an Oracle Forms trigger instead of multiple trigger steps, macros, or user exits. PL/SQL is the same in all environments. As soon as you master PL/SQL with one Oracle tool, you can transfer your knowledge to others, and so multiply the productivity gains. For example, scripts written with one tool can be used by other tools. Scalability PL/SQL stored procedures increase scalability by centralizing application processing on the server. Automatic dependency tracking helps you develop scalable applications. 1-4 Oracle Database Application Developer’s Guide - Fundamentals Overview of PL/SQL The shared memory facilities of the shared server enable Oracle Database to support many thousands of concurrent users on a single node. For more scalability, you can use the Oracle Connection Manager to multiplex network connections. Manageability After being validated, you can use a PL/SQL stored procedure in any number of applications. If its definition changes, then only the procedure is affected, not the applications that call it. This simplifies maintenance and enhancement. Also, maintaining a procedure on the Oracle Database is easier than maintaining copies on various client machines. Object-Oriented Programming Support You can use PL/SQL object types and collections for object-oriented programming. Object Types An object type is a user-defined composite datatype that encapsulates a data structure along with the functions and procedures needed to manipulate the data. The variables that form the data structure are called attributes. The functions and procedures that characterize the behavior of the object type are called methods, which you can implement in PL/SQL. Object types are an ideal object-oriented modeling tool, which you can use to reduce the cost and time required to build complex applications. Besides allowing you to create software components that are modular, maintainable, and reusable, object types allow different teams of programmers to develop software components concurrently. Collections A collection is an ordered group of elements, all of the same type (for example, the grades for a class of students). Each element has a unique subscript that determines its position in the collection. PL/SQL offers two kinds of collections: nested tables and varrays (variable-size arrays). Collections work like the set, queue, stack, and hash table data structures found in most third-generation programming languages. Collections can store instances of an object type and can also be attributes of an object type. Collections can be passed as parameters. So, you can use them to move columns of data into and out of database tables or between client-side applications and stored subprograms. You can define collection types in a PL/SQL package, then use the same types across many applications. Portability Applications written in PL/SQL can run on any operating system and hardware platform on which Oracle Database runs. You can write portable program libraries and reuse them in different environments. Security PL/SQL stored procedures enable you to divide application logic between the client and the server, which prevents client applications from manipulating sensitive Oracle Database data. Database triggers written in PL/SQL can prevent applications from making specified updates and can audit user queries. You can restrict access to Oracle Database data by allowing users to manipulate it only through stored procedures that have a restricted set of privileges. For example, you can grant users access to a procedure that updates a table but not grant them access to the table itself. Orientation to Oracle Programmatic Environments 1-5 Overview of Java Support Built Into the Database See Also: Oracle Database Security Guide for details on database security features Built-In Packages A package is an encapsulated collection of related program objects stored together in the database. Program objects are procedures, functions, variables, constants, cursors, and exceptions. The following packages are especially useful in application development for Oracle Database: ■ DBMS_PIPE is used to communicate between sessions. ■ DBMS_ALERT is used to broadcast alerts to users. ■ DBMS_LOCK and DBMS_TRANSACTION are used for lock and transaction management. ■ DBMS_AQ is used for Advanced Queuing. ■ DBMS_LOB is used to manipulate large objects. ■ DBMS_ROWID is used for employing ROWID values. ■ UTL_RAW is the RAW facility. ■ UTL_REF is for work with REF values. The following packages are useful for server manageability: ■ ■ ■ DBMS_SESSION is for session management by DBAs. DBMS_SPACE and DBMS_SHARED_POOL provide space information and reserve shared pool resources. DBMS_JOB is used to schedule jobs in the server. PL/SQL Web Development Tools Oracle Database provides built-in tools and technologies that enable you to deploy PL/SQL applications over the Web. Thus, PL/SQL serves as an alternative to Web application frameworks such as CGI. The PL/SQL Web Toolkit is a set of PL/SQL packages that you can use to develop stored procedures that can be invoked by a Web client. The PL/SQL Gateway enables an HTTP client to invoke a PL/SQL stored procedure through mod_plsql, which is a plug-in to Oracle HTTP Server. This module performs the following actions: 1. Translates a URL passed by a browser client 2. Calls an Oracle Database stored procedure with the parameters in the URL 3. Returns output (typically HTML) to the client See Also: Chapter 11, "Developing Applications with the PL/SQL Web Toolkit" to learn how to use PL/SQL in Web development Overview of Java Support Built Into the Database This section provides an overview of built-in database features that support Java applications. The database includes the core JDK libraries such as java.lang, java.io, and so on. The database supports client-side Java standards such as JDBC 1-6 Oracle Database Application Developer’s Guide - Fundamentals Overview of Java Support Built Into the Database and SQLJ, and provides server-side JDBC and SQLJ drivers that allow data-intensive Java code to run within the database. This section contains the following topics: ■ Overview of Oracle JVM ■ Overview of Oracle Extensions to JDBC ■ Overview of Oracle SQLJ ■ Overview of Oracle JPublisher ■ Overview of Java Stored Procedures ■ Overview of Oracle Database Web Services ■ Overview of Writing Procedures and Functions in Java See Also: ■ Oracle Database Concepts for background information about Java and how the database supports it ■ Oracle Database Java Developer's Guide ■ Oracle Database JDBC Developer's Guide and Reference ■ Oracle Database JPublisher User's Guide Overview of Oracle JVM Oracle JVM, the Java Virtual Machine provided with the Oracle Database, is compliant with the J2SE version 1.4.x specification and supports the database session architecture. Any database session can activate a dedicated JVM. All sessions share the same JVM code and statics; however, private states for any given session are held, and subsequently garbage collected, in an individual session space. This design provides the following benefits: ■ Java applications have the same session isolation and data integrity as SQL operations. ■ There is no need to run Java in a separate process for data integrity. ■ Oracle JVM is a robust JVM with a small memory footprint. ■ The JVM has the same linear Symmetric Multiprocessing (SMP) scalability as the database and can support thousands of concurrent Java sessions. Oracle JVM works consistently with every platform supported by Oracle Database. Java applications that you develop with Oracle JVM can easily be ported to any supported platform. Oracle JVM includes a deployment-time native compiler that enables Java code to be compiled once, stored in executable form, shared among users, and invoked more quickly and efficiently. Security features of the database are also available with Oracle JVM. Java classes must be loaded in a database schema (by using Oracle JDeveloper, a third-party IDE, SQL*Plus, or the loadjava utility) before they can be invoked. Java class invocation is secured and controlled through database authentication and authorization, Java 2 security, and invoker's or definer's rights. Orientation to Oracle Programmatic Environments 1-7 Overview of Java Support Built Into the Database Overview of Oracle Extensions to JDBC JDBC (Java Database Connectivity) is an API (Applications Programming Interface) that allows Java to send SQL statements to an object-relational database such as Oracle Database. The JDBC standard defines four types of JDBC drivers: ■ ■ ■ ■ Type 1. A JDBC-ODBC bridge. Software must be installed on client systems. Type 2. Native methods (calls C or C++) and Java methods. Software must be installed on the client. Type 3. Pure Java. The client uses sockets to call middleware on the server. Type 4. The most pure Java solution. Talks directly to the database by using Java sockets. JDBC is based on the X/Open SQL Call Level Interface, and complies with the SQL92 Entry Level standard. You can use JDBC to do dynamic SQL. In dynamic SQL, the embedded SQL statement to be executed is not known before the application is run and requires input to build the statement. The drivers that are implemented by Oracle have extensions to the capabilities in the JDBC standard that was defined by Sun Microsystems. Oracle's implementations of JDBC drivers are described in the following sections. Oracle Database support of and extensions to various levels of the JDBC standard are described in "Oracle Database Extensions to JDBC Standards" on page 1-9. JDBC Thin Driver The JDBC thin driver is a Type 4 (100% pure Java) driver that uses Java sockets to connect directly to a database server. It has its own implementation of a Two-Task Common (TTC), a lightweight implementation of TCP/IP from Oracle Net. It is written entirely in Java and is therefore platform-independent. The thin driver does not require Oracle software on the client side. It does need a TCP/IP listener on the server side. Use this driver in Java applets that are downloaded into a Web browser or in applications for which you do not want to install Oracle client software. The thin driver is self-contained, but it opens a Java socket, and thus can only run in a browser that supports sockets. JDBC OCI Driver The OCI driver is a Type 2 JDBC driver. It makes calls to the OCI (Oracle Call Interface) written in C to interact with Oracle Database, thus using native and Java methods. The OCI driver allows access to more features than the thin driver, such as Transparent Application Fail-Over, advanced security, and advanced LOB manipulation. The OCI driver provides the highest compatibility between different Oracle Database versions. It also supports all installed Oracle Net adapters, including IPC, named pipes, TCP/IP, and IPX/SPX. Because it uses native methods (a combination of Java and C) the OCI driver is platform-specific. It requires a client installation of version Oracle8i or later including Oracle Net, OCI libraries, CORE libraries, and all other dependent files. The OCI driver usually executes faster than the thin driver. 1-8 Oracle Database Application Developer’s Guide - Fundamentals Overview of Java Support Built Into the Database The OCI driver is not appropriate for Java applets, because it uses a C library that is platform-specific and cannot be downloaded into a Web browser. It is usable in J2EE components running in middle-tier application servers, such as Oracle Application Server. Oracle Application Server provides middleware services and tools that support access between applications and browsers. JDBC Server-Side Internal Driver The JDBC server-side internal driver is a Type 2 driver that runs inside the database server, reducing the number of round-trips needed to access large amounts of data. The driver, the Java server VM, the database, the Java native compiler which speeds execution by as much as 10 times, and the SQL engine all run within the same address space. This driver provides server-side support for any Java program used in the database: SQLJ stored procedures, functions, triggers, and Java stored procedures. You can also call PL/SQL stored procedures, functions, and triggers. The server driver fully supports the same features and extensions as the client-side drivers. Oracle Database Extensions to JDBC Standards Oracle Database includes the following extensions to the JDBC 1.22 standard: ■ Support for Oracle datatypes ■ Performance enhancement by row prefetching ■ Performance enhancement by execution batching ■ Specification of query column types to save round-trips ■ Control of DatabaseMetaData calls Oracle Database supports all APIs from the JDBC 2.0 standard, including the core APIs, optional packages, and numerous extensions. Some of the highlights include datasources, JTA, and distributed transactions. Oracle Database supports the following features from the JDBC 3.0 standard: ■ Support for JDK 1.4. ■ Toggling between local and global transactions. ■ Transaction savepoints. ■ Reuse of prepared statements by connection pools. Sample JDBC 2.0 Program The following example shows the recommended technique for looking up a data source using JNDI in JDBC 2.0: // import the JDBC packages import java.sql.*; import javax.sql.*; import oracle.jdbc.pool.*; ... InitialContext ictx = new InitialContext(); DataSource ds = (DataSource)ictx.lookup("jdbc/OracleDS"); Connection conn = ds.getConnection(); Statement stmt = conn.createStatement(); ResultSet rs = stmt.executeQuery("SELECT last_name FROM employees"); Orientation to Oracle Programmatic Environments 1-9 Overview of Java Support Built Into the Database while ( rs.next() ) { out.println( rs.getString("ename") + "
"); } conn.close(); Sample Pre-2.0 JDBC Program The following source code registers an Oracle JDBC thin driver, connects to the database, creates a Statement object, executes a query, and processes the result set. The SELECT statement retrieves and lists the contents of the last_name column of the hr.employees table. import import import import java.sql.* java.math.* java.io.* java.awt.* class JdbcTest { public static void main (String args []) throws SQLException { // Load Oracle driver DriverManager.registerDriver (new oracle.jdbc.OracleDriver()); // Connect to the local database Connection conn = DriverManager.getConnection ("jdbc:oracle:thin:@myhost:1521:orcl", "hr", "hr"); // Query the employee names Statement stmt = conn.createStatement (); ResultSet rset = stmt.executeQuery ("SELECT last_name FROM employees"); // Print the name out while (rset.next ()) System.out.println (rset.getString (1)); // Close the result set, statement, and the connection rset.close(); stmt.close(); conn.close(); } } One Oracle Database extension to the JDBC drivers is a form of the getConnection() method that uses a Properties object. The Properties object lets you specify user, password, and database information as well as row prefetching and execution batching. To use the OCI driver in this code, replace the Connection statement with the following, where MyHostString is an entry in the tnsnames.ora file: Connection conn = DriverManager.getConnection ("jdbc:oracle:oci8:@MyHostString", "hr", "hr"); If you are creating an applet, then the getConnection() and registerDriver() strings will be different. JDBC in SQLJ Applications JDBC code and SQLJ code (see "Overview of Oracle SQLJ" on page 1-11) interoperate, allowing dynamic SQL statements in JDBC to be used with both static and dynamic SQL statements in SQLJ. A SQLJ iterator class corresponds to the JDBC result set. 1-10 Oracle Database Application Developer’s Guide - Fundamentals Overview of Java Support Built Into the Database See Also: Oracle Database JDBC Developer's Guide and Reference for more information on JDBC Overview of Oracle SQLJ SQLJ is an ANSI SQL-1999 standard for embedding SQL statements in Java source code. SQLJ provides a simpler alternative to JDBC for both client-side and server-side SQL data access from Java. A SQLJ source file contains Java source with embedded SQL statements. Oracle SQLJ supports dynamic as well as static SQL. Support for dynamic SQL is an Oracle extension to the SQLJ standard. The term "SQLJ," when used in this manual, refers to the Oracle SQLJ implementation, including Oracle SQLJ extensions. Note: Oracle Database provides a translator and a run time driver to support SQLJ. The SQLJ translator is 100% pure Java and is portable to any JVM that is compliant with JDK version 1.1 or higher. The Oracle SQLJ translator performs the following tasks: ■ ■ ■ Translates SQLJ source to Java code with calls to the SQLJ run time driver. The SQLJ translator converts the source code to pure Java source code and can check the syntax and semantics of static SQL statements against a database schema and verify the type compatibility of host variables with SQL types. Compiles the generated Java code with the Java compiler. (Optional) Creates profiles for the target database. SQLJ generates "profile" files with customization specific to Oracle Database. Oracle Database supports SQLJ stored procedures, functions, and triggers which execute in the Oracle JVM. SQLJ is integrated with JDeveloper. Source-level debugging support for SQLJ is available in JDeveloper. The following is an example of a simple SQLJ executable statement, which returns one value because employee_id is unique in the employee table: String name; #sql { SELECT first_name INTO :name FROM employees WHERE employee_id=112 }; System.out.println("Name is " + name + ", employee number = " + employee_id); Each host variable (or qualified name or complex Java host expression) included in a SQL expression is preceded by a colon (:). Other SQLJ statements declare Java types. For example, you can declare an iterator (a construct related to a database cursor) for queries that retrieve many values, as follows: #sql iterator EmpIter (String EmpNam, int EmpNumb); See Also: ■ ■ For more examples and details on Oracle SQLJ syntax: Oracle Database JPublisher User's Guide Sample SQLJ code available on the Oracle Technology Network Web site: http://www.oracle.com/technology/ Orientation to Oracle Programmatic Environments 1-11 Overview of Java Support Built Into the Database Benefits of SQLJ Oracle SQLJ extensions to Java allow rapid development and easy maintenance of applications that perform database operations through embedded SQL. In particular, Oracle SQLJ does the following: ■ ■ ■ ■ ■ ■ ■ ■ Provides a concise, legible mechanism for database access from static SQL. Most SQL in applications is static. SQLJ provides more concise and less error-prone static SQL constructs than JDBC does. Provides an SQL Checker module for verification of syntax and semantics at translate time. Provides flexible deployment configurations, which makes it possible to implement SQLJ on the client, server, or middle tier. Supports a software standard. SQLJ is an effort of a group of vendors and will be supported by all of them. Applications can access multiple database vendors. Provides source code portability. Executables can be used with all of the vendor DBMSs if the code does not rely on any vendor-specific features. Enforces a uniform programming style for the clients and the servers. Integrates the SQLJ translator with Oracle JDeveloper, a graphical IDE that provides SQLJ translation, Java compilation, profile customizing, and debugging at the source code level, all in one step. Includes Oracle type extensions. Datatypes supported include: LOB datatypes, ROWID, REF CURSOR, VARRAY, nested table, user-defined object types, RAW, and NUMBER. Comparing SQLJ with JDBC JDBC provides a complete dynamic SQL interface from Java to databases. It gives developers full control over database operations. SQLJ simplifies Java database programming to improve development productivity. JDBC provides fine-grained control of the execution of dynamic SQL from Java, whereas SQLJ provides a higher-level binding to SQL operations in a specific database schema. Following are some differences between JDBC and SQLJ: ■ ■ ■ ■ ■ SQLJ source code is more concise than equivalent JDBC source code. SQLJ uses database connections to type-check static SQL code. JDBC, being a completely dynamic API, does not. SQLJ provides strong typing of query outputs and return parameters and allows type-checking on calls. JDBC passes values to and from SQL without compile-time type checking. SQLJ programs allow direct embedding of Java bind expressions within SQL statements. JDBC requires a separate get or set statement for each bind variable and specifies the binding by position number. SQLJ provides simplified rules for calling SQL stored procedures and functions. For example, the following JDBC excerpt requires a generic call to a stored procedure or function, in this case fun, to have the following syntax. (This example shows SQL92 and Oracle JDBC syntaxes. Both are allowed.) prepStmt.prepareCall("{call fun(?,?)}"); //stored procedure SQL92 prepStmt.prepareCall("{? = call fun(?,?)}"); //stored function SQL92 prepStmt.prepareCall("begin fun(:1,:2);end;"); //stored procedure Oracle prepStmt.prepareCall("begin :1 := fun(:2,:3);end;");//stored func Oracle 1-12 Oracle Database Application Developer’s Guide - Fundamentals Overview of Java Support Built Into the Database Following is the SQLJ equivalent: #sql {call fun(param_list) }; //Stored procedure // Declare x ... #sql x = {VALUES(fun(param_list)) }; // Stored function // where VALUES is the SQL construct The following benefits are common to SQLJ and JDBC: ■ ■ ■ SQLJ source files can contain JDBC calls. SQLJ and JDBC are interoperable. Oracle JPublisher generates custom Java classes to be used in your SQLJ or JDBC application for mappings to Oracle object types and collections. Java and PL/SQL stored procedures can be used interchangeably. SQLJ Stored Procedures in the Server SQLJ applications can be stored and executed in the server by using the following techniques: ■ ■ Translate, compile, and customize the SQLJ source code on a client and load the generated classes and resources into the server with the loadjava utility. The classes are typically stored in a Java archive (.jar) file. Load the SQLJ source code into the server, also using loadjava, where it is translated and compiled by the server's embedded translator. See Also: Oracle Database JPublisher User's Guide for more information on using stored procedures with Oracle SQLJ Overview of Oracle JPublisher Oracle JPublisher is a code generator that automates the process of creating database-centric Java classes by hand. Oracle JPublisher is a client-side utility and is built into the database system. You can run Oracle JPublisher from the command line or directly from the Oracle JDeveloper IDE. Oracle JPublisher inspects PL/SQL packages and database object types such as SQL object types, VARRAY types, and nested table types, and then generates a Java class that is a wrapper around the PL/SQL package with corresponding fields and methods. The generated Java class can be incorporated and used by Java clients or J2EE components to exchange and transfer object type instances to and from the database transparently. See Also: Oracle Database JPublisher User's Guide Overview of Java Stored Procedures Java stored procedures enable you to implement programs that run in the database server and which are independent of programs that run in the middle tier. Structuring applications in this way reduces complexity and increases reuse, security, performance, and scalability. For example, you can create a Java stored procedure that performs operations that require data persistence and a separate program to perform presentation or business logic operations. Orientation to Oracle Programmatic Environments 1-13 Overview of Java Support Built Into the Database Java stored procedures interface with SQL by using a similar execution model as PL/SQL. See Also: Oracle Database Java Developer's Guide Overview of Oracle Database Web Services Web services represent a distributed computing paradigm for Java application development that is an alternative to earlier Java protocols such as JDBC. It allows application-to-application interaction by means of the XML and Web protocols. For example, an electronics parts vendor can provide a Web-based programmatic interface to its suppliers for inventory management. The vendor can call a Web service as part of a program and automatically order new stock based on the data returned. The key technologies used in Web services are: ■ ■ ■ Web Services Description Language (WSDL), which is a standard format for creating an XML document. WSDL describes what a web service can do, where it resides, and how to invoke it. Specifically, it describes the operations and parameters, including parameter types, provided by a Web service. In addition, a WSDL document describes the location, the transport protocol, and the invocation style for the Web service. Simple Object Access Protocol (SOAP) messaging, which is an XML-based message protocol used by Web services. SOAP does not prescribe a specific transport mechanism such as HTTP, FTP, SMTP, or JMS; however, most Web services accept messages that use HTTP or HTTPS. Universal Description, Discovery, and Integration (UDDI) business registry, which is a directory that lists Web services on the internet. The UDDI registry is often compared to a telephone directory, listing unique identifiers (white pages), business categories (yellow pages), and instructions for binding to a service protocol (green pages). Web services can use a variety of techniques and protocols. For example: ■ ■ ■ Dispatching can occur in a synchronous (typical) or asynchronous manner. You can perform invocation in an RPC-style operation in which arguments are sent and a response returned, or in a message style such as a one-way SOAP document exchange. You can use different encoding rules: literal or encoded. You can invoke a Web service statically, in which case you may know everything about it beforehand, or dynamically, in which case you can discover its operations and transport endpoints on the fly. Oracle Database as a Web Service Provider Oracle Database can function as either a Web service provider or as a Web service consumer. When used as a provider, the database enables sharing and disconnected access to stored procedures, data, metadata, and other database resources such as the queuing and messaging systems. As a Web service provider, Oracle Database provides a disconnected and heterogeneous environment that: ■ ■ Exposes stored procedures independently of the language in which the procedures are written Exposes SQL Queries and XQuery 1-14 Oracle Database Application Developer’s Guide - Fundamentals Overview of Pro*C/C++ Overview of Writing Procedures and Functions in Java Functions and procedures are named blocks that encapsulate a sequence of statements. They are like building blocks that you can use to construct modular, maintainable applications. You write these named blocks and then define them by using the loadjava command or SQL CREATE FUNCTION, CREATE PROCEDURE, or CREATE PACKAGE statements. These Java methods can accept arguments and are callable from: ■ SQL CALL statements ■ Embedded SQL CALL statements ■ PL/SQL blocks, subprograms, and packages ■ DML statements (INSERT, UPDATE, DELETE, and SELECT) ■ Oracle development tools such as OCI, Pro*C/C++, and Oracle Developer ■ ■ Oracle Java interfaces such as JDBC, SQLJ statements, CORBA, and Enterprise Java Beans Method calls from object types Overview of Writing Database Triggers in Java A database trigger is a stored procedure that Oracle Database invokes ("fires") automatically when certain events occur, for example, when a DML operation modifies a certain table. Triggers enforce business rules, prevent incorrect values from being stored, and reduce the need to perform checking and cleanup operations in each application. Why Use Java for Stored Procedures and Triggers? ■ ■ ■ ■ ■ ■ ■ Stored procedures and triggers are compiled once, are easy to use and maintain, and require less memory and computing overhead. Network bottlenecks are avoided, and response time is improved. Distributed applications are easier to build and use. Computation-bound procedures run faster in the server. Data access can be controlled by letting users have only stored procedures and triggers that execute with their definer's privileges instead of invoker's rights. PL/SQL and Java stored procedures can call each other. Java in the server follows the Java language specification and can use the SQLJ standard, so that databases other than Oracle Database are also supported. Stored procedures and triggers can be reused in different applications as well as different geographic sites. Overview of Pro*C/C++ The Pro*C/C++ precompiler is a software tool that allows the programmer to embed SQL statements in a C or C++ source file. Pro*C/C++ reads the source file as input and outputs a C or C++ source file that replaces the embedded SQL statements with Oracle runtime library calls and is then compiled by the C or C++ compiler. When there are errors found during the precompilation or the subsequent compilation, modify your precompiler input file and re-run the two steps. Orientation to Oracle Programmatic Environments 1-15 Overview of Pro*C/C++ Implementing a Pro*C/C++ Application The following is a simple code fragment from a C source file that queries the table employees in the schema hr: ... #define UNAME_LEN 10 ... int emp_number; /* Define a host structure for the output values of a SELECT statement. */ /* No declare section needed if precompiler option MODE=ORACLE */ struct { VARCHAR last_name[UNAME_LEN]; float salary; float commission_pct; } emprec; /* Define an indicator structure to correspond to the host output structure. */ struct { short emp_name_ind; short sal_ind; short comm_ind; } emprec_ind; ... /* Select columns last_name, salary, and commission_pct given the user's input /* for employee_id. */ EXEC SQL SELECT last_name, salary, commission_pct INTO :emprec INDICATOR :emprec_ind FROM employees WHERE employee_id = :emp_number; ... The embedded SELECT statement is only slightly different from an interactive (SQL*Plus) SELECT statement. Every embedded SQL statement begins with EXEC SQL. The colon (:), precedes every host (C) variable. The returned values of data and indicators (set when the data value is NULL or character columns have been truncated) can be stored in structs (such as in the preceding code fragment), in arrays, or in arrays of structs. Multiple result set values are handled very simply in a manner that resembles the case shown, where there is only one result, because of the unique employee number. You use the actual names of columns and tables in embedded SQL. Use the default precompiler option values, or you can enter values which give you control over the use of resources, how errors are reported, the formatting of output, and how cursors (which correspond to a particular connection or SQL statement) are managed. Cursors are used when there are multiple result set values. Enter the options either in a configuration file, on the command line, or in-line inside your source code with a special statement that begins with EXEC ORACLE. If there are no errors found, you can then compile, link, and execute the output source file, like any other C program that you write. Use the precompiler to create server database access from clients that can be on many different platforms. Pro*C/C++ allows you the freedom to design your own user interfaces and to add database access to existing applications. Before writing your embedded SQL statements, you may want to test interactive versions of the SQL in SQL*Plus. You then make only minor changes to start testing your embedded SQL application. 1-16 Oracle Database Application Developer’s Guide - Fundamentals Overview of Pro*C/C++ Highlights of Pro*C/C++ Features The following is a short subset of the capabilities of Pro*C/C++. For complete details, refer to the Pro*C/C++ Precompiler Programmer's Guide. ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ You can write your application in either C or C++. You can write multithreaded programs if your platform supports a threads package. Concurrent connections are supported in either single-threaded or multithreaded applications. You can improve performance by embedding PL/SQL blocks. These blocks can call functions or procedures in Java or PL/SQL that are written by you or provided in Oracle Database packages. Using precompiler options, you can check the syntax and semantics of your SQL or PL/SQL statements during precompilation, as well as at runtime. You can call stored PL/SQL and Java subprograms. Modules written in COBOL or in C can be called from Pro*C/C++. External C procedures in shared libraries are callable by your program. You can conditionally precompile sections of your code so that they can execute in different environments. You can use arrays, or structures, or arrays of structures as host and indicator variables in your code to improve performance. You can deal with errors and warnings so that data integrity is guaranteed. As a programmer, you control how errors are handled. Your program can convert between internal datatypes and C language datatypes. The Oracle Call Interface (OCI) and Oracle C++ Call Interface (OCCI), lower-level C and C++ interfaces, are available for use in your precompiler source. Pro*C/C++ supports dynamic SQL, a technique that allows users to input variable values and statement syntax. Pro*C/C++ can use special SQL statements to manipulate tables containing user-defined object types. An Object Type Translator (OTT) will map the object types and named collection types in your database to structures and headers that you will then include in your source. Two kinds of collection types, nested tables and VARRAY, are supported with a set of SQL statements that allow a high degree of control over data. Large Objects are accessed by another set of SQL statements. A new ANSI SQL standard for dynamic SQL is supported for new applications, so that you can execute SQL statements with a varying number of host variables. An older technique for dynamic SQL is still usable by pre-existing applications. Globalization support lets you use multibyte characters and UCS2 Unicode data. Using scrollable cursors, you can move backward and forward through a result set. For example, you can fetch the last row of the result set, or jump forward or backward to an absolute or relative position within the result set. A connection pool is a group of physical connections to a database that can be shared by several named connections. Enabling the connection pool option can help to optimize the performance of Pro*C/C++ application. The connection pool option is not enabled by default. Orientation to Oracle Programmatic Environments 1-17 Overview of Pro*COBOL Overview of Pro*COBOL The Pro*COBOL precompiler is a software tool that allows the programmer to embed SQL statements in a COBOL source code file. Pro*COBOL reads the source file as input and outputs a COBOL source file that replaces the embedded SQL statements with Oracle Database runtime library calls, and is then compiled by the COBOL compiler. When there are errors found during the precompilation or the subsequent compilation, modify your precompiler input file and re-run the two steps. How You Implement a Pro*COBOL Application Here is a simple code fragment from a source file that queries the table employees in the schema hr: ... WORKING-STORAGE SECTION. * * DEFINE HOST INPUT AND OUTPUT HOST AND INDICATOR VARIABLES. * NO DECLARE SECTION NEEDED IF MODE=ORACLE. * 01 EMP-REC-VARS. 05 EMP-NAME PIC X(10) VARYING. 05 EMP-NUMBER PIC S9(4) COMP VALUE ZERO. 05 SALARY PIC S9(5)V99 COMP-3 VALUE ZERO. 05 COMMISSION PIC S9(5)V99 COMP-3 VALUE ZERO. 05 COMM-IND PIC S9(4) COMP VALUE ZERO. ... PROCEDURE DIVISION. ... EXEC SQL SELECT last_name, salary, commission_pct INTO :EMP-NAME, :SALARY, :COMMISSION:COMM-IND FROM employees WHERE employee_id = :EMP-NUMBER END-EXEC. ... The embedded SELECT statement is only slightly different from an interactive (SQL*Plus) SELECT statement. Every embedded SQL statement begins with EXEC SQL. The colon (:) precedes every host (COBOL) variable. The SQL statement is terminated by END-EXEC. The returned values of data and indicators (set when the data value is NULL or character columns have been truncated) can be stored in group items (such as in the preceding code fragment), in tables, or in tables of group items. Multiple result set values are handled very simply in a manner that resembles the case shown, where there is only one result, given the unique employee number. You use the actual names of columns and tables in embedded SQL. Use the default precompiler option values, or enter values that give you control over the use of resources, how errors are reported, the formatting of output, and how cursors are managed (cursors correspond to a particular connection or SQL statement). Enter the options in a configuration file, on the command line, or in-line inside your source code with a special statement that begins with EXEC ORACLE. If there are no errors found, you can then compile, link, and execute the output source file, like any other COBOL program that you write. 1-18 Oracle Database Application Developer’s Guide - Fundamentals Overview of OCI and OCCI Use the precompiler to create server database access from clients that can be on many different platforms. Pro*COBOL allows you the freedom to design your own user interfaces and to add database access to existing COBOL applications. The embedded SQL statements available conform to an ANSI standard, so that you can access data from many databases in a program, including remote servers networked through Oracle Net. Before writing your embedded SQL statements, you may want to test interactive versions of the SQL in SQL*Plus. You then make only minor changes to start testing your embedded SQL application. Highlights of Pro*COBOL Features The following is a short subset of the capabilities of Pro*COBOL. ■ ■ ■ ■ ■ ■ ■ You can call stored PL/SQL or Java subprograms. You can improve performance by embedding PL/SQL blocks. These blocks can call PL/SQL functions or procedures written by you or provided in Oracle Database packages. Precompiler options allow you to define how cursors, errors, syntax-checking, file formats, and so on, are handled. Using precompiler options, you can check the syntax and semantics of your SQL or PL/SQL statements during precompilation, as well as at runtime. You can conditionally precompile sections of your code so that they can execute in different environments. Use tables, or group items, or tables of group items as host and indicator variables in your code to improve performance. You can program how errors and warnings are handled, so that data integrity is guaranteed. Pro*COBOL supports dynamic SQL, a technique that allows users to input variable values and statement syntax. See Also: Pro*COBOL Programmer's Guide for complete details Overview of OCI and OCCI The Oracle Call Interface (OCI) and Oracle C++ Call Interface (OCCI) are application programming interfaces (APIs) that allow you to create applications that use native procedures or function calls of a third-generation language to access Oracle Database and control all phases of SQL statement execution. These APIs provide: ■ ■ Improved performance and scalability through the efficient use of system memory and network connectivity Consistent interfaces for dynamic session and transaction management in a two-tier client/server or multitier environment ■ N-tiered authentication ■ Comprehensive support for application development using Oracle objects ■ Access to external databases ■ Ability to develop applications that service an increasing number of users and requests without additional hardware investments Orientation to Oracle Programmatic Environments 1-19 Overview of OCI and OCCI OCI lets you manipulate data and schemas in a database using a host programming language, such as C. OCCI is an object-oriented interface suitable for use with C++. These APIs provide a library of standard database access and retrieval functions in the form of a dynamic runtime library (OCILIB) that can be linked in an application at runtime. This eliminates the need to embed SQL or PL/SQL within 3GL programs. See Also: For more information about OCI and OCCI calls: ■ Oracle Call Interface Programmer's Guide ■ Oracle C++ Call Interface Programmer's Guide ■ Oracle Streams Advanced Queuing User's Guide and Reference ■ Oracle Database Globalization Support Guide ■ Oracle Database Data Cartridge Developer's Guide Advantages of OCI OCI provides significant advantages over other methods of accessing Oracle Database: ■ More fine-grained control over all aspects of the application design. ■ High degree of control over program execution. ■ ■ ■ Use of familiar 3GL programming techniques and application development tools such as browsers and debuggers. Support of dynamic SQL, method 4. Availability on the broadest range of platforms of all the Oracle programmatic interfaces. ■ Dynamic bind and define using callbacks. ■ Describe functionality to expose layers of server metadata. ■ Asynchronous event notification for registered client applications. ■ Enhanced array data manipulation language (DML) capability for array INSERTs, UPDATEs, and DELETEs. ■ Ability to associate a commit request with an execute to reduce round-trips. ■ Optimization for queries using transparent prefetch buffers to reduce round-trips. ■ ■ Thread safety, so you do not have to implement mutual exclusion (mutex) locks on OCI handles. The server connection in nonblocking mode means that control returns to the OCI code when a call is still executing or could not complete. Parts of the OCI The OCI encompasses four main sets of functionality: ■ ■ ■ ■ OCI relational functions, for managing database access and processing SQL statements OCI navigational functions, for manipulating objects retrieved from an Oracle Database OCI datatype mapping and manipulation functions, for manipulating data attributes of Oracle types OCI external procedure functions, for writing C callbacks from PL/SQL 1-20 Oracle Database Application Developer’s Guide - Fundamentals Overview of OCI and OCCI Procedural and Non-Procedural Elements The Oracle Call Interface (OCI) enables you to develop applications that combine the non-procedural data access power of SQL with the procedural capabilities of most programming languages, including C and C++. Note the following characteristics of procedural and non-procedural languages: ■ ■ In a non-procedural language program, the set of data to be operated on is specified, but what operations will be performed and how the operations are to be carried out is not specified. The non-procedural nature of SQL makes it an easy language to learn and to use to perform database transactions. It is also the standard language used to access and manipulate data in modern relational and object-relational database systems. In a procedural language program, the execution of most statements depends on previous or subsequent statements and on control structures, such as loops or conditional branches, which are not available in SQL. The procedural nature of these languages makes them more complex than SQL, but it also makes them very flexible and powerful. The combination of both non-procedural and procedural language elements in an OCI program provides easy access to Oracle Database in a structured programming environment. The OCI supports all SQL data definition, data manipulation, query, and transaction control facilities that are available through Oracle Database. For example, an OCI program can run a query against Oracle Database. The queries can require the program to supply data to the database using input (bind) variables, as follows: SELECT name FROM employees WHERE empno = :empnumber In the preceding SQL statement,:empnumber is a placeholder for a value that will be supplied by the application. You can alternatively use PL/SQL, Oracle's procedural extension to SQL. The applications you develop can be more powerful and flexible than applications written in SQL alone. The OCI also provides facilities for accessing and manipulating objects in Oracle Database. Building an OCI Application As Figure 1–1 shows, you compile and link an OCI program in the same way that you compile and link a non-database application. There is no need for a separate preprocessing or precompilation step. Orientation to Oracle Programmatic Environments 1-21 Overview of Oracle Data Provider for .NET (ODP.NET) Figure 1–1 The OCI Development Process Source Files Host Language Compiler Object Files OCI Library Host Linker Application Object Server To properly link your OCI programs, it may be necessary on some platforms to include other libraries, in addition to the OCI library. Check your Oracle platform-specific documentation for further information about extra libraries that may be required. Note: Overview of Oracle Data Provider for .NET (ODP.NET) Oracle Data Provider for .NET (ODP.NET) is an implementation of a data provider for Oracle Database. ODP.NET uses APIs native to Oracle Database to offer fast and reliable access from any .NET application to database features and data. It also uses and inherits classes and interfaces available in the Microsoft .NET Framework Class Library. For programmers using Oracle Provider for OLE DB, ADO (ActiveX Data Objects) provides an automation layer that exposes an easy programming model. ADO.NET provides a similar programming model, but without the automation layer, for better performance. More importantly, the ADO.NET model allows native providers such as ODP.NET to expose specific features and datatypes specific to Oracle Database. See Also: Oracle Data Provider for .NET Developer's Guide Using ODP.NET in a Simple Application The following is a simple C# application that connects to Oracle Database and displays its version number before disconnecting. using System; using Oracle.DataAccess.Client; class Example { OracleConnection con; void Connect() 1-22 Oracle Database Application Developer’s Guide - Fundamentals Overview of Oracle Objects for OLE (OO4O) { con = new OracleConnection(); con.ConnectionString = "User Id=hr;Password=hr;Data Source=oracle"; con.Open(); Console.WriteLine("Connected to Oracle" + con.ServerVersion); } void Close() { con.Close(); con.Dispose(); } static void Main() { Example example = new Example(); example.Connect(); example.Close(); } } Note: Additional samples are provided in directory ORACLE_ BASE\ORACLE_HOME\ODP.NET\Samples. Overview of Oracle Objects for OLE (OO4O) Oracle Objects for OLE (OO4O) is a product designed to allow easy access to data stored in Oracle Database with any programming or scripting language that supports the Microsoft COM Automation and ActiveX technology. This includes Visual Basic, Visual C++, Visual Basic For Applications (VBA), IIS Active Server Pages (VBScript and JavaScript), and others. See the OO4O online help for detailed information about using OO4O. Oracle Objects for OLE consists of the following software layers: ■ OO4O "In-Process" Automation Server ■ Oracle Data Control ■ Oracle Objects for OLE C++ Class Library Figure 1–2, "Software Layers" illustrates the OO4O software components. Orientation to Oracle Programmatic Environments 1-23 Overview of Oracle Objects for OLE (OO4O) Figure 1–2 Software Layers Data Aware ActiveX Controls C++ Class Libraries Oracle Data Control Automation Controllers (VB, Excel, ASP) COM/DCOM OO4O In-Process Automation Server Oracle Client Libraries (OCI, CORE, NLS) Oracle Database OO4O Automation Server The OO4O Automation Server is a set of COM Automation objects for connecting to Oracle Database, executing SQL statements and PL/SQL blocks, and accessing the results. Unlike other COM-based database connectivity APIs, such as Microsoft ADO, the OO4O Automation Server has been developed and evolved specifically for use with Oracle Database. It provides an optimized API for accessing features that are unique to Oracle Database and are otherwise cumbersome or inefficient to use from ODBC or OLE database-specific components. OO4O provides key features for accessing Oracle Database efficiently and easily in environments ranging from the typical two-tier client/server applications, such as those developed in Visual Basic or Excel, to application servers deployed in multitiered application server environments such as Web server applications in Microsoft Internet Information Server (IIS) or Microsoft Transaction Server (MTS). Features include: ■ ■ ■ ■ Support for execution of PL/SQL and Java stored procedures, and PL/SQL anonymous blocks. This includes support for Oracle datatypes used as parameters to stored procedures, including PL/SQL cursors. Refer to "Support for Oracle LOB and Object Datatypes" on page 1-28. Support for scrollable and updatable cursors for easy and efficient access to result sets of queries. Thread-safe objects and Connection Pool Management Facility for developing efficient Web server applications. Full support for Oracle object-relational and LOB datatypes. 1-24 Oracle Database Application Developer’s Guide - Fundamentals Overview of Oracle Objects for OLE (OO4O) ■ Full support for Advanced Queuing. ■ Support for array inserts and updates. ■ Support for Microsoft Transaction Server (MTS). OO4O Object Model The Oracle Objects for OLE object model is illustrated in Figure 1–3, "Objects and Their Relations". Figure 1–3 Objects and Their Relations OraSession OraDatabase OraServer OraDynaset OraField OraMetaData OraMDAttribute OraParameter OraParameters OraParamArray OraSQLStmt OraAQ OraAQMsg OraSession An OraSession object manages collections of OraDatabase, OraConnection, and OraDynaset objects used within an application. Typically, a single OraSession object is created for each application, but you can create named OraSession objects for shared use within and between applications. The OraSession object is the top-most object for an application. It is the only object created by the CreateObject VB/VBA API and not by an Oracle Objects for OLE method. The following code fragment shows how to create an OraSession object: Dim OraSession as Object Set OraSession = CreateObject("OracleInProcServer.XOraSession") OraServer OraServer represents a physical network connection to Oracle Database. The OraServer interface is introduced to expose the connection-multiplexing feature provided in the Oracle Call Interface. After an OraServer object is created, multiple user sessions (OraDatabase) can be attached to it by invoking the OpenDatabase method. This feature is particularly useful for application components, such as Internet Information Server (IIS), that use Oracle Objects for OLE in n-tier distributed environments. Orientation to Oracle Programmatic Environments 1-25 Overview of Oracle Objects for OLE (OO4O) The use of connection multiplexing when accessing Oracle Database with a large number of user sessions active can help reduce server processing and resource requirements while improving server scalability. OraServer is used to share a single connection across multiple OraDatabase objects (multiplexing), whereas each OraDatabase obtained from an OraSession has its own physical connection. OraDatabase An OraDatabase interface adds additional methods for controlling transactions and creating interfaces representing of Oracle object types. Attributes of schema objects can be retrieved using the Describe method of the OraDatabase interface. In releases prior to Oracle8i, an OraDatabase object is created by invoking the OpenDatabase method of an OraSession interface. The Oracle Net alias, user name, and password are passed as arguments to this method. In Oracle8i and later, invocation of this method results in implicit creation of an OraServer object. An OraDatabase object can also be created using the OpenDatabase method of the OraServer interface. Transaction control methods are available at the OraDatabase (user session) level. Transactions may be started as Read-Write (default), Serializable, or Read-only. Transaction control methods include: ■ BeginTrans ■ CommitTrans ■ RollbackTrans For example: UserSession.BeginTrans(OO4O_TXN_READ_WRITE) UserSession.ExecuteSQL("delete emp where empno = 1234") UserSession.CommitTrans OraDynaset An OraDynaset object permits browsing and updating of data created from a SQL SELECT statement. The OraDynaset object can be thought of as a cursor, although in actuality several real cursors may be used to implement the semantics of OraDynaset. An OraDynaset object automatically maintains a local cache of data fetched from the server and transparently implements scrollable cursors within the browse data. Large queries may require significant local disk space; application developers are encouraged to refine queries to limit disk usage. OraField An OraField object represents a single column or data item within a row of a dynaset. If the current row is being updated, then the OraField object represents the currently updated value, although the value may not yet have been committed to the database. Assignment to the Value property of a field is permitted only if a record is being edited (using Edit) or a new record is being added (using AddNew). Other attempts to assign data to a field's Value property results in an error. 1-26 Oracle Database Application Developer’s Guide - Fundamentals Overview of Oracle Objects for OLE (OO4O) OraMetaData and OraMDAttribute An OraMetaData object is a collection of OraMDAttribute objects that represent the description information about a particular schema object in the database. The OraMetaData object can be visualized as a table with three columns: ■ Metadata Attribute Name ■ Metadata Attribute Value ■ Flag specifying whether the Value is another OraMetaData object The OraMDAttribute objects contained in the OraMetaData object can be accessed by subscripting using ordinal integers or by using the name of the property. Referencing a subscript that is not in the collection results in the return of a NULL OraMDAttribute object. OraParameters and OraParameter An OraParameter object represents a bind variable in a SQL statement or PL/SQL block. OraParameter objects are created, accessed, and removed indirectly through the OraParameters collection of an OraDatabase object. Each parameter has an identifying name and an associated value. You can automatically bind a parameter to SQL and PL/SQL statements of other objects (as noted in the object descriptions), by using the parameter name as a placeholder in the SQL or PL/SQL statement. Such use of parameters can simplify dynamic queries and increase program performance. OraParamArray An OraParamArray object represents an array-type bind variable in a SQL statement or PL/SQL block, as opposed to a scalar-type bind variable represented by the OraParameter object. OraParamArray objects are created, accessed, and removed indirectly through the OraParameters collection of an OraDatabase object. Each OraParamArray object has an identifying name and an associated value. OraSQLStmt An OraSQLStmt object represents a single SQL statement. Use the CreateSQL method to create an OraSQLStmt object from an OraDatabase object. During create and refresh, OraSQLStmt objects automatically bind all relevant, enabled input parameters to the specified SQL statement, using the parameter names as placeholders in the SQL statement. This can improve the performance of SQL statement execution without re-parsing the SQL statement. The OraSQLStmt object can be used later to execute the same query using a different value for the :SALARY placeholder. This is done as follows (updateStmt is the OraSQLStmt object here): OraDatabase.Parameters("SALARY").value = 200000 updateStmt.Parameters("ENAME").value = "KING" updateStmt.Refresh OraAQ An OraAQ object is instantiated by invoking the CreateAQ method of the OraDatabase interface. It represents a queue that is present in the database. Orientation to Oracle Programmatic Environments 1-27 Overview of Oracle Objects for OLE (OO4O) Oracle Objects for OLE provides interfaces for accessing Oracle Advanced Queuing (AQ) feature. It makes AQ accessible from popular COM-based development environments such as Visual Basic. For a detailed description of Oracle Advanced Queuing, refer to Oracle Streams Advanced Queuing User's Guide and Reference. OraAQMsg The OraAQMsg object encapsulates the message to be enqueued or dequeued. The message can be of any user-defined or raw type. For a detailed description of Oracle Advanced Queuing, refer to Oracle Streams Advanced Queuing User's Guide and Reference. OraAQAgent The OraAQAgent object represents a message recipient and is only valid for queues that allow multiple consumers. It is a child of OraAQMsg. An OraAQAgent object can be instantiated by invoking the AQAgent method. For example: Set agent = qMsg.AQAgent(name) An OraAQAgent object can also be instantiated by invoking the AddRecipient method. For example: Set agent = qMsg.AddRecipient(name, address, protocol). Support for Oracle LOB and Object Datatypes Oracle Objects for OLE provides full support for accessing and manipulating instances of object datatypes and LOBs in Oracle Database. Figure 1–4, "Supported Oracle Datatypes" illustrates the datatypes supported by OO4O. Instances of these types can be fetched from the database or passed as input or output variables to SQL statements and PL/SQL blocks, including stored procedures and functions. All instances are mapped to COM Automation Interfaces that provide methods for dynamic attribute access and manipulation. Figure 1–4 Supported Oracle Datatypes OraObject OraAttribute OraRef OraAttribute OraCollection Element Values OraField OraParameter OraParamArray OraBLOB OraCLOB OraBFILE Value of all other scalar types 1-28 Oracle Database Application Developer’s Guide - Fundamentals Overview of Oracle Objects for OLE (OO4O) OraBLOB and OraCLOB The OraBlob and OraClob interfaces in Oracle Objects for OLE provide methods for performing operations on large database objects of datatype BLOB, CLOB, and NCLOB. BLOB, CLOB, and NCLOB datatypes are also referred to here as LOB datatypes. LOB data is accessed using Read and the CopyToFile methods. LOB data is modified using Write, Append, Erase, Trim, Copy, CopyFromFile, and CopyFromBFile methods. Before modifying the content of a LOB column in a row, a row lock must be obtained. If the LOB column is a field of an OraDynaset, object, then the lock is obtained by invoking the Edit method. OraBFILE The OraBFile interface in Oracle Objects for OLE provides methods for performing operations on large database objects of datatype BFILE. BFILE objects are large binary data objects stored in operating system files outside of the database tablespaces. Oracle Data Control Oracle Data Control (ODC) is an ActiveX Control that is designed to simplify the exchange of data between Oracle Database and visual controls such edit, text, list, and grid controls in Visual Basic and other development tools that support custom controls. ODC acts as an agent to handle the flow of information from Oracle Database and a visual data-aware control, such as a grid control, that is bound to it. The data control manages various user interface (UI) tasks such as displaying and editing data. It also executes and manages the results of database queries. Oracle Data Control is compatible with the Microsoft data control included with Visual Basic. If you are familiar with the Visual Basic data control, learning to use Oracle Data Control is quick and easy. Communication between data-aware controls and a Data Control is governed by a protocol that has been specified by Microsoft. Oracle Objects for OLE C++ Class Library Oracle Objects for OLE C++ Class Library is a collection of C++ classes that provide programmatic access to the Oracle Object Server. Although the class library is implemented using OLE Automation, neither the OLE development kit nor any OLE development knowledge is necessary to use it. This library helps C++ developers avoid the chore of writing COM client code for accessing the OO4O interfaces. Additional Sources of Information For detailed information about Oracle Objects for OLE refer to the online help provided with the OO4O product: ■ Oracle Objects for OLE Help ■ Oracle Objects for OLE C++ Class Library Help To view examples of how to use Oracle Objects for OLE, refer to the samples located in the ORACLE_HOME\OO4O directory of the Oracle Database installation. Additional OO4O examples can be found in the following Oracle publications: ■ Oracle Database Application Developer's Guide - Large Objects Orientation to Oracle Programmatic Environments 1-29 Choosing a Programming Environment ■ ■ Oracle Streams Advanced Queuing User's Guide and Reference Oracle Database PL/SQL Packages and Types Reference Choosing a Programming Environment To choose a programming environment for a new development project: ■ ■ ■ ■ Review the preceding overviews and the manuals for each environment. Read the platform-specific manual that explains which compilers are approved for use with your platforms. If a particular language does not provide a feature you need, remember that PL/SQL and Java stored procedures can both be called from code written in any of the languages in this chapter. Stored procedures include triggers and object type methods. External procedures written in C can be called from OCI, Java, PL/SQL or SQL. The external procedure itself can call back into the database using either SQL, OCI, or Pro*C (but not C++). The following examples illustrate easy choices: ■ ■ Pro*COBOL does not support object types or collection types, while Pro*C/C++ does. SQLJ does not support dynamic SQL the way that JDBC does. Choosing Whether to Use OCI or a Precompiler Precompiler applications typically contain less code than equivalent OCI applications, which can help productivity. Some situations require detailed control of the database and are suited for OCI applications (either pure OCI or a precompiler application with embedded OCI calls): ■ ■ ■ ■ ■ ■ ■ OCI provides more detailed control over multiplexing and migrating sessions. OCI provides dynamic bind and define using callbacks that can be used for any arbitrary structure, including lists. OCI has many calls to handle metadata. OCI allows asynchronous event notifications to be received by a client application. It provides a means for clients to generate notifications for propagation to other clients. OCI allows DML statements to use arrays to complete as many iterations as possible before returning any error messages. OCI calls for special purposes include Advanced Queuing, globalization support, Data Cartridges, and support of the date and time datatypes. OCI calls can be embedded in a Pro*C/C++ application. Using Built-In Packages and Libraries Both Java and PL/SQL have built-in packages and libraries. PL/SQL and Java interoperate in the server. You can execute a PL/SQL package from Java or wrap a PL/SQL class with a Java wrapper so that it can be called from 1-30 Oracle Database Application Developer’s Guide - Fundamentals Choosing a Programming Environment distributed CORBA and EJB clients. The following table shows PL/SQL packages and their Java equivalents: Table 1–1 PL/SQL and Java Equivalent Software PL/SQL Package Java Equivalent DBMS_ALERT Call package with SQLJ or JDBC. DBMS_DDL JDBC has this functionality. DBMS_JOB Schedule a job that has a Java Stored procedure. DBMS_LOCK Call with SQLJ or JDBC. DBMS_MAIL Use JavaMail. DBMS_OUTPUT Use subclass oracle.aurora.rdbms.OracleDBMSOutputStream or Java stored procedure DBMS_JAVA.SET_STREAMS. DBMS_PIPE Call with SQLJ or JDBC. DBMS_SESSION Use JDBC to execute an ALTER SESSION statement. DBMS_SNAPSHOT Call with SQLJ or JDBC. DBMS_SQL Use JDBC. DBMS_TRANSACTION Use JDBC to execute an ALTER SESSION statement. DBMS_UTILITY Call with SQLJ or JDBC. UTL_FILE Grant the JAVAUSERPRIV privilege and then use Java I/O entry points. Java Compared to PL/SQL Both Java and PL/SQL can be used to build applications in the database. Here are some guidelines for their use: PL/SQL Is Optimized for Database Access PL/SQL uses the same datatypes as SQL. SQL datatypes are thus easier to use and SQL operations are faster than with Java, especially when a large amount of data is involved, when mostly database access is done, or when bulk operations are used. PL/SQL Is Integrated with the Database PL/SQL is an extension to SQL offering data encapsulation, information hiding, overloading, and exception-handling. Some advanced PL/SQL capabilities are not available for Java in Oracle9i. Examples are autonomous transactions and the dblink facility for remote databases. Code development is usually faster in PL/SQL than in Java. Both Java and PL/SQL Have Object-Oriented Features Java has inheritance, polymorphism, and component models for developing distributed systems. PL/SQL has inheritance and type evolution, the ability to change methods and attributes of a type while preserving subtypes and table data that use the type. Orientation to Oracle Programmatic Environments 1-31 Choosing a Programming Environment Java Is Used for Open Distributed Applications Java has a richer type system than PL/SQL and is an object-oriented language. Java can use CORBA (which can have many different computer languages in its clients) and EJB. PL/SQL packages can be called from CORBA or EJB clients. You can run XML tools, the Internet File System, or JavaMail from Java. Many Java-based development tools are available throughout the industry. 1-32 Oracle Database Application Developer’s Guide - Fundamentals Part I SQL for Application Developers This part contains the following chapters: ■ Chapter 2, "SQL Processing for Application Developers" ■ Chapter 3, "Using SQL Datatypes in Application Development" ■ Chapter 4, "Using Regular Expressions in Oracle Database" ■ Chapter 5, "Using Indexes in Application Development" ■ Chapter 6, "Maintaining Data Integrity in Application Development" 2 SQL Processing for Application Developers This chapter describes how Oracle Database processes SQL statements. Before reading this chapter you should read the section "SQL Processing" in Oracle Database Concepts. Topics include the following: ■ Grouping Operations into Transactions ■ Ensuring Repeatable Reads with Read-Only Transactions ■ Using Cursors within Applications ■ Locking Data Explicitly ■ About User Locks ■ Using Serializable Transactions for Concurrency Control ■ Autonomous Transactions ■ Resuming Execution After a Storage Error Condition Grouping Operations into Transactions This section contains the following topics: ■ Deciding How to Group Operations in Transactions ■ Improving Transaction Performance ■ Committing Transactions ■ Rolling Back Transactions ■ Defining Transaction Savepoints Deciding How to Group Operations in Transactions In general, only application designers who use the programming interfaces to Oracle Database are concerned with which types of actions should be grouped together as one transaction. You should use the following principles when deciding how to group transactions: ■ ■ ■ Transactions must be defined properly so that work is accomplished in logical units and data is kept consistent. Data in all referenced tables should be in a consistent state before the transaction begins and after it ends. Transactions should consist of only the SQL statements or PL/SQL blocks that comprise one consistent change to the data. SQL Processing for Application Developers 2-1 Grouping Operations into Transactions For example, suppose that you write a Web application that enables users to transfer funds between accounts. The transaction should include the debit to one account, which is executed by one SQL statement, and the credit to another account, which is executed by a second SQL statement. Both statements should fail or succeed together as a unit of work; the credit should not be committed without the debit. Other non-related actions, such as a new deposit to one account, should not be included in the same transaction. Improving Transaction Performance As an application developer, you should consider whether you can improve performance. Consider the following performance enhancements when designing and writing your application: ■ ■ ■ ■ ■ ■ ■ ■ Use the SET TRANSACTION command with the USE ROLLBACK SEGMENT clause to explicitly assign a transaction to a rollback segment. This technique can eliminate the need to allocate additional extents dynamically, which can reduce system performance. Note that this clause is relevant and valid only if you use rollback segments for undo. If you use automatic undo management, then Oracle Database ignores this clause. Establish standards for writing SQL statements so that you can take advantage of shared SQL areas. Oracle Database recognizes identical SQL statements and allows them to share memory areas. This reduces memory usage on the database server and increases system throughput. Use the ANALYZE command to collect statistics that can be used by Oracle Database to implement a cost-based approach to SQL statement optimization. You can supply additional "hints" to the optimizer as needed. Call the DBMS_APPLICATION_INFO.SET_ACTION procedure before beginning a transaction to register and name a transaction for later use when measuring performance across an application. You should specify which type of activity a transaction performs so that the system tuners can later see which transactions are taking up the most system resources. Increase user productivity and query efficiency by including user-written PL/SQL functions in SQL expressions as described in "Calling Stored Functions from SQL Expressions" on page 7-36. Create explicit cursors when writing a PL/SQL application. Reduce frequency of parsing and improve performance in precompiler programs by increasing the number of cursors with MAX_OPEN_CURSORS. Use the SET TRANSACTION command with the ISOLATION LEVEL set to SERIALIZABLE to get ANSI/ISO serializable transactions. See Also: ■ "How Serializable Transactions Interact" on page 2-15 ■ "Using Cursors within Applications" on page 2-6 ■ Oracle Database Concepts for more information about transaction tuning features Committing Transactions To commit a transaction, use the COMMIT statement. The following two statements are equivalent and commit the current transaction: 2-2 Oracle Database Application Developer’s Guide - Fundamentals Grouping Operations into Transactions COMMIT WORK; COMMIT; The COMMIT statements lets you include the COMMENT parameter along with a comment that provides information about the transaction being committed. This option is useful for including information about the origin of the transaction when you commit distributed transactions: COMMIT COMMENT 'Dallas/Accts_pay/Trans_type 10B'; Managing Commit Redo Behavior When a transaction updates the database, it generates a redo entry corresponding to this update. Oracle Database buffers this redo in memory until the completion of the transaction. When the transaction commits, the log writer (LGWR) process writes redo for the commit, along with the accumulated redo of all changes in the transaction, to disk. By default Oracle Database writes the redo to disk before the call returns to the client. This behavior introduces a latency in the commit because the application must wait for the redo to be persisted on disk. Suppose that you are writing an application that requires very high transaction throughput. If you are willing to trade commit durability for lower commit latency, then you can change the default COMMIT options so that the application does not need to wait for Oracle Database to write data to the online redo logs. Oracle Database enables you to change the handling of commit redo depending on the needs of your application. You can change the commit behavior in the following locations: ■ COMMIT_WRITE initialization parameter at the system or session level ■ COMMIT statement The options in the COMMIT statement override the current settings in the initialization parameter. Table 2–1 describes redo persistence options that you can set in either location. Table 2–1 Initialization Parameter and COMMIT Options for Managing Commit Redo Option Specifies that . . . WAIT The commit does not return as successful until the redo corresponding to the commit is persisted in the online redo logs (default). NOWAIT The commit should return to the application without waiting for the redo to be written to the online redo logs. IMMEDIATE The log writer process should write the redo for the commit immediately (default). In other words, this option forces a disk I/O. BATCH Oracle Database should buffer the redo. The log writer process is permitted to write the redo to disk in its own time. The following example shows how to set the commit behavior to BATCH and NOWAIT in the initialization parameter file: COMMIT_WRITE = BATCH, NOWAIT You can change the commit behavior at the system level by executing ALTER SYSTEM as in the following example: ALTER SYSTEM SET COMMIT_WRITE = BATCH, NOWAIT SQL Processing for Application Developers 2-3 Grouping Operations into Transactions After the initialization parameter is set, a COMMIT statement with no options conforms to the options specified in the parameter. Alternatively, you can override the current initialization parameter setting by specifying options directly on the COMMIT statement as in the following example: COMMIT WRITE BATCH NOWAIT In either case, your application specifies that log writer does not have to write the redo for the commit immediately to the online redo logs and should not wait for confirmation that the redo has been written to disk. You cannot change the default IMMEDIATE and WAIT behavior for distributed transactions. Note: If your application uses OCI, then you can modify redo behavior by setting the following flags in the OCITransCommit() function within your application: ■ OCI_TRANS_WRITEBATCH ■ OCI_TRANS_WRITENOWAIT ■ OCI_TRANS_WRITEIMMED ■ OCI_TRANS_WRITEWAIT Note that the specification of the NOWAIT and BATCH options allows a small window of vulnerability in which Oracle Database can roll back a transaction that your application view as committed. Your application must be able to tolerate the following scenarios: ■ ■ The database host crashes, which causes the database to lose redo that was buffered but not yet written to the online redo logs. A file I/O problem prevents log writer from writing buffered redo to disk. If the redo logs are not multiplexed, then the commit is lost. See Also: ■ ■ Oracle Database SQL Reference for information on the COMMIT statement Oracle Call Interface Programmer's Guide for information about the OCITransCommit() function Rolling Back Transactions To roll back an entire transaction, or to roll back part of a transaction to a savepoint, use the ROLLBACK statement. For example, either of the following statements rolls back the entire current transaction: ROLLBACK WORK; ROLLBACK; The WORK option of the ROLLBACK command has no function. To roll back to a savepoint defined in the current transaction, use the TO option of the ROLLBACK command. For example, either of the following statements rolls back the current transaction to the savepoint named POINT1: SAVEPOINT Point1; ... 2-4 Oracle Database Application Developer’s Guide - Fundamentals Ensuring Repeatable Reads with Read-Only Transactions ROLLBACK TO SAVEPOINT Point1; ROLLBACK TO Point1; Defining Transaction Savepoints To define a savepoint in a transaction, use the SAVEPOINT command. The following statement creates the savepoint named ADD_EMP1 in the current transaction: SAVEPOINT Add_emp1; If you create a second savepoint with the same identifier as an earlier savepoint, the earlier savepoint is erased. After creating a savepoint, you can roll back to the savepoint. There is no limit on the number of active savepoints for each session. An active savepoint is one that has been specified since the last commit or rollback. An Example of COMMIT, SAVEPOINT, and ROLLBACK Table 2–4 shows a series of SQL statements that illustrates the use of COMMIT, SAVEPOINT, and ROLLBACK statements within a transaction. Table 2–2 Use of COMMIT, SAVEPOINT, and ROLLBACK SQL Statement Results SAVEPOINT a; First savepoint of this transaction DELETE...; First DML statement of this transaction SAVEPOINT b; Second savepoint of this transaction INSERT INTO...; Second DML statement of this transaction SAVEPOINT c; Third savepoint of this transaction UPDATE...; Third DML statement of this transaction. ROLLBACK TO c; UPDATE statement is rolled back, savepoint C remains defined ROLLBACK TO b; INSERT statement is rolled back, savepoint C is lost, savepoint B remains defined ROLLBACK TO c; ORA-01086 error; savepoint C no longer defined INSERT INTO...; New DML statement in this transaction COMMIT; Commits all actions performed by the first DML statement (the DELETE statement) and the last DML statement (the second INSERT statement) All other statements (the second and the third statements) of the transaction were rolled back before the COMMIT. The savepoint A is no longer active. Ensuring Repeatable Reads with Read-Only Transactions By default, the consistency model for Oracle Database guarantees statement-level read consistency, but does not guarantee transaction-level read consistency (repeatable reads). If you want transaction-level read consistency, and if your transaction does not require updates, then you can specify a read-only transaction. After indicating that your transaction is read-only, you can execute as many queries as you like against any database table, knowing that the results of each query in the read-only transaction are consistent with respect to a single point in time. A read-only transaction does not acquire any additional data locks to provide transaction-level read consistency. The multi-version consistency model used for SQL Processing for Application Developers 2-5 Using Cursors within Applications statement-level read consistency is used to provide transaction-level read consistency; all queries return information with respect to the system change number (SCN) determined when the read-only transaction begins. Because no data locks are acquired, other transactions can query and update data being queried concurrently by a read-only transaction. Long-running queries sometimes fail because undo information required for consistent read operations is no longer available. This happens when committed undo blocks are overwritten by active transactions. Automatic undo management provides a way to explicitly control when undo space can be reused; that is, how long undo information is retained. Your database administrator can specify a retention period by using the parameter UNDO_RETENTION. See Also: Oracle Database Administrator's Guide for information on long-running queries and resumable space allocation For example, if UNDO_RETENTION is set to 30 minutes, then all committed undo information in the system is retained for at least 30 minutes. This ensures that all queries running for 30 minutes or less, under usual circumstances, do not encounter the OER error, "snapshot too old." A read-only transaction is started with a SET TRANSACTION statement that includes the READ ONLY option. For example: SET TRANSACTION READ ONLY; The SET TRANSACTION statement must be the first statement of a new transaction; if any DML statements (including queries) or other non-DDL statements (such as SET ROLE) precede a SET TRANSACTION READ ONLY statement, an error is returned. Once a SET TRANSACTION READ ONLY statement successfully executes, only SELECT (without a FOR UPDATE clause), COMMIT, ROLLBACK, or non-DML statements (such as SET ROLE, ALTER SYSTEM, LOCK TABLE) are allowed in the transaction. Otherwise, an error is returned. A COMMIT, ROLLBACK, or DDL statement terminates the read-only transaction; a DDL statement causes an implicit commit of the read-only transaction and commits in its own transaction. Using Cursors within Applications PL/SQL implicitly declares a cursor for all SQL data manipulation statements, including queries that return only one row. For queries that return more than one row, you can explicitly declare a cursor to process the rows individually. A cursor is a handle to a specific private SQL area. In other words, a cursor can be thought of as a name for a specific private SQL area. A PL/SQL cursor variable enables the retrieval of multiple rows from a stored procedure. Cursor variables allow you to pass cursors as parameters in your 3GL application. Cursor variables are described in Oracle Database PL/SQL User's Guide and Reference. Although most Oracle Database users rely on the automatic cursor handling of the database utilities, the programmatic interfaces offer application designers more control over cursors. In application development, a cursor is a named resource available to a program, which can be specifically used for parsing SQL statements embedded within the application. Declaring and Opening Cursors There is no absolute limit to the total number of cursors one session can have open at one time, subject to two constraints: 2-6 Oracle Database Application Developer’s Guide - Fundamentals Using Cursors within Applications ■ ■ Each cursor requires virtual memory, so a session's total number of cursors is limited by the memory available to that process. A systemwide limit of cursors for each session is set by the initialization parameter named OPEN_CURSORS found in the parameter file (such as INIT.ORA). See Also: Oracle Database Reference for descriptions of parameters Explicitly creating cursors for precompiler programs can offer some advantages in tuning those applications. For example, increasing the number of cursors can often reduce the frequency of parsing and improve performance. If you know how many cursors may be required at a given time, then you can make sure you can open that many simultaneously. Using a Cursor to Execute Statements Again After each stage of execution, the cursor retains enough information about the SQL statement to reexecute the statement without starting over, as long as no other SQL statement has been associated with that cursor. The statement can be reexecuted without including the parse stage. By opening several cursors, the parsed representation of several SQL statements can be saved. Repeated execution of the same SQL statements can thus begin at the describe, define, bind, or execute step, saving the repeated cost of opening cursors and parsing. To understand the performance characteristics of a cursor, a DBA can retrieve the text of the query represented by the cursor using the V$SQL catalog view. Because the results of EXPLAIN PLAN on the original query might differ from the way the query is actually processed, a DBA can get more precise information by examining the V$SQL_PLAN, V$SQL_PLAN_STATISTICS, and V$SQL_PLAN_STATISTICS_ALL catalog views.: ■ ■ ■ The V$SQL_PLAN view contains the execution plan information for each child cursor loaded in the library cache. The V$SQL_PLAN_STATISTICS view provides execution statistics at the row source level for each child cursor. The V$SQL_PLAN_STATISTICS_ALL view contains memory usage statistics for row sources that use SQL memory (sort or hash-join). This view concatenates information in V$SQL_PLAN with execution statistics from V$SQL_PLAN_STATISTICS and V$SQL_WORKAREA. See Also: Oracle Database Reference for details on these views Closing Cursors Closing a cursor means that the information currently in the associated private area is lost and its memory is deallocated. Once a cursor is opened, it is not closed until one of the following events occurs: ■ ■ The user program terminates its connection to the server. If the user program is an OCI program or precompiler application, then it explicitly closes any open cursor during the execution of that program. (However, when this program terminates, any cursors remaining open are implicitly closed.) SQL Processing for Application Developers 2-7 Locking Data Explicitly Cancelling Cursors Cancelling a cursor frees resources from the current fetch.The information currently in the associated private area is lost but the cursor remains open, parsed, and associated with its bind variables. Note: You cannot cancel cursors using Pro*C/C++ or PL/SQL. See Also: Oracle Call Interface Programmer's Guide for more information about cancelling cursors Locking Data Explicitly Oracle Database always performs necessary locking to ensure data concurrency, integrity, and statement-level read consistency. You can override these default locking mechanisms. For example, you might want to override the default locking of Oracle Database if: ■ ■ You want transaction-level read consistency or "repeatable reads"—where transactions query a consistent set of data for the duration of the transaction, knowing that the data has not been changed by any other transactions. This level of consistency can be achieved by using explicit locking, read-only transactions, serializable transactions, or overriding default locking for the system. A transaction requires exclusive access to a resource. To proceed with its statements, the transaction with exclusive access to a resource does not have to wait for other transactions to complete. The automatic locking mechanisms can be overridden at the transaction level. Transactions including the following SQL commands override Oracle Database's default locking: ■ LOCK TABLE ■ SELECT, including the FOR UPDATE clause ■ SET TRANSACTION with the READ ONLY or ISOLATION LEVEL SERIALIZABLE options Locks acquired by these statements are released after the transaction is committed or rolled back. The following sections describe each option available for overriding the default locking of Oracle Database. The initialization parameter DML_LOCKS determines the maximum number of DML locks allowed. See Also: Oracle Database Reference for a discussion of parameters Although the default value is usually enough, you might need to increase it if you use additional manual locks. Caution: If you override the default locking of Oracle Database at any level, be sure that the overriding locking procedures operate correctly: Ensure that data integrity is guaranteed, data concurrency is acceptable, and deadlocks are not possible or are appropriately handled. 2-8 Oracle Database Application Developer’s Guide - Fundamentals Locking Data Explicitly Choosing a Locking Strategy A transaction explicitly acquires the specified table locks when a LOCK TABLE statement is executed. A LOCK TABLE statement manually overrides default locking. When a LOCK TABLE statement is issued on a view, the underlying base tables are locked. The following statement acquires exclusive table locks for the EMP_TAB and DEPT_TAB tables on behalf of the containing transaction: LOCK TABLE Emp_tab, Dept_tab IN EXCLUSIVE MODE NOWAIT; You can specify several tables or views to lock in the same mode; however, only a single lock mode can be specified for each LOCK TABLE statement. When a table is locked, all rows of the table are locked. No other user can modify the table. Note: You can also indicate if you do or do not want to wait to acquire the lock. If you specify the NOWAIT option, then you only acquire the table lock if it is immediately available. Otherwise an error is returned to notify that the lock is not available at this time. In this case, you can attempt to lock the resource at a later time. If NOWAIT is omitted, then the transaction does not proceed until the requested table lock is acquired. If the wait for a table lock is excessive, then you might want to cancel the lock operation and retry at a later time; you can code this logic into your applications. When to Lock with ROW SHARE and ROW EXCLUSIVE Mode LOCK TABLE Emp_tab IN ROW SHARE MODE; LOCK TABLE Emp_tab IN ROW EXCLUSIVE MODE; ROW SHARE and ROW EXCLUSIVE table locks offer the highest degree of concurrency. You might use these locks if: ■ ■ Your transaction needs to prevent another transaction from acquiring an intervening share, share row, or exclusive table lock for a table before the table can be updated in your transaction. If another transaction acquires an intervening share, share row, or exclusive table lock, no other transactions can update the table until the locking transaction commits or rolls back. Your transaction needs to prevent a table from being altered or dropped before the table can be modified later in your transaction. When to Lock with SHARE Mode LOCK TABLE Emp_tab IN SHARE MODE; SHARE table locks are rather restrictive data locks. You might use these locks if: ■ ■ ■ Your transaction only queries the table, and requires a consistent set of the table data for the duration of the transaction. You can hold up other transactions that try to update the locked table, until all transactions that hold SHARE locks on the table either commit or roll back. Other transactions may acquire concurrent SHARE table locks on the same table, also allowing them the option of transaction-level read consistency. SQL Processing for Application Developers 2-9 Locking Data Explicitly Caution: Your transaction may or may not update the table later in the same transaction. However, if multiple transactions concurrently hold share table locks for the same table, no transaction can update the table (even if row locks are held as the result of a SELECT... FOR UPDATE statement). Therefore, if concurrent share table locks on the same table are common, updates cannot proceed and deadlocks are common. In this case, use share row exclusive or exclusive table locks instead. For example, assume that two tables, EMP_TAB and BUDGET_TAB, require a consistent set of data in a third table, DEPT_TAB. For a given department number, you want to update the information in both of these tables, and ensure that no new members are added to the department between these two transactions. Although this scenario is quite rare, it can be accommodated by locking the DEPT_TAB table in SHARE MODE, as shown in the following example. Because the DEPT_TAB table is rarely updated, locking it probably does not cause many other transactions to wait long. You may need to set up data structures similar to the following for certain examples to work: Note: CREATE TABLE dept_tab( deptno NUMBER(2) NOT NULL, dname VARCHAR2(14), loc VARCHAR2(13)); CREATE TABLE emp_tab ( empno NUMBER(4) NOT NULL, ename VARCHAR2(10), job VARCHAR2(9), mgr NUMBER(4), hiredate DATE, sal NUMBER(7,2), comm NUMBER(7,2), deptno NUMBER(2)); CREATE TABLE Budget_tab ( totsal NUMBER(7,2), deptno NUMBER(2) NOT NULL); LOCK TABLE Dept_tab IN SHARE MODE; UPDATE Emp_tab SET sal = sal * 1.1 WHERE deptno IN (SELECT deptno FROM Dept_tab WHERE loc = 'DALLAS'); UPDATE Budget_tab SET Totsal = Totsal * 1.1 WHERE Deptno IN (SELECT Deptno FROM Dept_tab WHERE Loc = 'DALLAS'); COMMIT; /* This releases the lock */ When to Lock with SHARE ROW EXCLUSIVE Mode LOCK TABLE Emp_tab IN SHARE ROW EXCLUSIVE MODE; 2-10 Oracle Database Application Developer’s Guide - Fundamentals Locking Data Explicitly You might use a SHARE ROW EXCLUSIVE table lock if: ■ ■ ■ Your transaction requires both transaction-level read consistency for the specified table and the ability to update the locked table. You do not care if other transactions acquire explicit row locks (using SELECT... FOR UPDATE), which might make UPDATE and INSERT statements in the locking transaction wait and might cause deadlocks. You only want a single transaction to have this behavior. When to Lock in EXCLUSIVE Mode LOCK TABLE Emp_tab IN EXCLUSIVE MODE; You might use an EXCLUSIVE table if: ■ ■ ■ Your transaction requires immediate update access to the locked table. When your transaction holds an exclusive table lock, other transactions cannot lock specific rows in the locked table. Your transaction also ensures transaction-level read consistency for the locked table until the transaction is committed or rolled back. You are not concerned about low levels of data concurrency, making transactions that request exclusive table locks wait in line to update the table sequentially. Privileges Required You can automatically acquire any type of table lock on tables in your schema. To acquire a table lock on a table in another schema, you must have the LOCK ANY TABLE system privilege or any object privilege (for example, SELECT or UPDATE) for the table. Letting Oracle Database Control Table Locking Letting Oracle Database control table locking means your application needs less programming logic, but also has less control, than if you manage the table locks yourself. Issuing the command SET TRANSACTION ISOLATION LEVEL SERIALIZABLE or ALTER SESSION ISOLATION LEVEL SERIALIZABLE preserves ANSI serializability without changing the underlying locking protocol. This technique allows concurrent access to the table while providing ANSI serializability. Getting table locks greatly reduces concurrency. See Also: ■ ■ Oracle Database SQL Reference for information on the SET TRANSACTION statement Oracle Database SQL Reference for information on the ALTER SESSION statements The settings for these parameters should be changed only when an instance is shut down. If multiple instances are accessing a single database, then all instances should use the same setting for these parameters. SQL Processing for Application Developers 2-11 Locking Data Explicitly Explicitly Acquiring Row Locks You can override default locking with a SELECT statement that includes the FOR UPDATE clause. This statement acquires exclusive row locks for selected rows (as an UPDATE statement does), in anticipation of updating the selected rows in a subsequent statement. You can use a SELECT... FOR UPDATE statement to lock a row without actually changing it. For example, several triggers in Chapter 9, "Coding Triggers", show how to implement referential integrity. In the EMP_DEPT_CHECK trigger (see "Foreign Key Trigger for Child Table"), the row that contains the referenced parent key value is locked to guarantee that it remains for the duration of the transaction; if the parent key is updated or deleted, referential integrity would be violated. SELECT... FOR UPDATE statements are often used by interactive programs that allow a user to modify fields of one or more specific rows (which might take some time); row locks are acquired so that only a single interactive program user is updating the rows at any given time. If a SELECT... FOR UPDATE statement is used when defining a cursor, the rows in the return set are locked when the cursor is opened (before the first fetch) rather than being locked as they are fetched from the cursor. Locks are only released when the transaction that opened the cursor is committed or rolled back, not when the cursor is closed. Each row in the return set of a SELECT... FOR UPDATE statement is locked individually; the SELECT... FOR UPDATE statement waits until the other transaction releases the conflicting row lock. If a SELECT... FOR UPDATE statement locks many rows in a table, and if the table experiences a lot of update activity, it might be faster to acquire an EXCLUSIVE table lock instead. The return set for a SELECT... FOR UPDATE may change while the query is running; for example, if columns selected by the query are updated or rows are deleted after the query started. When this happens, SELECT... FOR UPDATE acquires locks on the rows that did not change, gets a new read-consistent snapshot of the table using these locks, and then restarts the query to acquire the remaining locks. Note: This can cause a deadlock between sessions querying the table concurrently with DML operations when rows are locked in a non-sequential order. To prevent such deadlocks, design your application so that any concurrent DML on the table does not affect the return set of the query. If this is not feasible, you may want to serialize queries in your application. When acquiring row locks with SELECT... FOR UPDATE, you can specify the NOWAIT option to indicate that you are not willing to wait to acquire the lock. If you cannot acquire then lock immediately, an error is returned to signal that the lock is not possible at this time. You can try to lock the row again later. By default, the transaction waits until the requested row lock is acquired. If the wait for a row lock is too long, you can code logic into your application to cancel the lock operation and try again later. 2-12 Oracle Database Application Developer’s Guide - Fundamentals About User Locks About User Locks You can use Oracle Lock Management services for your applications by making calls to the DBMS_LOCK package. It is possible to request a lock of a specific mode, give it a unique name recognizable in another procedure in the same or another instance, change the lock mode, and release it. Because a reserved user lock is the same as an Oracle Database lock, it has all the features of a database lock, such as deadlock detection. Be certain that any user locks used in distributed transactions are released upon COMMIT, or an undetected deadlock can occur. See Also: Oracle Database PL/SQL Packages and Types Reference for detailed information on the DBMS_LOCK package When to Use User Locks User locks can help to: ■ Provide exclusive access to a device, such as a terminal ■ Provide application-level enforcement of read locks ■ Detect when a lock is released and cleanup after the application ■ Synchronize applications and enforce sequential processing Example of a User Lock The following Pro*COBOL precompiler example shows how locks can be used to ensure that there are no conflicts when multiple people need to access a single device. ***************************************************************** * Print Check * * Any cashier may issue a refund to a customer returning goods. * * Refunds under $50 are given in cash, more than $50 by check. * * This code prints the check. The one printer is opened by all * * the cashiers to avoid the overhead of opening and closing it * * for every check. This means that lines of output from multiple* * cashiers could become interleaved if we don't ensure exclusive* * access to the printer. The DBMS_LOCK package is used to * * ensure exclusive access. * ***************************************************************** CHECK-PRINT * Get the lock "handle" for the printer lock. MOVE "CHECKPRINT" TO LOCKNAME-ARR. MOVE 10 TO LOCKNAME-LEN. EXEC SQL EXECUTE BEGIN DBMS_LOCK.ALLOCATE_UNIQUE ( :LOCKNAME, :LOCKHANDLE ); END; END-EXEC. * Lock the printer in exclusive mode (default mode). EXEC SQL EXECUTE BEGIN DBMS_LOCK.REQUEST ( :LOCKHANDLE ); END; END-EXEC. * We now have exclusive use of the printer, print the check. ... * Unlock the printer so other people can use it EXEC SQL EXECUTE BEGIN DBMS_LOCK.RELEASE ( :LOCKHANDLE ); END; END-EXEC. SQL Processing for Application Developers 2-13 Using Serializable Transactions for Concurrency Control Viewing and Monitoring Locks Table 2–5 describes Oracle Database facilities to display locking information for ongoing transactions within an instance. Table 2–3 Ways to Display Locking Information Tool Description Oracle Enterprise Manager 10g Database Control From the Additional Monitoring Links section of the Database Performance page, click Database Locks to display user blocks, blocking locks, or the complete list of all database locks. Refer to Oracle Database 2 Day DBA for more information. UTLLOCKT.SQL The UTLLOCKT.SQL script displays a simple character lock wait-for graph in tree structured fashion. Using any ad hoc SQL tool (such as SQL*Plus) to execute the script, it prints the sessions in the system that are waiting for locks and the corresponding blocking locks. The location of this script file is operating system dependent. (You must have run the CATBLOCK.SQL script before using UTLLOCKT.SQL.) Using Serializable Transactions for Concurrency Control By default, Oracle Database permits concurrently executing transactions to modify, add, or delete rows in the same table, and in the same data block. Changes made by one transaction are not seen by another concurrent transaction until the transaction that made the changes commits. If a transaction A attempts to update or delete a row that has been locked by another transaction B (by way of a DML or SELECT... FOR UPDATE statement), then A's DML command blocks until B commits or rolls back. Once B commits, transaction A can see changes that B has made to the database. For most applications, this concurrency model is the appropriate one, because it provides higher concurrency and thus better performance. But some rare cases require transactions to be serializable. Serializable transactions must execute in such a way that they appear to be executing one at a time (serially), rather than concurrently. Concurrent transactions executing in serialized mode can only make database changes that they could have made if the transactions ran one after the other. Figure 2–1 shows a serializable transaction (B) interacting with another transaction (A). The ANSI/ISO SQL standard SQL92 defines three possible kinds of transaction interaction, and four levels of isolation that provide increasing protection against these interactions. These interactions and isolation levels are summarized in Table 2–4. Table 2–4 Summary of ANSI Isolation Levels Isolation Level Dirty Read1 Non-Repeatable Read2 Phantom Read3 READ UNCOMMITTED Possible Possible Possible READ COMMITTED Not possible Possible Possible REPEATABLE READ Not possible Not possible Possible SERIALIZABLE Not possible Not possible Not possible 1 2 3 A transaction can read uncommitted data changed by another transaction. A transaction rereads data committed by another transaction and sees the new data. A transaction can execute a query again, and discover new rows inserted by another committed transaction. 2-14 Oracle Database Application Developer’s Guide - Fundamentals Using Serializable Transactions for Concurrency Control The behavior of Oracle Database with respect to these isolation levels is summarized in Table 2–5. Table 2–5 ANSI Isolation Levels and Oracle Database Isolation Level Description READ UNCOMMITTED Oracle Database never permits "dirty reads." Although some other database products use this undesirable technique to improve thoughput, it is not required for high throughput with Oracle Database. READ COMMITTED Oracle Database meets the READ COMMITTED isolation standard. This is the default mode for all Oracle Database applications. Because an Oracle Database query only sees data that was committed at the beginning of the query (the snapshot time), Oracle Database actually offers more consistency than is required by the ANSI/ISO SQL92 standards for READ COMMITTED isolation. REPEATABLE READ Oracle Database does not normally support this isolation level, except as provided by SERIALIZABLE. SERIALIZABLE Oracle Database does not normally support this isolation level, except as provided by SERIALIZABLE. How Serializable Transactions Interact Figure 2–1 on page 2-16 shows how a serializable transaction (Transaction B) interacts with another transaction (A, which can be either SERIALIZABLE or READ COMMITTED). When a serializable transaction fails with an ORA-08177 error ("cannot serialize access"), the application can take any of several actions: ■ ■ ■ Commit the work executed to that point Execute additional, different, statements, perhaps after rolling back to a prior savepoint in the transaction Roll back the entire transaction and try it again Oracle Database stores control information in each data block to manage access by concurrent transactions. To use the SERIALIZABLE isolation level, you must use the INITRANS clause of the CREATE TABLE or ALTER TABLE command to set aside storage for this control information. To use serializable mode, INITRANS must be set to at least 3. SQL Processing for Application Developers 2-15 Using Serializable Transactions for Concurrency Control Figure 2–1 Time Line for Two Transactions TRANSACTION A (arbitrary) begin work update row 2 in block 1 insert row 4 Issue update "too recent" for B to see SET TRANSACTION ISOLATION LEVEL SERIALIZABLE read row 1 in block 1 Change other row in same block, see own changes update row 1 in block 1 read updated row 1 in block 1 Create possible "phantom" row Uncommitted changes invisible commit TRANSACTION B (serializable) read old row 2 in block 1 search for row 4 (notfound) Make changes visible to transactions that begin later Make changes after A commits update row 3 in block 1 B can see its own changes but not the committed changes of transaction A. re-read updated row 1 in block 1 search for row 4 (not found) read old row 2 in block 1 Failure on attempt to update row updated and committed since transaction B began update row 2 in block 1 FAILS; rollback and retry TIME Setting the Isolation Level of a Transaction You can change the isolation level of a transaction using the ISOLATION LEVEL clause of the SET TRANSACTION command, which must be the first command issued in a transaction. Use the ALTER SESSION command to set the transaction isolation level on a session-wide basis. Oracle Database Reference for the complete syntax of the SET TRANSACTION and ALTER SESSION commands See Also: The INITRANS Parameter Oracle Database stores control information in each data block to manage access by concurrent transactions. Therefore, if you set the transaction isolation level to 2-16 Oracle Database Application Developer’s Guide - Fundamentals Using Serializable Transactions for Concurrency Control SERIALIZABLE, then you must use the ALTER TABLE command to set INITRANS to at least 3. This parameter causes Oracle Database to allocate sufficient storage in each block to record the history of recent transactions that accessed the block. Higher values should be used for tables that will undergo many transactions updating the same blocks. Referential Integrity and Serializable Transactions Because Oracle Database does not use read locks, even in SERIALIZABLE transactions, data read by one transaction can be overwritten by another. Transactions that perform database consistency checks at the application level should not assume that the data they read will not change during the execution of the transaction (even though such changes are not visible to the transaction). Database inconsistencies can result unless such application-level consistency checks are coded carefully, even when using SERIALIZABLE transactions. Note, however, that the examples shown in this section are applicable for both READ COMMITTED and SERIALIZABLE transactions. Figure 2–2 on page 2-17 shows two different transactions that perform application-level checks to maintain the referential integrity parent/child relationship between two tables. One transaction checks that a row with a specific primary key value exists in the parent table before inserting corresponding child rows. The other transaction checks to see that no corresponding detail rows exist before deleting a parent row. In this case, both transactions assume (but do not ensure) that data they read will not change before the transaction completes. Figure 2–2 Referential Integrity Check B's query does not prevent this insert TRANSACTION A TRANSACTION B read parent (it exists) read child rows (not found) insert child row(s) delete parent commit work commit work A's query does not prevent this delete The read issued by transaction A does not prevent transaction B from deleting the parent row, and transaction B's query for child rows does not prevent transaction A from inserting child rows. This scenario leaves a child row in the database with no corresponding parent row. This result occurs even if both A and B are SERIALIZABLE transactions, because neither transaction prevents the other from making changes in the data it reads to check consistency. SQL Processing for Application Developers 2-17 Using Serializable Transactions for Concurrency Control As this example shows, sometimes you must take steps to ensure that the data read by one transaction is not concurrently written by another. This requires a greater degree of transaction isolation than defined by SQL92 SERIALIZABLE mode. Using SELECT FOR UPDATE Fortunately, it is straightforward in Oracle Database to prevent the anomaly described: ■ ■ Transaction A can use SELECT FOR UPDATE to query and lock the parent row and thereby prevent transaction B from deleting the row. Transaction B can prevent Transaction A from gaining access to the parent row by reversing the order of its processing steps. Transaction B first deletes the parent row, and then rolls back if its subsequent query detects the presence of corresponding rows in the child table. Referential integrity can also be enforced in Oracle Database using database triggers, instead of a separate query as in Transaction A. For example, an INSERT into the child table can fire a BEFORE INSERT row-level trigger to check for the corresponding parent row. The trigger queries the parent table using SELECT FOR UPDATE, ensuring that parent row (if it exists) remains in the database for the duration of the transaction inserting the child row. If the corresponding parent row does not exist, the trigger rejects the insert of the child row. SQL statements issued by a database trigger execute in the context of the SQL statement that caused the trigger to fire. All SQL statements executed within a trigger see the database in the same state as the triggering statement. Thus, in a READ COMMITTED transaction, the SQL statements in a trigger see the database as of the beginning of the triggering statement execution, and in a transaction executing in SERIALIZABLE mode, the SQL statements see the database as of the beginning of the transaction. In either case, the use of SELECT FOR UPDATE by the trigger correctly enforces referential integrity. READ COMMITTED and SERIALIZABLE Isolation Oracle Database gives you a choice of two transaction isolation levels with different characteristics. Both the READ COMMITTED and SERIALIZABLE isolation levels provide a high degree of consistency and concurrency. Both levels reduce contention, and are designed for deploying real-world applications. The rest of this section compares the two isolation modes and provides information helpful in choosing between them. Transaction Set Consistency A useful way to describe the READ COMMITTED and SERIALIZABLE isolation levels in Oracle Database is to consider: ■ A collection of database tables (or any set of data) ■ A sequence of reads of rows in those tables ■ The set of transactions committed at any moment An operation (a query or a transaction) is transaction set consistent if its read operations all return data written by the same set of committed transactions. When an operation is not transaction set consistent, some reads reflect the changes of one set of transactions, and other reads reflect changes made by other transactions. Such an operation sees the database in a state that reflects no single set of committed transactions. 2-18 Oracle Database Application Developer’s Guide - Fundamentals Using Serializable Transactions for Concurrency Control Oracle Database transactions executing in READ COMMITTED mode are transaction-set consistent on an individual-statement basis, because all rows read by a query must be committed before the query begins. Oracle Database transactions executing in SERIALIZABLE mode are transaction set consistent on an individual-transaction basis, because all statements in a SERIALIZABLE transaction execute on an image of the database as of the beginning of the transaction. In other database systems, a single query run in READ COMMITTED mode provides results that are not transaction set consistent. The query is not transaction set consistent, because it may see only a subset of the changes made by another transaction. For example, a join of a master table with a detail table could see a master record inserted by another transaction, but not the corresponding details inserted by that transaction, or vice versa. The READ COMMITTED mode avoids this problem, and so provides a greater degree of consistency than read-locking systems. In read-locking systems, at the cost of preventing concurrent updates, SQL92 REPEATABLE READ isolation provides transaction set consistency at the statement level, but not at the transaction level. The absence of phantom protection means two queries issued by the same transaction can see data committed by different sets of other transactions. Only the throughput-limiting and deadlock-susceptible SERIALIZABLE mode in these systems provides transaction set consistency at the transaction level. Comparison of READ COMMITTED and SERIALIZABLE Transactions Table 2–6 summarizes key similarities and differences between READ COMMITTED and SERIALIZABLE transactions. Table 2–6 Read Committed Versus Serializable Transaction Operation Read Committed Serializable Dirty write Not Possible Not Possible Dirty read Not Possible Not Possible Non-repeatable read Possible Not Possible Phantoms Possible Not Possible Compliant with ANSI/ISO SQL 92 Yes Yes Read snapshot time Statement Transaction Transaction set consistency Statement level Transaction level Row-level locking Yes Yes Readers block writers No No Writers block readers No No Different-row writers block writers No No Same-row writers block writers Yes Yes Waits for blocking transaction Yes Yes Subject to "can't serialize access" error No Yes Error after blocking transaction aborts No No Error after blocking transaction commits No Yes SQL Processing for Application Developers 2-19 Autonomous Transactions Choosing an Isolation Level for Transactions Choose an isolation level that is appropriate to the specific application and workload. You might choose different isolation levels for different transactions. The choice depends on performance and consistency needs, and consideration of application coding requirements. For environments with many concurrent users rapidly submitting transactions, you must assess transaction performance against the expected transaction arrival rate and response time demands, and choose an isolation level that provides the required degree of consistency while performing well. Frequently, for high performance environments, you must trade-off between consistency and concurrency (transaction throughput). Both Oracle Database isolation modes provide high levels of consistency and concurrency (and performance) through the combination of row-level locking and Oracle Database's multi-version concurrency control system. Because readers and writers do not block one another in Oracle Database, while queries still see consistent data, both READ COMMITTED and SERIALIZABLE isolation provide a high level of concurrency for high performance, without the need for reading uncommitted ("dirty") data. READ COMMITTED isolation can provide considerably more concurrency with a somewhat increased risk of inconsistent results (due to phantoms and non-repeatable reads) for some transactions. The SERIALIZABLE isolation level provides somewhat more consistency by protecting against phantoms and non-repeatable reads, and may be important where a read/write transaction executes a query more than once. However, SERIALIZABLE mode requires applications to check for the "can't serialize access" error, and can significantly reduce throughput in an environment with many concurrent transactions accessing the same data for update. Application logic that checks database consistency must take into account the fact that reads do not block writes in either mode. Application Tips for Transactions When a transaction runs in serializable mode, any attempt to change data that was changed by another transaction since the beginning of the serializable transaction causes an error: ORA-08177: Can't serialize access for this transaction. When you get this error, roll back the current transaction and execute it again. The transaction gets a new transaction snapshot, and the operation is likely to succeed. To minimize the performance overhead of rolling back transactions and executing them again, try to put DML statements that might conflict with other concurrent transactions near the beginning of your transaction. Autonomous Transactions This section gives a brief overview of autonomous transactions and what you can do with them. See Also: Oracle Database PL/SQL User's Guide and Reference and Chapter 9, "Coding Triggers" for detailed information on autonomous transactions 2-20 Oracle Database Application Developer’s Guide - Fundamentals Autonomous Transactions At times, you may want to commit or roll back some changes to a table independently of a primary transaction's final outcome. For example, in a stock purchase transaction, you may want to commit a customer's information regardless of whether the overall stock purchase actually goes through. Or, while running that same transaction, you may want to log error messages to a debug table even if the overall transaction rolls back. Autonomous transactions allow you to do such tasks. An autonomous transaction (AT) is an independent transaction started by another transaction, the main transaction (MT). It lets you suspend the main transaction, do SQL operations, commit or roll back those operations, then resume the main transaction. An autonomous transaction executes within an autonomous scope. An autonomous scope is a routine you mark with the pragma (compiler directive) AUTONOMOUS_TRANSACTION. The pragma instructs the PL/SQL compiler to mark a routine as autonomous (independent). In this context, the term routine includes: ■ Top-level (not nested) anonymous PL/SQL blocks ■ Local, standalone, and packaged functions and procedures ■ Methods of a SQL object type ■ PL/SQL triggers Figure 2–3 shows how control flows from the main routine (MT) to an autonomous routine (AT) and back again. As you can see, the autonomous routine can commit more than one transaction (AT1 and AT2) before control returns to the main routine. Figure 2–3 Transaction Control Flow Main Routine Autonomous Routine PROCEDURE proc1 IS emp_id NUMBER; BEGIN emp_id := 7788; INSERT ... SELECT ... proc2; DELETE ... COMMIT; END; PROCEDURE proc2 IS PRAGMA AUTON... dept_id NUMBER; BEGIN dept_id := 20; UPDATE ... INSERT ... UPDATE ... COMMIT; INSERT ... INSERT ... COMMIT; END; MT begins MT ends MT suspends AT1 begins AT1 ends AT2 begins AT2 ends MT resumes When you enter the executable section of an autonomous routine, the main routine suspends. When you exit the routine, the main routine resumes. COMMIT and ROLLBACK end the active autonomous transaction but do not exit the autonomous routine. As Figure 2–3 shows, when one transaction ends, the next SQL statement begins another transaction. A few more characteristics of autonomous transactions: ■ The changes autonomous transactions effect do not depend on the state or the eventual disposition of the main transaction. For example: – An autonomous transaction does not see any changes made by the main transaction. SQL Processing for Application Developers 2-21 Autonomous Transactions – ■ ■ When an autonomous transaction commits or rolls back, it does not affect the outcome of the main transaction. The changes an autonomous transaction effects are visible to other transactions as soon as that autonomous transaction commits. This means that users can access the updated information without having to wait for the main transaction to commit. Autonomous transactions can start other autonomous transactions. Figure 2–4 illustrates some of the possible sequences autonomous transactions can follow. Figure 2–4 Possible Sequences of Autonomous Transactions A main transaction scope (MT Scope) begins the main transaction, MTx. MTx invokes the first autonomous transaction scope (AT Scope1). MTx suspends. AT Scope 1 begins the transaction Tx1.1. At Scope 1 commits or rolls back Tx1.1, than ends. MTx resumes. MT Scope AT Scope 1 AT Scope 2 AT Scope 3 AT Scope 4 MTx Tx1.1 MTx MTx invokes AT Scope 2. MT suspends, passing control to AT Scope 2 which, initially, is performing queries. AT Scope 2 then begins Tx2.1 by, say, doing an update. AT Scope 2 commits or rolls back Tx2.1. Tx2.1 Later, AT Scope 2 begins a second transaction, Tx2.2, then commits or rolls it back. AT Scope 2 performs a few queries, then ends, passing control back to MTx. Tx2.2 MTx MTx invokes AT Scope 3. MTx suspends, AT Scope 3 begins. Tx3.1 AT Scope 3 begins Tx3.1 which, in turn, invokes AT Scope 4. Tx3.1 suspends, AT Scope 4 begins. Tx4.1 AT Scope 4 begins Tx4.1, commits or rolls it back, then ends. AT Scope 3 resumes. AT Scope 3 commits or rolls back Tx3.1, then ends. MTx resumes. Finally, MT Scope commits or rolls back MTx, then ends. Tx3.1 MTx Examples of Autonomous Transactions The two examples in this section illustrate some of the ways you can use autonomous transactions. 2-22 Oracle Database Application Developer’s Guide - Fundamentals Autonomous Transactions As these examples illustrate, there are four possible outcomes that can occur when you use autonomous and main transactions. Table 2–7 presents these possible outcomes. As you can see, there is no dependency between the outcome of an autonomous transaction and that of a main transaction. Table 2–7 Possible Transaction Outcomes Autonomous Transaction Main Transaction Commits Commits Commits Rolls back Rolls back Commits Rolls back Rolls back Entering a Buy Order In this example, illustrated by Figure 2–5, a customer enters a buy order. That customer's information (such as name, address, phone) is committed to a customer information table—even though the sale does not go through. Figure 2–5 Example: A Buy Order MT Scope begins the main transaction, MTx inserts the buy order into a table. MT Scope MTx invokes the autonomous transaction scope (AT Scope). When AT Scope begins, MT Scope suspends. AT Scope ATx ATx, updates the audit table with customer information. MTx seeks to validate the order, finds that the selected item is unavailable, and therefore rolls back the main transaction. MTx Example: Making a Bank Withdrawal In this example, a customer tries to make a withdrawal from a bank account. In the process, a main transaction calls one of two autonomous transaction scopes (AT Scope 1, and AT Scope 2). The following diagrams illustrate three possible scenarios for this transaction. ■ ■ ■ Scenario 1: There are sufficient funds to cover the withdrawal and therefore the bank releases the funds Scenario 2: There are insufficient funds to cover the withdrawal, but the customer has overdraft protection. The bank therefore releases the funds. Scenario 3: There are insufficient funds to cover the withdrawal, the customer does not have overdraft protection, and the bank therefore withholds the requested funds. Scenario 1 There are sufficient funds to cover the withdrawal and therefore the bank releases the funds. This is illustrated by Figure 2–6. SQL Processing for Application Developers 2-23 Autonomous Transactions Figure 2–6 Example: Bank Withdrawal—Sufficient Funds MTx generates a transaction ID. MT Scope AT Scope 1 AT Scope 2 MTx Tx1.1 inserts the transaction ID into the audit table and commits. MTx validates the balance on the account. Tx1.1 MTx Tx2.1, updates the audit table using the transaction ID generated above, then commits. MTx releases the funds. MT Scope ends. Tx2.1 MTx Scenario 2 There are insufficient funds to cover the withdrawal, but the customer has overdraft protection. The bank therefore releases the funds. This is illustrated by Figure 2–7. 2-24 Oracle Database Application Developer’s Guide - Fundamentals Autonomous Transactions Figure 2–7 Example: Bank Withdrawal—Insufficient Funds WITH Overdraft Protection MT Scope AT Scope 1 AT Scope 2 MTx Tx1.1 MTx discovers that there are insufficient funds to cover the withdrawal. It finds that the customer has overdraft protection and sets a flag to the appropriate value. MTx Tx2.1, updates the audit table. MTx, releases the funds. MT Scope ends. Tx2.1 MTx Scenario 3 There are insufficient funds to cover the withdrawal, the customer does not have overdraft protection, and the bank therefore withholds the requested funds. This is illustrated by Figure 2–8. SQL Processing for Application Developers 2-25 Autonomous Transactions Figure 2–8 Example: Bank Withdrawal—Insufficient Funds WITHOUT Overdraft Protection MT Scope AT Scope 1 AT Scope 2 MTx Tx1.1 MTx discovers that there are insufficient funds to cover the withdrawal. It finds that the customer does not have overdraft protection and sets a flag to the appropriate value. MTx Tx2.1, updates the audit table. MTx Scope rolls back MTx, denying the release of funds. MT Scope ends. Tx2.1 MTx Defining Autonomous Transactions This section is provided here to round out your general understanding of autonomous transactions. For a more thorough understanding of autonomous transactions, refer to Oracle Database PL/SQL User's Guide and Reference. Note: To define autonomous transactions, you use the pragma (compiler directive) AUTONOMOUS_TRANSACTION. The pragma instructs the PL/SQL compiler to mark the procedure, function, or PL/SQL block as autonomous (independent). You can code the pragma anywhere in the declarative section of a procedure, function, or PL/SQL block. But, for readability, code the pragma at the top of the section. The syntax follows: PRAGMA AUTONOMOUS_TRANSACTION; In the following example, you mark a packaged function as autonomous: CREATE OR REPLACE PACKAGE Banking AS FUNCTION Balance (Acct_id INTEGER) RETURN REAL; -- add additional functions and packages END Banking; CREATE OR REPLACE PACKAGE BODY Banking AS FUNCTION Balance (Acct_id INTEGER) RETURN REAL IS PRAGMA AUTONOMOUS_TRANSACTION; My_bal REAL; 2-26 Oracle Database Application Developer’s Guide - Fundamentals Resuming Execution After a Storage Error Condition BEGIN --add appropriate code END; -- add additional functions and packages... END Banking; Restrictions on Autonomous Transactions Note the following restrictions on autonomous transactions. ■ You cannot use the pragma to mark all subprograms in a package (or all methods in an object type) as autonomous. Only individual routines can be marked autonomous. For example, the following pragma is illegal: CREATE OR REPLACE PACKAGE Banking AS PRAGMA AUTONOMOUS_TRANSACTION; -- illegal FUNCTION Balance (Acct_id INTEGER) RETURN REAL; END Banking; ■ You cannot execute a PIPE ROW statement in your autonomous routine while your autonomous transaction is open. You must close the autonomous transaction before executing the PIPE ROW statement. This is normally accomplished by committing or rolling back the autonomous transaction before executing the PIPE ROW statement. See Also: Oracle Database PL/SQL User's Guide and Reference Resuming Execution After a Storage Error Condition When a long-running transaction is interrupted by an out-of-space error condition, your application can suspend the statement that encountered the problem and resume it after the space problem is corrected. This capability is known as resumable storage allocation. It lets you avoid time-consuming rollbacks, without the need to split the operation into smaller pieces and write your own code to track its progress. See Also: ■ Oracle Database Concepts ■ Oracle Database Administrator's Guide What Operations Can Be Resumed After an Error Condition? Queries, DML operations, and certain DDL operations can all be resumed if they encounter an out-of-space error. The capability applies if the operation is performed directly by a SQL statement, or if it is performed within a stored procedure, anonymous PL/SQL block, SQL*Loader, or an OCI call such as OCIStmtExecute(). Operations can be resumed after these kinds of error conditions: ■ Out of space errors, such as ORA-01653. ■ Space limit errors, such as ORA-01628. ■ Space quota errors, such as ORA-01536. Limitations on Resuming Operations After an Error Condition Certain storage errors cannot be handled using this technique. In dictionary-managed tablespaces, you cannot resume an operation if you run into the limit for rollback SQL Processing for Application Developers 2-27 Resuming Execution After a Storage Error Condition segments, or the maximum number of extents while creating an index or a table. Use locally managed tablespaces and automatic undo management in combination with this feature. Writing an Application to Handle Suspended Storage Allocation When an operation is suspended, your application does not receive the usual error code. Instead, perform any logging or notification by coding a trigger to detect the AFTER SUSPEND event and call the functions in the DBMS_RESUMABLE package to get information about the problem. Using this package, you can: ■ ■ Parse the error message with the DBMS_RESUMABLE.SPACE_ERROR_INFO function. For details about this function, refer to Oracle Database PL/SQL Packages and Types Reference. Set a new timeout value with the SET_TIMEOUT procedure. Within the body of the trigger, you can perform any notifications, such as sending a mail message to alert an operator to the space problem. Alternatively, the DBA can periodically check for suspended statements using the data dictionary views DBA_RESUMABLE, USER_RESUMABLE, and V$_SESSION_WAIT. When the space condition is corrected (usually by the DBA), the suspended statement automatically resumes execution. If it is not corrected before the timeout period expires, the operation causes a SERVERERROR exception. To reduce the chance of out-of-space errors within the trigger itself, you must declare it as an autonomous transaction so that it uses a rollback segment in the SYSTEM tablespace. If the trigger encounters a deadlock condition because of locks held by the suspended statement, the trigger is aborted and your application receives the original error condition, as if it was never suspended. If the trigger encounters an out-of-space condition, the trigger and the suspended statement are rolled back. You can prevent the rollback through an exception handler in the trigger, and just wait for the statement to be resumed. See Also: Oracle Database Reference for details on the DBA_RESUMABLE, USER_RESUMABLE, and V$_SESSION_WAIT data dictionary views Example of Resumable Storage Allocation This trigger handles applicable storage errors within the database. For some kinds of errors, it aborts the statement and alerts the DBA that this has happened through a mail message. For other errors that might be temporary, it specifies that the statement should wait for eight hours before resuming, with the expectation that the storage problem will be fixed by then. CREATE OR REPLACE TRIGGER suspend_example AFTER SUSPEND ON DATABASE DECLARE cur_sid NUMBER; cur_inst NUMBER; err_type VARCHAR2(64); object_owner VARCHAR2(64); object_type VARCHAR2(64); table_space_name VARCHAR2(64); object_name VARCHAR2(64); sub_object_name VARCHAR2(64); 2-28 Oracle Database Application Developer’s Guide - Fundamentals Resuming Execution After a Storage Error Condition msg_body VARCHAR2(64); ret_value boolean; error_txt varchar2(64); mail_conn utl_smtp.connection; BEGIN SELECT DISTINCT(sid) INTO cur_sid FROM v$mystat; cur_inst := userenv('instance'); ret_value := dbms_resumable.space_error_info(err_type, object_owner, object_type, table_space_name, object_name, sub_object_name); IF object_type = 'ROLLBACK SEGMENT' THEN INSERT INTO sys.rbs_error ( SELECT sql_text, error_msg, suspend_time FROM dba_resumable WHERE session_id = cur_sid AND instance_id = cur_inst); SELECT error_msg into error_txt FROM dba_resumable WHERE session_id = cur_sid AND instance_id = cur_inst; msg_body := 'Subject: Space error occurred: Space limit reached for rollback segment '|| object_name || ' on ' || to_char(SYSDATE, 'Month dd, YYYY, HH:MIam') || '. Error message was: ' || error_txt; mail_conn := utl_smtp.open_connection('localhost', 25); utl_smtp.helo(mail_conn, 'localhost'); utl_smtp.mail(mail_conn, 'sender@localhost'); utl_smtp.rcpt(mail_conn, 'recipient@localhost'); utl_smtp.data(mail_conn, msg_body); utl_smtp.quit(mail_conn); dbms_resumable.abort(cur_sid); ELSE dbms_resumable.set_timeout(3600*8); END IF; COMMIT; END; SQL Processing for Application Developers 2-29 Resuming Execution After a Storage Error Condition 2-30 Oracle Database Application Developer’s Guide - Fundamentals 3 Using SQL Datatypes in Application Development This chapter discusses how to use SQL datatypes in database applications. Topics include the following: ■ Representing Data with SQL Datatypes: Overview ■ Representing Character Data ■ Representing Numeric Data ■ Representing Datetime Data ■ Representing Specialized Data ■ Representing Conditional Expressions as Data ■ Identifying Rows by Address ■ How Oracle Database Converts Datatypes See Also: ■ ■ ■ Oracle Database Application Developer's Guide - Object-Relational Features for information about more complex types, such as object types, varrays, and nested tables Oracle Database Application Developer's Guide - Large Objects for information about LOB datatypes Oracle Database PL/SQL User's Guide and Reference to learn about the PL/SQL datatypes. Many SQL datatypes are the same or similar in PL/SQL. Representing Data with SQL Datatypes: Overview A datatype associates a fixed set of properties with the values that can be used in a column of a table or in an argument of a procedure or function. These properties cause Oracle Database to treat values of one datatype differently from values of another datatype. For example, Oracle Database can add values of NUMBER datatype, but not values of RAW datatype. Oracle Database provides a number of built-in datatypes as well as several categories for user-defined types that can be used as datatypes. The datatypes supported by Oracle Database can be divided into the following categories: Using SQL Datatypes in Application Development 3-1 Representing Character Data ■ ■ ■ ■ Oracle built-in datatypes, which include datatypes for characters, numbers, dates and times (known as datetime datatypes), raw data, large objects (LOBs), and row addresses (ROWIDs). ANSI datatypes and datatypes from the IBM products SQL/DS and DB2, which are usable in SQL statements that create tables and clusters User-defined types, which use Oracle built-in datatypes and other user-defined datatypes as the building blocks of object types that model the structure and behavior of data in applications Oracle-supplied types, which are SQL-based interfaces for defining new types The Oracle precompilers recognize other datatypes in embedded SQL programs. These datatypes are called external datatypes and are associated with host variables. You should not confuse Oracle Database built-in datatypes and user-defined types with external datatypes. See Also: ■ ■ ■ Oracle Database SQL Reference for complete reference information on the SQL datatypes Pro*COBOL Programmer's Guide and Pro*C/C++ Programmer's Guide for information on external datatypes, including how Oracle converts between them and built-in or user-defined types Oracle Database Concepts to learn about Oracle built-in datatypes Representing Character Data This section contains the following topics: ■ Representing Character Data: Overview ■ Specifying Column Lengths as Bytes or Characters ■ Choosing Between the CHAR and VARCHAR2 Datatypes ■ Using Character Literals in SQL Statements Representing Character Data: Overview You can use the following SQL datatypes to store alphanumeric data: ■ CHAR and NCHAR datatypes store fixed-length character literals. ■ VARCHAR2 and NVARCHAR2 datatypes store variable-length character literals. ■ NCHAR and NVARCHAR2 datatypes store Unicode character data only. ■ ■ CLOB and NCLOB datatypes store single-byte and multibyte character strings of up to (4 gigabytes - 1) * (the value obtained from DBMS_LOB.GETCHUNKSIZE). The LONG datatype stores variable-length character strings containing up to two gigabytes, but with many restrictions. This datatype is provided only for backward compatibility with existing applications. In general, new applications should use CLOB and NCLOB datatypes to store large amounts of character data, and BLOB and BFILE to store large amounts of binary data. 3-2 Oracle Database Application Developer’s Guide - Fundamentals Representing Character Data See Also: ■ ■ Oracle Database Application Developer's Guide - Large Objects for information on LOB datatypes (including CLOB and NCLOB datatypes) and migration from LONG to LOB datatypes Oracle Database SQL Reference for restrictions on LONG datatypes Specifying Column Lengths as Bytes or Characters You can specify the lengths of CHAR and VARCHAR2 columns as either bytes or characters. The lengths of NCHAR and NVARCHAR2 columns are always specified in characters, making them ideal for storing Unicode data, where a character might consist of multiple bytes. Consider the following list of column length specifications: ■ id VARCHAR2(32 BYTE) The id column contains only single-byte data, up to 32 bytes. ■ name VARCHAR2(32 CHAR) The name column contains data in the database character set. If the database character set allows multibyte characters, then the 32 characters can be stored as more than 32 bytes. ■ biography NVARCHAR2(2000) The biography column can represent 2000 characters in any Unicode-representable language. The encoding depends on the national character set, but the column can contain multibyte values even if the database character set is single-byte. ■ comment VARCHAR2(2000) The representation of comment as 2000 bytes or characters depends on the initialization parameter NLS_LENGTH_SEMANTICS. When using a multibyte database character encoding scheme, consider carefully the space required for tables with character columns. If the database character encoding scheme is single-byte, then the number of bytes and the number of characters in a column is the same. If it is multibyte, however, then there generally is no such correspondence. A character might consist of one or more bytes, depending upon the specific multibyte encoding scheme and whether shift-in/shift-out control codes are present. To avoid overflowing buffers, specify data as NCHAR or NVARCHAR2 if it might use a Unicode encoding that is different from the database character set. See Also: ■ Oracle Database Globalization Support Guide ■ Oracle Database SQL Reference Choosing Between the CHAR and VARCHAR2 Datatypes When deciding which datatype to use for a column that will store alphanumeric data in a table, consider the following points of distinction: ■ Space usage To store data more efficiently, use the VARCHAR2 datatype. The CHAR datatype blank-pads and stores trailing blanks up to a fixed column length for all column values, whereas the VARCHAR2 datatype does not add extra blanks. Using SQL Datatypes in Application Development 3-3 Representing Character Data ■ Comparison semantics Use the CHAR datatype when you require ANSI compatibility in comparison semantics (when trailing blanks are not important in string comparisons). Use the VARCHAR2 when trailing blanks are important in string comparisons. ■ Future compatibility The CHAR and VARCHAR2 datatypes are fully supported. At this time, the VARCHAR datatype automatically corresponds to the VARCHAR2 datatype and is reserved for future use. When an application interfaces with Oracle Database, there is a character set on the client and server side. Oracle Database uses the NLS_LANGUAGE parameter to automatically convert CHAR, VARCHAR2, and LONG data from the database character set to the character set defined for the user session, if these are different. In the section "Datatype Comparison Rules," Oracle Database SQL Reference explains the comparison semantics that Oracle Database uses to compare character data. Because Oracle Database blank-pads values stored in CHAR columns but not in VARCHAR2 columns, a value stored in a VARCHAR2 column can take up less space than the same value in a CHAR column. For this reason, a full table scan on a large table containing VARCHAR2 columns may read fewer data blocks than a full table scan on a table containing the same data stored in CHAR columns. If your application often performs full table scans on large tables containing character data, then you may be able to improve performance by storing data in VARCHAR2 rather than in CHAR columns. Performance is not the only factor to consider when deciding which datatype to use. Oracle Database uses different semantics to compare values of each datatype. You might choose one datatype over the other if your application is sensitive to the differences between these semantics. For example, if you want Oracle Database to ignore trailing blanks when comparing character values, then you must store these values in CHAR columns. Oracle Database SQL Reference for more information on comparison semantics for these datatypes See Also: Using Character Literals in SQL Statements Many SQL statements, functions, expressions, and conditions require you to specify character literal values. You can specify character literals with the following notations: ■ ■ Character literals with the 'text' notation, as in the literals 'users01.dbf' and 'Muthu''s computer'. National character literals with the N'text' or n'text' notation, where N or n specifies the literal using the national character set. For example, N'résumé' is a National character literal. Oracle Database translates N-quoted text into the national character set by way of the database character set. If client-side characters do not have corresponding encoding in the database character set, then Oracle Database converts them into question marks. To avoid the potential loss of data during the text literal conversion, set the environment variable $ORA_NCHAR_LITERAL_REPLACE to TRUE. This setting transparently replaces the N'text' internally and preserves the text literal for SQL processing. The UNISTR function provides support for Unicode character literals by enabling you to specify the Unicode encoding value of characters in the string, as in UNISTR('\1234'). This technique is useful, for example, when inserting data into 3-4 Oracle Database Application Developer’s Guide - Fundamentals Representing Numeric Data NCHAR columns. Because every character has a corresponding Unicode encoding, the client application can safely send character data to the server without data loss. Quoting Character Literals By default you must quote character literals in single-quotes, as in 'Hello'. This technique can sometimes be inconvenient if the text itself contains single quotes. In such cases, you can also use the Q-quote mechanism, which enables you to specify q or Q followed by a single quote and then another character to be used as the quote delimiter. For example, the literal q'#it's the "final" deadline#' uses the pound sign (#) as a quote delimiter for the string it's the "final" deadline. The Q-quote delimiter can be any single- or multibyte character except space, tab, and return. If the opening quote delimiter is a [, {, <, or ( character, then the closing quote delimiter must be the corresponding ], }, >, or ) character. In all other cases, the opening and closing delimiter must be the identical character. The following character literals use the alternative quoting mechanism: q'(name LIKE '%DBMS_%%')' q'<'Data,' he said, 'Make it so.'>' q'"name like '['"' nq'ïŸ1234ï' See Also: ■ ■ Oracle Database Globalization Support Guide to learn about national character sets Oracle Database SQL Reference to learn about character literals Representing Numeric Data This section contains the following topics: ■ What Are the Numeric Datatypes? ■ Using Floating-Point Number Formats ■ Using Comparison Operators for Native Floating-Point Datatypes ■ Performing Arithmetic Operations with Native Floating-Point Datatypes ■ Using Conversion Functions with Native Floating-Point Datatypes ■ Client Interfaces for Native Floating-Point Datatypes What Are the Numeric Datatypes? The following SQL datatypes store numeric data: ■ NUMBER ■ BINARY_FLOAT ■ BINARY_DOUBLE Use the NUMBER datatype to store real numbers in a fixed-point or floating-point format. Numbers using this datatype are guaranteed to be portable among different Oracle Database platforms, and offer up to 38 decimal digits of precision. You can store positive and negative numbers of magnitude 1 x 10-130 through 9.99 x10125, as well as zero, in a NUMBER column. Using SQL Datatypes in Application Development 3-5 Representing Numeric Data The BINARY_FLOAT and BINARY_DOUBLE datatypes store floating-point data in the 32-bit IEEE 754 format and the double precision 64-bit IEEE 754 format respectively. Compared to the Oracle NUMBER datatype, arithmetic operations on floating-point data are usually faster for BINARY_FLOAT and BINARY_DOUBLE. Also, high-precision values require less space when stored as BINARY_FLOAT and BINARY_DOUBLE. In client interfaces supported by Oracle Database, the native instruction set supplied by the hardware vendor performs arithmetic operations on BINARY_FLOAT and BINARY_DOUBLE datatypes. The term native floating-point datatypes refers to datatypes including BINARY_FLOAT and BINARY_DOUBLE and to all implementations of these types in supported client interfaces. The floating-point number system is a common way of representing and manipulating numeric values in computer systems. A floating-point number is characterized by the following components: ■ Binary-valued sign ■ Signed exponent ■ Significand ■ Base A floating-point value is the signed product of its significand and the base raised to the power of its exponent, as shown in the formula in Example 3–1. Example 3–1 Components of a Floating-Point Number (-1)sign . significand . base exponent For example, the number 4.31 can be represented in the following expression: (-1)0 . 431 . 10 -2 The components of the preceding expression are as follows: ■ 0 is the binary-valued sign ■ 431 is the significant ■ 10 is the base ■ -2 is the exponent See Also: ■ Oracle Database Concepts for information about the internal format for the NUMBER datatype Oracle Database SQL Reference for more information about the BINARY_FLOAT and BINARY_DOUBLE datatypes formats Using Floating-Point Number Formats A floating-point number format specifies how components of a floating-point number are represented. The choice of representation determines the range and precision of the values the format can represent. By definition, the range is the interval bounded by the smallest and the largest values the format can represent and the precision is the number of digits in the significand. Formats for floating-point values support neither infinite precision nor infinite range. There are a finite number of bits to represent a number and only a finite number of 3-6 Oracle Database Application Developer’s Guide - Fundamentals Representing Numeric Data values that a format can represent. A floating-point number that uses more precision than available with a given format is rounded. A floating-point number can be represented in a binary system (one that uses base 2), as in the IEEE 754 standard, or in a decimal system (one that uses base 10), such as Oracle NUMBER. The base affects many properties of the format, including how a numeric value is rounded. For a decimal floating-point number format like Oracle NUMBER, rounding is done to the nearest decimal place (for example. 1000, 10, or 0.01). The IEEE 754 formats use a binary format for floating-point values and round numbers to the nearest binary place (for example: 1024, 512, or 1/64). The native floating-point datatypes supported by the database round to the nearest binary place, so they are not satisfactory for applications that require decimal rounding. Use the Oracle NUMBER datatype for applications in which decimal rounding is required on floating-point data. Using a Floating-Point Binary Format The value of a floating-point number that uses a binary format is determined by the formula in Example 3–2. Example 3–2 Components of a Floating-Point Number in Binary Format (-1)s 2E (b0 b1 b2 ... bp-1) Table 3–1 describes the components of the formula. Table 3–1 Components of the Binary Format for Floating-Point Numbers Component Specifies . . . s 0 or 1 E Any integer between Emin and Emax, inclusive (see Table 3–2) bi 0 or 1, where the sequence of bits represents a number in base 2 (see Table 3–2) The leading bit of the significand, b0, must be set (1), except for subnormal numbers (explained later). Consequently, the leading bit is not actually stored, so the formats provide N bits of precision although only N-1 bits are stored. The IEEE 754 specification also defines extended single-precision and extended double-precision formats, which are not supported by Oracle Database. Note: The parameters for these formats are described in Table 3–2. Table 3–2 Summary of Binary Format Parameters Parameter Single-precision (32-bit) Double-precision (64-bit) p 24 53 Emin -126 -1022 Emax +127 +1023 Using SQL Datatypes in Application Development 3-7 Representing Numeric Data The storage parameters for the formats are described in Table 3–3. The in-memory formats for single-precision and double-precision datatypes are specified by IEEE 754. Table 3–3 Summary of Binary Format Storage Parameters Datatype Sign bits Exponent bits Significand bits Total bits Single-precision 1 8 24 (23 stored) 32 Double-precision 1 11 53 (52 stored) 64 A significand is normalized when the leading bit of the significand is set. IEEE 754 defines denormal or subnormal values as numbers that are too small to be represented with an implied leading set bit in the significand. The number is too small because its exponent would be too large if its significand were normalized to have an implied leading bit set. IEEE 754 formats support subnormal values. Subnormal values preserve the following property: if: x - y == 0.0 (using floating-point subtraction) then: x == y Table 3–4 shows the range and precision of the required formats in the IEEE 754 standard and those of Oracle NUMBER. Range limits are expressed here in terms of positive numbers; they also apply to the absolute value of a negative number. (The notation "number e exponent" used here stands for number multiplied by 10 raised to the exponent power: number . 10 exponent.) Table 3–4 Range and Precision of IEEE 754 formats Range and Precision Single-precision 32-bit1 Double-precision 64-bit1 Max positive normal number 3.40282347e+38 1.7976931348623157e+308 < 1.0e126 Min positive normal number 1.17549435e-38 2.2250738585072014e-308 1.0e-130 Max positive subnormal number 1.17549421e-38 2.2250738585072009e-308 not applicable Min positive subnormal number 1.40129846e-45 4.9406564584124654e-324 not applicable Precision (decimal digits) 6-9 15 - 17 1 Oracle NUMBER Datatype 38 - 40 These numbers are quoted from the IEEE Numerical Computation Guide. See Also: ■ ■ Oracle Database SQL Reference, section "Numeric Literals", for information about literal representation of numeric values Oracle Database SQL Reference for more information about floating-point formats Representing Special Values with Native Floating-Point Formats IEEE 754 allows special values to be represented. These special values are as follows: ■ Positive infinity (+INF) ■ Negative infinity (-INF) 3-8 Oracle Database Application Developer’s Guide - Fundamentals Representing Numeric Data ■ Not-a-number (NaN) ■ Positive zero (+0) ■ Negative zero (-0) NaN represent results of operations that are undefined. Many bit patterns in IEEE 754 represent NaN. Bit patterns can represent NaN with and without the sign bit set. IEEE 754 distinguishes between signalling NaNs and quiet NaNs. IEEE 754 specifies behavior for when exceptions are enabled and disabled. Oracle Database does not allow exceptions to be enabled; the database behavior is that specified by IEEE 754 for when exceptions are disabled. In particular, Oracle Database makes no distinction between signalling and quiet NaNs. Programmers who use OCI can retrieve NaN values from Oracle Database; whether a retrieved NaN value is signalling or quiet depends on the client platform and beyond the control of Oracle Database. IEEE 754 does not define the bit pattern for either type of NaN. Positive infinity, negative infinity, positive zero, and negative zero are each represented by a specific bit pattern. Ignoring signs, there are the following classes of values, with each of the classes except for NaN greater than the one preceding it in the list: ■ Zero ■ Subnormal ■ Normal ■ Infinity ■ NaN In IEEE 754, NaN is unordered with other classes of special values and with itself. Behavior of Special Values for Native Floating-Point Datatypes When used with the database, special values of native floating-point datatypes behave as follows: ■ All NaNs are quiet. ■ IEEE 754 exceptions are not raised. ■ NaN is ordered as follows: All non-NaN < NaN Any NaN == any other NaN ■ -0 is converted to +0. ■ All NaNs are converted to the same bit pattern. See Also: "Using Comparison Operators for Native Floating-Point Datatypes" on page 3-9 for more information on NaN compared to other values Using Comparison Operators for Native Floating-Point Datatypes Oracle Database defines the following comparison operators for operations involving floating-point datatypes: ■ Equal to ■ Not equal to Using SQL Datatypes in Application Development 3-9 Representing Numeric Data ■ Greater than ■ Greater than or equal to ■ Less than ■ Less than or equal to ■ Unordered Note the following special cases: ■ ■ Comparisons ignore the sign of zero (-0 is equal to, not less than, +0). In Oracle Database, NaN is equal to itself. NaN is greater than everything except itself. That is, NaN == NaN and NaN > x, unless x is NaN. See Also: "Behavior of Special Values for Native Floating-Point Datatypes" on page 3-9 for more information on comparison results, ordering, and other behaviors of special values Performing Arithmetic Operations with Native Floating-Point Datatypes Oracle Database defines operators for the following arithmetic operations: ■ Multiplication ■ Division ■ Addition ■ Subtraction ■ Remainder ■ Square root You can define the mode used to round the result of the operation. Exceptions can be raised when operations are performed. Exceptions can also be disabled. Formerly, Java required floating-point arithmetic to be exactly reproducible. IEEE 754 does not require such behavior. The standard allows for the result of operations, including arithmetic, to be delivered to a destination that uses a range greater than that used by the operands to the operation. You can compute the result of a double-precision multiplication at an extended double-precision destination. When this is done, the result must be rounded as if the destination were single-precision or double-precision. The range of the result, that is, the number of bits used for the exponent, can use the range supported by the wider (extended double-precision) destination. This occurrence may result in a double-rounding error in which the least significant bit of the result is incorrect. This state of affairs can only occur for double-precision multiplication and division on hardware that implements the IA-32 and IA-64 instruction set architecture. Thus, with the exception of this case, arithmetic for these datatypes is reproducible across platforms. When the result of a computation is NaN, all platforms produce a value for which IS NAN is true. However, all platforms do not have to use the same bit pattern. Using Conversion Functions with Native Floating-Point Datatypes Oracle Database defines functions that convert between floating-point and other formats, including string formats that use decimal precision (precision may be lost during the conversion). For example, you can use the following functions: 3-10 Oracle Database Application Developer’s Guide - Fundamentals Representing Numeric Data ■ ■ TO_BINARY_DOUBLE, which converts float to double, decimal (string) to double, and float or double to integer-valued double TO_BINARY_FLOAT, which converts double to float, decimal (string) to float, and float or double to integer-valued float ■ TO_CHAR, which converts float or double to decimal (string) ■ TO_NUMBER, which converts a float, double, or string to a number Oracle Database can raise exceptions during conversion. The IEEE 754 specification defines the following exceptions: ■ Invalid ■ Inexact ■ Divide by zero ■ Underflow ■ Overflow Oracle Database does not raise these exceptions for native floating-point datatypes. Generally, situations that would raise an exception produce the values described in Table 3–5. Table 3–5 Values Resulting from Exceptions Exception Value Underflow 0 Overflow -INF, +INF Invalid Operation NaN Divide by Zero -INF, +INF, NaN Inexact Any value – rounding was performed Client Interfaces for Native Floating-Point Datatypes Oracle Database has implemented support for native floating-point datatypes in the following client interfaces: ■ SQL ■ PL/SQL ■ OCI and OCCI ■ Pro*C/C++ ■ JDBC OCI Native Floating-Point Datatypes SQLT_BFLOAT and SQLT_BDOUBLE The OCI API implements the IEEE 754 single precision and double precision native floating-point datatypes with the datatypes SQLT_BFLOAT and SQLT_BDOUBLE respectively. Conversions between these types and the SQL types BINARY_FLOAT and BINARY_DOUBLE are exact on platforms that implement the IEEE 754 standard for the C datatypes FLOAT and DOUBLE. See Also: Oracle Call Interface Programmer's Guide Using SQL Datatypes in Application Development 3-11 Representing Datetime Data Native Floating-Point Datatypes Supported in Oracle OBJECT Types Oracle Database supports the SQL datatypes BINARY_FLOAT and BINARY_DOUBLE as attributes of Oracle OBJECT types. Pro*C/C++ Support for Native Floating-Point Datatypes Pro*C/C++ supports the native FLOAT and DOUBLE datatypes using the column datatypes BINARY_FLOAT and BINARY_DOUBLE. You can use these datatypes in the same way that Oracle NUMBER datatype is used. You can bind the native C/C++ datatypes FLOAT and DOUBLE to BINARY_FLOAT and BINARY_DOUBLE types respectively by setting the Pro*C/C++ precompiler command line option NATIVE_TYPES to Y (yes) when you compile your application. Representing Datetime Data This section contains the following topics: ■ Representing Datetime Data: Overview ■ Manipulating the Date Format ■ Manipulating the Time Format ■ Performing Date Arithmetic ■ Converting Between Datetime Types ■ Importing and Exporting Datetime Types Representing Datetime Data: Overview Oracle Database supports the following datetime datatypes: ■ DATE ■ TIMESTAMP ■ TIMESTAMP WITH TIME ZONE ■ TIMESTAMP WITH LOCAL TIME ZONE Using the DATE Datatype Use the DATE datatype to store point-in-time values (dates and times) in a table. The DATE datatype stores the century, year, month, day, hours, minutes, and seconds. Using the TIMESTAMP Datatype Use the TIMESTAMP datatype to store values that are precise to fractional seconds. For example, an application that must decide which of two events occurred first might use TIMESTAMP. An application that specifies the time for a job might use DATE. Using the TIMESTAMP WITH TIME ZONE Datatype Because TIMESTAMP WITH TIME ZONE can also store time zone information, it is particularly suited for recording date information that must be gathered or coordinated across geographic regions. 3-12 Oracle Database Application Developer’s Guide - Fundamentals Representing Datetime Data Using the TIMESTAMP WITH LOCAL TIME ZONE Datatype Use TIMESTAMP WITH LOCAL TIME ZONE when the time zone is not significant. For example, you might use it in an application that schedules teleconferences, where participants each see the start and end times for their own time zone. The TIMESTAMP WITH LOCAL TIME ZONE type is appropriate for two-tier applications in which you want to display dates and times that use the time zone of the client system. It is generally inappropriate in three-tier applications because data displayed in a Web browser is formatted according to the time zone of the Web server, not the time zone of the browser. The Web server is the database client, so its local time is used. Representing the Difference Between Datetime Values Use the INTERVAL DAY TO SECOND datatype to represent the precise difference between two datetime values. For example, you might use this value to set a reminder for a time 36 hours in the future or to record the time between the start and end of a race. To represent long spans of time with high precision, you can use a large value for the days portion. Use the INTERVAL YEAR TO MONTH datatype to represent the difference between two datetime values, where the only significant portions are the year and the month. For example, you might use this value to set a reminder for a date 18 months in the future, or check whether 6 months have elapsed since a particular date. Oracle Database stores dates in its own internal format. Date data is stored in fixed-length fields of seven bytes each, corresponding to century, year, month, day, hour, minute, and second. See Also: Oracle Call Interface Programmer's Guide for a complete description of the Oracle Database internal date format Manipulating the Date Format For input and output of dates, the standard Oracle Database default date format is DD-MON-RR. The RR datetime format element enables you store 20th century dates in the 21st century by specifying only the last two digits of the year. As explained in Oracle Database SQL Reference, the century of the return value varies according to the specified two-digit year and the last two digits of the current year. For example, the following format refers to the year 2004 in a query issued between 1950 and 2049, but to the year 2005 in a query issued between 2050 and 2099: '13-NOV-04' Changing the Default Date Format Use the following techniques to change the default date format: ■ To change on an instance-wide basis, use the NLS_DATE_FORMAT parameter. ■ To change during a session, use the ALTER SESSION statement. To enter dates that are not in the current default date format, use the TO_DATE function with a format mask. For example: SELECT TO_CHAR(TO_DATE('27-OCT-98', 'DD-MON-RR') ,'YYYY') "Year" FROM DUAL; Be careful when using a date format such as DD-MON-YY. The YY indicates the year in the current century. For example, 31-DEC-92 is December 31, 2092, not 1992 as Using SQL Datatypes in Application Development 3-13 Representing Datetime Data you might expect. If you want to indicate years in any century other than the current one, use a different format mask, such as the default RR. Oracle Database Concepts for information about Julian dates. Oracle Database Julian dates might not be compatible with Julian dates generated by other date algorithms. See Also: Displaying the Current Date and Time Use the SQL function SYSDATE to return the system date and time. You can use the FIXED_DATE initialization parameter to set SYSDATE to a constant, which can be useful for testing. By default, SYSDATE is printed without any BC or AD qualifier. You can add BC to the format string to print the date with BC or AD as appropriate: SELECT TO_CHAR(SYSDATE, 'DD-MON-YYYY BC') FROM DUAL; TO_CHAR(SYSDAT -------------24-JAN-2004 AD Manipulating the Time Format Time is stored in the following 24-hour format: HH24:MI:SS By default, the time in a DATE column is 12:00:00 A.M. (midnight) if no time portion is entered or if the DATE is truncated. In a time-only entry, the date portion defaults to the first day of the current month. To enter the time portion of a date, use the TO_DATE function with a format mask indicating the time portion, as shown in Example 3–3. Example 3–3 Indicating Time with the TO_DATE Function -- create test table CREATE TABLE birthdays ( Bname VARCHAR2(20), Bday DATE ); -- insert a row INSERT INTO birthdays (bname, bday) VALUES ( 'ANNIE', TO_DATE('13-NOV-92 10:56 A.M.','DD-MON-YY HH:MI A.M.') ); Performing Date Arithmetic Oracle Database provides a number of features to help with date arithmetic, so that you do not need to perform your own calculations on the number of seconds in a day, the number of days in each month, and so on. Some useful features include the following: ■ ADD_MONTHS function, which returns the date plus the specified number of months. 3-14 Oracle Database Application Developer’s Guide - Fundamentals Representing Datetime Data ■ ■ ■ ■ ■ ■ SYSDATE function, which returns the current date and time set for the operating system on which the database resides. SYSTIMESTAMP function, which returns the system date, including fractional seconds and time zone, of the system on which the database resides. TRUNC function, which when applied to a DATE value, trims off the time portion so that it represents the very beginning of the day (the stroke of midnight). By truncating two DATE values and comparing them, you can determine whether they refer to the same day. You can also use TRUNC along with a GROUP BY clause to produce daily totals. Arithmetic operators such as + and -. For example, SYSDATE-7 refers to 7 days before the current system date. INTERVAL datatypes, which enable you to represent constants when performing date arithmetic rather than performing your own calculations. For example, you can add or subtract INTERVAL constants from DATE values or subtract two DATE values and compare the result to an INTERVAL. Comparison operators such as >, <, =, and BETWEEN. Converting Between Datetime Types Oracle Database provides several useful functions that enable you to convert to a from datetime datatypes. Some useful functions include: ■ ■ ■ EXTRACT, which extracts and returns the value of a specified datetime field from a datetime or interval value expression NUMTODSINTERVAL, which converts a NUMBER or expression that can be implicitly converted to a NUMBER value to an INTERVAL DAY TO SECOND literal NUMTOYMINTERVAL, which converts a NUMBER or expression that can be implicitly converted to a NUMBER value to an INTERVAL YEAR TO MONTH literal ■ TO_DATE, which converts character data to a DATE datatype ■ TO_CHAR, which converts DATE data to character data ■ ■ ■ ■ TO_DSINTERVAL, which converts a character string to an INTERVAL DAY TO SECOND value TO_TIMESTAMP, which converts character data to a value of TIMESTAMP datatype TO_TIMESTAMP_TZ, which converts character data to a value of TIMESTAMP WITH TIME ZONE datatype TO_YMINTERVAL, which converts a character string to an INTERVAL YEAR TO MONTH type See Also: function Oracle Database SQL Reference for details about each Importing and Exporting Datetime Types TIMESTAMP WITH TIME ZONE and TIMESTAMP WITH LOCAL TIME ZONE values are always stored in normalized format, so that you can export, import, and compare them without worrying about time zone offsets. DATE and TIMESTAMP values do not store an associated time zone, and you must adjust them to account for any time zone differences between source and target databases. Using SQL Datatypes in Application Development 3-15 Representing Specialized Data Representing Specialized Data This section contains the following topics: ■ Representing Geographic Data ■ Representing Multimedia Data ■ Representing Large Amounts of Data ■ Representing Searchable Text ■ Representing XML ■ Representing Dynamically Typed Data ■ Representing Data with ANSI/ISO, DB2, and SQL/DS Datatypes Representing Geographic Data To represent Geographic Information System (GIS) or spatial data in the database, you can use Oracle Spatial features, including the type MDSYS.SDO_GEOMETRY. You can store the data in the database by using either an object-relational or a relational model. You can use a set of PL/SQL packages to query and manipulate the data. See Also: Oracle Spatial User's Guide and Reference to learn how to use MDSYS.SDO_GEOMETRY Representing Multimedia Data Oracle interMedia enables Oracle Database to store, manage, and retrieve images, audio, video, or other heterogeneous media data in an integrated fashion with other enterprise information. Oracle interMedia extends Oracle Database reliability, availability, and data management to multimedia content in traditional, Internet, electronic commerce, and media-rich applications. Whether you store such multimedia data inside the database as BLOB or BFILE values, or store it externally on a Web server or other kind of server, you can use interMedia to access the data using either an object-relational or a relational model, and manipulate and query the data using a set of object types. Oracle interMedia provides the ORDAudio, ORDDoc, ORDImage, ORDImageSignature, ORDVideo, and SI_StillImage object types and methods for the following purposes: ■ ■ ■ Extracting metadata and attributes from multimedia data Retrieving and managing multimedia data from Oracle interMedia, Web servers, file systems, and other servers Performing manipulation operations on image data See Also: Oracle interMedia Referenceto learn about the interMedia types Representing Large Amounts of Data Oracle Database provides several datatypes for representing large amounts of data. These datatypes are grouped under the general category of Large Objects (LOBs). Table 3–6 describes the different LOBs. 3-16 Oracle Database Application Developer’s Guide - Fundamentals Representing Specialized Data Table 3–6 Large Object Datatypes Datatype Name Description BLOB Binary large object Represents large amounts of binary data such as images, video, or other multimedia data. CLOB Character large object Represents large amounts of character data. CLOB types are stored by using the database character set. Note that the database stores a CLOB up to 4,000 bytes inline as a VARCHAR2. If the CLOB exceeds this length, then the database moves the CLOB out of line. NCLOB National character set Represents large amounts of character data in National large objects Character Set format. BFILE External large object Stores objects in the operating system's file system, outside of the database files or tablespace. Note that the BFILE type is read-only; other LOB types are read/write. BFILE objects are also sometimes referred to as external LOBs. An instance of type BLOB, CLOB, or NCLOB can exist as either a persistent LOB instance or a temporary LOB instance. Persistent and temporary instances differ as follows: ■ A temporary LOB instance is declared in the scope of your application. ■ A persistent LOB instance is created and stored in the database. With the exception of declaring, freeing, creating, and committing, operations on persistent and temporary LOB instances are performed the same way. See Also: Oracle Database Application Developer's Guide - Large Objects for more details on using LOBs in applications Using RAW and LONG RAW Datatypes The RAW and LONG RAW datatypes store data that is not interpreted by Oracle Database, that is, it is not converted when moving data between different systems. These datatypes are intended for binary data and byte strings. For example, LONG RAW can store graphics, sound, documents, and arrays of binary data; the interpretation is dependent on the use. Oracle Net and the Export and Import utilities do not perform character conversion when transmitting RAW or LONG RAW data. When Oracle Database automatically converts RAW or LONG RAW data to and from CHAR data, as is the case when entering RAW data as a literal in an INSERT statement, the database represents the data as one hexadecimal character representing the bit pattern for every four bits of RAW data. For example, one byte of RAW data with bits 11001011 is displayed and entered as CB. You cannot index LONG RAW data, but you can index RAW data. In earlier releases, the LONG and LONG RAW datatypes were typically used to store large amounts of data. Use of these types is no longer recommended for new development. If your application still uses these types, migrate your application to use LOB types. Oracle recommends that you convert LONG RAW columns to binary LOB (BLOB) columns and convert LONG columns to character LOB (CLOB or NCLOB) columns. LOB columns are subject to far fewer restrictions than LONG and LONG RAW columns. Using SQL Datatypes in Application Development 3-17 Representing Specialized Data See Also: ■ ■ See Oracle Database Application Developer's Guide - Large Objects for information about the BLOB and BFILE datatypes See the Oracle Database SQL Reference for restrictions on LONG and LONG RAW datatypes Representing Searchable Text Rather than writing low-level code to do full-text searches, you can use Oracle Text. It stores the search data in a special kind of index, and lets you query the data with operators and PL/SQL packages. This technology enables you to create your own search engine using data from tables, files, or URLs, and combine the search logic with relational queries. You can also search XML data this way with the XPath notation. See Also: Oracle Text Application Developer's Guide for more information Representing XML If you have information stored as files in XML format, or if you want to take an object type and store it as XML, then you can use the XMLType built-in type. XMLType columns store their data as CLOBs. You can take an existing CLOB, VARCHAR2, or any object type, and call the XMLType constructor to turn it into an XML object. When an XML object is inside the database, you can use queries to traverse it (using the XML XPath notation) and extract all or part of its data. You can also produce XML output from existing relational data and split XML documents across relational tables and columns. You can use the following packages to transfer XML data into and out of relational tables: ■ ■ ■ DBMS_XMLQUERY, which provides database-to-XMLType functionality DBMS_XMLGEN, which converts the results of a SQL query to a canonical XML format DBMS_XMLSAVE, which provides XML to database-type functionality You can use the following SQL functions to process XML: ■ ■ ■ ■ ■ ■ EXTRACT, which applies a VARCHAR2 XPath string and returns an XMLType instance containing an XML fragment SYS_XMLAGG, which aggregates all of the XML documents or fragments represented by an expression and produces a single XML document SYS_XMLGEN, which takes an expression that evaluates to a particular row and column of the database, and returns an instance of type XMLType containing an XML document UPDATEXML, which takes as arguments an XMLType instance and an XPath-value pair and returns an XMLType instance with the updated value XMLAGG, which takes a collection of XML fragments and returns an aggregated XML document XMLCOLATTVAL, which creates an XML fragment and then expands the resulting XML so that each XML fragment has the name column with the attribute name 3-18 Oracle Database Application Developer’s Guide - Fundamentals Representing Specialized Data ■ ■ ■ ■ XMLCONCAT, which takes as input a series of XMLType instances, concatenates the series of elements for each row, and returns the concatenated series XMLELEMENT, which takes an element name for identifier, an optional collection of attributes for the element, and arguments that make up the content of the element XMLFOREST, which converts each of its argument parameters to XML, and then returns an XML fragment that is the concatenation of these converted arguments XMLSEQUENCE, which either takes as input an XMLType instance and returns a varray of the top-level nodes in the XMLType, or takes as input a REFCURSOR instance, with an optional instance of the XMLFormat object, and returns as an XMLSequence type an XML document for each row of the cursor XMLTRANSFORM, which takes as arguments an XMLType instance and an XSL style sheet, applies the style sheet to the instance, and returns an XMLType See Also: ■ ■ ■ Oracle XML DB Developer's Guide for details about the XMLType datatype Oracle XML Developer's Kit Programmer's Guide for information about client-side programming with XML Oracle Database SQL Reference for information about XML functions Representing Dynamically Typed Data Some languages allow datatypes to change at runtime or let a program check the type of a variable. For example, C has the union keyword and the void * pointer, while Java has the typeof operator and wrapper types such as Number. Oracle Database includes features that enable you to create variables and columns that can hold data of any type and test such data values to determine their underlying representation. Using these features, a single table column can represent a numeric value in one row, a string value in another row, and an object in another row. You can use the built-in type SYS.ANYDATA to represent values of any scalar or object type. This type is an object type with methods to bring in a scalar value of any type, and turn the value back into a scalar or object. In the same way, you can use the built-in type SYS.ANYDATASET to represent values of any collection type. To manipulate and check type information, you can use SYS.ANYTYPE in combination with the DBMS_TYPES package. The program in Example 3–4 represents data of different underlying types in a table, then interprets the underlying type of each row and processes each value appropriately. Example 3–4 Accessing Information in a SYS.ANYDATA Column -- This example defines and executes a PL/SQL procedure that -- uses methods built into SYS.ANYDATA to access information about -- data stored in a SYS.ANYDATA table column. DROP TYPE Employee_type FORCE; DROP TABLE mytab; CREATE OR REPLACE TYPE Employee_type AS OBJECT ( empno NUMBER, ename VARCHAR2(10) ); / CREATE TABLE mytab ( id NUMBER, data SYS.ANYDATA ); INSERT INTO mytab VALUES (1, SYS.ANYDATA.ConvertNumber(5)); INSERT INTO mytab VALUES (2, Using SQL Datatypes in Application Development 3-19 Representing Specialized Data SYS.ANYDATA.ConvertObject(Employee_type(5555, 'john'))); COMMIT; CREATE OR REPLACE PROCEDURE p IS CURSOR cur IS SELECT id, data FROM mytab; v_id mytab.id%TYPE; v_data mytab.data%TYPE; v_type SYS.ANYTYPE; v_typecode PLS_INTEGER; v_typename VARCHAR2(60); v_dummy PLS_INTEGER; v_n NUMBER; v_employee Employee_type; non_null_anytype_for_NUMBER exception; unknown_typename exception; BEGIN OPEN cur; LOOP FETCH cur INTO v_id, v_data; EXIT WHEN cur%NOTFOUND; /* The typecode is a number that signifies what type is represented by v_data. GetType also produces a value of type SYS.AnyType with methods you can call to find precision and scale of a number, length of a string, and so on. */ v_typecode := v_data.GetType ( v_type /* OUT */ ); /* Now we compare the typecode against constants from DBMS_TYPES to see what kind of data we have, and decide how to display it. */ CASE v_typecode WHEN DBMS_TYPES.TYPECODE_NUMBER THEN IF v_type IS NOT NULL -- This condition should never happen, but we check just in case. THEN RAISE non_null_anytype_for_NUMBER; END IF; -- For each type, there is a Get method. v_dummy := v_data.GetNUMBER ( v_n /* OUT */ ); DBMS_OUTPUT.PUT_LINE ( TO_CHAR(v_id) || ': NUMBER = ' || To_Char(v_n) ); WHEN DBMS_TYPES.TYPECODE_OBJECT THEN v_typename := v_data.GetTypeName(); -- An object type's name is qualified with the schema name. IF v_typename NOT IN ( 'HR.EMPLOYEE_TYPE' ) -- If we encounter any object type besides EMPLOYEE_TYPE, raise an exception. THEN RAISE unknown_typename; END IF; v_dummy := v_data.GetObject ( v_employee /* OUT */ ); DBMS_OUTPUT.PUT_LINE ( To_Char(v_id) || ': user-defined type = ' || v_typename || ' ( ' || v_employee.empno || ', ' || v_employee.ename || ' )' ); END CASE; END LOOP; CLOSE cur; EXCEPTION WHEN non_null_anytype_for_NUMBER THEN RAISE_Application_Error ( -20000, 'Paradox: the return AnyType instance FROM GetType ' || 'should be NULL for all but user-defined types' ); WHEN unknown_typename THEN RAISE_Application_Error ( -20000, 'Unknown user-defined type ' || v_typename || ' - program written to handle only HR.EMPLOYEE_TYPE' ); 3-20 Oracle Database Application Developer’s Guide - Fundamentals Representing Specialized Data END; / The query and procedure in Example 3–4 produce output like that shown in Example 3–5. Example 3–5 Sample Output for Example 3–4 SQL> SELECT t.data.gettypename() AS "Type Name" FROM mytab t; Type Name -------------------------------------------------------------------------------SYS.NUMBER HR.EMPLOYEE_TYPE SQL> EXEC p; 1: NUMBER = 5 2: user-defined type = HR.EMPLOYEE_TYPE ( 5555, john ) You can access the same features through the OCI interface by using the OCIType, OCIAnyData, and OCIAnyDataSet interfaces. See Also: ■ ■ ■ Oracle Database PL/SQL Packages and Types Reference for details about the DBMS_TYPES package Oracle Database Application Developer's Guide - Object-Relational Features for information and examples using the ANYDATA, ANYDATASET, and ANYTYPE types Oracle Call Interface Programmer's Guide for details about the OCI interfaces Representing Data with ANSI/ISO, DB2, and SQL/DS Datatypes You can define columns of tables in Oracle Database by means of ANSI/ISO, DB2, and SQL/DS datatypes. Oracle Database internally converts such datatypes to Oracle datatypes. The ANSI datatype conversions are shown in Table 3–7. The ANSI/ISO datatypes NUMERIC, DECIMAL, and DEC can specify only fixed-point numbers. For these datatypes, s defaults to 0. Table 3–7 ANSI Datatype Conversions to Oracle Datatypes ANSI SQL Datatype Oracle Datatype CHARACTER (n), CHAR (n) CHAR (n) NUMERIC (p,s), DECIMAL (p,s), DEC (p,s) NUMBER (p,s) INTEGER, INT, SMALLINT NUMBER (38) FLOAT (p) FLOAT (p) REAL FLOAT (63) DOUBLE PRECISION FLOAT (126) CHARACTER VARYING(n), CHAR VARYING(n) VARCHAR2 (n) TIMESTAMP TIMESTAMP TIMESTAMP WITH TIME ZONE TIMESTAMP WITH TIME ZONE Using SQL Datatypes in Application Development 3-21 Representing Conditional Expressions as Data Table 3–8 shows the DB2 and SQL/DS conversions. Table 3–8 SQL/DS, DB2 Datatype Conversions to Oracle Datatypes DB2 or SQL/DS Datatype Oracle Datatype CHARACTER (n) CHAR (n) VARCHAR (n) VARCHAR2 (n) LONG VARCHAR LONG DECIMAL (p,s) NUMBER (p,s) INTEGER, SMALLINT NUMBER (38) FLOAT (p) FLOAT (p) DATE DATE TIMESTAMP TIMESTAMP The datatypes TIME, GRAPHIC, VARGRAPHIC, and LONG VARGRAPHIC of IBM products SQL/DS and DB2 have no corresponding Oracle datatype, and they cannot be used. Representing Conditional Expressions as Data The Oracle Expression Filter feature enables you to store conditional expressions as data in the database. The Expression Filter provides a mechanism that you can use to place a constraint on a VARCHAR2 column to ensure that the values stored are valid SQL WHERE clause expressions. This mechanism also identifies the set of attributes that are legal to reference in the conditional expressions. For example, suppose you create a traders table in which row holds data for a stock trading account holder. You want to define a column that stores information about stocks each trader is interested in as a conditional expression. You follow these steps: 1. Create a table traders holds data for a stock trading account holder: CREATE TABLE traders ( name VARCHAR2(50), email VARCHAR2(50), interest VARCHAR2(50) ); 2. Create the user-defined datatype ticker with attributes for the trading symbol, limit price, and amount of change in the stock price: CREATE OR REPLACE TYPE ticker AS OBJECT ( symbol VARCHAR2(20), price NUMBER, change NUMBER ); 3. Use the following PL/SQL block to create an attribute set ticker based on the ticker datatype: BEGIN DBMS_EXPFIL.CREATE_ATTRIBUTE_SET( attr_set => 'ticker', from_type => 'YES' ); END; 3-22 Oracle Database Application Developer’s Guide - Fundamentals Identifying Rows by Address 4. Associate the attribute set with the expression set stored in the database column trader.interest as follows: BEGIN DBMS_EXPFIL.ASSIGN_ATTRIBUTE_SET (attr_set => 'ticker', expr_tab => 'traders', expr_col => 'interest'); END; The preceding code places a constraint on the interest column that ensures the column stores valid conditional expressions. 5. Populate the table with trader names, email addresses and conditional expressions that represents a stock the trader is interested in at a particular price: INSERT INTO traders (name, email, interest) VALUES ('Vishu', 'vishu@abc.com', 'symbol = ''ABC'' AND price > 25'); 6. Use the EVALUATE operator to identify the conditional expressions that evaluate to TRUE for a given data item. For example, the following query returns traders who are interested in a given stock quote (symbol='ABC', price=31, change=5.2): SELECT Name, Email FROM Traders WHERE EVALUATE ( interest, 'symbol=>''ABC'', price=>31, change=>5.2' ) = 1; To speed up this type of query, you can optionally create an Oracle Expression Filter index on the interest column. See Also: Oracle Database Application Developer's Guide - Rules Manager and Expression Filter for details on Oracle Expression Filter Identifying Rows by Address Each row in a database table has an address called a rowid. You can examine a row address by querying the pseudocolumn ROWID, whose values are strings representing the address of each row. These strings have the datatype ROWID or UROWID. You can also create tables and clusters that contain actual columns having the ROWID datatype. Oracle Database does not guarantee that the values of such columns are valid rowids. Rowid values are important for application development for the following reasons: ■ They are the fastest way to access a single row. ■ They can show you how the rows in a table are stored. ■ They are unique identifiers for rows in a table. See Also: ■ ■ Oracle Database Concepts for general information about the ROWID pseudocolumn and the ROWID datatype Oracle Database SQL Reference to learn about the ROWID pseudocolumn Using SQL Datatypes in Application Development 3-23 Identifying Rows by Address Querying the ROWID Pseudocolumn Each table in Oracle Database has a pseudocolumn named ROWID. If the row is too large to fit within a single data block, then ROWID identifies the initial row piece. Although rowids are usually unique, different rows can have the same rowid if they are in the same data block but in different clustered tables. The following SQL statements return the ROWID pseudocolumn of the row of the hr.employees table that satisfies the query, and inserts it into the t_tab table: CREATE TABLE t_tab (col1 ROWID); INSERT INTO t_tab SELECT ROWID FROM hr.employees WHERE employee_id = 7499; Note: Although you can use the ROWID pseudocolumn in the SELECT and WHERE clause of a query, these pseudocolumn values are not actually stored in the database. You cannot insert, update, or delete a value of the ROWID pseudocolumn. Accessing the ROWID Datatype In tables that are not index-organized and foreign tables, the values of the ROWID pseudocolumn have the datatype ROWID. The format of this datatype is either extended or restricted. Restricted ROWID Internally, the ROWID is a structure that holds information that the database server needs to access a row. The restricted internal ROWID is 6 bytes on most platforms. Each restricted rowid includes the following data: ■ Datafile identifier ■ Block identifier ■ Row identifier The restricted ROWID pseudocolumn is returned to client applications in the form of an 18-character string with a hexadecimal encoding of the datablock, row, and datafile components of the ROWID. Extended ROWID The extended ROWID datatype includes the data in the restricted rowid plus a data object number. The data object number is an identification number assigned to every database segment. The extended internal ROWID is 10 bytes on most platforms. Data in an extended ROWID pseudocolumn is returned to the client application in the form of an 18-character string (for example, "AAAA8mAALAAAAQkAAA"), which represents a base 64 encoding of the components of the extended ROWID in a four-piece format, OOOOOOFFFBBBBBBRRR. Extended rowids are not available directly. You can use a supplied package, DBMS_ROWID, to interpret extended rowid contents. The package functions extract and provide information that would be available directly from a restricted rowid as well as information specific to extended rowids. Oracle Database PL/SQL Packages and Types Reference for information about the DBMS_ROWID package See Also: 3-24 Oracle Database Application Developer’s Guide - Fundamentals How Oracle Database Converts Datatypes External Binary ROWID Some client applications use a binary form of the ROWID. For example, OCI and some precompiler applications can map the ROWID datatype to a 3GL structure on bind or define calls. The size of the binary ROWID is the same for extended and restricted ROWIDs. The information for the extended ROWID is included in an unused field of the restricted ROWID structure. The format of the extended binary ROWID, expressed as a C struct, is as follows: struct riddef { ub4 ridobjnum; /* data obj#--this field is unused in restricted ROWIDs */ ub2 ridfilenum; ub1 filler; ub4 ridblocknum; ub2 ridslotnum; } Accessing the UROWID Datatype The rows of some tables have addresses that are not physical or permanent or were not generated by Oracle Database. For example, the row addresses of index-organized tables are stored in index leaves, which can move. Oracle provides these tables with logical row identifiers, called logical rowids. Rowids of foreign tables, such as DB2 tables accessed through a gateway, are not standard Oracle Database rowids. Oracle provides foreign tables with identifiers called foreign rowids. Oracle Database uses universal rowids (urowids) to store the addresses of index-organized and foreign tables. Both types of urowid are stored in the ROWID pseudocolumn, as are the physical rowids of heap-organized tables. Oracle creates logical rowids based on the primary key of the table. The logical rowids do not change as long as the primary key does not change. The ROWID pseudocolumn of an index-organized table has a datatype of UROWID. You can access this pseudocolumn as you would the ROWID pseudocolumn of a heap-organized table (that is, using a SELECT ... ROWID statement). If you want to store the rowids of an index-organized table, then you can define a column of type UROWID for the table and retrieve the value of the ROWID pseudocolumn into that column. How Oracle Database Converts Datatypes In some cases, Oracle Database allows data of one datatype where it expects data of a different datatype. Generally, an expression cannot contain values with different datatypes. However, Oracle Database can use various SQL functions to automatically convert data to the expected datatype. See Also: Oracle Database SQL Reference for details about datatype conversion Datatype Conversion During Assignments The datatype conversion for an assignment succeeds if Oracle Database can convert the datatype of the value used in the assignment to that of the assignment target. For the examples in the following list, assume a package with a public variable and a table declared as in the following statements: CREATE PACKAGE Test_Pack AS var1 CHAR(5); END; CREATE TABLE Table1_tab (col1 NUMBER); Using SQL Datatypes in Application Development 3-25 How Oracle Database Converts Datatypes ■ variable := expression The datatype of expression must be either the same as, or convertible to, the datatype of variable. For example, Oracle Database automatically converts the data provided in the following assignment within the body of a stored procedure: VAR1 := 0; ■ INSERT INTO Table1_tab VALUES (expression1, expression2, ...) The datatypes of expression1, expression2, and so on, must be either the same as, or convertible to, the datatypes of the corresponding columns in Table1_tab. For example, Oracle Database automatically converts the data provided in the following INSERT statement for Table1_tab: INSERT INTO Table1_tab VALUES ( ' 19 ' ); ■ UPDATE Table1_tab SET column = expression The datatype of expression must be either the same as, or convertible to, the datatype of column. For example, Oracle Database automatically converts the data provided in the following UPDATE statement issued against Table1_tab: UPDATE Table1_tab SET col1 = ' 30 ' ; ■ SELECT column INTO variable FROM Table1_tab The datatype of column must be either the same as, or convertible to, the datatype of variable. For example, Oracle Database automatically converts data selected from the table before assigning it to the variable in the following statement: SELECT Col1 INTO Var1 FROM Table1_tab WHERE Col1 = 30; Datatype Conversion During Expression Evaluation For expression evaluation, Oracle Database can automatically perform the same conversions as for assignments. An expression is converted to a type based on its context. For example, operands to arithmetic operators are converted to NUMBER, and operands to string functions are converted to VARCHAR2. Oracle Database can automatically convert the following: ■ VARCHAR2 or CHAR to NUMBER ■ VARCHAR2 or CHAR to DATE Character to NUMBER conversions succeed only if the character string represents a valid number. Character to DATE conversions succeed only if the character string satisfies the session default format, which is specified by the initialization parameter NLS_DATE_FORMAT. Some common types of expressions follow: ■ Simple expressions, such as: 3-26 Oracle Database Application Developer’s Guide - Fundamentals How Oracle Database Converts Datatypes commission + '500' ■ Boolean expressions, such as: bonus > salary / '10' ■ Function and procedure calls, such as: MOD (counter, '2') ■ WHERE clause conditions, such as: WHERE hiredate = TO_DATE('1997-01-01','yyyy-mm-dd') ■ WHERE clause conditions, such as: WHERE rowid = 'AAAAaoAATAAAADAAA' In general, Oracle Database uses the rule for expression evaluation when a datatype conversion is needed in places not covered by the rule for assignment conversions. In assignments of the form: variable := expression Oracle Database first evaluates expression using the conversion rules for expressions; expression can be as simple or complex as desired. If it succeeds, then the evaluation of expression results in a single value and datatype. Then, Oracle Database tries to assign this value to the target variable using the conversion rules for assignments. Using SQL Datatypes in Application Development 3-27 How Oracle Database Converts Datatypes 3-28 Oracle Database Application Developer’s Guide - Fundamentals 4 Using Regular Expressions in Oracle Database This chapter introduces regular expression support for Oracle Database. This chapter covers the following topics: ■ Using Regular Expressions with Oracle Database: Overview ■ Regular Expression Metacharacters in Oracle Database ■ Using Regular Expressions in SQL Statements: Scenarios See Also: ■ ■ ■ ■ Oracle Database SQL Reference for information about Oracle Database SQL functions for regular expressions Oracle Database Globalization Support Guide for details on using SQL regular expression functions in a multilingual environment Oracle Regular Expressions Pocket Reference by Jonathan Gennick, O'Reilly & Associates Mastering Regular Expressions by Jeffrey E. F. Friedl, O'Reilly & Associates Using Regular Expressions with Oracle Database: Overview This section contains the following topics: ■ What Are Regular Expressions? ■ How Are Oracle Database Regular Expressions Useful? ■ Oracle Database Implementation of Regular Expressions ■ Oracle Database Support for the POSIX Regular Expression Standard What Are Regular Expressions? Regular expressions enable you to search for patterns in string data by using standardized syntax conventions. You specify a regular expression by means of the following types of characters: ■ Metacharacters, which are operators that specify search algorithms ■ Literals, which are the characters for which you are searching Using Regular Expressions in Oracle Database 4-1 Using Regular Expressions with Oracle Database: Overview A regular expression can specify complex patterns of character sequences. For example, the following regular expression searches for the literals f or ht, the t literal, the p literal optionally followed by the s literal, and finally the colon (:) literal: (f|ht)tps?: The parentheses are metacharacters that group a series of pattern elements to a single element; the pipe symbol (|) matches one of the alternatives in the group. The question mark (?) is a metacharacter indicating that the preceding pattern, in this case the s character, is optional. Thus, the preceding regular expression matches the http:, https:, ftp:, and ftps: strings. How Are Oracle Database Regular Expressions Useful? Regular expressions are a powerful text processing component of programming languages such as Perl and Java. For example, a Perl script can process each HTML file in a directory, read its contents into a scalar variable as a single string, and then use regular expressions to search for URLs in the string. One reason that many developers write in Perl is for its robust pattern matching functionality. Oracle's support of regular expressions enables developers to implement complex match logic in the database. This technique is useful for the following reasons: ■ ■ ■ By centralizing match logic in Oracle Database, you avoid intensive string processing of SQL results sets by middle-tier applications. For example, life science customers often rely on Perl to do pattern analysis on bioinformatics data stored in huge databases of DNAs and proteins. Previously, finding a match for a protein sequence such as [AG].{4}GK[ST] would be handled in the middle tier. The SQL regular expression functions move the processing logic closer to the data, thereby providing a more efficient solution. Prior to Oracle Database 10g, developers often coded data validation logic on the client, requiring the same validation logic to be duplicated for multiple clients. Using server-side regular expressions to enforce constraints solves this problem. The built-in SQL and PL/SQL regular expression functions and conditions make string manipulations more powerful and less cumbersome than in previous releases of Oracle Database. Oracle Database Implementation of Regular Expressions Oracle Database implements regular expression support with a set of Oracle Database SQL functions and conditions that enable you to search and manipulate string data. You can use these functions in any environment that supports Oracle Database SQL. You can use these functions on a text literal, bind variable, or any column that holds character data such as CHAR, NCHAR, CLOB, NCLOB, NVARCHAR2, and VARCHAR2 (but not LONG). Table 4–1 gives a brief description of the regular expression functions and conditions. Table 4–1 SQL Regular Expression Functions and Conditions SQL Element Category Description REGEXP_LIKE Condition Searches a character column for a pattern. Use this function in the WHERE clause of a query to return rows matching a regular expression. The condition is also valid in a constraint or as a PL/SQL function returning a boolean. The following WHERE clause filters employees with a first name of Steven or Stephen: WHERE REGEXP_LIKE(first_name, '^Ste(v|ph)en$') 4-2 Oracle Database Application Developer’s Guide - Fundamentals Using Regular Expressions with Oracle Database: Overview Table 4–1 (Cont.) SQL Regular Expression Functions and Conditions SQL Element Category Description REGEXP_REPLACE Function Searches for a pattern in a character column and replaces each occurrence of that pattern with the specified string. The following function puts a space after each character in the country_name column: REGEXP_REPLACE(country_name, '(.)', '\1 ') REGEXP_INSTR Function Searches a string for a given occurrence of a regular expression pattern and returns an integer indicating the position in the string where the match is found. You specify which occurrence you want to find and the start position. For example, the following performs a boolean test for a valid email address in the email column: REGEXP_INSTR(email, '\w+@\w+(\.\w+)+') > 0 REGEXP_SUBSTR Function Returns the substring matching the regular expression pattern that you specify. The following function uses the x flag to match the first string by ignoring spaces in the regular expression: REGEXP_SUBSTR('oracle', 'o r a c l e', 1, 1, 'x') A string literal in a REGEXP function or condition conforms to the rules of SQL text literals. By default, regular expressions must be enclosed in single quotes. If your regular expression includes the single quote character, then enter two single quotation marks to represent one single quotation mark within the expression. This technique ensures that the entire expression is interpreted by the SQL function and improves the readability of your code. You can also use the q-quote syntax to define your own character to terminate a text literal. For example, you could delimit your regular expression with the pound sign (#) and then use a single quote within the expression. If your expression comes from a column or a bind variable, then the same rules for quoting do not apply. Note: See Also: ■ ■ Oracle Database SQL Reference for syntax, descriptions, and examples of the REGEXP functions and conditions Oracle Database SQL Reference for a discussion of character literals Oracle Database Support for the POSIX Regular Expression Standard Oracle's implementation of regular expressions conforms to the following standards: ■ IEEE Portable Operating System Interface (POSIX) standard draft 1003.2/D11.2 ■ Unicode Regular Expression Guidelines of the Unicode Consortium Oracle Database follows the exact syntax and matching semantics for these operators as defined in the POSIX standard for matching ASCII (English language) data. You can find the POSIX standard draft at the following URL: http://www.opengroup.org/onlinepubs/007908799/xbd/re.html Oracle Database enhances regular expression support in the following ways: ■ Extends the matching capabilities for multilingual data beyond what is specified in the POSIX standard. Using Regular Expressions in Oracle Database 4-3 Regular Expression Metacharacters in Oracle Database ■ Adds support for the common Perl regular expression extensions that are not included in the POSIX standard but do not conflict with it. Oracle Database provides built-in support for some of the most heavily used Perl regular expression operators, for example, character class shortcuts, the non-greedy modifier, and so on. Oracle Database supports a set of common metacharacters used in regular expressions. The behavior of supported metacharacters and related features is described in "Regular Expression Metacharacters in Oracle Database" on page 4-4. The interpretation of metacharacters differs between tools that support regular expressions. If you are porting regular expressions from another environment to Oracle Database, ensure that the regular expression syntax is supported and the behavior is what you expect. Note: Regular Expression Metacharacters in Oracle Database This section contains the following topics: ■ POSIX Metacharacters in Oracle Database Regular Expressions ■ Regular Expression Operator Multilingual Enhancements ■ Perl-Influenced Extensions in Oracle Regular Expressions POSIX Metacharacters in Oracle Database Regular Expressions Table 4–2 lists the list of metacharacters supported for use in regular expressions passed to SQL regular expression functions and conditions. These metacharacters conform to the POSIX standard; any differences in behavior from the standard are noted in the "Description" column. Table 4–2 POSIX Metacharacters in Oracle Database Regular Expressions Syntax Operator Name Description Example . Any Character — Dot Matches any character in the database character set. The expression a.b matches the If the n flag is set, it matches the newline character. strings abb, acb, and adb, but does The newline is recognized as the linefeed character not match acc. (\x0a) on UNIX and Windows or the carriage return character (\x0d) on Macintosh platforms. Note: In the POSIX standard, this operator matches any English character except NULL and the newline character. + One or More — Plus Quantifier Matches one or more occurrences of the preceding subexpression. The expression a+ matches the strings a, aa, and aaa, but does not match bbb. ? Zero or One — Question Mark Quantifier Matches zero or one occurrence of the preceding subexpression. The expression ab?c matches the strings abc and ac, but does not match abbc. * Zero or More — Star Quantifier Matches zero or more occurrences of the preceding subexpression. By default, a quantifier match is greedy because it matches as many times as possible while still allowing the rest of the match to succeed. The expression ab*c matches the strings ac, abc, and abbc, but does not match abb. {m} Interval—Exact Count Matches exactly m occurrences of the preceding subexpression. The expression a{3} matches the strings aaa, but does not match aa. {m,} Interval—At Least Count Matches at least m occurrences of the preceding subexpression. The expression a{3,} matches the strings aaa and aaaa, but does not match aa. 4-4 Oracle Database Application Developer’s Guide - Fundamentals Regular Expression Metacharacters in Oracle Database Table 4–2 (Cont.) POSIX Metacharacters in Oracle Database Regular Expressions Syntax Operator Name Description Example {m,n} Interval—Betwee n Count Matches at least m, but not more than n occurrences of the preceding subexpression. The expression a{3,5} matches the strings aaa, aaaa, and aaaaa, but does not match aa. [ ... ] Matching Character List Matches any single character in the list within the brackets. The following operators are allowed within the list, but other metacharacters included are treated as literals: The expression [abc] matches the first character in the strings all, bill, and cold, but does not match any characters in doll. ■ Range operator: - ■ POSIX character class: [: :] ■ POSIX collation element: [. .] ■ POSIX character equivalence class: [= =] A dash (-) is a literal when it occurs first or last in the list, or as an ending range point in a range expression, as in [#--]. A right bracket (]) is treated as a literal if it occurs first in the list. Note: In the POSIX standard, a range includes all collation elements between the start and end of the range in the linguistic definition of the current locale. Thus, ranges are linguistic rather than byte values ranges; the semantics of the range expression are independent of character set. In Oracle Database, the linguistic range is determined by the NLS_SORT initialization parameter. [^ ... ] Non-Matching Character List Matches any single character not in the list within the brackets. Characters not in the non-matching character list are returned as a match. Refer to the description of the Matching Character List operator for an account of metacharacters allowed in the character list. The expression [^abc] matches the character d in the string abcdef, but not the character a, b, or c. The expression [^abc]+ matches the sequence def in the string abcdef, but not a, b, or c. The expression [^a-i] excludes any character between a and i from the search result. This expression matches the character j in the string hij, but does not match any characters in the string abcdefghi. | Or Matches one of the alternatives. The expression a|b matches character a or character b. ( ... ) Subexpression or Grouping Treats the expression within parentheses as a unit. The subexpression can be a string of literals or a complex expression containing operators. The expression (abc)?def matches the optional string abc, followed by def. Thus, the expression matches abcdefghi and def, but does not match ghi. \n Backreference Matches the nth preceding subexpression, that is, whatever is grouped within parentheses, where n is an integer from 1 to 9. The parentheses cause an expression to be remembered; a backreference refers to it. A backreference counts subexpressions from left to right, starting with the opening parenthesis of each preceding subexpression. The expression is invalid if the source string contains fewer than n subexpressions preceding the \n. The expression (abc|def)xy\1 matches the strings abcxyabc and defxydef, but does not match abcxydef or abcxy. Treats the subsequent metacharacter in the expression as a literal. Use a backslash (\) to search for a character that is normally treated as a metacharacter. Use consecutive backslashes (\\) to match the backslash literal itself. The expression \+ searches for the plus character (+). It matches the plus character in the string abc+def, but does not match abcdef. A backreference enables you to search for a repeated string without knowing the actual string ahead of time. For example, the expression ^(.*)\1$ matches a line Oracle supports the backreference expression in the consisting of two adjacent instances of the same string. regular expression pattern and the replacement string of the REGEXP_REPLACE function. \ Escape Character Using Regular Expressions in Oracle Database 4-5 Regular Expression Metacharacters in Oracle Database Table 4–2 (Cont.) POSIX Metacharacters in Oracle Database Regular Expressions Syntax Operator Name Description Example ^ Beginning of Line Matches the beginning of a string (default). In Anchor multiline mode, it matches the beginning of any line within the source string. The expression ^def matches def in the string defghi but does not match def in abcdef. $ End of Line Anchor Matches the end of a string (default). In multiline mode, it matches the beginning of any line within the source string. The expression def$ matches def in the string abcdef but does not match def in the string defghi. [:class:] POSIX Character Class Matches any character belonging to the specified POSIX character class. You can use this operator to search for characters with specific formatting such as uppercase characters, or you can search for special characters such as digits or punctuation characters. The full set of POSIX character classes is supported. The expression [[:upper:]]+ searches for one or more consecutive uppercase characters. This expression matches DEF in the string abcDEFghi but does not match the string abcdefghi. Note: In English regular expressions, range expressions often indicate a character class. For example, [a-z] indicates any lowercase character. This convention is not useful in multilingual environments, where the first and last character of a given character class may not be the same in all languages. Oracle supports the character classes in Table 4–3 based on character class definitions in Globalization classification data. [.element.] POSIX Collating Specifies a collating element to use in the regular Element Operator expression. The element must be a defined collating element in the current locale. Use any collating element defined in the locale, including single-character and multicharacter elements. The NLS_SORT initialization parameter determines supported collation elements. The expression [[.ch.]] searches for the collating element ch and matches ch in string chabc, but does not match cdefg. The expression [a-[.ch.]] specifies the range a to ch. This operator lets you use a multicharacter collating element in cases where only one character would otherwise be allowed. For example, you can ensure that the collating element ch, when defined in a locale such as Traditional Spanish, is treated as one character in operations that depend on the ordering of characters. [=character=] POSIX Character Equivalence Class Matches all characters that are members of the same The expression [[=n=]] searches for characters equivalent to n in a character equivalence class in the current locale as Spanish locale. It matches both N the specified character. and ñ in the string El Niño. The character equivalence class must occur within a character list, so the character equivalence class is always nested within the brackets for the character list in the regular expression. Usage of character equivalents depends on how canonical rules are defined for your database locale. Refer to the Oracle Database Globalization Support Guide for more information on linguistic sorting and string searching. Oracle Database SQL Reference for syntax, descriptions, and examples of the REGEXP functions and conditions See Also: Regular Expression Operator Multilingual Enhancements When applied to multilingual data, Oracle's implementation of the POSIX operators extends beyond the matching capabilities specified in the POSIX standard. Table 4–3 shows the relationship of the operators in the context of the POSIX standard. ■ ■ The first column lists the supported operators. The second column indicates whether the POSIX standard for Basic Regular Expression (BRE) defines the operator. 4-6 Oracle Database Application Developer’s Guide - Fundamentals Regular Expression Metacharacters in Oracle Database ■ ■ The third column indicates whether the POSIX standard for Extended Regular Expression (ERE) defines the operator. The fourth column indicates whether the Oracle Database implementation extends the operator's semantics for handling multilingual data. Oracle Database lets you enter multibyte characters directly, if you have a direct input method, or use functions to compose the multibyte characters. You cannot use the Unicode hexadecimal encoding value of the form \xxxx. Oracle evaluates the characters based on the byte values used to encode the character, not the graphical representation of the character. Table 4–3 POSIX and Multilingual Operator Relationships Operator POSIX BRE syntax POSIX ERE Syntax Multilingual Enhancement \ Yes Yes -- * Yes Yes -- + -- Yes -- ? -- Yes -- | -- Yes -- ^ Yes Yes Yes $ Yes Yes Yes . Yes Yes Yes [ ] Yes Yes Yes ( ) Yes Yes -- {m} Yes Yes -- {m,} Yes Yes -- {m,n} Yes Yes -- \n Yes Yes Yes [..] Yes Yes Yes [::] Yes Yes Yes [==] Yes Yes Yes Perl-Influenced Extensions in Oracle Regular Expressions Table 4–4 describes Perl-influenced metacharacters supported in Oracle Database regular expression functions and conditions. These metacharacters are not in the POSIX standard, but are common at least partly due to the popularity of Perl. Note that Perl character class matching is based on the locale model of the operating system, whereas Oracle Database regular expressions are based on the language-specific data of the database. In general, a regular expression involving locale data cannot be expected to produce the same results between Perl and Oracle Database. Using Regular Expressions in Oracle Database 4-7 Regular Expression Metacharacters in Oracle Database Table 4–4 Perl-Influenced Extensions in Oracle Regular Expressions Reg. Exp. Matches . . . Example \d A digit character. It is equivalent to the POSIX class [[:digit:]]. The expression ^\(\d{3}\) \d{3}-\d{4}$ matches (650) 555-1212 but does not match 650-555-1212. \D A non-digit character. It is equivalent to the POSIX class [^[:digit:]]. The expression \w\d\D matches b2b and b2_ but does not match b22. \w A word character, which is defined as an alphanumeric or underscore (_) character. It is equivalent to the POSIX class [[:alnum:]_]. Note that if you do not want to include the underscore character, you can use the POSIX class [[:alnum:]]. The expression \w+@\w+(\.\w+)+ matches the string jdoe@company.co.uk but not the string jdoe@company. \W A non-word character. It is equivalent to the POSIX class [^[:alnum:]_]. The expression \w+\W\s\w+ matches the string to: bill but not the string to bill. \s A whitespace character. It is equivalent to the POSIX class [[:space:]]. The expression \(\w\s\w\s\) matches the string (a b ) but not the string (ab). \S A non-whitespace character. It is equivalent to the POSIX class [^[:space:]]. The expression \(\w\S\w\S\) matches the string (abde) but not the string (a b d e). \A Only at the beginning of a string. In multi-line mode, that is, when embedded newline characters in a string are considered the termination of a line, \A does not match the beginning of each line. The expression \AL matches only the first L character in the string Line1\nLine2\n, regardless of whether the search is in single-line or multi-line mode. \Z Only at the end of string or before a newline ending a string. In multi-line mode, that is, when embedded newline characters in a string are considered the termination of a line, \Z does not match the end of each line. In the expression \s\Z, the \s matches the last space in the string L i n e \n, regardless of whether the search is in single-line or multi-line mode. \z Only at the end of a string. In the expression \s\z, the \s matches the newline in the string L i n e \n, regardless of whether the search is in single-line or multi-line mode. *? The expression \w*?x\w is "non-greedy" and so The preceding pattern element 0 or more times (non-greedy). Note that this quantifier matches abxc in the string abxcxd. The expression matches the empty string whenever possible. \w*x\w is "greedy" and so matches abxcxd in the string abxcxd. The expression \w*?x\w also matches the string xa. +? The preceding pattern element 1 or more times (non-greedy). ?? The expression a??aa is "non-greedy" and matches aa The preceding pattern element 0 or 1 time in the string aaaa. The expression a?aa is "greedy" (non-greedy). Note that this quantifier matches the empty string whenever possible. and so matches aaa in the string aaaa. {n}? The preceding pattern element exactly n times (non-greedy). In this case {n}? is equivalent to {n}. The expression (a|aa){2}? matches aa in the string aaaa. {n,}? The preceding pattern element at least n times (non-greedy). The expression a{2,}? is "non-greedy" and matches aa in the string aaaaa. The expression a{2,} is "greedy" and so matches aaaaa. {n,m}? At least n but not more than m times The expression a{2,4}? is "non-greedy" and matches (non-greedy). Note that {0,m}? matches the aa in the string aaaaa. The expression a{2,4} is empty string whenever possible. "greedy" and so matches aaaa. The expression \w+?x\w is "non-greedy" and so matches abxc in the string abxcxd. The expression \w+x\w is "greedy" and so matches abxcxd in the string abxcxd. The expression \w+?x\w does not match the string xa, but does match the string axa. 4-8 Oracle Database Application Developer’s Guide - Fundamentals Using Regular Expressions in SQL Statements: Scenarios The Oracle Database regular expression functions and conditions support the pattern matching modifiers described in Table 4–5. Table 4–5 Pattern Matching Modifiers Mod. Description i Example Specifies case-insensitive matching. The following regular expression returns AbCd: REGEXP_SUBSTR('AbCd', 'abcd', 1, 1, 'i') c Specifies case-sensitive matching. The following regular expression fails to match: REGEXP_SUBSTR('AbCd', 'abcd', 1, 1, 'c') n m x Allows the period (.), which by default does not match newlines, to match the newline character. Performs the search in multi-line mode. The metacharacter ^ and $ signify the start and end, respectively, of any line anywhere in the source string, rather than only at the start or end of the entire source string. Ignores whitespace characters in the regular expression. By default, whitespace characters match themselves. The following regular expression matches the string, but would not match if the n flag were not specified: REGEXP_SUBSTR('a'||CHR(10)||'d', 'a.d', 1, 1, 'n') The following regular expression returns ac: REGEXP_SUBSTR('ab'||CHR(10)||'ac', '^a.', 1, 2, 'm') The following regular expression returns abcd: REGEXP_SUBSTR('abcd', 'a b c d', 1, 1, 'x') Using Regular Expressions in SQL Statements: Scenarios This section contains the following scenarios: ■ Using an Integrity Constraint to Enforce a Phone Number Format ■ Using Back References to Reposition Characters Using an Integrity Constraint to Enforce a Phone Number Format Regular expressions are a useful way to enforce integrity constraints. For example, suppose that you want to ensure that phone numbers are entered into the database in a standard format. Example 4–1 creates a contacts table and adds a check constraint to the p_number column to enforce the following format mask: (XXX) XXX-XXXX Example 4–1 Enforcing a Phone Number Format with Regular Expressions CREATE TABLE contacts ( l_name VARCHAR2(30), p_number VARCHAR2(30) CONSTRAINT p_number_format CHECK ( REGEXP_LIKE ( p_number, '^\(\d{3}\) \d{3}-\d{4}$' ) ) ); Table 4–6 explains the elements of the regular expression. Table 4–6 Explanation of the Regular Expression Elements in Example 4–1 Regular Expression Element Matches . . . ^ The beginning of the string. Using Regular Expressions in Oracle Database 4-9 Using Regular Expressions in SQL Statements: Scenarios Table 4–6 (Cont.) Explanation of the Regular Expression Elements in Example 4–1 Regular Expression Element Matches . . . \( A left parenthesis. The backward slash (\) is an escape character that indicates that the left parenthesis following it is a literal rather than a grouping expression. \d{3} Exactly three digits. \) A right parenthesis. The backward slash (\) is an escape character that indicates that the right parenthesis following it is a literal rather than a grouping expression. (space character) A space character. \d{3} Exactly three digits. - A hyphen. \d{4} Exactly four digits. $ The end of the string. Example 4–2 shows a SQL script that attempts to insert seven phone numbers into the contacts table. Only the first two INSERT statements use a format that conforms to the p_number_format constraint; the remaining statements generate check constraint errors. Example 4–2 insert_contacts.sql -- first two statements use valid phone number format INSERT INTO contacts (p_number) VALUES( '(650) 555-5555' ); INSERT INTO contacts (p_number) VALUES( '(215) 555-3427' ); -- remaining statements generate check contraint errors INSERT INTO contacts (p_number) VALUES( '650 555-5555' ); INSERT INTO contacts (p_number) VALUES( '650 555 5555' ); INSERT INTO contacts (p_number) VALUES( '650-555-5555' ); INSERT INTO contacts (p_number) VALUES( '(650)555-5555' ); INSERT INTO contacts (p_number) VALUES( ' (650) 555-5555' ); / Using Back References to Reposition Characters As explained in Table 4–2, back references store matched subexpressions in a temporary buffer, thereby enabling you to reposition characters. You access buffers with the \n notation, where \n is a number between 1 and 9. Each subexpression is contained in parentheses and is numbered from left to right. Example 4–3 creates a famous_people table and populates the famous_people.names column with names in different formats. Example 4–3 Using Back References to Reposition Characters CREATE TABLE famous_people ( names VARCHAR2(30) ); 4-10 Oracle Database Application Developer’s Guide - Fundamentals Using Regular Expressions in SQL Statements: Scenarios -- populate table with data INSERT INTO famous_people VALUES ('John Quincy Adams'); INSERT INTO famous_people VALUES ('Harry S. Truman'); INSERT INTO famous_people VALUES ('John Adams'); INSERT INTO famous_people VALUES (' John Quincy Adams'); INSERT INTO famous_people VALUES ('John_Quincy_Adams'); COMMIT; Example 4–4 shows a query that repositions names in the format "first middle last" to the format "last, first middle". It ignores names not in the format "first middle last". Example 4–4 Using Back References to Reposition Characters SELECT names "names", REGEXP_REPLACE(names, '^(\S+)\s(\S+)\s(\S+)$', '\3, \1 \2') AS "names after regexp" FROM famous_people; Table 4–7 explains the elements of the regular expression. Table 4–7 Explanation of the Regular Expression Elements in Example 4–4 Regular Expression Element Description ^ Matches the beginning of the string. $ Matches the end of the string. (\S+) Matches one or more non-space characters. The parentheses are not escaped so they function as a grouping expression. \s Matches a whitespace character. \1 Substitutes the first subexpression, that is, the first group of parentheses in the matching pattern. \2 Substitutes the second subexpression, that is, the second group of parentheses in the matching pattern. \3 Substitutes the third subexpression, that is, the third group of parentheses in the matching pattern. , Inserts a comma character. Example 4–5 shows the result set of the query in Example 4–4. The regular expression matched only the first two rows. Example 4–5 Result Set of Regular Expression Query names -----------------------------names after regexp -----------------------------John Quincy Adams Adams, John Quincy Using Regular Expressions in Oracle Database 4-11 Using Regular Expressions in SQL Statements: Scenarios Harry S. Truman Truman, Harry S. John Adams John Adams John Quincy Adams John Quincy Adams John_Quincy_Adams John_Quincy_Adams 4-12 Oracle Database Application Developer’s Guide - Fundamentals 5 Using Indexes in Application Development This chapter discusses considerations for using the different types of indexes in an application. The topics include: ■ Guidelines for Application-Specific Indexes ■ Creating Indexes: Basic Examples ■ When to Use Domain Indexes ■ When to Use Function-Based Indexes See Also: ■ ■ ■ ■ Oracle Database Administrator's Guide for information about creating and managing indexes Oracle Database Performance Tuning Guide for detailed information about using indexes Oracle Database SQL Reference for the syntax of commands to work with indexes Oracle Database Administrator's Guide for information on creating hash clusters to improve performance, as an alternative to indexing Guidelines for Application-Specific Indexes You can create indexes on columns to speed up queries. Indexes provide faster access to data for operations that return a small portion of a table's rows. In general, you should create an index on a column in any of the following situations: ■ The column is queried frequently. ■ A referential integrity constraint exists on the column. ■ A UNIQUE key integrity constraint exists on the column. You can create an index on any column; however, if the column is not used in any of these situations, creating an index on the column does not increase performance and the index takes up resources unnecessarily. Although the database creates an index for you on a column with an integrity constraint, explicitly creating an index on such a column is recommended. You can use the following techniques to determine which columns are best candidates for indexing: Using Indexes in Application Development 5-1 Guidelines for Application-Specific Indexes ■ ■ Use the EXPLAIN PLAN feature to show a theoretical execution plan of a given query statement. Use the V$SQL_PLAN view to determine the actual execution plan used for a given query statement. Sometimes, if an index is not being used by default and it would be most efficient to use that index, you can use a query hint so that the index is used. The following sections explain how to create, alter, and drop indexes using SQL commands, and give guidelines for managing indexes. See Also: Oracle Database Performance Tuning Guide for information on using the V$SQL_PLAN view, the EXPLAIN PLAN statement, query hints, and measuring the performance benefits of indexes Create Indexes After Inserting Table Data Typically, you insert or load data into a table (using SQL*Loader or Import) before creating indexes. Otherwise, the overhead of updating the index slows down the insert or load operation. The exception to this rule is that you must create an index for a cluster before you insert any data into the cluster. Switch Your Temporary Tablespace to Avoid Space Problems Creating Indexes When you create an index on a table that already has data, Oracle Database must use sort space to create the index. The database uses the sort space in memory allocated for the creator of the index (the amount for each user is determined by the initialization parameter SORT_AREA_SIZE), but the database must also swap sort information to and from temporary segments allocated on behalf of the index creation. If the index is extremely large, it can be beneficial to complete the following steps: 1. Create a new temporary tablespace using the CREATE TABLESPACE command. 2. Use the TEMPORARY TABLESPACE option of the ALTER USER command to make this your new temporary tablespace. 3. Create the index using the CREATE INDEX command. 4. Drop this tablespace using the DROP TABLESPACE command. Then use the ALTER USER command to reset your temporary tablespace to your original temporary tablespace. Under certain conditions, you can load data into a table with the SQL*Loader "direct path load", and an index can be created as data is loaded. See Also: Oracle Database Utilities for information on direct path load Index the Correct Tables and Columns Use the following guidelines for determining when to create an index: ■ ■ Create an index if you frequently want to retrieve less than about 15% of the rows in a large table. This threshold percentage varies greatly, however, according to the relative speed of a table scan and how clustered the row data is about the index key. The faster the table scan, the lower the percentage; the more clustered the row data, the higher the percentage. Index columns that are used for joins to improve join performance. 5-2 Oracle Database Application Developer’s Guide - Fundamentals Guidelines for Application-Specific Indexes ■ ■ Primary and unique keys automatically have indexes, but you might want to create an index on a foreign key; see Chapter 6, "Maintaining Data Integrity in Application Development" for more information. Small tables do not require indexes; if a query is taking too long, then the table might have grown from small to large. Some columns are strong candidates for indexing. Columns with one or more of the following characteristics are good candidates for indexing: ■ Values are unique in the column, or there are few duplicates. ■ There is a wide range of values (good for regular indexes). ■ There is a small range of values (good for bitmap indexes). ■ The column contains many nulls, but queries often select all rows having a value. In this case, a comparison that matches all the non-null values, such as: WHERE COL_X >= -9.99 *power(10,125) is preferable to WHERE COL_X IS NOT NULL This is because the first uses an index on COL_X (assuming that COL_X is a numeric column). Columns with the following characteristics are less suitable for indexing: ■ There are many nulls in the column and you do not search on the non-null values. LONG and LONG RAW columns cannot be indexed. The size of a single index entry cannot exceed roughly one-half (minus some overhead) of the available space in the data block. Consult with the database administrator for assistance in determining the space required by an index. Limit the Number of Indexes for Each Table The more indexes, the more overhead is incurred as the table is altered. When rows are inserted or deleted, all indexes on the table must be updated. When a column is updated, all indexes on the column must be updated. You must weigh the performance benefit of indexes for queries against the performance overhead of updates. For example, if a table is primarily read-only, you might use more indexes; but, if a table is heavily updated, you might use fewer indexes. Choose the Order of Columns in Composite Indexes Although you can specify columns in any order in the CREATE INDEX command, the order of columns in the CREATE INDEX statement can affect query performance. In general, you should put the column expected to be used most often first in the index. You can create a composite index (using several columns), and the same index can be used for queries that reference all of these columns, or just some of them. For example, assume the columns of the VENDOR_PARTS table are as shown in Figure 5–1. Using Indexes in Application Development 5-3 Guidelines for Application-Specific Indexes Figure 5–1 The VENDOR_PARTS Table Table VENDOR_PARTS VEND ID 1012 1012 1012 1010 1010 1220 1012 1292 PART NO UNIT COST 10–440 10–441 457 10–440 457 08–300 08–300 457 .25 .39 4.95 .27 5.10 1.33 1.19 5.28 Assume that there are five vendors, and each vendor has about 1000 parts. Suppose that the VENDOR_PARTS table is commonly queried by SQL statements such as the following: SELECT * FROM vendor_parts WHERE part_no = 457 AND vendor_id = 1012; To increase the performance of such queries, you might create a composite index putting the most selective column first; that is, the column with the most values: CREATE INDEX ind_vendor_id ON vendor_parts (part_no, vendor_id); Composite indexes speed up queries that use the leading portion of the index. So in this example, queries with WHERE clauses using only the PART_NO column also note a performance gain. Because there are only five distinct values, placing a separate index on VENDOR_ID would serve no purpose. Gather Statistics to Make Index Usage More Accurate The database can use indexes more effectively when it has statistical information about the tables involved in the queries. You can gather statistics when the indexes are created by including the keywords COMPUTE STATISTICS in the CREATE INDEX statement. As data is updated and the distribution of values changes, you or the DBA can periodically refresh the statistics by calling procedures like DBMS_ STATS.GATHER_TABLE_STATISTICS and DBMS_STATS.GATHER_SCHEMA_ STATISTICS. Drop Indexes That Are No Longer Required You might drop an index if: ■ It does not speed up queries. The table might be very small, or there might be many rows in the table but very few index entries. ■ The queries in your applications do not use the index. ■ The index must be dropped before being rebuilt. When you drop an index, all extents of the index's segment are returned to the containing tablespace and become available for other objects in the tablespace. Use the SQL command DROP INDEX to drop an index. For example, the following statement drops a specific named index: DROP INDEX Emp_ename; 5-4 Oracle Database Application Developer’s Guide - Fundamentals Creating Indexes: Basic Examples If you drop a table, then all associated indexes are dropped. To drop an index, the index must be contained in your schema or you must have the DROP ANY INDEX system privilege. Privileges Required to Create an Index When using indexes in an application, you might need to request that the DBA grant privileges or make changes to initialization parameters. To create a new index, you must own, or have the INDEX object privilege for, the corresponding table. The schema that contains the index must also have a quota for the tablespace intended to contain the index, or the UNLIMITED TABLESPACE system privilege. To create an index in another user's schema, you must have the CREATE ANY INDEX system privilege. Creating Indexes: Basic Examples You can create an index for a table to improve the performance of queries issued against the corresponding table. You can also create an index for a cluster. You can create a composite index on multiple columns up to a maximum of 32 columns. A composite index key cannot exceed roughly one-half (minus some overhead) of the available space in the data block. Oracle Database automatically creates an index to enforce a UNIQUE or PRIMARY KEY integrity constraint. In general, it is better to create such constraints to enforce uniqueness, instead of using the obsolete CREATE UNIQUE INDEX syntax. Use the SQL command CREATE INDEX to create an index. In this example, an index is created for a single column, to speed up queries that test that column: CREATE INDEX emp_ename ON emp_tab(ename); In this example, several storage settings are explicitly specified for the index: CREATE INDEX emp_ename ON emp_tab(ename) TABLESPACE users STORAGE (INITIAL 20K NEXT 20k PCTINCREASE 75) PCTFREE 0 COMPUTE STATISTICS; In this example, the index applies to two columns, to speed up queries that test either the first column or both columns: CREATE INDEX emp_ename ON emp_tab(ename, empno) COMPUTE STATISTICS; In this example, the query is going to sort on the function UPPER(ENAME). An index on the ENAME column itself would not speed up this operation, and it might be slow to call the function for each result row. A function-based index precomputes the result of the function for each column value, speeding up queries that use the function for searching or sorting: CREATE INDEX emp_upper_ename ON emp_tab(UPPER(ename)) COMPUTE STATISTICS; Using Indexes in Application Development 5-5 When to Use Domain Indexes When to Use Domain Indexes Domain indexes are appropriate for special-purpose applications implemented using data cartridges. The domain index helps to manipulate complex data, such as spatial, audio, or video data. If you need to develop such an application, refer to Oracle Database Data Cartridge Developer's Guide. Oracle Database supplies a number of specialized data cartridges to help manage these kinds of complex data. So, if you need to create a search engine, or a geographic information system, you can do much of the work simply by creating the right kind of index. When to Use Function-Based Indexes A function-based index is an index built on an expression. It extends your indexing capabilities beyond indexing on a column. A function-based index increases the variety of ways in which you can access data. Note: ■ ■ The index is more effective if you gather statistics for the table or schema, using the procedures in the DBMS_STATS package. The index cannot contain any null values. Either make sure the appropriate columns contain no null values, or use the NVL function in the index expression to substitute some other value for nulls. The expression indexed by a function-based index can be an arithmetic expression or an expression that contains a PL/SQL function, package function, C callout, or SQL function. Function-based indexes also support linguistic sorts based on collation keys, efficient linguistic collation of SQL statements, and case-insensitive sorts. Like other indexes, function-based indexes improve query performance. For example, if you need to access a computationally complex expression often, then you can store it in an index. Then when you need to access the expression, it is already computed. You can find a detailed description of the advantages of function-based indexes in "Advantages of Function-Based Indexes" on page 5-6. Function-based indexes have all of the same properties as indexes on columns. Unlike indexes on columns which can be used by both cost-based and rule-based optimization, however, function-based indexes can be used by only by cost-based optimization. Other restrictions on function-based indexes are described in "Restrictions for Function-Based Indexes" on page 5-8. See Also: ■ ■ Oracle Database Concepts Oracle Database Administrator's Guide for information on creating function-based indexes Advantages of Function-Based Indexes Function-based indexes: ■ Increase the number of situations where the optimizer can perform a range scan instead of a full table scan. For example, consider the expression in this WHERE clause: 5-6 Oracle Database Application Developer’s Guide - Fundamentals When to Use Function-Based Indexes CREATE INDEX Idx ON Example_tab(Column_a + Column_b); SELECT * FROM Example_tab WHERE Column_a + Column_b < 10; The optimizer can use a range scan for this query because the index is built on (column_a + column_b). Range scans typically produce fast response times if the predicate selects less than 15% of the rows of a large table. The optimizer can estimate how many rows are selected by expressions more accurately if the expressions are materialized in a function-based index. (Expressions of function-based indexes are represented as virtual columns and ANALYZE can build histograms on such columns.) ■ ■ ■ Precompute the value of a computationally intensive function and store it in the index. An index can store computationally intensive expression that you access often. When you need to access a value, it is already computed, greatly improving query execution performance. Create indexes on object columns and REF columns. Methods that describe objects can be used as functions on which to build indexes. For example, you can use the MAP method to build indexes on an object type column. Create more powerful sorts. You can perform case-insensitive sorts with the UPPER and LOWER functions, descending order sorts with the DESC keyword, and linguistic-based sorts with the NLSSORT function. Note: Oracle Database sorts columns with the DESC keyword in descending order. Such indexes are treated as function-based indexes. Descending indexes cannot be bitmapped or reverse, and cannot be used in bitmapped optimizations. To get the DESC functionality prior to Oracle Database version 8, remove the DESC keyword from the CREATE INDEX statement. Another function-based index calls the object method distance_from_equator for each city in the table. The method is applied to the object column Reg_Obj. A query could use this index to quickly find cities that are more than 1000 miles from the equator: CREATE INDEX Distance_index ON Weatherdata_tab (Distance_from_equator (Reg_obj)); SELECT * FROM Weatherdata_tab WHERE (Distance_from_equator (Reg_Obj)) > '1000'; Another index stores the temperature delta and the maximum temperature. The result of the delta is sorted in descending order. A query could use this index to quickly find table rows where the temperature delta is less than 20 and the maximum temperature is greater than 75. CREATE INDEX compare_index ON Weatherdata_tab ((Maxtemp - Mintemp) DESC, Maxtemp); SELECT * FROM Weatherdata_tab WHERE ((Maxtemp - Mintemp) < '20' AND Maxtemp > '75'); Examples of Function-Based Indexes This section presents several examples of function-based indexes. Using Indexes in Application Development 5-7 When to Use Function-Based Indexes Example: Function-Based Index for Case-Insensitive Searches The following command allows faster case-insensitive searches in table EMP_TAB. CREATE INDEX Idx ON Emp_tab (UPPER(Ename)); The SELECT command uses the function-based index on UPPER(e_name) to return all of the employees with name like :KEYCOL. SELECT * FROM Emp_tab WHERE UPPER(Ename) like :KEYCOL; Example: Precomputing Arithmetic Expressions with a Function-Based Index The following command computes a value for each row using columns A, B, and C, and stores the results in the index. CREATE INDEX Idx ON Fbi_tab (A + B * (C - 1), A, B); The SELECT statement can either use index range scan (since the expression is a prefix of index IDX) or index fast full scan (which may be preferable if the index has specified a high parallel degree). SELECT a FROM Fbi_tab WHERE A + B * (C - 1) < 100; Example: Function-Based Index for Language-Dependent Sorting This example demonstrates how a function-based index can be used to sort based on the collation order for a national language. The NLSSORT function returns a sort key for each name, using the collation sequence GERMAN. CREATE INDEX Nls_index ON Nls_tab (NLSSORT(Name, 'NLS_SORT = German')); The SELECT statement selects all of the contents of the table and orders it by NAME. The rows are ordered using the German collation sequence. The Globalization Support parameters are not needed in the SELECT statement, because in a German session, NLS_SORT is set to German and NLS_COMP is set to ANSI. SELECT * FROM Nls_tab WHERE Name IS NOT NULL ORDER BY Name; Restrictions for Function-Based Indexes Note the following restrictions for function-based indexes: ■ ■ Only cost-based optimization can use function-based indexes. Remember to call DBMS_STATS.GATHER_TABLE_STATISTICS or DBMS_STATS.GATHER_ SCHEMA_STATISTICS, for the function-based index to be effective. Any top-level or package-level PL/SQL functions that are used in the index expression must be declared as DETERMINISTIC. That is, they always return the same result given the same input, for example, the UPPER function. You must ensure that the subprogram really is deterministic, because Oracle Database does not check that the assertion is true. The following semantic rules demonstrate how to use the keyword DETERMINISTIC: ■ ■ You can declare a top level subprogram as DETERMINISTIC. You can declare a PACKAGE level subprogram as DETERMINISTIC in the PACKAGE specification but not in the PACKAGE BODY. Errors are raised if DETERMINISTIC is used inside a PACKAGE BODY. 5-8 Oracle Database Application Developer’s Guide - Fundamentals When to Use Function-Based Indexes ■ ■ ■ ■ You can declare a private subprogram (declared inside another subprogram or a PACKAGE BODY) as DETERMINISTIC. A DETERMINISTIC subprogram can call another subprogram whether the called program is declared as DETERMINISTIC or not. If you change the semantics of a DETERMINISTIC function and recompile it, then existing function-based indexes and materialized views report results for the prior version of the function. Thus, if you change the semantics of a function, you must manually rebuild any dependent function-based indexes and materialized views. Expressions in a function-based index cannot contain any aggregate functions. The expressions should reference only columns in a row in the table. ■ You must analyze the table or index before the index is used. ■ Bitmap optimizations cannot use descending indexes. ■ Function-based indexes are not used when OR-expansion is done. ■ ■ The index function cannot be marked NOT NULL. To avoid a full table scan, you must ensure that the query cannot fetch null values. Function-based indexes cannot use expressions that return VARCHAR2 or RAW data types of unknown length from PL/SQL functions. A workaround is to limit the size of the function's output by indexing a substring of known length: -- INITIALS() might return 1 letter, 2 letters, 3 letters, and so on. -- We limit the return value to 10 characters for purposes of the index. CREATE INDEX func_substr_index ON emp_tab(substr(initials(ename),1,10); -- Call SUBSTR both when creating the index and when referencing -- the function in queries. SELECT SUBSTR(initials(ename),1,10) FROM emp_tab; See Also: Oracle Database SQL Reference for an account of CREATE FUNCTION restrictions Using Indexes in Application Development 5-9 When to Use Function-Based Indexes 5-10 Oracle Database Application Developer’s Guide - Fundamentals 6 Maintaining Data Integrity in Application Development This chapter explains how to enforce the business rules associated with your database and prevent the entry of invalid information into tables by using integrity constraints. Topics include the following: ■ Overview of Integrity Constraints ■ Enforcing Referential Integrity with Constraints ■ Managing Constraints That Have Associated Indexes ■ Guidelines for Indexing Foreign Keys ■ About Referential Integrity in a Distributed Database ■ When to Use CHECK Integrity Constraints ■ Examples of Defining Integrity Constraints ■ Enabling and Disabling Integrity Constraints ■ Altering Integrity Constraints ■ Dropping Integrity Constraints ■ Managing FOREIGN KEY Integrity Constraints ■ Viewing Definitions of Integrity Constraints Overview of Integrity Constraints You can define integrity constraints to enforce business rules on data in your tables. Business rules specify conditions and relationships that must always be true, or must always be false. Because each company defines its own policies about things like salaries, employee numbers, inventory tracking, and so on, you can specify a different set of rules for each database table. When an integrity constraint applies to a table, all data in the table must conform to the corresponding rule. When you issue a SQL statement that modifies data in the table, Oracle Database ensures that the new data satisfies the integrity constraint, without the need to do any checking within your program. When to Enforce Business Rules with Integrity Constraints You can enforce rules by defining integrity constraints more reliably than by adding logic to your application. Oracle Database can check that all the data in a table obeys an integrity constraint faster than an application can. Maintaining Data Integrity in Application Development 6-1 Overview of Integrity Constraints Example of an Integrity Constraint for a Business Rule To ensure that each employee works for a valid department, first create a rule that all values in the department table are unique: ALTER TABLE Dept_tab ADD PRIMARY KEY (Deptno); Then, create a rule that every department listed in the employee table must match one of the values in the department table: ALTER TABLE Emp_tab ADD FOREIGN KEY (Deptno) REFERENCES Dept_tab(Deptno); When you add a new employee record to the table, Oracle Database automatically checks that its department number appears in the department table. To enforce this rule without integrity constraints, you can use a trigger to query the department table and test that each new employee's department is valid. But this method is less reliable than the integrity constraint. SELECT in Oracle Database uses "consistent read", so the query might miss uncommitted changes from other transactions. When to Enforce Business Rules in Applications You might enforce business rules through application logic as well as through integrity constraints, if you can filter out bad data before attempting an insert or update. This might let you provide instant feedback to the user, and reduce the load on the database. This technique is appropriate when you can determine that data values are wrong or out of range without checking against any data already in the table. Creating Indexes for Use with Constraints All enabled unique and primary keys require corresponding indexes. You should create these indexes by hand, rather than letting the database create them for you. Note that: ■ ■ Constraints use existing indexes where possible, rather than creating new ones. Unique and primary keys can use non-unique as well as unique indexes. They can even use just the first few columns of non-unique indexes. ■ At most one unique or primary key can use each non-unique index. ■ The column orders in the index and the constraint do not need to match. ■ If you need to check whether an index is used by a constraint, for example when you want to drop the index, the object number of the index used by a unique or primary key constraint is stored in CDEF$.ENABLED for that constraint. It is not shown in any catalog view. You should almost always index foreign keys; the database does not do this for you automatically. When to Use NOT NULL Integrity Constraints By default, all columns can contain nulls. Only define NOT NULL constraints for columns of a table that absolutely require values at all times. For example, a new employee's manager or hire date might be temporarily omitted. Some employees might not have a commission. Columns like these should not have 6-2 Oracle Database Application Developer’s Guide - Fundamentals Overview of Integrity Constraints NOT NULL integrity constraints. However, an employee name might be required from the very beginning, and you can enforce this rule with a NOT NULL integrity constraint. NOT NULL constraints are often combined with other types of integrity constraints to further restrict the values that can exist in specific columns of a table. Use the combination of NOT NULL and UNIQUE key integrity constraints to force the input of values in the UNIQUE key; this combination of data integrity rules eliminates the possibility that any new row' data will ever attempt to conflict with an existing row's data. Because Oracle Database indexes do not store keys that are all null, if you want to allow index-only scans of the table or some other operation that requires indexing all rows, you must put a NOT NULL constraint on at least one indexed column. See Also: "Defining Relationships Between Parent and Child Tables" on page 6-8 A NOT NULL constraint is specified like this: ALTER TABLE emp MODIFY ename NOT NULL; Figure 6–1 shows an example of a table with NOT NULL integrity constraints. Figure 6–1 Table with NOT NULL Integrity Constraints Table EMPLOYEES ID LNAME JOB MGR HIREDATE 100 101 102 103 King Kochhar De Hann Hunold AD_PRES AD_VP AD_VP IT_PROG 100 100 102 17–JUN–87 21–SEP–89 13–JAN–93 03–JAN–90 SAL 24000 17000 17000 9000 COMM DEPTNO 90 90 90 60 Absence of NOT NULL Constraint (any row can contain a null for this column) NOT NULL Constraint (no row may contain a null value for this column) When to Use Default Column Values Assign default values to columns that contain a typical value. For example, in the DEPT_TAB table, if most departments are located at one site, then the default value for the LOC column can be set to this value (such as NEW YORK). Default values can help avoid errors where there is a number, such as zero, that applies to a column that has no entry. For example, a default value of zero can simplify testing, by changing a test like this: IF sal IS NOT NULL AND sal < 50000 to the simpler form: IF sal < 50000 Depending upon your business rules, you might use default values to represent zero or false, or leave the default values as NULL to signify an unknown value. Defaults are also useful when you use a view to make a subset of a table's columns visible. For example, you might allow users to insert rows through a view. The base table might also have a column named INSERTER, not included in the definition of the Maintaining Data Integrity in Application Development 6-3 Overview of Integrity Constraints view, to log the user that inserts each row. To record the user name automatically, define a default value that calls the USER function: CREATE TABLE ( value1 value2 inserter ); audit_trail NUMBER, VARCHAR2(32), VARCHAR2(30) DEFAULT USER Setting Default Column Values Default values can be defined using any literal, or almost any expression, including calls to the following: ■ SYSDATE ■ SYS_CONTEXT ■ USER ■ USERENV ■ UID Default values cannot include expressions that refer to a sequence, PL/SQL function, column, LEVEL, ROWNUM, or PRIOR. The datatype of a default literal or expression must match or be convertible to the column datatype. Sometimes the default value is the result of a SQL function. For example, a call to SYS_CONTEXT can set a different default value depending on conditions such as the user name. To be used as a default value, a SQL function must have parameters that are all literals, cannot reference any columns, and cannot call any other functions. If you do not explicitly define a default value for a column, the default for the column is implicitly set to NULL. You can use the keyword DEFAULT within an INSERT statement instead of a literal value, and the corresponding default value is inserted. Choosing a Table's Primary Key Each table can have one primary key, which uniquely identifies each row in a table and ensures that no duplicate rows exist. Use the following guidelines when selecting a primary key: ■ ■ ■ ■ ■ Whenever practical, use a column containing a sequence number. It is a simple way to satisfy all the other guidelines. Choose a column whose data values are unique, because the purpose of a primary key is to uniquely identify each row of the table. Choose a column whose data values are never changed. A primary key value is only used to identify a row in the table, and its data should never be used for any other purpose. Therefore, primary key values should rarely or never be changed. Choose a column that does not contain any nulls. A PRIMARY KEY constraint, by definition, does not allow any row to contain a null in any column that is part of the primary key. Choose a column that is short and numeric. Short primary keys are easy to type. You can use sequence numbers to easily generate numeric primary keys. 6-4 Oracle Database Application Developer’s Guide - Fundamentals Overview of Integrity Constraints Minimize your use of composite primary keys. Although composite primary keys are allowed, they do not satisfy all of the other recommendations. For example, composite primary key values are long and cannot be assigned by sequence numbers. ■ When to Use UNIQUE Key Integrity Constraints Choose columns for unique keys carefully. The purpose of these constraints is different from that of primary keys. Unique key constraints are appropriate for any column where duplicate values are not allowed. Primary keys identify each row of the table uniquely, and typically contain values that have no significance other than being unique. Figure 6–2 shows an example of a table with a unique key constraint. Figure 6–2 Table with a UNIQUE Key Constraint Table DEPARTMENTS DEPID DNAME LOC 10 20 30 40 Administration Marketing Purchasing Human Resources 1700 1800 1700 2400 UNIQUE Key Constraint (no row may duplicate a value in the constraint's column) INSERT INTO 50 MARKETING 60 1700 This row violates the UNIQUE key constraint, because "MARKETING" is already present in another row; therefore, it is not allowed in the table. 2400 This row is allowed because a null value is entered for the DNAME column; however, if a NOT NULL constraint is also defined on the DNAME column, this row is not allowed. You cannot have identical values in the non-null columns of a composite UNIQUE key constraint (UNIQUE key constraints allow NULL values). Note: Some examples of good unique keys include: ■ ■ ■ ■ An employee social security number (the primary key might be the employee number) A truck license plate number (the primary key might be the truck number) A customer phone number, consisting of the two columns AREA_CODE and LOCAL_PHONE (the primary key might be the customer number) A department name and location (the primary key might be the department number) Constraints On Views: for Performance, Not Data Integrity The constraints discussed throughout this chapter apply to tables, not views. Although you can declare constraints on views, such constraints do not help maintain data integrity. Instead, they are used to enable query rewrites on queries involving Maintaining Data Integrity in Application Development 6-5 Enforcing Referential Integrity with Constraints views, which helps performance with materialized views and other data warehousing features. Such constraints are always declared with the DISABLE keyword, and you cannot use the VALIDATE keyword. The constraints are never enforced, and there is no associated index. See Also: Oracle Database Data Warehousing Guide for information on query rewrite, materialized views, and the performance reasons for declaring constraints on views Enforcing Referential Integrity with Constraints Whenever two tables contain one or more common columns, Oracle Database can enforce the relationship between the two tables through a referential integrity constraint. Define a PRIMARY or UNIQUE key constraint on the column in the parent table (the one that has the complete set of column values). Define a FOREIGN KEY constraint on the column in the child table (the one whose values must refer to existing values in the other table). "Defining Relationships Between Parent and Child Tables" on page 6-8 for information on defining additional integrity constraints, including the foreign key See Also: Figure 6–3 shows a foreign key defined on the department number. It guarantees that every value in this column must match a value in the primary key of the department table. This constraint prevents erroneous department numbers from getting into the employee table. Foreign keys can be comprised of multiple columns. Such a composite foreign key must reference a composite primary or unique key of the exact same structure, with the same number of columns and the same datatypes. Because composite primary and unique keys are limited to 32 columns, a composite foreign key is also limited to 32 columns. 6-6 Oracle Database Application Developer’s Guide - Fundamentals Enforcing Referential Integrity with Constraints Figure 6–3 Tables with Referential Integrity Constraints Parent Key Primary key of referenced table Table DEPARTMENTS DEPID DNAME LOC 10 20 30 40 Administration Marketing Purchasing Human Resources 1700 1800 1700 2400 Foreign Key (values in dependent table must match a value in unique key or primary key of referenced table) Referenced or Parent Table Table EMPLOYEES ID LNAME JOB MGR HIREDATE 100 101 102 103 King Kochhar De Hann Hunold AD_PRES AD_VP AD_VP IT_PROG 100 100 102 17–JUN–87 21–SEP–89 13–JAN–93 03–JAN–90 SAL COMM DEPTNO 24000 17000 17000 9000 90 90 90 60 25 Dependent or Child Table INSERT INTO 556 CRICKET PU_CLERK 31–OCT–96 5000 556 CRICKET PU_CLERK 31–OCT–96 5000 This row violates the referential constraint because "50" is not present in the referenced table's primary key; therefore, the row is not allowed in the table. This row is allowed in the table because a null value is entered in the DEPTNO column; however, if a not null constraint is also defined for this column, this row is not allowed. About Nulls and Foreign Keys Foreign keys allow key values that are all NULL, even if there are no matching PRIMARY or UNIQUE keys. ■ ■ By default (without any NOT NULL or CHECK clauses), the FOREIGN KEY constraint enforces the "match none" rule for composite foreign keys in the ANSI/ISO standard. To enforce the match full rule for NULL values in composite foreign keys, which requires that all components of the key be NULL or all be non-NULL, define a CHECK constraint that allows only all nulls or all non-nulls in the composite foreign key. For example, with a composite key comprised of columns A, B, and C: CHECK ((A IS NULL AND B IS NULL AND C IS NULL) OR (A IS NOT NULL AND B IS NOT NULL AND C IS NOT NULL)) ■ In general, it is not possible to use declarative referential integrity to enforce the match partial rule for NULL values in composite foreign keys, which requires the Maintaining Data Integrity in Application Development 6-7 Enforcing Referential Integrity with Constraints non-NULL portions of the key to appear in the corresponding portions in the primary or unique key of a single row in the referenced table. You can often use triggers to handle this case, as described in Chapter 9, "Coding Triggers". Defining Relationships Between Parent and Child Tables Several relationships between parent and child tables can be determined by the other types of integrity constraints defined on the foreign key in the child table. No Constraints on the Foreign Key When no other constraints are defined on the foreign key, any number of rows in the child table can reference the same parent key value. This model allows nulls in the foreign key. This model establishes a one-to-many relationship between the parent and foreign keys that allows undetermined values (nulls) in the foreign key. An example of such a relationship is shown in Figure 6–3, "Tables with Referential Integrity Constraints" between the employee and department tables. Each department (parent key) has many employees (foreign key), and some employees might not be in a department (nulls in the foreign key). NOT NULL Constraint on the Foreign Key When nulls are not allowed in a foreign key, each row in the child table must explicitly reference a value in the parent key because nulls are not allowed in the foreign key. Any number of rows in the child table can reference the same parent key value, so this model establishes a one-to-many relationship between the parent and foreign keys. However, each row in the child table must have a reference to a parent key value; the absence of a value (a null) in the foreign key is not allowed. The same example in the previous section can be used to illustrate such a relationship. However, in this case, employees must have a reference to a specific department. When a UNIQUE constraint is defined on the foreign key, only one row in the child table can reference a given parent key value. This model allows nulls in the foreign key. UNIQUE Constraint on the Foreign Key This model establishes a one-to-one relationship between the parent and foreign keys that allows undetermined values (nulls) in the foreign key. For example, assume that the employee table had a column named MEMBERNO, referring to an employee membership number in the company insurance plan. Also, a table named INSURANCE has a primary key named MEMBERNO, and other columns of the table keep respective information relating to an employee insurance policy. The MEMBERNO in the employee table should be both a foreign key and a unique key: ■ ■ To enforce referential integrity rules between the EMP_TAB and INSURANCE tables (the FOREIGN KEY constraint) To guarantee that each employee has a unique membership number (the UNIQUE key constraint) When both UNIQUE and NOT NULL constraints are defined on the foreign key, only one row in the child table can reference a given parent key value, and because NULL values are not allowed in the foreign key, each row in the child table must explicitly reference a value in the parent key. UNIQUE and NOT NULL Constraints on the Foreign Key This model establishes a one-to-one relationship between the parent and foreign keys that does not allow undetermined values (nulls) in the foreign key. If you expand the previous example by adding a NOT NULL constraint on the MEMBERNO column of the 6-8 Oracle Database Application Developer’s Guide - Fundamentals Enforcing Referential Integrity with Constraints employee table, in addition to guaranteeing that each employee has a unique membership number, you also ensure that no undetermined values (nulls) are allowed in the MEMBERNO column of the employee table. Rules for Multiple FOREIGN KEY Constraints Oracle Database allows a column to be referenced by multiple FOREIGN KEY constraints; there is no limit on the number of dependent keys. This situation might be present if a single column is part of two different composite foreign keys. Deferring Constraint Checks When Oracle Database checks a constraint, it signals an error if the constraint is not satisfied. You can use the SET CONSTRAINTS statement to defer checking the validity of constraints until the end of a transaction. Note: You cannot issue a SET CONSTRAINTS statement inside a trigger. The SET CONSTRAINTS setting lasts for the duration of the transaction, or until another SET CONSTRAINTS statement resets the mode. See Also: Oracle Database SQL Reference Guidelines for Deferring Constraint Checks Consider the following guidelines when deferring constraint checks. Select Appropriate Data You may wish to defer constraint checks on UNIQUE and FOREIGN keys if the data you are working with has any of the following characteristics: ■ ■ ■ Tables are snapshots. Some tables contain a large amount of data being manipulated by another application, which may or may not return the data in the same order. Update cascade operations on foreign keys. Ensure Constraints Are Created Deferrable After you have identified and selected the appropriate tables, make sure their FOREIGN, UNIQUE and PRIMARY key constraints are created deferrable. You can do so by issuing statements similar to the following: CREATE TABLE dept ( deptno NUMBER PRIMARY KEY, dname VARCHAR2 (30) ); CREATE TABLE emp ( empno NUMBER, ename VARCHAR2 (30), deptno NUMBER REFERENCES (dept), CONSTRAINT epk PRIMARY KEY (empno) DEFERRABLE, CONSTRAINT efk FOREIGN KEY (deptno) REFERENCES (dept.deptno) DEFERRABLE); INSERT INTO dept VALUES (10, 'Accounting'); INSERT INTO dept VALUES (20, 'SALES'); INSERT INTO emp VALUES (1, 'Corleone', 10); INSERT INTO emp VALUES (2, 'Costanza', 20); COMMIT; Maintaining Data Integrity in Application Development 6-9 Managing Constraints That Have Associated Indexes SET CONSTRAINT efk DEFERRED; UPDATE dept SET deptno = deptno + 10 WHERE deptno = 20; SELECT * from emp ORDER BY deptno; EMPNO ENAME DEPTNO ------------------ ------1 Corleone 10 2 Costanza 20 UPDATE emp SET deptno = deptno + 10 WHERE deptno = 20; SELECT * FROM emp ORDER BY deptno; EMPNO ----1 2 COMMIT; ENAME -------------Corleone Costanza DEPTNO ------10 30 Set All Constraints Deferred Within the application that manipulates the data, you must set all constraints deferred before you begin processing any data. Use the following DML statement to set all constraints deferred: SET CONSTRAINTS ALL DEFERRED; Note: The SET CONSTRAINTS statement applies only to the current transaction. The defaults specified when you create a constraint remain as long as the constraint exists. The ALTER SESSION SET CONSTRAINTS statement applies for the current session only. Check the Commit (Optional) You can check for constraint violations before committing by issuing the SET CONSTRAINTS ALL IMMEDIATE statement just before issuing the COMMIT. If there are any problems with a constraint, this statement will fail and the constraint causing the error will be identified. If you commit while constraints are violated, the transaction will be rolled back and you will receive an error message. Managing Constraints That Have Associated Indexes When you create a UNIQUE or PRIMARY key, Oracle Database checks to see if an existing index can be used to enforce uniqueness for the constraint. If there is no such index, the database creates one. Minimizing Space and Time Overhead for Indexes Associated with Constraints When Oracle Database uses a unique index to enforce a constraint, and constraints associated with the unique index are dropped or disabled, the index is dropped. To preserve the statistics associated with the index (because, for example, it would take a long time to re-create it), you can specify the KEEP INDEX clause on the DROP command for the constraint. While enabled foreign keys reference a PRIMARY or UNIQUE key, you cannot disable or drop the PRIMARY or UNIQUE key constraint or the index. UNIQUE and PRIMARY keys with deferrable constraints must all use non-unique indexes. Note: 6-10 Oracle Database Application Developer’s Guide - Fundamentals When to Use CHECK Integrity Constraints To reuse existing indexes when creating unique and primary key constraints, you can include USING INDEX in the constraint clause. For example: CREATE TABLE b ( b1 INTEGER, b2 INTEGER, CONSTRAINT unique1 (b1, b2) USING INDEX (CREATE UNIQUE INDEX b_index on b(b1, b2), CONSTRAINT unique2 (b1, b2) USING INDEX b_index ); Guidelines for Indexing Foreign Keys You should almost always index foreign keys. The only exception is when the matching unique or primary key is never updated or deleted. Oracle Database Concepts for information on locking mechanisms involving indexes and keys See Also: About Referential Integrity in a Distributed Database The declaration of a referential integrity constraint cannot specify a foreign key that references a primary or unique key of a remote table. However, you can maintain parent/child table relationships across nodes using triggers. See Also: Chapter 9, "Coding Triggers" for more information about triggers that enforce referential integrity If you decide to define referential integrity across the nodes of a distributed database using triggers, be aware that network failures can make both the parent table and the child table inaccessible. Note: For example, assume that the child table is in the SALES database, and the parent table is in the HQ database. If the network connection between the two databases fails, then some DML statements against the child table (those that insert rows or update a foreign key value) cannot proceed, because the referential integrity triggers must have access to the parent table in the HQ database. When to Use CHECK Integrity Constraints Use CHECK constraints when you need to enforce integrity rules based on logical expressions, such as comparisons. Never use CHECK constraints when any of the other types of integrity constraints can provide the necessary checking. See Also: "Choosing Between CHECK and NOT NULL Integrity Constraints" on page 6-13 Examples of CHECK constraints include the following: ■ A CHECK constraint on employee salaries so that no salary value is greater than 10000. Maintaining Data Integrity in Application Development 6-11 When to Use CHECK Integrity Constraints ■ ■ A CHECK constraint on department locations so that only the locations "BOSTON", "NEW YORK", and "DALLAS" are allowed. A CHECK constraint on the salary and commissions columns to prevent the commission from being larger than the salary. Restrictions on CHECK Constraints A CHECK integrity constraint requires that a condition be true or unknown for every row of the table. If a statement causes the condition to evaluate to false, then the statement is rolled back. The condition of a CHECK constraint has the following limitations: ■ ■ ■ ■ The condition must be a boolean expression that can be evaluated using the values in the row being inserted or updated. The condition cannot contain subqueries or sequences. The condition cannot include the SYSDATE, UID, USER, or USERENV SQL functions. The condition cannot contain the pseudocolumns LEVEL, PRIOR, or ROWNUM. Oracle Database SQL Reference for an explanation of these pseudocolumns See Also: ■ The condition cannot contain a user-defined SQL function. Designing CHECK Constraints When using CHECK constraints, remember that a CHECK constraint is violated only if the condition evaluates to false; true and unknown values (such as comparisons with nulls) do not violate a check condition. Make sure that any CHECK constraint that you define is specific enough to enforce the rule. For example, consider the following CHECK constraint: CHECK (Sal > 0 OR Comm >= 0) At first glance, this rule may be interpreted as "do not allow a row in the employee table unless the employee salary is greater than zero or the employee commission is greater than or equal to zero." But if a row is inserted with a null salary, that row does not violate the CHECK constraint, regardless of whether or not the commission value is valid, because the entire check condition is evaluated as unknown. In this case, you can prevent such violations by placing NOT NULL integrity constraints on both the SAL and COMM columns. Note: If you are not sure when unknown values result in NULL conditions, review the truth tables for the logical operators AND and OR in Oracle Database SQL Reference Rules for Multiple CHECK Constraints A single column can have multiple CHECK constraints that reference the column in its definition. There is no limit to the number of CHECK constraints that can be defined that reference a column. The order in which the constraints are evaluated is not defined, so be careful not to rely on the order or to define multiple constraints that conflict with each other. 6-12 Oracle Database Application Developer’s Guide - Fundamentals Examples of Defining Integrity Constraints Choosing Between CHECK and NOT NULL Integrity Constraints According to the ANSI/ISO standard, a NOT NULL integrity constraint is an example of a CHECK integrity constraint, where the condition is the following: CHECK (Column_name IS NOT NULL) Therefore, NOT NULL integrity constraints for a single column can, in practice, be written in two forms: using the NOT NULL constraint or a CHECK constraint. For ease of use, you should always choose to define NOT NULL integrity constraints, instead of CHECK constraints with the IS NOT NULL condition. In the case where a composite key can allow only all nulls or all values, you must use a CHECK integrity constraint. For example, the following expression of a CHECK integrity constraint allows a key value in the composite key made up of columns C1 and C2 to contain either all nulls or all values: CHECK ((C1 IS NULL AND C2 IS NULL) OR (C1 IS NOT NULL AND C2 IS NOT NULL)) Examples of Defining Integrity Constraints Here are some examples showing how to create simple constraints during the prototype phase of your database design. Each constraint is given a name in these examples. Naming the constraints prevents the database from creating multiple copies of the same constraint, with different system-generated names, if the DDL is run multiple times. See Also: Oracle Database Administrator's Guide for information on creating and maintaining constraints for a large production database Example: Defining Integrity Constraints with the CREATE TABLE Command The following examples of CREATE TABLE statements show the definition of several integrity constraints: CREATE TABLE Dept_tab ( Deptno NUMBER(3) CONSTRAINT Dept_pkey PRIMARY KEY, Dname VARCHAR2(15), Loc VARCHAR2(15), CONSTRAINT Dname_ukey UNIQUE (Dname, Loc), CONSTRAINT Loc_check1 CHECK (loc IN ('NEW YORK', 'BOSTON', 'CHICAGO'))); CREATE TABLE Empno Ename Job Mgr Hiredate Sal Comm Deptno Emp_tab ( NUMBER(5) CONSTRAINT Emp_pkey PRIMARY KEY, VARCHAR2(15) NOT NULL, VARCHAR2(10), NUMBER(5) CONSTRAINT Mgr_fkey REFERENCES Emp_tab, DATE, NUMBER(7,2), NUMBER(5,2), NUMBER(3) NOT NULL CONSTRAINT dept_fkey REFERENCES Dept_tab ON DELETE CASCADE); Maintaining Data Integrity in Application Development 6-13 Enabling and Disabling Integrity Constraints Example: Defining Constraints with the ALTER TABLE Command You can also define integrity constraints using the constraint clause of the ALTER TABLE command. For example: CREATE UNIQUE INDEX I_dept ON Dept_tab(deptno); ALTER TABLE Dept_tab ADD CONSTRAINT Dept_pkey PRIMARY KEY (deptno); ALTER TABLE Emp_tab ADD CONSTRAINT Dept_fkey FOREIGN KEY (Deptno) REFERENCES Dept_tab; ALTER TABLE Emp_tab MODIFY (Ename VARCHAR2(15) NOT NULL); You cannot create a validated constraint on a table if the table already contains any rows that would violate the constraint. Privileges Required to Create Constraints The creator of a constraint must have the ability to create tables (the CREATE TABLE or CREATE ANY TABLE system privilege), or the ability to alter the table (the ALTER object privilege for the table or the ALTER ANY TABLE system privilege) with the constraint. Additionally, UNIQUE and PRIMARY KEY integrity constraints require that the owner of the table have either a quota for the tablespace that contains the associated index or the UNLIMITED TABLESPACE system privilege. FOREIGN KEY integrity constraints also require some additional privileges. See Also: "Privileges Required to Create FOREIGN KEY Integrity Constraints" on page 6-20 Naming Integrity Constraints Assign names to constraints NOT NULL, UNIQUE KEY, PRIMARY KEY, FOREIGN KEY, and CHECK using the CONSTRAINT option of the constraint clause. This name must be unique with respect to other constraints that you own. If you do not specify a constraint name, one is assigned automatically by Oracle Database. Picking your own name makes error messages for constraint violations more understandable, and prevents the creation of duplicate constraints with different names if the SQL statements are run more than once. See the previous examples of the CREATE TABLE and ALTER TABLE statements for examples of the CONSTRAINT option of the constraint clause. Note that the name of each constraint is included with other information about the constraint in the data dictionary. "Viewing Definitions of Integrity Constraints" on page 6-21 for examples of data dictionary views See Also: Enabling and Disabling Integrity Constraints This section explains the mechanisms and procedures for manually enabling and disabling integrity constraints. enabled constraint. When a constraint is enabled, the corresponding rule is enforced on the data values in the associated columns. The definition of the constraint is stored in the data dictionary. disabled constraint. When a constraint is disabled, the corresponding rule is not enforced. The definition of the constraint is still stored in the data dictionary. 6-14 Oracle Database Application Developer’s Guide - Fundamentals Enabling and Disabling Integrity Constraints An integrity constraint represents an assertion about the data in a database. This assertion is always true when the constraint is enabled. The assertion may or may not be true when the constraint is disabled, because data that violates the integrity constraint can be in the database. Why Disable Constraints? During day-to-day operations, constraints should always be enabled. In certain situations, temporarily disabling the integrity constraints of a table makes sense for performance reasons. For example: ■ ■ ■ When loading large amounts of data into a table using SQL*Loader When performing batch operations that make massive changes to a table (such as changing each employee number by adding 1000 to the existing number) When importing or exporting one table at a time Temporarily turning off integrity constraints can speed up these operations. About Exceptions to Integrity Constraints If a row of a table disobeys an integrity constraint, then this row is in violation of the constraint and is called an exception to the constraint. If any exceptions exist, then the constraint cannot be enabled. The rows that violate the constraint must be updated or deleted before the constraint can be enabled. You can identify exceptions for a specific integrity constraint as you try to enable the constraint. See Also: "Fixing Constraint Exceptions" on page 6-17 for more information on this procedure Enabling Constraints When you define an integrity constraint in a CREATE TABLE or ALTER TABLE statement, Oracle Database automatically enables the constraint by default. For code clarity, you can explicitly enable the constraint by including the ENABLE clause in its definition. Use this technique when creating tables that start off empty, and are populated a row at a time by individual transactions. In such cases, you want to ensure that data is consistent at all times, and the performance overhead of each DML operation is small. The following CREATE TABLE and ALTER TABLE statements both define and enable integrity constraints: CREATE TABLE Emp_tab ( Empno NUMBER(5) PRIMARY KEY); ALTER TABLE Emp_tab ADD PRIMARY KEY (Empno); An ALTER TABLE statement that tries to enable an integrity constraint will fail if any existing rows of the table violate the integrity constraint. The statement is rolled back and the constraint definition is not stored and not enabled. See Also: "Fixing Constraint Exceptions" on page 6-17 for more information about rows that violate integrity constraints Maintaining Data Integrity in Application Development 6-15 Enabling and Disabling Integrity Constraints Creating Disabled Constraints The following CREATE TABLE and ALTER TABLE statements both define and disable integrity constraints: CREATE TABLE Emp_tab ( Empno NUMBER(5) PRIMARY KEY DISABLE); ALTER TABLE Emp_tab ADD PRIMARY KEY (Empno) DISABLE; Use this technique when creating tables that will be loaded with large amounts of data before anybody else accesses them, particularly if you need to cleanse data after loading it, or need to fill empty columns with sequence numbers or parent/child relationships. An ALTER TABLE statement that defines and disables an integrity constraints never fails, because its rule is not enforced. Enabling and Disabling Existing Integrity Constraints Use the ALTER TABLE command to: ■ Enable a disabled constraint, using the ENABLE clause. ■ Disable an enabled constraint, using the DISABLE clause. Enabling Existing Constraints Once you have finished cleansing data and filling empty columns, you can enable constraints that were disabled during data loading. The following statements are examples of statements that enable disabled integrity constraints: ALTER TABLE Dept_tab ENABLE CONSTRAINT Dname_ukey; ALTER TABLE Dept_tab ENABLE PRIMARY KEY ENABLE UNIQUE (Dname) ENABLE UNIQUE (Loc); An ALTER TABLE statement that attempts to enable an integrity constraint fails if any of the table rows violate the integrity constraint. The statement is rolled back and the constraint is not enabled. See Also: "Fixing Constraint Exceptions" on page 6-17 for more information about rows that violate integrity constraints Disabling Existing Constraints If you need to perform a large load or update when a table already contains data, you can temporarily disable constraints to improve performance of the bulk operation. The following statements are examples of statements that disable enabled integrity constraints: ALTER TABLE Dept_tab DISABLE CONSTRAINT Dname_ukey; ALTER TABLE Dept_tab 6-16 Oracle Database Application Developer’s Guide - Fundamentals Altering Integrity Constraints DISABLE PRIMARY KEY DISABLE UNIQUE (Dname) DISABLE UNIQUE (Loc); Tip: Using the Data Dictionary to Find Constraints The preceding examples require that you know the relevant constraint names and which columns they affect. To find this information, you can query one of the data dictionary views defined for constraints, USER_CONSTRAINTS or USER_CONS_ COLUMNS. For more information about these views, refer to "Viewing Definitions of Integrity Constraints" on page 6-21 and Oracle Database Reference. Guidelines for Enabling and Disabling Key Integrity Constraints When enabling or disabling UNIQUE, PRIMARY KEY, and FOREIGN KEY integrity constraints, you should be aware of several important issues and prerequisites. UNIQUE key and PRIMARY KEY constraints are usually managed by the database administrator. See Also: Oracle Database Administrator's Guide and "Managing FOREIGN KEY Integrity Constraints" on page 6-19 Fixing Constraint Exceptions When you try to create or enable a constraint, and the statement fails because integrity constraint exceptions exist, the statement is rolled back. You cannot enable the constraint until all exceptions are either updated or deleted. To determine which rows violate the integrity constraint, include the EXCEPTIONS option in the ENABLE clause of a CREATE TABLE or ALTER TABLE statement. See Also: Oracle Database Administrator's Guide for more information about responding to constraint exceptions Altering Integrity Constraints Starting with Oracle8i, you can alter the state of an existing constraint with the MODIFY CONSTRAINT clause. Oracle Database SQL Reference for information on the parameters you can modify See Also: MODIFY CONSTRAINT Example #1 The following commands show several alternatives for whether the CHECK constraint is enforced, and when the constraint checking is done: CREATE TABLE X1_tab (a1 NUMBER CONSTRAINT y CHECK (a1>3) DEFERRABLE DISABLE); ALTER ALTER ALTER ALTER TABLE TABLE TABLE TABLE X1_tab X1_tab X1_tab X1_tab MODIFY MODIFY MODIFY MODIFY CONSTRAINT CONSTRAINT CONSTRAINT CONSTRAINT Y_cnstrt Y_cnstrt Y_cnstrt Y_cnstrt ENABLE; RELY; INITIALLY DEFERRED; ENABLE NOVALIDATE; MODIFY CONSTRAINT Example #2 The following commands show several alternatives for whether the NOT NULL constraint is enforced, and when the checking is done: Maintaining Data Integrity in Application Development 6-17 Altering Integrity Constraints CREATE TABLE X1_tab (A1 NUMBER CONSTRAINT Y_cnstrt NOT NULL DEFERRABLE INITIALLY DEFERRED NORELY DISABLE); ALTER TABLE X1_tab ADD CONSTRAINT One_cnstrt UNIQUE(A1) DEFERRABLE INITIALLY IMMEDIATE RELY USING INDEX PCTFREE = 30 ENABLE VALIDATE; ALTER TABLE X1_tab MODIFY UNIQUE(A1) INITIALLY DEFERRED NORELY USING INDEX PCTFREE = 40 ENABLE NOVALIDATE; ALTER TABLE X1_tab MODIFY CONSTRAINT One_cnstrt INITIALLY IMMEDIATE RELY; Modify Constraint Example #3 The following commands show several alternatives for whether the primary key constraint is enforced, and when the checking is done: CREATE TABLE T1_tab (A1 INT, B1 INT); ALTER TABLE T1_tab add CONSTRAINT P1_cnstrt PRIMARY KEY(a1) DISABLE; ALTER TABLE T1_tab MODIFY PRIMARY KEY INITIALLY IMMEDIATE USING INDEX PCTFREE = 30 ENABLE NOVALIDATE; ALTER TABLE T1_tab MODIFY PRIMARY KEY USING INDEX PCTFREE = 35 ENABLE; ALTER TABLE T1_tab MODIFY PRIMARY KEY ENABLE NOVALIDATE; Renaming Integrity Constraints Because constraint names must be unique, even across multiple schemas, you can encounter problems when you want to clone a table and all its constraints, because the constraint name for the new table conflicts with the one for the original table. Or, you might create a constraint with a default system-generated name, and later realize that you want to give the constraint a name that is easy to remember, so that you can easily enable and disable it. One of the properties you can alter for a constraint is its name. The following SQL*Plus script finds the system-generated name for a constraint and changes it: prompt Enter table name to find its primary key: accept table_name select constraint_name from user_constraints where table_name = upper('&table_name.') and constraint_type = 'P'; prompt Enter new name for its primary key: accept new_constraint set serveroutput on declare -- USER_CONSTRAINTS.CONSTRAINT_NAME is declared as VARCHAR2(30). -- Using %TYPE here protects us if the length changes in a future release. constraint_name user_constraints.constraint_name%type; begin select constraint_name into constraint_name from user_constraints where table_name = upper('&table_name.') and constraint_type = 'P'; 6-18 Oracle Database Application Developer’s Guide - Fundamentals Managing FOREIGN KEY Integrity Constraints dbms_output.put_line('The primary key for ' || upper('&table_name.') || ' is: ' || constraint_name); execute immediate 'alter table &table_name. rename constraint ' || constraint_name || ' to &new_constraint.'; end; / Dropping Integrity Constraints Drop an integrity constraint if the rule that it enforces is no longer true or if the constraint is no longer needed. Drop an integrity constraint using the ALTER TABLE command and the DROP clause. For example, the following statements drop integrity constraints: ALTER TABLE Dept_tab DROP UNIQUE (Dname); ALTER TABLE Dept_tab DROP UNIQUE (Loc); ALTER TABLE Emp_tab DROP PRIMARY KEY, DROP CONSTRAINT Dept_fkey; DROP TABLE Emp_tab CASCADE CONSTRAINTS; When dropping UNIQUE, PRIMARY KEY, and FOREIGN KEY integrity constraints, you should be aware of several important issues and prerequisites. UNIQUE and PRIMARY KEY constraints are usually managed by the database administrator. See Also: Oracle Database Administrator's Guide and "Managing FOREIGN KEY Integrity Constraints" on page 6-19 Managing FOREIGN KEY Integrity Constraints General information about defining, enabling, disabling, and dropping all types of integrity constraints is given in section "Dropping Integrity Constraints". The present section supplements this information, focusing specifically on issues regarding FOREIGN KEY integrity constraints, which enforce relationships between columns in different tables. FOREIGN KEY integrity constraints cannot be enabled if the constraint of the referenced primary or unique key is not present or not enabled. Note: Datatypes and Names for Foreign Key Columns You must use the same datatype for corresponding columns in the dependent and referenced tables. The column names do not need to match. Limit on Columns in Composite Foreign Keys Because foreign keys reference primary and unique keys of the parent table, and PRIMARY KEY and UNIQUE key constraints are enforced using indexes, composite foreign keys are limited to 32 columns. Maintaining Data Integrity in Application Development 6-19 Managing FOREIGN KEY Integrity Constraints Foreign Key References Primary Key by Default If the column list is not included in the REFERENCES option when defining a FOREIGN KEY constraint (single column or composite), then Oracle Database assumes that you intend to reference the primary key of the specified table. Alternatively, you can explicitly specify the column(s) to reference in the parent table within parentheses. Oracle Database automatically checks to verify that this column list references a primary or unique key of the parent table. If it does not, then an informative error is returned. Privileges Required to Create FOREIGN KEY Integrity Constraints To create a FOREIGN KEY constraint, the creator of the constraint must have privileged access to the parent and child tables. ■ ■ The Parent Table The creator of the referential integrity constraint must own the parent table or have REFERENCES object privileges on the columns that constitute the parent key of the parent table. The Child Table The creator of the referential integrity constraint must have the ability to create tables (that is, the CREATE TABLE or CREATE ANY TABLE system privilege) or the ability to alter the child table (that is, the ALTER object privilege for the child table or the ALTER ANY TABLE system privilege). In both cases, necessary privileges cannot be obtained through a role; they must be explicitly granted to the creator of the constraint. These restrictions allow: ■ ■ The owner of the child table to explicitly decide which constraints are enforced and which other users can create constraints The owner of the parent table to explicitly decide if foreign keys can depend on the primary and unique keys in her tables Choosing How Foreign Keys Enforce Referential Integrity Oracle Database allows different types of referential integrity actions to be enforced, as specified with the definition of a FOREIGN KEY constraint: ■ Prevent Delete or Update of Parent Key The default setting prevents the deletion or update of a parent key if there is a row in the child table that references the key. For example: CREATE TABLE Emp_tab ( FOREIGN KEY (Deptno) REFERENCES Dept_tab); ■ Delete Child Rows When Parent Key Deleted The ON DELETE CASCADE action allows parent key data that is referenced from the child table to be deleted, but not updated. When data in the parent key is deleted, all rows in the child table that depend on the deleted parent key values are also deleted. To specify this referential action, include the ON DELETE CASCADE option in the definition of the FOREIGN KEY constraint. For example: CREATE TABLE Emp_tab ( FOREIGN KEY (Deptno) REFERENCES Dept_tab ON DELETE CASCADE); ■ Set Foreign Keys to Null When Parent Key Deleted The ON DELETE SET NULL action allows data that references the parent key to be deleted, but not updated. When referenced data in the parent key is deleted, all rows in the child table that 6-20 Oracle Database Application Developer’s Guide - Fundamentals Viewing Definitions of Integrity Constraints depend on those parent key values have their foreign keys set to null. To specify this referential action, include the ON DELETE SET NULL option in the definition of the FOREIGN KEY constraint. For example: CREATE TABLE Emp_tab ( FOREIGN KEY (Deptno) REFERENCES Dept_tab ON DELETE SET NULL); Viewing Definitions of Integrity Constraints The data dictionary contains the following views that relate to integrity constraints: ■ ALL_CONSTRAINTS ■ ALL_CONS_COLUMNS ■ USER_CONSTRAINTS ■ USER_CONS_COLUMNS ■ DBA_CONSTRAINTS ■ DBA_CONS_COLUMNS You can query these views to find the names of constraints, what columns they affect, and other information to help you manage constraints. See Also: Oracle Database Reference for information on each view Examples of Defining Integrity Constraints The following CREATE TABLE statements define a number of integrity constraints: CREATE TABLE Dept_tab ( Deptno NUMBER(3) PRIMARY KEY, Dname VARCHAR2(15), Loc VARCHAR2(15), CONSTRAINT Dname_ukey UNIQUE (Dname, Loc), CONSTRAINT LOC_CHECK1 CHECK (Loc IN ('NEW YORK', 'BOSTON', 'CHICAGO'))); CREATE TABLE Emp_tab ( Empno NUMBER(5) PRIMARY KEY, Ename VARCHAR2(15) NOT NULL, Job VARCHAR2(10), Mgr NUMBER(5) CONSTRAINT Mgr_fkey REFERENCES Emp_tab ON DELETE CASCADE, Hiredate DATE, Sal NUMBER(7,2), Comm NUMBER(5,2), Deptno NUMBER(3) NOT NULL CONSTRAINT Dept_fkey REFERENCES Dept_tab); Example 1: Listing All of Your Accessible Constraints The following query lists all constraints defined on all tables accessible to the user: SELECT Constraint_name, Constraint_type, Table_name, R_constraint_name Maintaining Data Integrity in Application Development 6-21 Viewing Definitions of Integrity Constraints FROM User_constraints; Considering the example statements at the beginning of this section, a list similar to this is returned: CONSTRAINT_NAME --------------SYS_C00275 DNAME_UKEY LOC_CHECK1 SYS_C00278 SYS_C00279 SYS_C00280 MGR_FKEY DEPT_FKEY C P U C C C P R R TABLE_NAME ---------DEPT_TAB DEPT_TAB DEPT_TAB EMP_TAB EMP_TAB EMP_TAB EMP_TAB EMP_TAB R_CONSTRAINT_NAME ----------------- SYS_C00280 SYS_C00275 Notice the following: ■ ■ Some constraint names are user specified (such as DNAME_UKEY), while others are system specified (such as SYS_C00275). Each constraint type is denoted with a different character in the CONSTRAINT_ TYPE column. The following table summarizes the characters used for each constraint type. Constraint Type Character PRIMARY KEY P UNIQUE KEY U FOREIGN KEY R CHECK, NOT NULL C An additional constraint type is indicated by the character "V" in the CONSTRAINT_TYPE column. This constraint type corresponds to constraints created using the WITH CHECK OPTION for views. Note: Example 2: Distinguishing NOT NULL Constraints from CHECK Constraints In the previous example, several constraints are listed with a constraint type of C. To distinguish which constraints are NOT NULL constraints and which are CHECK constraints in the EMP_TAB and DEPT_TAB tables, submit the following query: SELECT Constraint_name, Search_condition FROM User_constraints WHERE (Table_name = 'DEPT_TAB' OR Table_name = 'EMP_TAB') AND Constraint_type = 'C'; Considering the example CREATE TABLE statements at the beginning of this section, a list similar to this is returned: CONSTRAINT_NAME --------------LOC_CHECK1 SYS_C00278 SYS_C00279 SEARCH_CONDITION ---------------------------------------loc IN ('NEW YORK', 'BOSTON', 'CHICAGO') ENAME IS NOT NULL DEPTNO IS NOT NULL Notice that the following are explicitly listed in the SEARCH_CONDITION column: 6-22 Oracle Database Application Developer’s Guide - Fundamentals Viewing Definitions of Integrity Constraints ■ NOT NULL constraints ■ The conditions for user-defined CHECK constraints The following query lists all columns that constitute the constraints defined on all tables accessible to you, the user: Example 3: Listing Column Names that Constitute an Integrity Constraint SELECT Constraint_name, Table_name, Column_name FROM User_cons_columns; Considering the example statements at the beginning of this section, a list similar to this is returned: CONSTRAINT_NAME --------------DEPT_FKEY DNAME_UKEY DNAME_UKEY LOC_CHECK1 MGR_FKEY SYS_C00275 SYS_C00278 SYS_C00279 SYS_C00280 TABLE_NAME ----------EMP_TAB DEPT_TAB DEPT_TAB DEPT_TAB EMP_TAB DEPT_TAB EMP_TAB EMP_TAB EMP_TAB COLUMN_NAME --------------DEPTNO DNAME LOC LOC MGR DEPTNO ENAME DEPTNO EMPNO Maintaining Data Integrity in Application Development 6-23 Viewing Definitions of Integrity Constraints 6-24 Oracle Database Application Developer’s Guide - Fundamentals Part II PL/SQL for Application Developers This part contains the following chapters: ■ Chapter 7, "Coding PL/SQL Procedures and Packages" ■ Chapter 8, "Coding Dynamic SQL" ■ Chapter 9, "Coding Triggers" ■ Chapter 10, "Developing Flashback Applications" ■ Chapter 11, "Developing Applications with the PL/SQL Web Toolkit" ■ Chapter 12, "Developing PL/SQL Server Pages" ■ Chapter 13, "Developing Applications with Database Change Notification" 7 Coding PL/SQL Procedures and Packages This chapter describes some of the procedural capabilities of Oracle Database for application development, including: ■ Overview of PL/SQL Program Units ■ Compiling PL/SQL Procedures for Native Execution ■ Remote Dependencies ■ Cursor Variables ■ Handling PL/SQL Compile-Time Errors ■ Handling Run-Time PL/SQL Errors ■ Debugging Stored Procedures ■ Calling Stored Procedures ■ Calling Remote Procedures ■ Calling Stored Functions from SQL Expressions ■ Returning Large Amounts of Data from a Function ■ Coding Your Own Aggregate Functions Overview of PL/SQL Program Units PL/SQL is a modern, block-structured programming language. It provides several features that make developing powerful database applications very convenient. For example, PL/SQL provides procedural constructs, such as loops and conditional statements, that are not available in standard SQL. You can directly enter SQL data manipulation language (DML) statements inside PL/SQL blocks, and you can use procedures supplied by Oracle to perform data definition language (DDL) statements. PL/SQL code runs on the server, so using PL/SQL lets you centralize significant parts of your database applications for increased maintainability and security. It also enables you to achieve a significant reduction of network overhead in client/server applications. Some Oracle tools, such as Oracle Forms, contain a PL/SQL engine that lets you run PL/SQL locally. Note: Coding PL/SQL Procedures and Packages 7-1 Overview of PL/SQL Program Units You can even use PL/SQL for some database applications in place of 3GL programs that use embedded SQL or Oracle Call Interface (OCI). PL/SQL program units include: ■ Anonymous Blocks ■ Stored Program Units (Procedures, Functions, and Packages) ■ Triggers See Also: ■ ■ Oracle Database PL/SQL User's Guide and Reference for syntax and examples of operations on PL/SQL packages Oracle Database PL/SQL Packages and Types Reference for information about the PL/SQL packages that come with Oracle Database Anonymous Blocks An anonymous block is a PL/SQL program unit that has no name. An anonymous block consists of an optional declarative part, an executable part, and one or more optional exception handlers. The declarative part declares PL/SQL variables, exceptions, and cursors. The executable part contains PL/SQL code and SQL statements, and can contain nested blocks. Exception handlers contain code that is called when the exception is raised, either as a predefined PL/SQL exception (such as NO_DATA_FOUND or ZERO_DIVIDE) or as an exception that you define. The following short example of a PL/SQL anonymous block prints the names of all employees in department 20 in the hr.employees table by using the DBMS_OUTPUT package: DECLARE Last_name Cursor VARCHAR2(10); c1 IS SELECT last_name FROM employees WHERE department_id = 20; BEGIN OPEN c1; LOOP FETCH c1 INTO Last_name; EXIT WHEN c1%NOTFOUND; DBMS_OUTPUT.PUT_LINE(Last_name); END LOOP; END; / If you test this block using SQL*Plus, then enter the statement SET SERVEROUTPUT ON so that output using the DBMS_OUTPUT procedures (for example, PUT_LINE) is activated. Also, end the example with a slash (/) to activate it. Note: Exceptions let you handle Oracle Database error conditions within PL/SQL program logic. This allows your application to prevent the server from issuing an error that could cause the client application to end. The following anonymous block handles the 7-2 Oracle Database Application Developer’s Guide - Fundamentals Overview of PL/SQL Program Units predefined Oracle Database exception NO_DATA_FOUND (which would result in an ORA-01403 error if not handled): DECLARE Emp_number INTEGER := 9999; Emp_name VARCHAR2(10); BEGIN SELECT Ename INTO Emp_name FROM Emp_tab WHERE Empno = Emp_number; -- no such number DBMS_OUTPUT.PUT_LINE('Employee name is ' || Emp_name); EXCEPTION WHEN NO_DATA_FOUND THEN DBMS_OUTPUT.PUT_LINE('No such employee: ' || Emp_number); END; You can also define your own exceptions, declare them in the declaration part of a block, and define them in the exception part of the block. An example follows: DECLARE Emp_name VARCHAR2(10); Emp_number INTEGER; Empno_out_of_range EXCEPTION; BEGIN Emp_number := 10001; IF Emp_number > 9999 OR Emp_number < 1000 THEN RAISE Empno_out_of_range; ELSE SELECT Ename INTO Emp_name FROM Emp_tab WHERE Empno = Emp_number; DBMS_OUTPUT.PUT_LINE('Employee name is ' || Emp_name); END IF; EXCEPTION WHEN Empno_out_of_range THEN DBMS_OUTPUT.PUT_LINE('Employee number ' || Emp_number || ' is out of range.'); END; Anonymous blocks are usually used interactively from a tool, such as SQL*Plus, or in a precompiler, OCI, or SQL*Module application. They are usually used to call stored procedures or to open cursor variables. See Also: ■ ■ ■ Oracle Database PL/SQL Packages and Types Reference for complete information about the DBMS_OUTPUT package Oracle Database PL/SQL User's Guide and Reference and "Handling Run-Time PL/SQL Errors" on page 7-26 "Cursor Variables" on page 7-22 Stored Program Units (Procedures, Functions, and Packages) A stored procedure, function, or package is a PL/SQL program unit that: ■ Has a name. ■ Can take parameters, and can return values. ■ Is stored in the data dictionary. ■ Can be called by many users. Coding PL/SQL Procedures and Packages 7-3 Overview of PL/SQL Program Units The term stored procedure is sometimes used generically for both stored procedures and stored functions. The only difference between procedures and functions is that functions always return a single value to the caller, while procedures do not return a value to the caller. Note: Naming Procedures and Functions Because a procedure or function is stored in the database, it must be named. This distinguishes it from other stored procedures and makes it possible for applications to call it. Each publicly-visible procedure or function in a schema must have a unique name, and the name must be a legal PL/SQL identifier. If you plan to call a stored procedure using a stub generated by SQL*Module, then the stored procedure name must also be a legal identifier in the calling host 3GL language, such as Ada or C. Note: Parameters for Procedures and Functions Stored procedures and functions can take parameters. The following example shows a stored procedure that is similar to the anonymous block in "Anonymous Blocks" on page 7-2. To execute the following, use CREATE OR REPLACE PROCEDURE... Caution: PROCEDURE Get_emp_names (Dept_num IN NUMBER) IS Emp_name VARCHAR2(10); CURSOR c1 (Depno NUMBER) IS SELECT Ename FROM Emp_tab WHERE deptno = Depno; BEGIN OPEN c1(Dept_num); LOOP FETCH c1 INTO Emp_name; EXIT WHEN C1%NOTFOUND; DBMS_OUTPUT.PUT_LINE(Emp_name); END LOOP; CLOSE c1; END; In this stored procedure example, the department number is an input parameter which is used when the parameterized cursor c1 is opened. The formal parameters of a procedure have three major attributes, described in Table 7–1. Table 7–1 Attributes of Procedure Parameters Parameter Attribute Description Name This must be a legal PL/SQL identifier. Mode This indicates whether the parameter is an input-only parameter (IN), an output-only parameter (OUT), or is both an input and an output parameter (IN OUT). If the mode is not specified, then IN is assumed. 7-4 Oracle Database Application Developer’s Guide - Fundamentals Overview of PL/SQL Program Units Table 7–1 (Cont.) Attributes of Procedure Parameters Parameter Attribute Description Datatype This is a standard PL/SQL datatype. Parameter Modes Parameter modes define the behavior of formal parameters. The three parameter modes, IN (the default), OUT, and IN OUT, can be used with any subprogram. However, avoid using the OUT and IN OUT modes with functions. The purpose of a function is to take no arguments and return a single value. It is poor programming practice to have a function return multiple values. Also, functions should be free from side effects, which change the values of variables not local to the subprogram. Table 7–2 summarizes the information about parameter modes. Table 7–2 Parameter Modes IN OUT IN OUT The default. Must be specified. Must be specified. Passes values to a subprogram. Returns values to the caller. Passes initial values to a subprogram; returns updated values to the caller. Formal parameter acts like a Formal parameter acts like an constant. uninitialized variable. Formal parameter acts like an initialized variable. Formal parameter cannot be Formal parameter cannot be assigned a value. used in an expression; must be assigned a value. Formal parameter should be assigned a value. Actual parameter can be a Actual parameter must be a constant, initialized variable, variable. literal, or expression. Actual parameter must be a variable. See Also: Oracle Database PL/SQL User's Guide and Reference for details about parameter modes Parameter Datatypes The datatype of a formal parameter consists of one of the following: ■ An unconstrained type name, such as NUMBER or VARCHAR2. ■ A type that is constrained using the %TYPE or %ROWTYPE attributes. Note: Numerically constrained types such as NUMBER(2) or VARCHAR2(20) are not allowed in a parameter list. %TYPE and %ROWTYPE Attributes Use the type attributes %TYPE and %ROWTYPE to constrain the parameter. For example, the Get_emp_names procedure specification in "Parameters for Procedures and Functions" on page 7-4 could be written as the following: PROCEDURE Get_emp_names(Dept_num IN Emp_tab.Deptno%TYPE) This has the Dept_num parameter take the same datatype as the Deptno column in the Emp_tab table. The column and table must be available when a declaration using %TYPE (or %ROWTYPE) is elaborated. Coding PL/SQL Procedures and Packages 7-5 Overview of PL/SQL Program Units Using %TYPE is recommended, because if the type of the column in the table changes, then it is not necessary to change the application code. If the Get_emp_names procedure is part of a package, then you can use previously-declared public (package) variables to constrain a parameter datatype. For example: Dept_number number(2); ... PROCEDURE Get_emp_names(Dept_num IN Dept_number%TYPE); Use the %ROWTYPE attribute to create a record that contains all the columns of the specified table. The following example defines the Get_emp_rec procedure, which returns all the columns of the Emp_tab table in a PL/SQL record for the given empno: To execute the following, use CREATE OR REPLACE PROCEDURE... Caution: PROCEDURE Get_emp_rec (Emp_number IN Emp_tab.Empno%TYPE, Emp_ret OUT Emp_tab%ROWTYPE) IS BEGIN SELECT Empno, Ename, Job, Mgr, Hiredate, Sal, Comm, Deptno INTO Emp_ret FROM Emp_tab WHERE Empno = Emp_number; END; You could call this procedure from a PL/SQL block as follows: DECLARE Emp_row Emp_tab%ROWTYPE; BEGIN Get_emp_rec(7499, Emp_row); DBMS_OUTPUT.PUT(Emp_row.Ename DBMS_OUTPUT.PUT(' ' DBMS_OUTPUT.PUT(' ' DBMS_OUTPUT.PUT(' ' DBMS_OUTPUT.NEW_LINE; END; -- declare a record matching a -- row in the Emp_tab table -|| || || || call for Emp_tab# 7499 ' ' || Emp_row.Empno); Emp_row.Job || ' ' || Emp_row.Mgr); Emp_row.Hiredate || ' ' || Emp_row.Sal); Emp_row.Comm || ' '|| Emp_row.Deptno); Stored functions can also return values that are declared using %ROWTYPE. For example: FUNCTION Get_emp_rec (Dept_num IN Emp_tab.Deptno%TYPE) RETURN Emp_tab%ROWTYPE IS ... Tables and Records You can pass PL/SQL tables as parameters to stored procedures and functions. You can also pass tables of records as parameters. When passing a user defined type, such as a PL/SQL table or record to a remote procedure, to make PL/SQL use the same definition so that the type checker can verify the source, you must create a redundant loop back DBLINK so that when the PL/SQL compiles, both sources pull from the same location. Note: 7-6 Oracle Database Application Developer’s Guide - Fundamentals Overview of PL/SQL Program Units Default Parameter Values Parameters can take default values. Use the DEFAULT keyword or the assignment operator to give a parameter a default value. For example, the specification for the Get_emp_names procedure could be written as the following: PROCEDURE Get_emp_names (Dept_num IN NUMBER DEFAULT 20) IS ... or PROCEDURE Get_emp_names (Dept_num IN NUMBER := 20) IS ... When a parameter takes a default value, it can be omitted from the actual parameter list when you call the procedure. When you do specify the parameter value on the call, it overrides the default value. Unlike in an anonymous PL/SQL block, you do not use the keyword DECLARE before the declarations of variables, cursors, and exceptions in a stored procedure. In fact, it is an error to use it. Note: Creating Stored Procedures and Functions Use a text editor to write the procedure or function. At the beginning of the procedure, place the following statement: CREATE PROCEDURE Procedure_name AS ... For example, to use the example in "%TYPE and %ROWTYPE Attributes" on page 7-5, create a text (source) file called get_emp.sql containing the following code: CREATE PROCEDURE Get_emp_rec (Emp_number IN Emp_tab.Empno%TYPE, Emp_ret OUT Emp_tab%ROWTYPE) AS BEGIN SELECT Empno, Ename, Job, Mgr, Hiredate, Sal, Comm, Deptno INTO Emp_ret FROM Emp_tab WHERE Empno = Emp_number; END; / Then, using an interactive tool such as SQL*Plus, load the text file containing the procedure by entering the following statement: SQL> @get_emp This loads the procedure into the current schema from the get_emp.sql file (.sql is the default file extension). Note the slash (/) at the end of the code. This is not part of the code; it just activates the loading of the procedure. Use the CREATE [OR REPLACE] FUNCTION... statement to store functions. Caution: When developing a new procedure, it is usually much more convenient to use the CREATE OR REPLACE PROCEDURE statement. This replaces any previous version of that procedure in the same schema with the newer version, but note that this is done without warning. You can use either the keyword IS or AS after the procedure parameter list. Coding PL/SQL Procedures and Packages 7-7 Overview of PL/SQL Program Units Oracle Database Reference for the complete syntax of the CREATE PROCEDURE and CREATE FUNCTION statements See Also: Privileges to Create Procedures and Functions To create a standalone procedure or function, or package specification or body, you must meet the following prerequisites: ■ You must have the CREATE PROCEDURE system privilege to create a procedure or package in your schema, or the CREATE ANY PROCEDURE system privilege to create a procedure or package in another user's schema. To create without errors (to compile the procedure or package successfully) requires the following additional privileges: Note: ■ ■ The owner of the procedure or package must be explicitly granted the necessary object privileges for all objects referenced within the body of the code. The owner cannot obtain required privileges through roles. If the privileges of the owner of a procedure or package change, then the procedure must be reauthenticated before it is run. If a necessary privilege to a referenced object is revoked from the owner of the procedure or package, then the procedure cannot be run. The EXECUTE privilege on a procedure gives a user the right to run a procedure owned by another user. Privileged users run the procedure under the security domain of the owner of the procedure. Therefore, users never need to be granted the privileges to the objects referenced by a procedure. This allows for more disciplined and efficient security strategies with database applications and their users. Furthermore, all procedures and packages are stored in the data dictionary (in the SYSTEM tablespace). No quota controls the amount of space available to a user who creates procedures and packages. Package creation requires a sort. So the user creating the package should be able to create a sort segment in the temporary tablespace with which the user is associated. Note: See Also: "Privileges Required to Execute a Procedure" on page 7-33 Altering Stored Procedures and Functions To alter a stored procedure or function, you must first drop it using the DROP PROCEDURE or DROP FUNCTION statement, then re-create it using the CREATE PROCEDURE or CREATE FUNCTION statement. Alternatively, use the CREATE OR REPLACE PROCEDURE or CREATE OR REPLACE FUNCTION statement, which first drops the procedure or function if it exists, then re-creates it as specified. Caution: The procedure or function is dropped without any warning. Dropping Procedures and Functions A standalone procedure, a standalone function, a package body, or an entire package can be dropped using the SQL statements DROP PROCEDURE, DROP FUNCTION, DROP 7-8 Oracle Database Application Developer’s Guide - Fundamentals Overview of PL/SQL Program Units PACKAGE BODY, and DROP PACKAGE, respectively. A DROP PACKAGE statement drops both the specification and body of a package. The following statement drops the Old_sal_raise procedure in your schema: DROP PROCEDURE Old_sal_raise; Privileges to Drop Procedures and Functions To drop a procedure, function, or package, the procedure or package must be in your schema, or you must have the DROP ANY PROCEDURE privilege. An individual procedure within a package cannot be dropped; the containing package specification and body must be re-created without the procedures to be dropped. External Procedures A PL/SQL procedure executing on an Oracle Database instance can call an external procedure written in a 3GL. The 3GL procedure runs in a separate address space from that of the database. Chapter 14, "Calling External Procedures" for information about external procedures See Also: PL/SQL Packages A package is an encapsulated collection of related program objects (for example, procedures, functions, variables, constants, cursors, and exceptions) stored together in the database. Using packages is an alternative to creating procedures and functions as standalone schema objects. Packages have many advantages over standalone procedures and functions. For example, they: ■ Let you organize your application development more efficiently. ■ Let you grant privileges more efficiently. ■ Let you modify package objects without recompiling dependent schema objects. ■ Enable Oracle Database to read multiple package objects into memory at once. ■ ■ Can contain global variables and cursors that are available to all procedures and functions in the package. Let you overload procedures or functions. Overloading a procedure means creating multiple procedures with the same name in the same package, each taking arguments of different number or datatype. Oracle Database PL/SQL User's Guide and Reference for more information about subprogram name overloading See Also: The specification part of a package declares the public types, variables, constants, and subprograms that are visible outside the immediate scope of the package. The body of a package defines the objects declared in the specification, as well as private objects that are not visible to applications outside the package. Example of a PL/SQL Package Specification and Body The following example shows a package specification for a package named Employee_management. The package contains one stored function and two stored procedures. The body for this package defines the function and the procedures: CREATE PACKAGE BODY Employee_management AS FUNCTION Hire_emp (Name VARCHAR2, Job VARCHAR2, Coding PL/SQL Procedures and Packages 7-9 Overview of PL/SQL Program Units Mgr NUMBER, Hiredate DATE, Sal NUMBER, Comm NUMBER, Deptno NUMBER) RETURN NUMBER IS New_empno NUMBER(10); ------ This function accepts all arguments for the fields in the employee table except for the employee number. A value for this field is supplied by a sequence. The function returns the sequence number generated by the call to this function. BEGIN SELECT Emp_sequence.NEXTVAL INTO New_empno FROM dual; INSERT INTO Emp_tab VALUES (New_empno, Name, Job, Mgr, Hiredate, Sal, Comm, Deptno); RETURN (New_empno); END Hire_emp; PROCEDURE fire_emp(emp_id IN NUMBER) AS -- This procedure deletes the employee with an employee -- number that corresponds to the argument Emp_id. If -- no employee is found, then an exception is raised. BEGIN DELETE FROM Emp_tab WHERE Empno = Emp_id; IF SQL%NOTFOUND THEN Raise_application_error(-20011, 'Invalid Employee Number: ' || TO_CHAR(Emp_id)); END IF; END fire_emp; PROCEDURE Sal_raise (Emp_id IN NUMBER, Sal_incr IN NUMBER) AS ------ This procedure accepts two arguments. Emp_id is a number that corresponds to an employee number. SAL_INCR is the amount by which to increase the employee's salary. If employee exists, then update salary with increase. BEGIN UPDATE Emp_tab SET Sal = Sal + Sal_incr WHERE Empno = Emp_id; IF SQL%NOTFOUND THEN Raise_application_error(-20011, 'Invalid Employee Number: ' || TO_CHAR(Emp_id)); END IF; END Sal_raise; END Employee_management; If you want to try this example, then first create the sequence number Emp_sequence. Do this with the following SQL*Plus statement: Note: SQL> CREATE SEQUENCE Emp_sequence > START WITH 8000 INCREMENT BY 10; 7-10 Oracle Database Application Developer’s Guide - Fundamentals Overview of PL/SQL Program Units PL/SQL Object Size Limitation The size limitation for PL/SQL stored database objects such as procedures, functions, triggers, and packages is the size of the DIANA (Descriptive Intermediate Attributed Notation for Ada) code in the shared pool in bytes. The UNIX limit on the size of the flattened DIANA/pcode size is 64K but the limit may be 32K on desktop platforms. The most closely related number that a user can access is the PARSED_SIZE in the data dictionary view USER_OBJECT_SIZE. That gives the size of the DIANA in bytes as stored in the SYS.IDL_xxx$ tables. This is not the size in the shared pool. The size of the DIANA part of PL/SQL code (used during compilation) is significantly larger in the shared pool than it is in the system table. Creating Packages Each part of a package is created with a different statement. Create the package specification using the CREATE PACKAGE statement. The CREATE PACKAGE statement declares public package objects. To create a package body, use the CREATE PACKAGE BODY statement. The CREATE PACKAGE BODY statement defines the procedural code of the public procedures and functions declared in the package specification. You can also define private, or local, package procedures, functions, and variables in a package body. These objects can only be accessed by other procedures and functions in the body of the same package. They are not visible to external users, regardless of the privileges they hold. It is often more convenient to add the OR REPLACE clause in the CREATE PACKAGE or CREATE PACKAGE BODY statements when you are first developing your application. The effect of this option is to drop the package or the package body without warning. The CREATE statements would then be the following: CREATE OR REPLACE PACKAGE Package_name AS ... and CREATE OR REPLACE PACKAGE BODY Package_name AS ... Creating Packaged Objects The body of a package can contain: ■ Procedures and functions declared in the package specification. ■ Definitions of cursors declared in the package specification. ■ Local procedures and functions, not declared in the package specification. ■ Local variables. Procedures, functions, cursors, and variables that are declared in the package specification are global. They can be called, or used, by external users that have EXECUTE permission for the package or that have EXECUTE ANY PROCEDURE privileges. When you create the package body, make sure that each procedure that you define in the body has the same parameters, by name, datatype, and mode, as the declaration in the package specification. For functions in the package body, the parameters and the return type must agree in name and type. Privileges to Create or Drop Packages The privileges required to create or drop a package specification or package body are the same as those required to create or drop a standalone procedure or function. Coding PL/SQL Procedures and Packages 7-11 Overview of PL/SQL Program Units See Also: ■ "Privileges to Create Procedures and Functions" on page 7-8 ■ "Privileges to Drop Procedures and Functions" on page 7-9 Naming Packages and Package Objects The names of a package and all public objects in the package must be unique within a given schema. The package specification and its body must have the same name. All package constructs must have unique names within the scope of the package, unless overloading of procedure names is desired. Package Invalidations and Session State Each session that references a package object has its own instance of the corresponding package, including persistent state for any public and private variables, cursors, and constants. If any of the session's instantiated packages (specification or body) are invalidated, then all package instances in the session are invalidated and recompiled. As a result, the session state is lost for all package instances in the session. When a package in a given session is invalidated, the session receives the following error the first time it attempts to use any object of the invalid package instance: ORA-04068: existing state of packages has been discarded The second time a session makes such a package call, the package is reinstantiated for the session without error. For optimal performance, Oracle Database returns this error message only once—each time the package state is discarded. Note: If you handle this error in your application, ensure that your error handling strategy can accurately handle this error. For example, when a procedure in one package calls a procedure in another package, your application should be aware that the session state is lost for both packages. In most production environments, DDL operations that can cause invalidations are usually performed during inactive working hours; therefore, this situation might not be a problem for end-user applications. However, if package invalidations are common in your system during working hours, then you might want to code your applications to handle this error when package calls are made. Packages Supplied With Oracle Database There are many packages provided with Oracle Database, either to extend the functionality of the database or to give PL/SQL access to SQL features. You can call these packages from your application. See Also: Oracle Database PL/SQL Packages and Types Reference for an overview of these Oracle Database packages Overview of Bulk Binds Oracle Database uses two engines to run PL/SQL blocks and subprograms. The PL/SQL engine runs procedural statements, while the SQL engine runs SQL statements. During execution, every SQL statement causes a context switch between the two engines, resulting in performance overhead. 7-12 Oracle Database Application Developer’s Guide - Fundamentals Overview of PL/SQL Program Units Performance can be improved substantially by minimizing the number of context switches required to run a particular block or subprogram. When a SQL statement runs inside a loop that uses collection elements as bind variables, the large number of context switches required by the block can cause poor performance. Collections include the following: ■ Varrays ■ Nested tables ■ Index-by tables ■ Host arrays Binding is the assignment of values to PL/SQL variables in SQL statements. Bulk binding is binding an entire collection at once. Bulk binds pass the entire collection back and forth between the two engines in a single operation. Typically, using bulk binds improves performance for SQL statements that affect four or more database rows. The more rows affected by a SQL statement, the greater the performance gain from bulk binds. This section provides an overview of bulk binds to help you decide if you should use them in your PL/SQL applications. For detailed information about using bulk binds, including ways to handle exceptions that occur in the middle of a bulk bind operation, refer to the Oracle Database PL/SQL User's Guide and Reference. Note: When to Use Bulk Binds If you have scenarios like these in your applications, consider using bulk binds to improve performance. DML Statements that Reference Collections The FORALL keyword can improve the performance of INSERT, UPDATE, or DELETE statements that reference collection elements. For example, the following PL/SQL block increases the salary for employees whose manager's ID number is 7902, 7698, or 7839, both with and without using bulk binds: DECLARE TYPE Numlist IS VARRAY (100) OF NUMBER; Id NUMLIST := NUMLIST(7902, 7698, 7839); BEGIN -- Efficient method, using a bulk bind FORALL i IN Id.FIRST..Id.LAST -- bulk-bind the VARRAY UPDATE Emp_tab SET Sal = 1.1 * Sal WHERE Mgr = Id(i); -- Slower method, running the UPDATE statements within a regular loop FOR i IN Id.FIRST..Id.LAST LOOP UPDATE Emp_tab SET Sal = 1.1 * Sal WHERE Mgr = Id(i); END LOOP; END; Without the bulk bind, PL/SQL sends a SQL statement to the SQL engine for each employee that is updated, leading to context switches that hurt performance. Coding PL/SQL Procedures and Packages 7-13 Overview of PL/SQL Program Units If you have a set of rows prepared in a PL/SQL table, you can bulk-insert or bulk-update the data using a loop like: FORALL i in Emp_Data.FIRST..Emp_Data.LAST INSERT INTO Emp_tab VALUES(Emp_Data(i)); SELECT Statements that Reference Collections The BULK COLLECT INTO clause can improve the performance of queries that reference collections. For example, the following PL/SQL block queries multiple values into PL/SQL tables, both with and without bulk binds: -- Find all employees whose manager's ID number is 7698. DECLARE TYPE Var_tab IS TABLE OF VARCHAR2(20) INDEX BY BINARY_INTEGER; Empno VAR_TAB; Ename VAR_TAB; Counter NUMBER; CURSOR C IS SELECT Empno, Ename FROM Emp_tab WHERE Mgr = 7698; BEGIN -- Efficient method, using a bulk bind SELECT Empno, Ename BULK COLLECT INTO Empno, Ename FROM Emp_Tab WHERE Mgr = 7698; -- Slower method, assigning each collection element within a loop. counter := 1; FOR rec IN C LOOP Empno(Counter) := rec.Empno; Ename(Counter) := rec.Ename; Counter := Counter + 1; END LOOP; END; You can use BULK COLLECT INTO with tables of scalar values, or tables of %TYPE values. Without the bulk bind, PL/SQL sends a SQL statement to the SQL engine for each employee that is selected, leading to context switches that hurt performance. FOR Loops that Reference Collections and the Returning Into Clause You can use the FORALL keyword along with the BULK COLLECT INTO keywords to improve the performance of FOR loops that reference collections and return DML. For example, the following PL/SQL block updates the Emp_tab table by computing bonuses for a collection of employees; then it returns the bonuses in a column called Bonlist. The actions are performed both with and without using bulk binds: DECLARE TYPE Emplist IS VARRAY(100) OF NUMBER; Empids EMPLIST := EMPLIST(7369, 7499, 7521, 7566, 7654, 7698); TYPE Bonlist IS TABLE OF Emp_tab.sal%TYPE; Bonlist_inst BONLIST; BEGIN Bonlist_inst := BONLIST(1,2,3,4,5); FORALL i IN Empids.FIRST..empIDs.LAST 7-14 Oracle Database Application Developer’s Guide - Fundamentals Remote Dependencies UPDATE Emp_tab SET Bonus = 0.1 * Sal WHERE Empno = Empids(i) RETURNING Sal BULK COLLECT INTO Bonlist; FOR i IN Empids.FIRST..Empids.LAST LOOP UPDATE Emp_tab Set Bonus = 0.1 * sal WHERE Empno = Empids(i) RETURNING Sal INTO BONLIST(i); END LOOP; END; Without the bulk bind, PL/SQL sends a SQL statement to the SQL engine for each employee that is updated, leading to context switches that hurt performance. Triggers A trigger is a special kind of PL/SQL anonymous block. You can define triggers to fire before or after SQL statements, either on a statement level or for each row that is affected. You can also define INSTEAD OF triggers or system triggers (triggers on DATABASE and SCHEMA). See Also: Chapter 9, "Coding Triggers" Compiling PL/SQL Procedures for Native Execution You can speed up PL/SQL procedures by compiling them into native code residing in shared libraries. The procedures are translated into C code, then compiled with your usual C compiler and linked into the Oracle Database process. You can use this technique with both the supplied Oracle Database PL/SQL packages, and procedures you write yourself. You can use the ALTER SYSTEM or ALTER SESSION command, or update your initialization file, to set the parameter PLSQL_CODE_TYPE to the value NATIVE (the default setting is the value INTERPRETED). Because this technique cannot do much to speed up SQL statements called from these procedures, it is most effective for compute-intensive procedures that do not spend much time executing SQL. With Java, you can use the ncomp tool to compile your own packages and classes. See Also: ■ ■ Oracle Database PL/SQL User's Guide and Reference for details on PL/SQL native compilation Oracle Database Java Developer's Guide for details on Java native compilation Remote Dependencies Dependencies among PL/SQL program units can be handled in two ways: ■ Timestamps ■ Signatures Coding PL/SQL Procedures and Packages 7-15 Remote Dependencies Timestamps If timestamps are used to handle dependencies among PL/SQL program units, then whenever you alter a program unit or a relevant schema object, all of its dependent units are marked as invalid and must be recompiled before they can be run. Each program unit carries a timestamp that is set by the server when the unit is created or recompiled. Figure 7–1 demonstrates this graphically. Procedures P1 and P2 call stored procedure P3. Stored procedure P3 references table T1. In this example, each of the procedures is dependent on table T1. P3 depends upon T1 directly, while P1 and P2 depend upon T1 indirectly. Figure 7–1 P1 Dependency Relationships P3 T1 P2 If P3 is altered, then P1 and P2 are marked as invalid immediately, if they are on the same server as P3. The compiled states of P1 and P2 contain records of the timestamp of P3. Therefore, if the procedure P3 is altered and recompiled, then the timestamp on P3 no longer matches the value that was recorded for P3 during the compilation of P1 and P2. If P1 and P2 are on a client system, or on another Oracle Database instance in a distributed environment, then the timestamp information is used to mark them as invalid at runtime. Disadvantages of the Timestamp Model The disadvantage of this dependency model is that it is unnecessarily restrictive. Recompilation of dependent objects across the network are often performed when not strictly necessary, leading to performance degradation. Furthermore, on the client side, the timestamp model can lead to situations that block an application from running at all, if the client-side application is built using PL/SQL version 2. Earlier releases of tools, such as Oracle Forms, that used PL/SQL version 1 on the client side did not use this dependency model, because PL/SQL version 1 had no support for stored procedures. For releases of Oracle Forms that are integrated with PL/SQL version 2 on the client side, the timestamp model can present problems. For example, during the installation of the application, the application is rendered invalid unless the client-side PL/SQL procedures that it uses are recompiled at the client site. Also, if a client-side procedure depends on a server procedure, and if the server procedure is changed or automatically recompiled, then the client-side PL/SQL procedure must then be recompiled. Yet in many application environments (such as Forms runtime applications), there is no PL/SQL compiler available on the client. This blocks the application from running at all. The client application developer must then redistribute new versions of the application to all customers. Signatures To alleviate some of the problems with the timestamp-only dependency model, Oracle Database provides the additional capability of remote dependencies using signatures. 7-16 Oracle Database Application Developer’s Guide - Fundamentals Remote Dependencies The signature capability affects only remote dependencies. Local (same server) dependencies are not affected, as recompilation is always possible in this environment. A signature is associated with each compiled stored program unit. It identifies the unit using the following criteria: ■ The name of the unit (the package, procedure, or function name). ■ The types of each of the parameters of the subprogram. ■ The modes of the parameters (IN, OUT, IN OUT). ■ The number of parameters. ■ The type of the return value for a function. The user has control over whether signatures or timestamps govern remote dependencies. See Also: "Controlling Remote Dependencies" on page 7-20 When the signature dependency model is used, a dependency on a remote program unit causes an invalidation of the dependent unit if the dependent unit contains a call to a subprogram in the parent unit, and if the signature of this subprogram has been changed in an incompatible manner. For example, consider a procedure get_emp_name stored on a server in Boston (BOSTON_SERVER). The procedure is defined as the following: You may need to set up data structures, similar to the following, for certain examples to work: Note: CONNECT system/manager CREATE PUBLIC DATABASE LINK boston_server USING 'inst1_alias'; CONNECT scott/tiger CREATE OR REPLACE PROCEDURE get_emp_name ( emp_number IN NUMBER, hire_date OUT VARCHAR2, emp_name OUT VARCHAR2) AS BEGIN SELECT ename, to_char(hiredate, 'DD-MON-YY') INTO emp_name, hire_date FROM emp WHERE empno = emp_number; END; When get_emp_name is compiled on BOSTON_SERVER, its signature, as well as its timestamp, is recorded. Suppose that on another server in California, some PL/SQL code calls get_emp_name identifying it using a DBlink called BOSTON_SERVER, as follows: CREATE OR REPLACE PROCEDURE print_ename (emp_number IN NUMBER) AS hire_date VARCHAR2(12); ename VARCHAR2(10); BEGIN get_emp_name@BOSTON_SERVER(emp_number, hire_date, ename); dbms_output.put_line(ename); dbms_output.put_line(hire_date); END; Coding PL/SQL Procedures and Packages 7-17 Remote Dependencies When this California server code is compiled, the following actions take place: ■ A connection is made to the Boston server. ■ The signature of get_emp_name is transferred to the California server. ■ The signature is recorded in the compiled state of print_ename. At runtime, during the remote procedure call from the California server to the Boston server, the recorded signature of get_emp_name that was saved in the compiled state of print_ename gets sent to the Boston server, regardless of whether or not there were any changes. If the timestamp dependency mode is in effect, then a mismatch in timestamps causes an error status to be returned to the calling procedure. However, if the signature mode is in effect, then any mismatch in timestamps is ignored, and the recorded signature of get_emp_name in the compiled state of Print_ename on the California server is compared with the current signature of get_emp_name on the Boston server. If they match, then the call succeeds. If they do not match, then an error status is returned to the print_name procedure. Note that the get_emp_name procedure on the Boston server could have been changed. Or, its timestamp could be different from that recorded in the print_name procedure on the California server, possibly due to the installation of a new release of the server. As long as the signature remote dependency mode is in effect on the California server, a timestamp mismatch does not cause an error when get_emp_name is called. Note: DETERMINISTIC, PARALLEL_ENABLE, and purity information do not show in the signature mode. Optimizations based on these settings are not automatically reconsidered if a function on a remote system is redefined with different settings. This may lead to incorrect query results when calls to the remote function occur, even indirectly, in a SQL statement, or if the remote function is used, even indirectly, in a function-based index. When Does a Signature Change? Here is information on when a signature changes. Switching Datatype Classes A signature changes when you switch from one class of datatype to another. Within each datatype class, there can be several types. Changing a parameter datatype from one type to another within a class does not cause the signature to change. Datatypes that are not listed in the following table, such as NCHAR or TIMESTAMP, are not part of any class; changing their type always causes a signature mismatch. VARCHAR types: VARCHAR2, VARCHAR, STRING, LONG, ROWID Character types: CHARACTER, CHAR Raw types: RAW, LONG RAW Integer types: BINARY_INTEGER, PLS_INTEGER, BOOLEAN, NATURAL, POSITIVE, POSITIVEN, NATURALN Number types: NUMBER, INTEGER, INT, SMALLINT, DECIMAL, DEC, REAL, FLOAT, NUMERIC, DOUBLE PRECISION, DOUBLE PRECISION, NUMERIC 7-18 Oracle Database Application Developer’s Guide - Fundamentals Remote Dependencies Date types: DATE, TIMESTAMP, TIMESTAMP WITH TIME ZONE, TIMESTAMP WITH LOCAL TIME ZONE, INTERVAL YEAR TO MONTH, INTERVAL DAY TO SECOND Modes Changing to or from an explicit specification of the default parameter mode IN does not change the signature of a subprogram. For example, changing between: PROCEDURE P1 (Param1 NUMBER); PROCEDURE P1 (Param1 IN NUMBER); does not change the signature. Any other change of parameter mode does change the signature. Default Parameter Values Changing the specification of a default parameter value does not change the signature. For example, procedure P1 has the same signature in the following two examples: PROCEDURE P1 (Param1 IN NUMBER := 100); PROCEDURE P1 (Param1 IN NUMBER := 200); An application developer who requires that callers get the new default value must recompile the called procedure, but no signature-based invalidation occurs when a default parameter value assignment is changed. Examples of Changing Procedure Signatures Using the Get_emp_names procedure defined in "Parameters for Procedures and Functions" on page 7-4, if the procedure body is changed to the following: DECLARE Emp_number NUMBER; Hire_date DATE; BEGIN -- date format model changes SELECT Ename, To_char(Hiredate, 'DD/MON/YYYY') INTO Emp_name, Hire_date FROM Emp_tab WHERE Empno = Emp_number; END; The specification of the procedure has not changed, so its signature has not changed. But if the procedure specification is changed to the following: CREATE OR REPLACE PROCEDURE Get_emp_name ( Emp_number IN NUMBER, Hire_date OUT DATE, Emp_name OUT VARCHAR2) AS And if the body is changed accordingly, then the signature changes, because the parameter Hire_date has a different datatype. However, if the name of that parameter changes to When_hired, and the datatype remains VARCHAR2, and the mode remains OUT, the signature does not change. Changing the name of a formal parameter does not change the signature of the unit. Consider the following example: CREATE OR REPLACE PACKAGE Emp_package AS TYPE Emp_data_type IS RECORD ( Emp_number NUMBER, Coding PL/SQL Procedures and Packages 7-19 Remote Dependencies Hire_date VARCHAR2(12), Emp_name VARCHAR2(10)); PROCEDURE Get_emp_data (Emp_data IN OUT Emp_data_type); END; CREATE OR REPLACE PACKAGE BODY Emp_package AS PROCEDURE Get_emp_data (Emp_data IN OUT Emp_data_type) IS BEGIN SELECT Empno, Ename, TO_CHAR(Hiredate, 'DD/MON/YY') INTO Emp_data FROM Emp_tab WHERE Empno = Emp_data.Emp_number; END; END; If the package specification is changed so that the record's field names are changed, but the types remain the same, then this does not affect the signature. For example, the following package specification has the same signature as the previous package specification example: CREATE OR REPLACE PACKAGE Emp_package AS TYPE Emp_data_type IS RECORD ( Emp_num NUMBER, -- was Emp_number Hire_dat VARCHAR2(12), -- was Hire_date Empname VARCHAR2(10)); -- was Emp_name PROCEDURE Get_emp_data (Emp_data IN OUT Emp_data_type); END; Changing the name of the type of a parameter does not cause a change in the signature if the type remains the same as before. For example, the following package specification for Emp_package is the same as the first one: CREATE OR REPLACE PACKAGE Emp_package AS TYPE Emp_data_record_type IS RECORD ( Emp_number NUMBER, Hire_date VARCHAR2(12), Emp_name VARCHAR2(10)); PROCEDURE Get_emp_data (Emp_data IN OUT Emp_data_record_type); END; Controlling Remote Dependencies The dynamic initialization parameter REMOTE_DEPENDENCIES_MODE controls whether the timestamp or the signature dependency model is in effect. ■ If the initialization parameter file contains the following specification: REMOTE_DEPENDENCIES_MODE = TIMESTAMP Then only timestamps are used to resolve dependencies (if this is not explicitly overridden dynamically). ■ If the initialization parameter file contains the following parameter specification: REMOTE_DEPENDENCIES_MODE = SIGNATURE Then signatures are used to resolve dependencies (if this not explicitly overridden dynamically). 7-20 Oracle Database Application Developer’s Guide - Fundamentals Remote Dependencies ■ You can alter the mode dynamically by using the DDL statements. For example, this example alters the dependency model for the current session: ALTER SESSION SET REMOTE_DEPENDENCIES_MODE = {SIGNATURE | TIMESTAMP} Thise example alters the dependency model systemwide after startup: ALTER SYSTEM SET REMOTE_DEPENDENCIES_MODE = {SIGNATURE | TIMESTAMP} If the REMOTE_DEPENDENCIES_MODE parameter is not specified, either in the init.ora parameter file or using the ALTER SESSION or ALTER SYSTEM DDL statements, then timestamp is the default value. Therefore, unless you explicitly use the REMOTE_DEPENDENCIES_MODE parameter, or the appropriate DDL statement, your server is operating using the timestamp dependency model. When you use REMOTE_DEPENDENCIES_MODE=SIGNATURE: ■ ■ ■ If you change the default value of a parameter of a remote procedure, then the local procedure calling the remote procedure is not invalidated. If the call to the remote procedure does not supply the parameter, then the default value is used. In this case, because invalidation/recompilation does not automatically occur, the old default value is used. If you want to see the new default values, then you must recompile the calling procedure manually. If you add a new overloaded procedure in a package (a new procedure with the same name as an existing one), then local procedures that call the remote procedure are not invalidated. If it turns out that this overloading results in a rebinding of existing calls from the local procedure under the timestamp mode, then this rebinding does not happen under the signature mode, because the local procedure does not get invalidated. You must recompile the local procedure manually to achieve the new rebinding. If the types of parameters of an existing packaged procedure are changed so that the new types have the same shape as the old ones, then the local calling procedure is not invalidated or recompiled automatically. You must recompile the calling procedure manually to get the semantics of the new type. Dependency Resolution When REMOTE_DEPENDENCIES_MODE = TIMESTAMP (the default value), dependencies among program units are handled by comparing timestamps at runtime. If the timestamp of a called remote procedure does not match the timestamp of the called procedure, then the calling (dependent) unit is invalidated and must be recompiled. In this case, if there is no local PL/SQL compiler, then the calling application cannot proceed. In the timestamp dependency mode, signatures are not compared. If there is a local PL/SQL compiler, then recompilation happens automatically when the calling procedure is run. When REMOTE_DEPENDENCIES_MODE = SIGNATURE, the recorded timestamp in the calling unit is first compared to the current timestamp in the called remote unit. If they match, then the call proceeds. If the timestamps do not match, then the signature of the called remote subprogram, as recorded in the calling subprogram, is compared with the current signature of the called subprogram. If they do not match (using the criteria described in the section "When Does a Signature Change?" on page 7-18), then an error is returned to the calling session. Coding PL/SQL Procedures and Packages 7-21 Cursor Variables Suggestions for Managing Dependencies Follow these guidelines for setting the REMOTE_DEPENDENCIES_MODE parameter: ■ ■ ■ ■ Server-side PL/SQL users can set the parameter to TIMESTAMP (or let it default to that) to get the timestamp dependency mode. Server-side PL/SQL users can choose to use the signature dependency mode if they have a distributed system and they want to avoid possible unnecessary recompilations. Client-side PL/SQL users should set the parameter to SIGNATURE. This allows: – Installation of new applications at client sites, without the need to recompile procedures. – Ability to upgrade the server, without encountering timestamp mismatches. When using signature mode on the server side, add new procedures to the end of the procedure (or function) declarations in a package specification. Adding a new procedure in the middle of the list of declarations can cause unnecessary invalidation and recompilation of dependent procedures. Cursor Variables A cursor is a static object; a cursor variable is a pointer to a cursor. Because cursor variables are pointers, they can be passed and returned as parameters to procedures and functions. A cursor variable can also refer to different cursors in its lifetime. Some additional advantages of cursor variables include: ■ ■ ■ Encapsulation Queries are centralized in the stored procedure that opens the cursor variable. Ease of maintenance If you need to change the cursor, then you only need to make the change in one place: the stored procedure. There is no need to change each application. Convenient security The user of the application is the username used when the application connects to the server. The user must have EXECUTE permission on the stored procedure that opens the cursor. But, the user does not need to have READ permission on the tables used in the query. This capability can be used to limit access to the columns in the table, as well as access to other stored procedures. See Also: Oracle Database PL/SQL User's Guide and Reference for details on cursor variables Declaring and Opening Cursor Variables Memory is usually allocated for a cursor variable in the client application using the appropriate ALLOCATE statement. In Pro*C, use the EXEC SQL ALLOCATEstatement. In OCI, use the Cursor Data Area. You can also use cursor variables in applications that run entirely in a single server session. You can declare cursor variables in PL/SQL subprograms, open them, and use them as parameters for other PL/SQL subprograms. 7-22 Oracle Database Application Developer’s Guide - Fundamentals Cursor Variables Examples of Cursor Variables This section includes several examples of cursor variable usage in PL/SQL. For additional cursor variable examples that use the programmatic interfaces, refer to the following manuals: ■ Pro*C/C++ Programmer's Guide ■ Pro*COBOL Programmer's Guide ■ Oracle Call Interface Programmer's Guide ■ Oracle SQL*Module for Ada Programmer's Guide Fetching Data The following package defines a PL/SQL cursor variable type Emp_val_cv_type, and two procedures. The first procedure, Open_emp_cv, opens the cursor variable using a bind variable in the WHERE clause. The second procedure, Fetch_emp_data, fetches rows from the Emp_tab table using the cursor variable. CREATE OR REPLACE PACKAGE Emp_data AS TYPE Emp_val_cv_type IS REF CURSOR RETURN Emp_tab%ROWTYPE; PROCEDURE Open_emp_cv (Emp_cv IN OUT Emp_val_cv_type, Dept_number IN INTEGER); PROCEDURE Fetch_emp_data (emp_cv IN Emp_val_cv_type, emp_row OUT Emp_tab%ROWTYPE); END Emp_data; CREATE OR REPLACE PACKAGE BODY Emp_data AS PROCEDURE Open_emp_cv (Emp_cv IN OUT Emp_val_cv_type, Dept_number IN INTEGER) IS BEGIN OPEN emp_cv FOR SELECT * FROM Emp_tab WHERE deptno = dept_number; END open_emp_cv; PROCEDURE Fetch_emp_data (Emp_cv IN Emp_val_cv_type, Emp_row OUT Emp_tab%ROWTYPE) IS BEGIN FETCH Emp_cv INTO Emp_row; END Fetch_emp_data; END Emp_data; The following example shows how to call the Emp_data package procedures from a PL/SQL block: DECLARE -- declare a cursor variable Emp_curs Emp_data.Emp_val_cv_type; Dept_number Dept_tab.Deptno%TYPE; Emp_row Emp_tab%ROWTYPE; BEGIN Dept_number := 20; -- open the cursor using a variable Emp_data.Open_emp_cv(Emp_curs, Dept_number); -- fetch the data and display it LOOP Emp_data.Fetch_emp_data(Emp_curs, Emp_row); EXIT WHEN Emp_curs%NOTFOUND; DBMS_OUTPUT.PUT(Emp_row.Ename || ' '); DBMS_OUTPUT.PUT_LINE(Emp_row.Sal); END LOOP; Coding PL/SQL Procedures and Packages 7-23 Handling PL/SQL Compile-Time Errors END; Implementing Variant Records The power of cursor variables comes from their ability to point to different cursors. In the following package example, a discriminant is used to open a cursor variable to point to one of two different cursors: CREATE OR REPLACE PACKAGE Emp_dept_data AS TYPE Cv_type IS REF CURSOR; PROCEDURE Open_cv (Cv IN OUT cv_type, Discrim IN POSITIVE); END Emp_dept_data; CREATE OR REPLACE PACKAGE BODY Emp_dept_data AS PROCEDURE Open_cv (Cv IN OUT cv_type, Discrim IN POSITIVE) IS BEGIN IF Discrim = 1 THEN OPEN Cv FOR SELECT * FROM Emp_tab WHERE Sal > 2000; ELSIF Discrim = 2 THEN OPEN Cv FOR SELECT * FROM Dept_tab; END IF; END Open_cv; END Emp_dept_data; You can call the Open_cv procedure to open the cursor variable and point it to either a query on the Emp_tab table or the Dept_tab table. The following PL/SQL block shows how to fetch using the cursor variable, and then use the ROWTYPE_MISMATCH predefined exception to handle either fetch: DECLARE Emp_rec Emp_tab%ROWTYPE; Dept_rec Dept_tab%ROWTYPE; Cv Emp_dept_data.CV_TYPE; BEGIN Emp_dept_data.open_cv(Cv, 1); -- Open Cv For Emp_tab Fetch Fetch cv INTO Dept_rec; -- but fetch into Dept_tab record -- which raises ROWTYPE_MISMATCH DBMS_OUTPUT.PUT(Dept_rec.Deptno); DBMS_OUTPUT.PUT_LINE(' ' || Dept_rec.Loc); EXCEPTION WHEN ROWTYPE_MISMATCH THEN BEGIN DBMS_OUTPUT.PUT_LINE ('Row type mismatch, fetching Emp_tab data...'); FETCH Cv INTO Emp_rec; DBMS_OUTPUT.PUT(Emp_rec.Deptno); DBMS_OUTPUT.PUT_LINE(' ' || Emp_rec.Ename); END; Handling PL/SQL Compile-Time Errors When you use SQL*Plus to submit PL/SQL code, and when the code contains errors, you receive notification that compilation errors have occurred, but there is no immediate indication of what the errors are. For example, if you submit a standalone (or stored) procedure PROC1 in the file proc1.sql as follows: 7-24 Oracle Database Application Developer’s Guide - Fundamentals Handling PL/SQL Compile-Time Errors SQL> @proc1 And, if there are one or more errors in the code, then you receive a notice such as the following: MGR-00072: Warning: Procedure proc1 created with compilation errors In this case, use the SHOW ERRORS statement in SQL*Plus to get a list of the errors that were found. SHOW ERRORS with no argument lists the errors from the most recent compilation. You can qualify SHOW ERRORS using the name of a procedure, function, package, or package body: SQL> SHOW ERRORS PROC1 SQL> SHOW ERRORS PROCEDURE PROC1 SQL*Plus User's Guide and Reference for complete information about the SHOW ERRORS statement See Also: Note: Before issuing the SHOW ERRORS statement, use the SET LINESIZE statement to get long lines on output. The value 132 is usually a good choice. For example: SET LINESIZE 132 Assume that you want to create a simple procedure that deletes records from the employee table using SQL*Plus: CREATE OR REPLACE PROCEDURE Fire_emp(Emp_id NUMBER) AS BEGIN DELETE FROM Emp_tab WHER Empno = Emp_id; END / Notice that the CREATE PROCEDURE statement has two errors: the DELETE statement has an error (the E is absent from WHERE), and the semicolon is missing after END. After the CREATE PROCEDURE statement is entered and an error is returned, a SHOW ERRORS statement returns the following lines: SHOW ERRORS; ERRORS FOR PROCEDURE Fire_emp: LINE/COL ERROR -------------- -------------------------------------------3/27 PL/SQL-00103: Encountered the symbol "EMPNO" wh. . . 5/0 PL/SQL-00103: Encountered the symbol "END" when . . . 2 rows selected. Notice that each line and column number where errors were found is listed by the SHOW ERRORS statement. Alternatively, you can query the following data dictionary views to list errors when using any tool or application: ■ USER_ERRORS ■ ALL_ERRORS ■ DBA_ERRORS The error text associated with the compilation of a procedure is updated when the procedure is replaced, and it is deleted when the procedure is dropped. Coding PL/SQL Procedures and Packages 7-25 Handling Run-Time PL/SQL Errors Original source code can be retrieved from the data dictionary using the following views: ALL_SOURCE, USER_SOURCE, and DBA_SOURCE. Oracle Database Reference for more information about these data dictionary views See Also: Handling Run-Time PL/SQL Errors Oracle Database allows user-defined errors in PL/SQL code to be handled so that user-specified error numbers and messages are returned to the client application. After received, the client application can handle the error based on the user-specified error number and message returned by Oracle Database. User-specified error messages are returned using the RAISE_APPLICATION_ERROR procedure. For example: RAISE_APPLICATION_ERROR(Error_number, 'text', Keep_error_stack) This procedure stops procedure execution, rolls back any effects of the procedure, and returns a user-specified error number and message (unless the error is trapped by an exception handler). ERROR_NUMBER must be in the range of -20000 to -20999. Error number -20000 should be used as a generic number for messages where it is important to relay information to the user, but having a unique error number is not required. Text must be a character expression, 2 Kbytes or less (longer messages are ignored). Keep_error_stack can be TRUE if you want to add the error to any already on the stack, or FALSE if you want to replace the existing errors. By default, this option is FALSE. Some of the Oracle Database packages, such as DBMS_OUTPUT, DBMS_DESCRIBE, and DBMS_ALERT, use application error numbers in the range -20000 to -20005. Refer to the descriptions of these packages for more information. Note: The RAISE_APPLICATION_ERROR procedure is often used in exception handlers or in the logic of PL/SQL code. For example, the following exception handler selects the string for the associated user-defined error message and calls the RAISE_APPLICATION_ERROR procedure: ... WHEN NO_DATA_FOUND THEN SELECT Error_string INTO Message FROM Error_table, V$NLS_PARAMETERS V WHERE Error_number = -20101 AND Lang = v.value AND v.parameter = "NLS_LANGUAGE"; Raise_application_error(-20101, Message); ... See Also: "Handling Errors in Remote Procedures" on page 7-28 for information on exception handling when calling remote procedures The following section includes an example of passing a user-specified error number from a trigger to a procedure. 7-26 Oracle Database Application Developer’s Guide - Fundamentals Handling Run-Time PL/SQL Errors Declaring Exceptions and Exception Handling Routines User-defined exceptions are explicitly defined and signaled within the PL/SQL block to control processing of errors specific to the application. When an exception is raised (signaled), the usual execution of the PL/SQL block stops, and a routine called an exception handler is called. Specific exception handlers can be written to handle any internal or user-defined exception. Application code can check for a condition that requires special attention using an IF statement. If there is an error condition, then two options are available: ■ ■ Enter a RAISE statement that names the appropriate exception. A RAISE statement stops the execution of the procedure, and control passes to an exception handler (if any). Call the RAISE_APPLICATION_ERROR procedure to return a user-specified error number and message. You can also define an exception handler to handle user-specified error messages. For example, Figure 7–2 on page 7-27 illustrates the following: ■ ■ ■ An exception and associated exception handler in a procedure A conditional statement that checks for an error (such as transferring funds not available) and enters a user-specified error number and message within a trigger How user-specified error numbers are returned to the calling environment (in this case, a procedure), and how that application can define an exception that corresponds to the user-specified error number Declare a user-defined exception in a procedure or package body (private exceptions), or in the specification of a package (public exceptions). Define an exception handler in the body of a procedure (standalone or package). Figure 7–2 Exceptions and User-Defined Errors Procedure fire_emp(empid NUMBER) IS invalid_empid EXCEPTION; PRAGMA EXCEPTION_INIT(invalid_empid, –20101); BEGIN DELETE FROM emp WHERE empno = empid; EXCEPTION WHEN invlid_empid THEN INSERT INTO emp_audit VALUES (empid, ’Fired before probation ended’); END; Table EMP Error number returned to calling environment TRIGGER emp_probation BEFORE DELETE ON emp FOR EACH ROW BEGIN IF (sysdate–:old.hiredate)<30 THEN raise_application_error(20101, ’Employee’||old.ename||’ on probation’) END IF; END; Coding PL/SQL Procedures and Packages 7-27 Handling Run-Time PL/SQL Errors Unhandled Exceptions In database PL/SQL program units, an unhandled user-error condition or internal error condition that is not trapped by an appropriate exception handler causes the implicit rollback of the program unit. If the program unit includes a COMMIT statement before the point at which the unhandled exception is observed, then the implicit rollback of the program unit can only be completed back to the previous COMMIT. Additionally, unhandled exceptions in database-stored PL/SQL program units propagate back to client-side applications that call the containing program unit. In such an application, only the application program unit call is rolled back (not the entire application program unit), because it is submitted to the database as a SQL statement. If unhandled exceptions in database PL/SQL program units are propagated back to database applications, then the database PL/SQL code should be modified to handle the exceptions. Your application can also trap for unhandled exceptions when calling database program units and handle such errors appropriately. Handling Errors in Distributed Queries You can use a trigger or a stored procedure to create a distributed query. This distributed query is decomposed by the local Oracle Database instance into a corresponding number of remote queries, which are sent to the remote nodes for execution. The remote nodes run the queries and send the results back to the local node. The local node then performs any necessary post-processing and returns the results to the user or application. If a portion of a distributed statement fails, possibly due to an integrity constraint violation, then Oracle Database returns error number ORA-02055. Subsequent statements, or procedure calls, return error number ORA-02067 until a rollback or a rollback to savepoint is entered. You should design your application to check for any returned error messages that indicates that a portion of the distributed update has failed. If you detect a failure, then you should rollback the entire transaction (or rollback to a savepoint) before allowing the application to proceed. Handling Errors in Remote Procedures When a procedure is run locally or at a remote location, four types of exceptions can occur: ■ PL/SQL user-defined exceptions, which must be declared using the keyword EXCEPTION. ■ PL/SQL predefined exceptions, such as NO_DATA_FOUND. ■ SQL errors, such as ORA-00900 and ORA-02015. ■ Application exceptions, which are generated using the RAISE_APPLICATION_ERROR() procedure. When using local procedures, all of these messages can be trapped by writing an exception handler, such as shown in the following example: EXCEPTION WHEN ZERO_DIVIDE THEN /* ...handle the exception */ 7-28 Oracle Database Application Developer’s Guide - Fundamentals Debugging Stored Procedures Notice that the WHEN clause requires an exception name. If the exception that is raised does not have a name, such as those generated with RAISE_APPLICATION_ERROR, then one can be assigned using PRAGMA_EXCEPTION_INIT, as shown in the following example: DECLARE ... Null_salary EXCEPTION; PRAGMA EXCEPTION_INIT(Null_salary, -20101); BEGIN ... RAISE_APPLICATION_ERROR(-20101, 'salary is missing'); ... EXCEPTION WHEN Null_salary THEN ... When calling a remote procedure, exceptions are also handled by creating a local exception handler. The remote procedure must return an error number to the local calling procedure, which then handles the exception, as shown in the previous example. Because PL/SQL user-defined exceptions always return ORA-06510 to the local procedure, these exceptions cannot be handled. All other remote exceptions can be handled in the same manner as local exceptions. Debugging Stored Procedures Compiling a stored procedure involves fixing any syntax errors in the code. You might need to do additional debugging to make sure that the procedure works correctly, performs well, and recovers from errors. Such debugging might involve: ■ ■ Adding extra output statements to verify execution progress and check data values at certain points within the procedure. Running a separate debugger to analyze execution in greater detail. Oracle JDeveloper Recent releases of Oracle JDeveloper have extensive features for debugging PL/SQL, Java, and multi-language programs. You can get Oracle JDeveloper as part of various Oracle product suites. Often, a more recent release is available as a download at http://www.oracle.com/technology/. Oracle Procedure Builder and TEXT_IO Package Oracle Procedure Builder is an advanced client/server debugger that transparently debugs your database applications. It lets you run PL/SQL procedures and triggers in a controlled debugging environment, and you can set breakpoints, list the values of variables, and perform other debugging tasks. Oracle Procedure Builder is part of the Oracle Developer tool set. It also provides the TEXT_IO package that is useful for printing debug information. DBMS_OUTPUT Package You can also debug stored procedures and triggers using the DBMS_OUTPUT supplied package. Put PUT and PUT_LINE statements in your code to output the value of variables and expressions to your terminal. Coding PL/SQL Procedures and Packages 7-29 Debugging Stored Procedures Privileges for Debugging PL/SQL and Java Stored Procedures Starting with Oracle Database 10g, a new privilege model applies to debugging PL/SQL and Java code running within the database. This model applies whether you are using Oracle JDeveloper, Oracle Developer, or any of the various third-party PL/SQL or Java development environments, and it affects both the DBMS_DEBUG and DBMS_DEBUG_JDWP APIs. For a session to connect to a debugger, the effective user at the time of the connect operation must have the DEBUG CONNECT SESSION system privilege. This effective user may be the owner of a definer's rights routine involved in making the connect call. When a debugger becomes connected to a session, the session login user and the currently enabled session-level roles are fixed as the privilege environment for that debugging connection. Any DEBUG or EXECUTE privileges needed for debugging must be granted to that combination of user and roles. ■ ■ To be able to display and change Java public variables or variables declared in a PL/SQL package specification, the debugging connection must be granted either EXECUTE or DEBUG privilege on the relevant code. To be able to either display and change private variables or breakpoint and execute code lines step by step, the debugging connection must be granted DEBUG privilege on the relevant code Caution: The DEBUG privilege effectively allows a debugging session to do anything that the procedure being debugged could have done if that action had been included in its code. In addition to these privilege requirements, the ability to stop on individual code lines and debugger access to variables are allowed only in code compiled with debug information generated. The PLSQL_DEBUG parameter and the DEBUG keyword on commands such as ALTER PACKAGE can be used to control whether the PL/SQL compiler includes debug information in its results. If it does not, variables will not be accessible, and neither stepping nor breakpoints will stop on code lines. The PL/SQL compiler will never generate debug information for code that has been obfuscated using the PL/SQL wrap utility. See Also: Oracle Database PL/SQL User's Guide and Reference, "Obfuscating PL/SQL Source Code" The DEBUG ANY PROCEDURE system privilege is equivalent to the DEBUG privilege granted on all objects in the database. Objects owned by SYS are included if the value of the O7_DICTIONARY_ACCESSIBILITY parameter is TRUE. A debug role mechanism is available to carry privileges needed for debugging that are not normally enabled in the session. Refer to the documentation on the DBMS_DEBUG and DBMS_DEBUG_JDWP packages for details on how to specify a debug role and any necessary related password. The JAVADEBUGPRIV role carries the DEBUG CONNECT SESSION and DEBUG ANY PROCEDURE privileges. Grant it only with the care those privileges warrant. Caution: Granting DEBUG ANY PROCEDURE privilege, or granting DEBUG privilege on any object owned by SYS, means granting complete rights to the database. 7-30 Oracle Database Application Developer’s Guide - Fundamentals Calling Stored Procedures Writing Low-Level Debugging Code If you are actually writing code that will be part of a debugger, you might need to use packages such as DBMS_DEBUG_JDWP or DBMS_DEBUG. DBMS_DEBUG_JDWP Package The DBMS_DEBUG_JDWP package, provided starting with Oracle9i Release 2, provides a framework for multi-language debugging that is expected to supersede the DBMS_DEBUG package over time. It is especially useful for programs that combine PL/SQL and Java. DBMS_DEBUG Package The DBMS_DEBUG package, provided starting with Oracle8i, implements server-side debuggers and provides a way to debug server-side PL/SQL program units. Several of the debuggers available, such as Oracle Procedure Builder and various third-party vendor solutions, use this API. See Also: ■ ■ ■ ■ ■ Oracle Procedure Builder Developer's Guide Oracle Database PL/SQL Packages and Types Reference for more information about the DBMS_DEBUG and DBMS_OUTPUT packages and associated privileges The Oracle JDeveloper documentation for information on using package DBMS_DEBUG_JDWP Oracle Database SQL Reference for more details on privileges The PL/SQL page at http://www.oracle.com/technology/ for information about writing low-level debug code Calling Stored Procedures You may need to set up data structures, similar to the following, for certain examples to work: Note: CREATE TABLE Emp_tab ( Empno NUMBER(4) NOT NULL, Ename VARCHAR2(10), Job VARCHAR2(9), Mgr NUMBER(4), Hiredate DATE, Sal NUMBER(7,2), Comm NUMBER(7,2), Deptno NUMBER(2)); CREATE OR REPLACE PROCEDURE fire_emp1(Emp_id NUMBER) AS BEGIN DELETE FROM Emp_tab WHERE Empno = Emp_id; END; VARIABLE Empnum NUMBER; Procedures can be called from many different environments. For example: ■ A procedure can be called within the body of another procedure or a trigger. Coding PL/SQL Procedures and Packages 7-31 Calling Stored Procedures ■ ■ ■ A procedure can be interactively called by a user using an Oracle Database tool. A procedure can be explicitly called within an application, such as a SQL*Forms or a precompiler application. A stored function can be called from a SQL statement in a manner similar to calling a built-in SQL function, such as LENGTH or ROUND. This section includes some common examples of calling procedures from within these environments. See Also: "Calling Stored Functions from SQL Expressions" on page 7-36 A Procedure or Trigger Calling Another Procedure A procedure or trigger can call another stored procedure. For example, included in the body of one procedure might be the following line: . . . Sal_raise(Emp_id, 200); . . . This line calls the Sal_raise procedure. Emp_id is a variable within the context of the procedure. Recursive procedure calls are allowed within PL/SQL: A procedure can call itself. Interactively Calling Procedures From Oracle Database Tools A procedure can be called interactively from an Oracle Database tool, such as SQL*Plus. For example, to call a procedure named SAL_RAISE, owned by you, you can use an anonymous PL/SQL block, as follows: BEGIN Sal_raise(7369, 200); END; Interactive tools, such as SQL*Plus, require you to follow these lines with a slash (/) to run the PL/SQL block. Note: An easier way to run a block is to use the SQL*Plus statement EXECUTE, which wraps BEGIN and END statements around the code you enter. For example: EXECUTE Sal_raise(7369, 200); Some interactive tools allow session variables to be created. For example, when using SQL*Plus, the following statement creates a session variable: VARIABLE Assigned_empno NUMBER After defined, any session variable can be used for the duration of the session. For example, you might run a function and capture the return value using a session variable: EXECUTE :Assigned_empno := Hire_emp('JSMITH', 'President', 1032, SYSDATE, 5000, NULL, 10); PRINT Assigned_empno; ASSIGNED_EMPNO -------------2893 7-32 Oracle Database Application Developer’s Guide - Fundamentals Calling Stored Procedures See Also: ■ ■ SQL*Plus User's Guide and Reference Your tools documentation for information about performing similar operations using your development tool Calling Procedures within 3GL Applications A 3GL database application, such as a precompiler or an OCI application, can include a call to a procedure within the code of the application. To run a procedure within a PL/SQL block in an application, simply call the procedure. The following line within a PL/SQL block calls the Fire_emp procedure: Fire_emp1(:Empnun); In this case, :Empno is a host (bind) variable within the context of the application. To run a procedure within the code of a precompiler application, you must use the EXEC call interface. For example, the following statement calls the Fire_emp procedure in the code of a precompiler application: EXEC SQL EXECUTE BEGIN Fire_emp1(:Empnum); END; END-EXEC; See Also: For information about calling PL/SQL procedures from within 3GL applications: ■ Oracle Call Interface Programmer's Guide ■ Pro*C/C++ Programmer's Guide ■ Oracle SQL*Module for Ada Programmer's Guide Name Resolution When Calling Procedures References to procedures and packages are resolved according to the algorithm described in the "Rules for Name Resolution in SQL Statements" section of Chapter 2, "Designing Schema Objects". Privileges Required to Execute a Procedure If you are the owner of a standalone procedure or package, then you can run the standalone procedure or packaged procedure, or any public procedure or packaged procedure at any time, as described in the previous sections. If you want to run a standalone or packaged procedure owned by another user, then the following conditions apply: ■ ■ You must have the EXECUTE privilege for the standalone procedure or package containing the procedure, or you must have the EXECUTE ANY PROCEDURE system privilege. If you are executing a remote procedure, then you must be granted the EXECUTE privilege or EXECUTE ANY PROCEDURE system privilege directly, not through a role. You must include the name of the owner in the call. For example:1 EXECUTE Jward.Fire_emp (1043); EXECUTE Jward.Hire_fire.Fire_emp (1043); Coding PL/SQL Procedures and Packages 7-33 Calling Remote Procedures ■ ■ If the procedure is a definer's-rights procedure, then it runs with the privileges of the procedure owner. The owner must have all the necessary object privileges for any referenced objects. If the procedure is an invoker's-rights procedure, then it runs with your privileges (as the invoker). In this case, you also need privileges on all referenced objects; that is, all objects accessed by the procedure through external references that are resolved in your schema. You may hold these privileges directly or through a role. Roles are enabled unless an invoker's-rights procedure is called directly or indirectly by a definer's-rights procedure. Specifying Values for Procedure Arguments When you call a procedure, specify a value or parameter for each of the procedure's arguments. Identify the argument values using either of the following methods, or a combination of both: ■ List the values in the order the arguments appear in the procedure declaration. ■ Specify the argument names and corresponding values, in any order. For example, these statements each call the procedure Sal_raise to increase the salary of employee number 7369 by 500: Sal_raise(7369, 500); Sal_raise(Sal_incr=>500, Emp_id=>7369); Sal_raise(7369, Sal_incr=>500); The first statement identifies the argument values by listing them in the order in which they appear in the procedure specification. The second statement identifies the argument values by name and in an order different from that of the procedure specification. If you use argument names, then you can list the arguments in any order. The third statement identifies the argument values using a combination of these methods. If you use a combination of order and argument names, then values identified in order must precede values identified by name. If you used the DEFAULT option to define default values for IN parameters to a subprogram (see the Oracle Database PL/SQL User's Guide and Reference),then you can pass different numbers of actual parameters to the first subprogram, accepting or overriding the default values as you please. If an actual value is not passed, then the corresponding default value is used. If you want to assign a value to an argument that occurs after an omitted argument (for which the corresponding default is used), then you must explicitly designate the name of the argument, as well as its value. Calling Remote Procedures Call remote procedures using an appropriate database link and the procedure name. The following SQL*Plus statement runs the procedure Fire_emp located in the database and pointed to by the local database link named BOSTON_SERVER: 1 You may need to set up the following data structures for certain examples to work: CONNECT SYS/password AS SYSDBA;CREATE USER Jward IDENTIFIED BY Jward;GRANT CREATE ANY PACKAGE TO Jward;GRANT CREATE SESSION TO Jward;GRANT EXECUTE ANY PROCEDURE TO Jward;CONNECT Scott/Tiger 7-34 Oracle Database Application Developer’s Guide - Fundamentals Calling Remote Procedures EXECUTE fire_emp1@boston_server(1043); See Also: "Handling Errors in Remote Procedures" on page 7-28 for information on exception handling when calling remote procedures Remote Procedure Calls and Parameter Values You must explicitly pass values to all remote procedure parameters, even if there are defaults. You cannot access remote package variables and constants. Referencing Remote Objects Remote objects can be referenced within the body of a locally defined procedure. The following procedure deletes a row from the remote employee table: CREATE OR REPLACE PROCEDURE fire_emp(emp_id NUMBER) IS BEGIN DELETE FROM emp@boston_server WHERE empno = emp_id; END; The following list explains how to properly call remote procedures, depending on the calling environment. ■ Remote procedures (standalone and packaged) can be called from within a procedure, an OCI application, or a precompiler application by specifying the remote procedure name, a database link, and the arguments for the remote procedure. CREATE OR REPLACE PROCEDURE local_procedure(arg IN NUMBER) AS BEGIN fire_emp1@boston_server(arg); END; ■ In the previous example, you could create a synonym for FIRE_EMP1@BOSTON_SERVER. This would enable you to call the remote procedure from an Oracle Database tool application, such as a SQL*Forms application, as well from within a procedure, OCI application, or precompiler application. CREATE SYNONYM synonym1 for fire_emp1@boston_server; CREATE OR REPLACE PROCEDURE local_procedure(arg IN NUMBER) AS BEGIN synonym1(arg); END; ■ If you do not want to use a synonym, then you could write a local cover procedure to call the remote procedure. DECLARE arg NUMBER; BEGIN local_procedure(arg); END; Here, local_procedure is defined as in the first item of this list. See Also: "Synonyms for Procedures and Packages" on page 7-36 Coding PL/SQL Procedures and Packages 7-35 Calling Stored Functions from SQL Expressions Caution: Unlike stored procedures, which use compile-time binding, runtime binding is used when referencing remote procedures. The user account to which you connect depends on the database link. All calls to remotely stored procedures are assumed to perform updates; therefore, this type of referencing always requires two-phase commit of that transaction (even if the remote procedure is read-only). Furthermore, if a transaction that includes a remote procedure call is rolled back, then the work done by the remote procedure is also rolled back. A procedure called remotely can usually execute a COMMIT, ROLLBACK, or SAVEPOINT statement, the same as a local procedure. However, there are some differences in behavior: ■ ■ ■ If the transaction was originated by a non-Oracle database, as may be the case in XA applications, these operations are not allowed in the remote procedure. After doing one of these operations, the remote procedure cannot start any distributed transactions of its own. If the remote procedure does not commit or roll back its work, the commit is done implicitly when the database link is closed. In the meantime, further calls to the remote procedure are not allowed because it is still considered to be performing a transaction. A distributed update modifies data on two or more databases. A distributed update is possible using a procedure that includes two or more remote updates that access data on different databases. Statements in the construct are sent to the remote databases, and the execution of the construct succeeds or fails as a unit. If part of a distributed update fails and part succeeds, then a rollback (of the entire transaction or to a savepoint) is required to proceed. Consider this when creating procedures that perform distributed updates. Pay special attention when using a local procedure that calls a remote procedure. If a timestamp mismatch is found during execution of the local procedure, then the remote procedure is not run, and the local procedure is invalidated. Synonyms for Procedures and Packages Synonyms can be created for standalone procedures and packages to do the following: ■ ■ Hide the identity of the name and owner of a procedure or package. Provide location transparency for remotely stored procedures (standalone or within a package). When a privileged user needs to call a procedure, an associated synonym can be used. Because the procedures defined within a package are not individual objects (the package is the object), synonyms cannot be created for individual procedures within a package. Calling Stored Functions from SQL Expressions You can include user-written PL/SQL functions in SQL expressions. (You must be using PL/SQL release 2.1 or higher.) By using PL/SQL functions in SQL statements, you can do the following: 7-36 Oracle Database Application Developer’s Guide - Fundamentals Calling Stored Functions from SQL Expressions ■ ■ ■ ■ Increase user productivity by extending SQL. Expressiveness of the SQL statement increases where activities are too complex, too awkward, or unavailable with SQL. Increase query efficiency. Functions used in the WHERE clause of a query can filter data using criteria that would otherwise need to be evaluated by the application. Manipulate character strings to represent special datatypes (for example, latitude, longitude, or temperature). Provide parallel query execution: If the query is parallelized, then SQL statements in your PL/SQL function may also be run in parallel (using the parallel query option). Using PL/SQL Functions PL/SQL functions must be created as top-level functions or declared within a package specification before they can be named within a SQL statement. Stored PL/SQL functions are used in the same manner as built-in Oracle functions (such as SUBSTR or ABS). PL/SQL functions can be placed wherever an Oracle function can be placed within a SQL statement, or, wherever expressions can occur in SQL. For example, they can be called from the following: ■ The select list of the SELECT statement. ■ The condition of the WHERE and HAVING clause. ■ The CONNECT BY, START WITH, ORDER BY, and GROUP BY clauses. ■ The VALUES clause of the INSERT statement. ■ The SET clause of the UPDATE statement. You cannot call stored PL/SQL functions from a CHECK constraint clause of a CREATE or ALTER TABLE statement or use them to specify a default value for a column. These situations require an unchanging definition. Unlike functions, which are called as part of an expression, procedures are called as statements. Therefore, PL/SQL procedures are not directly callable from SQL statements. However, functions called from a PL/SQL statement or referenced in a SQL expression can call a PL/SQL procedure. Note: Syntax for SQL Calling a PL/SQL Function Use the following syntax to reference a PL/SQL function from SQL: [[schema.]package.]function_name[@dblink][(param_1...param_n)] For example, to reference a function you created that is called My_func, in the My_funcs_pkg package, in the Scott schema, that takes two numeric parameters, you could call the following: SELECT Scott.My_funcs_pkg.My_func(10,20) FROM dual; Naming Conventions If only one of the optional schema or package names is given, then the first identifier can be either a schema name or a package name. For example, to determine whether Coding PL/SQL Procedures and Packages 7-37 Calling Stored Functions from SQL Expressions Payroll in the reference Payroll.Tax_rate is a schema or package name, Oracle Database proceeds as follows: ■ ■ ■ Oracle Database first checks for the Payroll package in the current schema. If the PAYROLL package is found in the current schema, then Oracle Database looks for a Tax_rate function in the Payroll package. If a Tax_rate function is not found in the Payroll package, then an error message is returned. If a Payroll package is not found, then Oracle Database looks for a schema named Payroll that contains a top-level Tax_rate function. If the Tax_rate function is not found in the Payroll schema, then an error message is returned. You can also refer to a stored top-level function using any synonym that you have defined for it. Name Precedence In SQL statements, the names of database columns take precedence over the names of functions with no parameters. For example, if schema Scott creates the following two objects: CREATE TABLE Emp_tab(New_sal NUMBER ...); CREATE FUNCTION New_sal RETURN NUMBER IS ...; Then, in the following two statements, the reference to New_sal refers to the column Emp_tab.New_sal: SELECT New_sal FROM Emp_tab; SELECT Emp_tab.New_sal FROM Emp_tab; To access the function new_sal, enter the following: SELECT Scott.New_sal FROM Emp_tab; Example of Calling a PL/SQL Function from SQL For example, to call the Tax_rate PL/SQL function from schema Scott, run it against the Ss_no and sal columns in Tax_table, and place the results in the variable Income_tax, specify the following: You may need to set up data structures similar to the following for certain examples to work: Note: CREATE TABLE Tax_table ( Ss_no NUMBER, Sal NUMBER); CREATE OR REPLACE FUNCTION tax_rate (ssn IN NUMBER, salary IN NUMBER) RETURN NUMBER IS sal_out NUMBER; BEGIN sal_out := salary * 1.1; END; DECLARE Tax_id NUMBER; Income_tax NUMBER; BEGIN SELECT scott.tax_rate (Ss_no, Sal) INTO Income_tax FROM Tax_table 7-38 Oracle Database Application Developer’s Guide - Fundamentals Calling Stored Functions from SQL Expressions WHERE Ss_no = Tax_id; END; These sample calls to PL/SQL functions are allowed in SQL expressions: Circle_area(Radius) Payroll.Tax_rate(Empno) scott.Payroll.Tax_rate@boston_server(Dependents, Empno) Arguments To pass any number of arguments to a function, supply the arguments within the parentheses. You must use positional notation; named notation is not supported. For functions that do not accept arguments, use (). Using Default Values The stored function Gross_pay initializes two of its formal parameters to default values using the DEFAULT clause. For example: CREATE OR REPLACE FUNCTION Gross_pay (Emp_id IN NUMBER, St_hrs IN NUMBER DEFAULT 40, Ot_hrs IN NUMBER DEFAULT 0) RETURN NUMBER AS ... When calling Gross_pay from a procedural statement, you can always accept the default value of St_hrs. This is because you can use named notation, which lets you skip parameters. For example: IF Gross_pay(Eenum, Ot_hrs => Otime) > Pay_limit THEN ... However, when calling Gross_pay from a SQL expression, you cannot accept the default value of St_hrs, unless you accept the default value of Ot_hrs. This is because you cannot use named notation. Privileges To call a PL/SQL function from SQL, you must either own or have EXECUTE privileges on the function. To select from a view defined with a PL/SQL function, you must have SELECT privileges on the view. No separate EXECUTE privileges are necessary to select from the view. Requirements for Calling PL/SQL Functions from SQL Expressions To be callable from SQL expressions, a user-defined PL/SQL function must meet the following basic requirements: ■ ■ ■ ■ ■ It must be a stored function, not a function defined within a PL/SQL block or subprogram. It must be a row function, not a column (group) function; in other words, it cannot take an entire column of data as its argument. All its formal parameters must be IN parameters; none can be an OUT or IN OUT parameter. The datatypes of its formal parameters must be Oracle built-in types, such as CHAR, DATE, or NUMBER, not PL/SQL types, such as BOOLEAN, RECORD, or TABLE. Its return type (the datatype of its result value) must be an Oracle built-in type. Coding PL/SQL Procedures and Packages 7-39 Calling Stored Functions from SQL Expressions For example, the following stored function meets the basic requirements: You may need to set up the following data structures for certain examples to work: Note: CREATE TABLE Payroll( Srate NUMBER, Orate NUMBER, Acctno NUMBER); CREATE FUNCTION Gross_pay (Emp_id IN NUMBER, St_hrs IN NUMBER DEFAULT 40, Ot_hrs IN NUMBER DEFAULT 0) RETURN NUMBER AS St_rate NUMBER; Ot_rate NUMBER; BEGIN SELECT Srate, Orate INTO St_rate, Ot_rate FROM Payroll WHERE Acctno = Emp_id; RETURN St_hrs * St_rate + Ot_hrs * Ot_rate; END Gross_pay; Controlling Side Effects The purity of a stored subprogram (function or procedure) refers to the side effects of that subprogram on database tables or package variables. Side effects can prevent the parallelization of a query, yield order-dependent (and therefore, indeterminate) results, or require that package state be maintained across user sessions. Various side effects are not allowed when a subprogram is called from a SQL query or DML statement. In releases prior to Oracle8i, Oracle Database leveraged the PL/SQL compiler to enforce restrictions during the compilation of a stored subprogram or a SQL statement. Starting with Oracle8i, the compile-time restrictions were relaxed, and a smaller set of restrictions are enforced during execution. This change provides uniform support for stored subprograms written in PL/SQL, Java, and C, and it allows programmers the most flexibility possible. Restrictions When a SQL statement is run, checks are made to see if it is logically embedded within the execution of an already running SQL statement. This occurs if the statement is run from a trigger or from a subprogram that was in turn called from the already running SQL statement. In these cases, further checks occur to determine if the new SQL statement is safe in the specific context. The following restrictions are enforced on subprograms: ■ ■ ■ A subprogram called from a query or DML statement may not end the current transaction, create or rollback to a savepoint, or ALTER the system or session. A subprogram called from a query (SELECT) statement or from a parallelized DML statement may not execute a DML statement or otherwise modify the database. A subprogram called from a DML statement may not read or modify the particular table being modified by that DML statement. 7-40 Oracle Database Application Developer’s Guide - Fundamentals Calling Stored Functions from SQL Expressions These restrictions apply regardless of what mechanism is used to run the SQL statement inside the subprogram or trigger. For example: ■ ■ ■ They apply to a SQL statement called from PL/SQL, whether embedded directly in a subprogram or trigger body, run using the native dynamic mechanism (EXECUTE IMMEDIATE), or run using the DBMS_SQL package. They apply to statements embedded in Java with SQLJ syntax or run using JDBC. They apply to statements run with OCI using the callback context from within an "external" C function. You can avoid these restrictions if the execution of the new SQL statement is not logically embedded in the context of the already running statement. PL/SQL's autonomous transactions provide one escape (see "Autonomous Transactions" on page 2-20). Another escape is available using Oracle Call Interface (OCI) from an external C function, if you create a new connection rather than using the handle available from the OCIExtProcContext argument. Declaring a Function You can use the keywords DETERMINISTIC and PARALLEL_ENABLE in the syntax for declaring a function. These are optimization hints that inform the query optimizer and other software components about the following: ■ Functions that need not be called redundantly ■ Functions permitted within a parallelized query or parallelized DML statement Only functions that are DETERMINISTIC are allowed in function-based indexes and in certain snapshots and materialized views. A deterministic function depends solely on the values passed into it as arguments and does not reference or modify the contents of package variables or the database or have other side-effects. Such a function produces the same result value for any combination of argument values passed into it. You place the DETERMINISTIC keyword after the return value type in a declaration of the function. For example: CREATE FUNCTION F1 (P1 NUMBER) RETURN NUMBER DETERMINISTIC IS BEGIN RETURN P1 * 2; END; You may place this keyword in the following places: ■ On a function defined in a CREATE FUNCTION statement ■ In a function declaration in a CREATE PACKAGE statement ■ On a method declaration in a CREATE TYPE statement You should not repeat the keyword on the function or method body in a CREATE PACKAGE BODY or CREATE TYPE BODY statement. Certain performance optimizations occur on calls to functions that are marked DETERMINISTIC without any other action being required. The following features require that any function used with them be declared DETERMINISTIC: ■ ■ Any user-defined function used in a function-based index. Any function used in a materialized view, if that view is to qualify for Fast Refresh or is marked ENABLE QUERY REWRITE. Coding PL/SQL Procedures and Packages 7-41 Calling Stored Functions from SQL Expressions The preceding functions features attempt to use previously calculated results rather than calling the function when it is possible to do so. Functions that fall in the following categories should typically be DETERMINISTIC: ■ Functions used in a WHERE, ORDER BY, or GROUP BY clause ■ Functions that MAP or ORDER methods of a SQL type ■ Functions that in any other way help determine whether or where a row should appear in a result set Oracle Database cannot require that you should explicitly declare functions in the preceding categories as DETERMINISTIC without breaking existing applications, but the use of the keyword might be a wise choice of style within your application. Keep the following points in mind when you create DETERMINISTIC functions: ■ ■ The database cannot recognize if the behavior of the function is indeed deterministic. If the DETERMINISTIC keyword is applied to a function whose behavior is not truly deterministic, then the result of queries involving that function is unpredictable. If you change the semantics of a DETERMINISTIC function and recompile it, then existing function-based indexes and materialized views report results for the prior version of the function. Thus, if you change the semantics of a function, you must manually rebuild any dependent function-based indexes and materialized views. See Also: Oracle Database SQL Reference for an account of CREATE FUNCTION restrictions Parallel Query and Parallel DML Oracle Database's parallel execution feature divides the work of executing a SQL statement across multiple processes. Functions called from a SQL statement which is run in parallel may have a separate copy run in each of these processes, with each copy called for only the subset of rows that are handled by that process. Each process has its own copy of package variables. When parallel execution begins, these are initialized based on the information in the package specification and body as if a new user is logging into the system; the values in package variables are not copied from the original login session. And changes made to package variables are not automatically propagated between the various sessions or back to the original session. Java STATIC class attributes are similarly initialized and modified independently in each process. Because a function can use package (or Java STATIC) variables to accumulate some value across the various rows it encounters, Oracle Database cannot assume that it is safe to parallelize the execution of all user-defined functions. For query (SELECT) statements in Oracle Database versions prior to 8.1.5, the parallel query optimization looked to see if a function was noted as RNPS and WNPS in a PRAGMA RESTRICT_REFERENCES declaration; those functions that were marked as both RNPS and WNPS could be run in parallel. Functions defined with a CREATE FUNCTION statement had their code implicitly examined to determine if they were pure enough; parallelized execution might occur even though a pragma cannot be specified on these functions. See Also: "PRAGMA RESTRICT_REFERENCES – for Backward Compatibility" on page 7-43 For DML statements in Oracle Database versions prior to 8.1.5, the parallelization optimization looked to see if a function was noted as having all four of RNDS, WNDS, 7-42 Oracle Database Application Developer’s Guide - Fundamentals Calling Stored Functions from SQL Expressions RNPS and WNPS specified in a PRAGMA RESTRICT_REFERENCES declaration; those functions that were marked as neither reading nor writing to either the database or package variables could run in parallel. Again, those functions defined with a CREATE FUNCTION statement had their code implicitly examined to determine if they were actually pure enough; parallelized execution might occur even though a pragma cannot be specified on these functions. Oracle Database versions 8.1.5 and later continue to parallelize those functions that earlier versions recognize as parallelizable. The PARALLEL_ENABLE keyword is the preferred way to mark your code as safe for parallel execution. This keyword is syntactically similar to DETERMINISTIC as described in "Declaring a Function" on page 7-41; it is placed after the return value type in a declaration of the function, as in: CREATE FUNCTION F1 (P1 NUMBER) RETURN NUMBER PARALLEL_ENABLE IS BEGIN RETURN P1 * 2; END; A PL/SQL function defined with CREATE FUNCTION may still be run in parallel without any explicit declaration that it is safe to do so, if the system can determine that it neither reads nor writes package variables nor calls any function that might do so. A Java method or C function is never seen by the system as safe to run in parallel, unless the programmer explicitly indicates PARALLEL_ENABLE on the "call specification", or provides a PRAGMA RESTRICT_REFERENCES indicating that the function is sufficiently pure. An additional runtime restriction is imposed on functions run in parallel as part of a parallelized DML statement. Such a function is not permitted to in turn execute a DML statement; it is subject to the same restrictions that are enforced on functions that are run inside a query (SELECT) statement. See Also: "Restrictions" on page 7-40 PRAGMA RESTRICT_REFERENCES – for Backward Compatibility In Oracle Database versions prior to 8.1.5 (Oracle8i), programmers used the pragma RESTRICT_REFERENCES to assert the purity level of a subprogram. In subsequent versions, use the hints parallel-enable and deterministic, instead, to communicate subprogram purity to Oracle Database. You can remove RESTRICT_REFERENCES from your code. However, this pragma remains available for backward compatibility in situations where one of the following is true: ■ ■ It is impossible or impractical to edit existing code to remove RESTRICT_REFERENCES completely. If you do not remove it from a subprogram S1 that depends on another subprogram S2, then RESTRICT_REFERENCES might also be needed in S2, so that S1 will compile. Replacing RESTRICT_REFERENCES in existing code with hints parallel-enable and deterministic would negatively affect the behavior of new, dependent code. Use RESTRICT_REFERENCES to preserve the behavior of the existing code. An existing PL/SQL application can thus continue using the pragma even on new functionality, to ease integration with the existing code. Do not use the pragma in a wholly new application. If you use the pragma RESTRICT_REFERENCES, place it in a package specification, not in a package body. It must follow the declaration of a subprogram (function or Coding PL/SQL Procedures and Packages 7-43 Calling Stored Functions from SQL Expressions procedure), but it need not follow immediately. Only one pragma can reference a given subprogram declaration. The pragma RESTRICT_REFERENCES applies to both functions and procedures. Purity levels are important for functions, but also for procedures that are called by functions. Note: To code the pragma RESTRICT_REFERENCES, use the following syntax: PRAGMA RESTRICT_REFERENCES ( Function_name, WNDS [, WNPS] [, RNDS] [, RNPS] [, TRUST] ); Where: Keyword Description WNDS The subprogram writes no database state (does not modify database tables). RNDS The subprogram reads no database state (does not query database tables). WNPS The subprogram writes no package state (does not change the values of packaged variables). RNPS The subprogram reads no package state (does not reference the values of packaged variables). TRUST The other restrictions listed in the pragma are not enforced; they are simply assumed to be true. This allows easy calling from functions that have RESTRICT_REFERENCES declarations to those that do not. You can pass the arguments in any order. If any SQL statement inside the subprogram body violates a rule, then you get an error when the statement is parsed. In the following example, the function compound neither reads nor writes database or package state; therefore, you can assert the maximum purity level. Always assert the highest purity level that a subprogram allows. That way, the PL/SQL compiler never rejects the subprogram unnecessarily. You may need to set up the following data structures for certain examples here to work: Note: CREATE TABLE Accts ( Yrs NUMBER, Amt NUMBER, Acctno NUMBER, Rte NUMBER); CREATE PACKAGE Finance AS -- package specification FUNCTION Compound (Years IN NUMBER, Amount IN NUMBER, Rate IN NUMBER) RETURN NUMBER; PRAGMA RESTRICT_REFERENCES (Compound, WNDS, WNPS, RNDS, RNPS); END Finance; CREATE PACKAGE BODY Finance AS FUNCTION Compound (Years IN NUMBER, Amount IN NUMBER, --package body 7-44 Oracle Database Application Developer’s Guide - Fundamentals Calling Stored Functions from SQL Expressions Rate IN NUMBER) RETURN NUMBER IS BEGIN RETURN Amount * POWER((Rate / 100) + 1, Years); END Compound; -- no pragma in package body END Finance; Later, you might call compound from a PL/SQL block, as follows: DECLARE Interest NUMBER; Acct_id NUMBER; BEGIN SELECT Finance.Compound(Yrs, Amt, Rte) INTO Interest FROM Accounts WHERE Acctno = Acct_id; -- function call Using the Keyword TRUST The keyword TRUST in the RESTRICT_REFERENCES syntax allows easy calling from functions that have RESTRICT_REFERENCES declarations to those that do not. When TRUST is present, the restrictions listed in the pragma are not actually enforced, but rather are simply assumed to be true. When calling from a section of code that is using pragmas to one that is not, there are two likely usage styles. One is to place a pragma on the routine to be called, for example on a "call specification" for a Java method. Then, calls from PL/SQL to this method will complain if the method is less restricted than the calling subprogram. For example: CREATE OR REPLACE PACKAGE P1 IS FUNCTION F1 (P1 NUMBER) RETURN NUMBER IS LANGUAGE JAVA NAME 'CLASS1.METHODNAME(int) return int'; PRAGMA RESTRICT_REFERENCES(F1,WNDS,TRUST); FUNCTION F2 (P1 NUMBER) RETURN NUMBER; PRAGMA RESTRICT_REFERENCES(F2,WNDS); END; CREATE OR REPLACE PACKAGE BODY P1 IS FUNCTION F2 (P1 NUMBER) RETURN NUMBER IS BEGIN RETURN F1(P1); END; END; Here, F2 can call F1, as F1 has been declared to be WNDS. The other approach is to mark only the caller, which may then make a call to any subprogram without complaint. For example: CREATE OR REPLACE PACKAGE P1a IS FUNCTION F1 (P1 NUMBER) RETURN NUMBER IS LANGUAGE JAVA NAME 'CLASS1.METHODNAME(int) return int'; FUNCTION F2 (P1 NUMBER) RETURN NUMBER; PRAGMA RESTRICT_REFERENCES(F2,WNDS,TRUST); END; CREATE OR REPLACE PACKAGE BODY P1a IS FUNCTION F2 (P1 NUMBER) RETURN NUMBER IS BEGIN RETURN F1(P1); Coding PL/SQL Procedures and Packages 7-45 Calling Stored Functions from SQL Expressions END; END; Here, F2 can call F1 because while F2 is promised to be WNDS (because TRUST is specified), the body of F2 is not actually examined to determine if it truly satisfies the WNDS restriction. Because F2 is not examined, its call to F1 is allowed, even though there is no PRAGMA RESTRICT_REFERENCES for F1. Differences between Static and Dynamic SQL Statements. Static INSERT, UPDATE, and DELETE statements do not violate RNDS if these statements do not explicitly read any database states, such as columns of a table. However, dynamic INSERT, UPDATE, and DELETE statements always violate RNDS, regardless of whether or not the statements explicitly read database states. The following INSERT violates RNDS if it is executed dynamically, but it does not violate RNDS if it is executed statically. INSERT INTO my_table values(3, 'SCOTT'); The following UPDATE always violates RNDS statically and dynamically, because it explicitly reads the column name of my_table. UPDATE my_table SET id=777 WHERE name='SCOTT'; Overloading Packaged PL/SQL Functions PL/SQL lets you overload packaged (but not standalone) functions: You can use the same name for different functions if their formal parameters differ in number, order, or datatype family. However, a RESTRICT_REFERENCES pragma can apply to only one function declaration. Therefore, a pragma that references the name of overloaded functions always applies to the nearest preceding function declaration. In this example, the pragma applies to the second declaration of valid: CREATE PACKAGE Tests AS FUNCTION Valid (x NUMBER) RETURN CHAR; FUNCTION Valid (x DATE) RETURN CHAR; PRAGMA RESTRICT_REFERENCES (valid, WNDS); END; Serially Reusable PL/SQL Packages PL/SQL packages usually consume user global area (UGA) memory corresponding to the number of package variables and cursors in the package. This limits scalability, because the memory increases linearly with the number of users. The solution is to allow some packages to be marked as SERIALLY_REUSABLE (using pragma syntax). For serially reusable packages, the package global memory is not kept in the UGA for each user; rather, it is kept in a small pool and reused for different users. This means that the global memory for such a package is only used within a unit of work. At the end of that unit of work, the memory can therefore be released to the pool to be reused by another user (after running the initialization code for all the global variables). The unit of work for serially reusable packages is implicitly a call to the server; for example, an OCI call to the server, or a PL/SQL RPC call from a client to a server, or an RPC call from a server to another server. 7-46 Oracle Database Application Developer’s Guide - Fundamentals Calling Stored Functions from SQL Expressions Package States The state of a nonreusable package (one not marked SERIALLY_REUSABLE) persists for the lifetime of a session. A package state includes global variables, cursors, and so on. The state of a serially reusable package persists only for the lifetime of a call to the server. On a subsequent call to the server, if a reference is made to the serially reusable package, then Oracle Database creates a new instantiation of the serially reusable package and initializes all the global variables to NULL or to the default values provided. Any changes made to the serially reusable package state in the previous calls to the server are not visible. Note: Creating a new instantiation of a serially reusable package on a call to the server does not necessarily imply that Oracle Database allocates memory or configures the instantiation object. Oracle Database looks for an available instantiation work area (which is allocated and configured) for this package in a least-recently used (LRU) pool in the SGA. At the end of the call to the server, this work area is returned back to the LRU pool. The reason for keeping the pool in the SGA is that the work area can be reused across users who have requests for the same package. Why Serially Reusable Packages? Because the state of a non-reusable package persists for the lifetime of the session, this locks up UGA memory for the whole session. In applications, such as Oracle Office, a log-on session can typically exist for days together. Applications often need to use certain packages only for certain localized periods in the session and would ideally like to de-instantiate the package state in the middle of the session, after they are done using the package. With SERIALLY_REUSABLE packages, application developers have a way of modelling their applications to manage their memory better for scalability. Package state that they care about only for the duration of a call to the server should be captured in SERIALLY_REUSABLE packages. Syntax of Serially Reusable Packages A package can be marked serially reusable by a pragma. The syntax of the pragma is: PRAGMA SERIALLY_REUSABLE; A package specification can be marked serially reusable, whether or not it has a corresponding package body. If the package has a body, then the body must have the serially reusable pragma, if its corresponding specification has the pragma; it cannot have the serially reusable pragma unless the specification also has the pragma. Semantics of Serially Reusable Packages A package that is marked SERIALLY_REUSABLE has the following properties: ■ Its package variables are meant for use only within the work boundaries, which correspond to calls to the server (either OCI call boundaries or PL/SQL RPC calls to the server). Coding PL/SQL Procedures and Packages 7-47 Calling Stored Functions from SQL Expressions If the application programmer makes a mistake and depends on a package variable that is set in a previous unit of work, then the application program can fail. PL/SQL cannot check for such cases. Note: ■ ■ ■ A pool of package instantiations is kept, and whenever a "unit of work" needs this package, one of the instantiations is "reused", as follows: – The package variables are reinitialized (for example, if the package variables have default values, then those values are reinitialized). – The initialization code in the package body is run again. At the "end work" boundary, cleanup is done. – If any cursors were left open, then they are silently closed. – Some non-reusable secondary memory is freed (such as memory for collection variables or long VARCHAR2s). – This package instantiation is returned back to the pool of reusable instantiations kept for this package. Serially reusable packages cannot be accessed from database triggers or other PL/SQL subprograms that are called from SQL statements. If you try, then Oracle Database generates an error. Examples of Serially Reusable Packages This section presents a few examples of serially reusable packages. Example 1: How Package Variables Act Across Call Boundaries This example has a serially reusable package specification (there is no body). CONNECT Scott/Tiger CREATE OR REPLACE PACKAGE Sr_pkg IS PRAGMA SERIALLY_REUSABLE; N NUMBER := 5; -- default initialization END Sr_pkg; Suppose your Enterprise Manager (or SQL*Plus) application issues the following: CONNECT Scott/Tiger # first CALL to server BEGIN Sr_pkg.N := 10; END; # second CALL to server BEGIN DBMS_OUTPUT.PUT_LINE(Sr_pkg.N); END; This program prints: 5 7-48 Oracle Database Application Developer’s Guide - Fundamentals Calling Stored Functions from SQL Expressions If the package had not had the pragma SERIALLY_REUSABLE, the program would have printed '10'. Note: Example 2: How Package Variables Act Across Call Boundaries This example has both a package specification and package body, which are serially reusable. CONNECT Scott/Tiger DROP PACKAGE Sr_pkg; CREATE OR REPLACE PACKAGE Sr_pkg IS PRAGMA SERIALLY_REUSABLE; TYPE Str_table_type IS TABLE OF VARCHAR2(200) INDEX BY BINARY_INTEGER; Num NUMBER := 10; Str VARCHAR2(200) := 'default-init-str'; Str_tab STR_TABLE_TYPE; PROCEDURE Print_pkg; PROCEDURE Init_and_print_pkg(N NUMBER, V VARCHAR2); END Sr_pkg; CREATE OR REPLACE PACKAGE BODY Sr_pkg IS -- the body is required to have the pragma because the -- specification of this package has the pragma PRAGMA SERIALLY_REUSABLE; PROCEDURE Print_pkg IS BEGIN DBMS_OUTPUT.PUT_LINE('num: ' || Sr_pkg.Num); DBMS_OUTPUT.PUT_LINE('str: ' || Sr_pkg.Str); DBMS_OUTPUT.PUT_LINE('number of table elems: ' || Sr_pkg.Str_tab.Count); FOR i IN 1..Sr_pkg.Str_tab.Count LOOP DBMS_OUTPUT.PUT_LINE(Sr_pkg.Str_tab(i)); END LOOP; END; PROCEDURE Init_and_print_pkg(N NUMBER, V VARCHAR2) IS BEGIN -- init the package globals Sr_pkg.Num := N; Sr_pkg.Str := V; FOR i IN 1..n LOOP Sr_pkg.Str_tab(i) := V || ' ' || i; END LOOP; -- print the package Print_pkg; END; END Sr_pkg; SET SERVEROUTPUT ON; Rem SR package access in a CALL: BEGIN -- initialize and print the package DBMS_OUTPUT.PUT_LINE('Initing and printing pkg state..'); Sr_pkg.Init_and_print_pkg(4, 'abracadabra'); -- print it in the same call to the server. -- we should see the initialized values. DBMS_OUTPUT.PUT_LINE('Printing package state in the same CALL...'); Sr_pkg.Print_pkg; END; Coding PL/SQL Procedures and Packages 7-49 Calling Stored Functions from SQL Expressions Initing and printing pkg state.. num: 4 str: abracadabra number of table elems: 4 abracadabra 1 abracadabra 2 abracadabra 3 abracadabra 4 Printing package state in the same CALL... num: 4 str: abracadabra number of table elems: 4 abracadabra 1 abracadabra 2 abracadabra 3 abracadabra 4 REM SR package access in subsequent CALL: BEGIN -- print the package in the next call to the server. -- We should that the package state is reset to the initial (default) values. DBMS_OUTPUT.PUT_LINE('Printing package state in the next CALL...'); Sr_pkg.Print_pkg; END; Statement processed. Printing package state in the next CALL... num: 10 str: default-init-str number of table elems: 0 Example 3: Open Cursors in Serially Reusable Packages at Call Boundaries This example demonstrates that any open cursors in serially reusable packages get closed automatically at the end of a work boundary (which is a call). Also, in a new call, these cursors need to be opened again. REM REM REM REM REM REM For serially reusable pkg: At the end work boundaries (which is currently the OCI call boundary) all open cursors will be closed. Because the cursor is closed - every time we fetch we will start at the first row again. CONNECT Scott/Tiger DROP PACKAGE Sr_pkg; DROP TABLE People; CREATE TABLE People (Name VARCHAR2(20)); INSERT INTO People VALUES ('ET'); INSERT INTO People VALUES ('RAMBO'); CREATE OR REPLACE PACKAGE Sr_pkg IS PRAGMA SERIALLY_REUSABLE; CURSOR C IS SELECT Name FROM People; END Sr_pkg; SQL> SET SERVEROUTPUT ON; SQL> CREATE OR REPLACE PROCEDURE Fetch_from_cursor IS Name VARCHAR2(200); BEGIN IF (Sr_pkg.C%ISOPEN) THEN DBMS_OUTPUT.PUT_LINE('cursor is already open.'); ELSE 7-50 Oracle Database Application Developer’s Guide - Fundamentals Returning Large Amounts of Data from a Function DBMS_OUTPUT.PUT_LINE('cursor is closed; opening now.'); OPEN Sr_pkg.C; END IF; -- fetching from cursor. FETCH sr_pkg.C INTO name; DBMS_OUTPUT.PUT_LINE('fetched: ' || Name); FETCH Sr_pkg.C INTO name; DBMS_OUTPUT.PUT_LINE('fetched: ' || Name); -- Oops forgot to close the cursor (Sr_pkg.C). -- But, because it is a Serially Reusable pkg's cursor, -- it will be closed at the end of this CALL to the server. END; EXECUTE fetch_from_cursor; cursor is closed; opening now. fetched: ET fetched: RAMBO Returning Large Amounts of Data from a Function In a data warehousing environment, you might use a PL/SQL function to transform large amounts of data. Perhaps the data is passed through a series of transformations, each performed by a different function. PL/SQL table functions let you perform such transformations without significant memory overhead or the need to store the data in tables between each transformation stage. These functions can accept and return multiple rows, can return rows as they are ready rather than all at once, and can be parallelized. In this technique: ■ ■ ■ ■ ■ The producer function uses the PIPELINED keyword in its declaration. The producer function uses an OUT parameter that is a record, corresponding to a row in the result set. As each output record is completed, it is sent to the consumer function using the PIPE ROW keyword. The producer function ends with a RETURN statement that does not specify any return value. The consumer function or SQL statement uses the TABLE keyword to treat the resulting rows like a regular table. For example: CREATE FUNCTION StockPivot(p refcur_pkg.refcur_t) RETURN TickerTypeSet PIPELINED IS out_rec TickerType := TickerType(NULL,NULL,NULL); in_rec p%ROWTYPE; BEGIN LOOP -- Function accepts multiple rows through a REF CURSOR argument. FETCH p INTO in_rec; EXIT WHEN p%NOTFOUND; -- Return value is a record type that matches the table definition. out_rec.ticker := in_rec.Ticker; out_rec.PriceType := 'O'; out_rec.price := in_rec.OpenPrice; -- Once a result row is ready, we send it back to the calling program, -- and continue processing. PIPE ROW(out_rec); Coding PL/SQL Procedures and Packages 7-51 Coding Your Own Aggregate Functions -- This function outputs twice as many rows as it receives as input. out_rec.PriceType := 'C'; out_rec.Price := in_rec.ClosePrice; PIPE ROW(out_rec); END LOOP; CLOSE p; -- The function ends with a RETURN statement that does not specify any value. RETURN; END; / -- Here we use the result of this function in a SQL query. SELECT * FROM TABLE(StockPivot(CURSOR(SELECT * FROM StockTable))); -- Here we use the result of this function in a PL/SQL block. DECLARE total NUMBER := 0; price_type VARCHAR2(1); BEGIN FOR item IN (SELECT * FROM TABLE(StockPivot(CURSOR(SELECT * FROM StockTable)))) LOOP -- Access the values of each output row. -- We know the column names based on the declaration of the output type. -- This computation is just for illustration. total := total + item.price; price_type := item.price_type; END LOOP; END; / Coding Your Own Aggregate Functions To analyze a set of rows and compute a result value, you can code your own aggregate function that works the same as a built-in aggregate like SUM: ■ ■ ■ ■ Define a SQL object type that defines these member functions: ■ ODCIAggregateInitialize ■ ODCIAggregateIterate ■ ODCIAggregateMerge ■ ODCIAggregateTerminate Code the member functions. In particular, ODCIAggregateIterate accumulates the result as it is called once for each row that is processed. Store any intermediate results using the attributes of the object type. Create the aggregate function, and associate it with the new object type. Call the aggregate function from SQL queries, DML statements, or other places that you might use the built-in aggregates. You can include typical options such as DISTINCT and ALL in the calls to the aggregate function. See Also: Oracle Database Data Cartridge Developer's Guide for details of this process and the requirements for the member functions 7-52 Oracle Database Application Developer’s Guide - Fundamentals 8 Coding Dynamic SQL This chapter covers the following topics: ■ What Is Dynamic SQL? ■ Why Use Dynamic SQL? ■ Developing with Native Dynamic SQL: Scenario ■ Choosing Between Native Dynamic SQL and the DBMS_SQL Package ■ Programming with Dynamic SQL ■ Avoiding SQL Injection in PL/SQL What Is Dynamic SQL? Unlike static SQL, which remains the same in each execution, dynamic SQL enables you to build SQL statements as character strings at runtime. The strings contain the text of a SQL statement or PL/SQL block and can also contain placeholders for bind arguments. You can create general purpose, flexible applications with dynamic SQL because the full text of a SQL statement may be unknown at compilation. You can use dynamic SQL in several different development environments, including PL/SQL, Pro*C/C++, and Java. For an example of an application that uses dynamic SQL, suppose that a reporting application in a data warehouse environment does not know a table name until runtime. These tables are named according to the starting month and year of the quarter, for example, inv_01_2003, inv_04_2003, inv_07_2003, inv_10_2003, inv_01_2004, and so on. You can use dynamic SQL in your reporting application to specify the table name at runtime. In a different example, suppose that you want to run a complex query with a user-selectable sort order. Instead of coding the query twice with a different ORDER BY clause in each query, you can construct the query dynamically to include a specified ORDER BY clause. Programming with Dynamic SQL For the sake of consistency, this chapter discusses dynamic SQL mainly from the perspective of PL/SQL. To process most dynamic SQL statements, you use the EXECUTE IMMEDIATE statement. To process a multi-row query in a PL/SQL procedure, you use the OPEN-FOR, FETCH, and CLOSE statements. Coding Dynamic SQL 8-1 Why Use Dynamic SQL? Oracle Database enables you to implement dynamic SQL in a PL/SQL application in the following ways: ■ ■ Using native dynamic SQL, which involves placing dynamic SQL statements directly into PL/SQL blocks Calling procedures in the DBMS_SQL package Although this chapter discusses PL/SQL support for dynamic SQL, you can call dynamic SQL from other languages: ■ ■ ■ If you use C/C++, you can call dynamic SQL with the Oracle Call Interface (OCI), or you can use the Pro*C/C++ precompiler to add dynamic SQL extensions to your C code. If you use COBOL, you can use the Pro*COBOL precompiler to add dynamic SQL extensions to your COBOL code. If you use Java, you can develop applications that use dynamic SQL with JDBC. If you have a program that uses OCI, Pro*C/C++, or Pro*COBOL to execute dynamic SQL, consider switching to native dynamic SQL inside PL/SQL stored procedures and functions. The network round-trips required to perform dynamic SQL operations from client-side applications might hurt performance. Stored procedures can reside on the server, eliminating network overhead. You can call the PL/SQL stored procedures and stored functions from the OCI, Pro*C/C++, or Pro*COBOL application. See Also: Oracle Database PL/SQL Packages and Types Reference for details about the DBMS_SQL package. To learn about calling Oracle Database stored procedures and stored functions from languages other than PL/SQL, consult the following resources: ■ Oracle Call Interface Programmer's Guide ■ Pro*C/C++ Programmer's Guide ■ Pro*COBOL Programmer's Guide Why Use Dynamic SQL? Dynamic SQL and static SQL both have advantages and disadvantages. The full text of static SQL statements is known at compilation, which provides the following benefits: ■ ■ Successful compilation verifies that the SQL statements reference valid database objects and that the necessary privileges are in place to access the objects. Performance of static SQL is generally better than dynamic SQL. Despite these advantages, static SQL has limitations that can be overcome with dynamic SQL, as in the following cases: ■ ■ ■ You do not know the full text of the SQL statements that must be executed in a PL/SQL procedure. These SQL statements may depend on user input or on processing work performed by the program. You want to execute DDL statements and other SQL statements that are not supported in purely static SQL programs. You want to write a program that can handle changes in data definitions without the need to recompile. Dynamic SQL is more flexible than static SQL because it enables you to write reusable code that can be adapted for different environments. 8-2 Oracle Database Application Developer’s Guide - Fundamentals Why Use Dynamic SQL? As a general rule you should use dynamic SQL only if you cannot use static SQL to accomplish your goals, or if using static SQL is too cumbersome. The following sections describe typical situations in which you should use dynamic SQL: ■ Executing DDL and SCL Statements in PL/SQL ■ Executing Dynamic Queries ■ Referencing Database Objects that Do Not Exist at Compilation ■ Optimizing Execution Dynamically ■ Executing Dynamic PL/SQL Blocks ■ Performing Dynamic Operations Using Invoker's Rights Executing DDL and SCL Statements in PL/SQL Only dynamic SQL can execute the following types of statements within PL/SQL program units: ■ Data definition language (DDL) statements such as CREATE, DROP, GRANT, and REVOKE ■ Session control language (SCL) statements such as ALTER SESSION and SET ROLE ■ The TABLE clause in the SELECT statement The following native dynamic SQL example uses a SELECT statement with the TABLE clause. Example 8–1 Using SELECT . . . TABLE in Dynamic SQL -- Create an object t_emp and a datatype t_emplist as a table of type t_emp CREATE TYPE t_emp AS OBJECT (id NUMBER, name VARCHAR2(20)) / CREATE TYPE t_emplist AS TABLE OF t_emp / -- Create a table with a nested table of type t_emplist CREATE TABLE dept_new (id NUMBER, emps t_emplist) NESTED TABLE emps STORE AS emp_table; -- Populate the dept_new table with data INSERT INTO dept_new VALUES ( 10, t_emplist ( t_emp(1, 'SCOTT'), t_emp(2, 'BRUCE') ) ); -- Write a PL/SQL block that queries table dept_new and nested table emps -- SELECT ... FROM ... TABLE is not allowed in static SQL in PL/SQL DECLARE v_deptid NUMBER; v_ename VARCHAR2(20); BEGIN EXECUTE IMMEDIATE 'SELECT d.id, e.name FROM dept_new d, TABLE(d.emps) e WHERE e.id = 1' INTO v_deptid, v_ename; END; / Coding Dynamic SQL 8-3 Why Use Dynamic SQL? Oracle Database SQL Reference for information about DDL and SCL statements See Also: Executing Dynamic Queries You can use dynamic SQL to create applications that execute dynamic queries whose full text is not available until runtime. Many types of applications need to use dynamic queries, including applications that do the following: ■ Allow users to input or choose query search or sorting criteria at runtime ■ Allow users to input or choose optimizer hints at run time ■ Query a database where the data definitions of tables are constantly changing ■ Query a database where new tables are created often "Querying with Dynamic SQL: Example" on page 8-14 and "Developing with Native Dynamic SQL: Scenario" on page 8-7 See Also: Referencing Database Objects that Do Not Exist at Compilation Many applications must interact with data that is generated periodically. For example, an application knows the table definitions at compile time, but not the table names. Dynamic SQL enables you to specify table names at runtime. In the sample data warehouse application discussed in "What Is Dynamic SQL?" on page 8-1, the system generates new tables every quarter. You could allow a user to specify the name of the table at runtime with a dynamic SQL query similar to the following sample procedure. Example 8–2 Dynamically Specifying a Table Name CREATE OR REPLACE PROCEDURE query_invoice (p_month VARCHAR2, p_year VARCHAR2) IS TYPE cur_typ IS REF CURSOR; v_inv_cursor cur_typ; -- Declare a cursor variable v_inv_query VARCHAR2(200); v_inv_num NUMBER; v_inv_cust VARCHAR2(20); v_inv_amt NUMBER; BEGIN -- Write dynamic query receiving month and year as parameters -- and using these values to form the table name, for example, inv_APR_2004 v_inv_query := 'SELECT num, cust, amt FROM inv_' || p_month || '_' || p_year || ' WHERE v_inv_num = :g_id'; -- Open a cursor variable OPEN v_inv_cursor FOR v_inv_query USING v_inv_num; -- Fetch row into variables LOOP FETCH v_inv_cursor INTO v_inv_num, v_inv_cust, v_inv_amt; EXIT WHEN v_inv_cursor%NOTFOUND; -- process row here END LOOP; CLOSE v_inv_cursor; END; 8-4 Oracle Database Application Developer’s Guide - Fundamentals Why Use Dynamic SQL? / Optimizing Execution Dynamically You can use dynamic SQL to build a SQL statement that optimizes execution by concatenating hints into a dynamic SQL statement. This technique enables you change the hints based on your current database statistics without recompiling. The following sample procedure uses a variable called p_hint to allow users to pass a hint option to the SELECT statement. Example 8–3 Concatenating Hints CREATE OR REPLACE PROCEDURE query_emp (p_hint VARCHAR2) IS TYPE cur_typ IS REF CURSOR; v_emp_cursor cur_typ; BEGIN OPEN v_emp_cursor FOR 'SELECT ' || p_hint ||' empno, ename, sal, job FROM emp WHERE empno = 7566'; -- process ... CLOSE v_emp_cursor; END; / In Example 8–3, the user can pass values such as the following for p_hint: p_hint = '/*+ ALL_ROWS */' p_hint = '/*+ FIRST_ROWS */' p_hint = '/*+ CHOOSE */' See Also: Oracle Database Performance Tuning Guide to learn more about using hints Executing Dynamic PL/SQL Blocks You can use the EXECUTE IMMEDIATE statement to execute anonymous PL/SQL blocks. In this way you can add flexibility by constructing the contents of the block at runtime. For example, suppose that you want to write an application that takes an event number and dispatches it to a handler for the event. The name of the handler is in the form EVENT_HANDLER_event_num, where event_num is the number of the event. One approach is to implement the dispatcher as a switch statement, where the code handles each event by making a static call to its appropriate handler. This code is not very extensible because the dispatcher code must be updated whenever a handler for a new event is added. Example 8–4 Event Dispatching with Static SQL CREATE OR REPLACE PROCEDURE event_handler_1 (p_handle NUMBER) IS BEGIN -- process event 1 RETURN; END; / Coding Dynamic SQL 8-5 Why Use Dynamic SQL? CREATE OR REPLACE PROCEDURE event_handler_2 (p_handle NUMBER) IS BEGIN -- process event 2 RETURN; END; / CREATE OR REPLACE PROCEDURE event_handler_3 (p_handle NUMBER) IS BEGIN -- process event 3 RETURN; END; / CREATE OR REPLACE PROCEDURE event_dispatcher (p_event_num NUMBER, p_handle NUMBER) IS BEGIN IF (p_event_num = 1) THEN EVENT_HANDLER_1(p_handle); ELSIF (p_event_num = 2) THEN EVENT_HANDLER_2(p_handle); ELSIF (p_event_num = 3) THEN EVENT_HANDLER_3(p_handle); END IF; END; / By using native dynamic SQL, you can write a smaller, more flexible event dispatcher as shown in the following example. Example 8–5 Event Dispatching with Native Dynamic SQL CREATE OR REPLACE PROCEDURE event_dispatcher (p_event_num NUMBER, p_handle NUMBER) IS BEGIN EXECUTE IMMEDIATE 'BEGIN EVENT_HANDLER_' || TO_CHAR(p_event_num) || '(:1); END;' USING p_handle; END; / Performing Dynamic Operations Using Invoker's Rights By using the invoker's rights feature with dynamic SQL, you can build applications that issue dynamic SQL statements under the privileges and schema of the invoker. These two features—invoker's rights and dynamic SQL—enable you to build reusable application subcomponents that can operate on and access the invoker's data and modules. See Also: Oracle Database PL/SQL User's Guide and Reference to learn about invoker's rights and native dynamic SQL 8-6 Oracle Database Application Developer’s Guide - Fundamentals Developing with Native Dynamic SQL: Scenario Developing with Native Dynamic SQL: Scenario This scenario shows you how to perform the following operations with native dynamic SQL: ■ Execute DDL and DML operations. ■ Execute single row and multiple row queries. The database in this scenario is used for human resources. A master table named offices contains the list of all company locations. The offices table has the following definition: Column Name LOCATION Null? NOT_NULL Type VARCHAR2(200) Multiple emp_location tables contain the employee information, where location is the name of city where the office is located. For example, a table named emp_houston contains employee information for the company's Houston office, whereas a table named emp_boston contains employee information for the company's Boston office. Each emp_location table has the following definition: Column Name EMPNO ENAME JOB SAL DEPTNO Null? NOT_NULL NOT_NULL NOT_NULL NOT_NULL NOT_NULL Type NUMBER(4) VARCHAR2(10) VARCHAR2(9) NUMBER(7,2) NUMBER(2) The following sections describe various native dynamic SQL operations that can be performed on the data in the hr database. Sample DML Operation Using Native Dynamic SQL The following native dynamic SQL procedure gives a raise to all employees with a particular job title. Example 8–6 Performing DML in Native Dynamic SQL CREATE OR REPLACE PROCEDURE salary_raise (p_raise_percent NUMBER, p_job VARCHAR2) IS TYPE loc_array_type IS TABLE OF VARCHAR2(40) INDEX BY binary_integer; v_dml_str VARCHAR2(200); v_loc_array loc_array_type; BEGIN -- bulk fetch the list of office locations SELECT location BULK COLLECT INTO v_loc_array FROM offices; -- for each location, give a raise to employees with the given 'job' FOR i IN v_loc_array.first..v_loc_array.last LOOP v_dml_str := 'UPDATE emp_' || v_loc_array(i) || ' SET sal = sal * (1+(:p_raise_percent/100))' || ' WHERE p_job = :g_job_title'; EXECUTE IMMEDIATE v_dml_str USING p_raise_percent, p_job; END LOOP; END; / Coding Dynamic SQL 8-7 Developing with Native Dynamic SQL: Scenario SHOW ERRORS; Sample DDL Operation Using Native Dynamic SQL The EXECUTE IMMEDIATE statement can perform DDL operations. For example, the following procedure adds an office location. Example 8–7 Creating a Table with Native Dynamic SQL CREATE OR REPLACE PROCEDURE add_location (p_loc VARCHAR2) IS BEGIN -- Insert new location in master table, for example, detroit INSERT INTO offices VALUES (p_loc); -- Create an employee information table EXECUTE IMMEDIATE -- Use the parameter value to concatenate the table name, as in emp_detroit 'CREATE TABLE emp_' || p_loc || '( empno NUMBER(4) NOT NULL, ename VARCHAR2(10), job VARCHAR2(9), sal NUMBER(7,2), deptno NUMBER(2) )'; END; / SHOW ERRORS; The following procedure uses the same concatenation technique to drop a table. Example 8–8 Dropping a Table with Native Dynamic SQL CREATE OR REPLACE PROCEDURE drop_location (p_loc VARCHAR2) IS BEGIN -- Drop the employee table for location 'p_loc', for example, emp_detroit EXECUTE IMMEDIATE 'DROP TABLE ' || 'emp_' || p_loc; -- Remove location from master table DELETE FROM offices WHERE location = p_loc; END; / SHOW ERRORS; Sample Single-Row Query Using Native Dynamic SQL The EXECUTE IMMEDIATE statement can perform dynamic single-row queries. You can specify bind variables in the USING clause and fetch the resulting row into the target specified in the INTO clause of the statement. The following function retrieves the number of employees at a particular location performing a specified job. Example 8–9 Performing Single-Row Queries in Native Dynamic SQL CREATE OR REPLACE FUNCTION get_num_of_employees (p_loc VARCHAR2, p_job VARCHAR2) RETURN NUMBER IS v_query_str VARCHAR2(1000); 8-8 Oracle Database Application Developer’s Guide - Fundamentals Choosing Between Native Dynamic SQL and the DBMS_SQL Package v_num_of_employees NUMBER; BEGIN -- Use concatenation to form the table name in the SELECT statement v_query_str := 'SELECT COUNT(*) FROM emp_' || p_loc || ' WHERE job = :1'; -- Execute the query and put the result row in a variable EXECUTE IMMEDIATE v_query_str INTO v_num_of_employees USING p_job; RETURN v_num_of_employees; END; / SHOW ERRORS; Sample Multiple-Row Query with Native Dynamic SQL The OPEN-FOR, FETCH, and CLOSE statements can perform dynamic multiple-row queries. For example, the following procedure lists all of the employees with a particular job at a specified location. Example 8–10 Performing Multiple-Row Queries with Dynamic SQL CREATE OR REPLACE PROCEDURE list_employees (p_loc VARCHAR2, p_job VARCHAR2) IS TYPE cur_typ IS REF CURSOR; -- Define a cursor variable v_emp_cursor cur_typ; v_query_str VARCHAR2(1000); v_emp_name VARCHAR2(20); v_emp_num NUMBER; BEGIN -- Use concatenation to form the SELECT statement v_query_str := 'SELECT ename, empno FROM emp_' || p_loc || ' WHERE job = :g_job_title'; -- Open a cursor variable for the query OPEN v_emp_cursor FOR v_query_str USING p_job; -- Loop through each row to find employees who perform the specified job LOOP -- Fetch the employee name and ID into variables FETCH v_emp_cursor INTO v_emp_name, v_emp_num; EXIT WHEN v_emp_cursor%NOTFOUND; -- Process row here END LOOP; CLOSE v_emp_cursor; END; / SHOW ERRORS; Choosing Between Native Dynamic SQL and the DBMS_SQL Package Oracle Database provides two methods for using dynamic SQL within PL/SQL: native dynamic SQL and the DBMS_SQL package. Each method has advantages and disadvantages. The following sections provide detailed information about the advantages of both methods. Coding Dynamic SQL 8-9 Choosing Between Native Dynamic SQL and the DBMS_SQL Package Native dynamic SQL enables you to place dynamic SQL statements directly into PL/SQL code. These dynamic statements include the following: ■ Queries and DML statements ■ PL/SQL anonymous blocks ■ DDL statements ■ Transaction control statements ■ Session control statements To process most native dynamic SQL statements, use the EXECUTE IMMEDIATE statement. To process a multi-row SELECT statement, use OPEN-FOR, FETCH, and CLOSE statements. Note: To use native dynamic SQL, you must set the COMPATIBLE initialization parameter to 8.1.0 or higher. As an alternative to native dynamic SQL, the DBMS_SQL package offers a PL/SQL API to execute dynamic SQL statements. For example, the DBMS_SQL package contains procedures to do the following: ■ Open a cursor ■ Parse a cursor ■ Supply binds Programs that use the DBMS_SQL package make calls to this package to perform dynamic SQL operations. See Also: ■ ■ ■ Oracle Database PL/SQL User's Guide and Reference to learn about native dynamic SQL Oracle Database PL/SQL Packages and Types Reference to learn about the DBMS_SQL package Oracle Database Upgrade Guide to learn about the COMPATIBLE initialization parameter Advantages of Native Dynamic SQL Native dynamic SQL provides the following advantages over the DBMS_SQL package: ■ Native Dynamic SQL is Easy to Use ■ Native Dynamic SQL is Faster than DBMS_SQL ■ Native Dynamic SQL Supports User-Defined Types ■ Native Dynamic SQL Supports Fetching into Records Native Dynamic SQL is Easy to Use Because native dynamic SQL is integrated with SQL, you can use it in the same way that you use static SQL within PL/SQL code. Native dynamic SQL code is typically more compact and readable than equivalent code that uses the DBMS_SQL package. 8-10 Oracle Database Application Developer’s Guide - Fundamentals Choosing Between Native Dynamic SQL and the DBMS_SQL Package With the DBMS_SQL package you must call many procedures and functions in a strict sequence, which means that even simple operations require extensive code. You can avoid this complexity by using native dynamic SQL instead. Table 8–1 illustrates the difference in the amount of code required to perform the same operation with the DBMS_SQL package and native dynamic SQL. Table 8–1 Code Comparison of DBMS_SQL Package and Native Dynamic SQL DBMS_SQL Package Native Dynamic SQL CREATE OR REPLACE PROCEDURE insert_into_table (p_table_name VARCHAR2, p_deptnumber NUMBER, p_deptname VARCHAR2, p_location VARCHAR2) IS v_cur_hdl INTEGER; v_stmt_str VARCHAR2(200); v_rows_processed BINARY_INTEGER; CREATE OR REPLACE PROCEDURE insert_into_table (p_table_name VARCHAR2, p_deptnumber NUMBER, p_deptname VARCHAR2, p_location VARCHAR2) IS v_stmt_str VARCHAR2(200); BEGIN v_stmt_str := 'INSERT INTO ' || p_table_name || ' VALUES (:g_deptno, :g_dname, :g_loc)'; v_cur_hdl := DBMS_SQL.OPEN_CURSOR; DBMS_SQL.PARSE(v_cur_hdl, v_stmt_str, DBMS_SQL.NATIVE); DBMS_SQL.BIND_VARIABLE (v_cur_hdl,':g_deptno', p_deptnumber); DBMS_SQL.BIND_VARIABLE (v_cur_hdl, ':g_dname', p_deptname); DBMS_SQL.BIND_VARIABLE (v_cur_hdl, ':g_loc', p_location); v_rows_processed := DBMS_SQL.EXECUTE(v_cur_hdl); DBMS_SQL.CLOSE_CURSOR(v_cur_hdl); END; / BEGIN v_stmt_str := 'INSERT INTO ' || p_table_name || ' VALUES (:g_deptno, :g_dname, :g_loc)'; EXECUTE IMMEDIATE v_stmt_str USING p_deptnumber, p_deptname, p_location; END; / Native Dynamic SQL is Faster than DBMS_SQL Native dynamic SQL in PL/SQL performs comparably to static SQL because the PL/SQL interpreter has built-in support. Programs that use native dynamic SQL are much faster than programs that use the DBMS_SQL package. Typically, native dynamic SQL statements perform 1.5 to 3 times better than equivalent DBMS_SQL calls. Of course, performance gains may vary depending on your application. Native dynamic SQL bundles the statement preparation, binding, and execution steps into a single operation, which minimizes the data copying and procedure call overhead and improves performance. The DBMS_SQL package is based on a procedural API and incurs high procedure call and data copy overhead. Each time you bind a variable, the DBMS_SQL package copies the PL/SQL bind variable into its space for use during execution. Each time you execute a fetch, the data is copied into the space managed by the DBMS_SQL package and then the fetched data is copied, one column at a time, into the appropriate PL/SQL variables, resulting in substantial overhead. Improving Performance Through Bind Variables When using either native dynamic SQL or the DBMS_SQL package, you can improve performance by using bind variables because bind variables allow Oracle Database to share a single cursor for multiple SQL statements. Coding Dynamic SQL 8-11 Choosing Between Native Dynamic SQL and the DBMS_SQL Package In Example 8–11 the native dynamic SQL code uses a parameter instead of a bind variable to construct the SQL statement. Example 8–11 Using Native Dynamic SQL Without Bind Variables CREATE OR REPLACE PROCEDURE del_dept (p_department_id departments.department_id%TYPE) IS BEGIN EXECUTE IMMEDIATE 'DELETE FROM departments WHERE department_id = ' || TO_CHAR(p_department_id); END; / SHOW ERRORS; For each distinct p_department_id parameter, the procedure creates a new cursor, which causes resource contention and poor performance. Instead, you can construct the SQL statement by using a bind variable, as shown in Example 8–12. Example 8–12 Using Native Dynamic SQL with Bind Variables CREATE OR REPLACE PROCEDURE del_dept (p_department_id departments.department_id%TYPE) IS BEGIN EXECUTE IMMEDIATE 'DELETE FROM departments WHERE department_id = :1' USING p_department_id; END; / SHOW ERRORS; In Example 8–12 the same cursor is reused for different values of the bind my_deptno, which improves performance and scalability. Native Dynamic SQL Supports User-Defined Types Native dynamic SQL supports all of the types supported by static SQL in PL/SQL, including user-defined types such as user-defined objects, collections, and REFs. The DBMS_SQL package does not support these user-defined types. The DBMS_SQL package provides limited support for arrays. Refer to the Oracle Database PL/SQL Packages and Types Reference for information. Note: Native Dynamic SQL Supports Fetching into Records Native dynamic SQL and static SQL both support fetching into records, but the DBMS_SQL package does not. With native dynamic SQL, the rows resulting from a query can be directly fetched into PL/SQL records. In Example 8–13 the rows from a query are fetched into the v_emp_rec variable. Example 8–13 Using Native Dynamic SQL to Fetch into Records DECLARE TYPE EmpCurTyp v_emp_cursor emp_record v_stmt_str v_e_job IS REF CURSOR; EmpCurTyp; emp%ROWTYPE; VARCHAR2(200); emp.job%TYPE; 8-12 Oracle Database Application Developer’s Guide - Fundamentals Choosing Between Native Dynamic SQL and the DBMS_SQL Package BEGIN v_stmt_str := 'SELECT * FROM emp WHERE job = :1'; -- in a multi-row query OPEN v_emp_cursor FOR v_stmt_str USING 'MANAGER'; LOOP FETCH v_emp_cursor INTO emp_record; EXIT WHEN v_emp_cursor%NOTFOUND; END LOOP; CLOSE v_emp_cursor; -- in a single-row query EXECUTE IMMEDIATE v_stmt_str INTO emp_record USING 'PRESIDENT'; END; / Advantages of the DBMS_SQL Package The DBMS_SQL package provides the following advantages over native dynamic SQL: ■ DBMS_SQL is Supported in Client-Side Programs ■ DBMS_SQL Supports Statements with Unknown Number of Inputs or Outputs ■ DBMS_SQL Supports SQL Statements Larger than 32 KB ■ DBMS_SQL Lets You Reuse SQL Statements DBMS_SQL is Supported in Client-Side Programs The DBMS_SQL package is supported in client-side programs, but native dynamic SQL is not. Every call to the DBMS_SQL package from the client-side program translates to a PL/SQL remote procedure call (RPC). These calls occur when you need to do any of the following: ■ Bind a variable ■ Define a variable ■ Execute a statement DBMS_SQL Supports Statements with Unknown Number of Inputs or Outputs Native dynamic SQL does not support statements with an unknown number of inputs or outputs. The DBMS_SQL package does not have this limitation. One consequence is that you can use the DESCRIBE_COLUMNS procedure in the DBMS_SQL package to describe columns for a cursor opened and parsed through DBMS_SQL. This feature is similar to the DESCRIBE command in SQL*Plus. Native dynamic SQL does not have a DESCRIBE facility. See Also: Oracle Database PL/SQL Packages and Types Reference for an example of using DESCRIBE_COLUMNS to create a query in a situation where the SELECT list is not known until runtime DBMS_SQL Supports SQL Statements Larger than 32 KB The DBMS_SQL package supports SQL statements larger than 32 KB. Native dynamic SQL does not. DBMS_SQL Lets You Reuse SQL Statements The PARSE procedure in the DBMS_SQL package parses a SQL statement once. After the initial parsing, you can use the statement multiple times with different sets of bind arguments. Coding Dynamic SQL 8-13 Choosing Between Native Dynamic SQL and the DBMS_SQL Package Native dynamic SQL prepares a SQL statement each time the statement is used, which typically involves parsing, optimization, and plan generation. Although the extra prepare operations incur a small performance penalty, the decrease in speed is typically outweighed by the performance benefits of native dynamic SQL. Examples of DBMS_SQL Package Code and Native Dynamic SQL Code The following examples illustrate the code differences necessary to complete operations with the DBMS_SQL package and native dynamic SQL. Specifically, the following types of examples are presented: ■ Query ■ DML operation ■ DML returning operation In general, the native dynamic SQL code is more readable and compact, which can improve developer productivity. Querying with Dynamic SQL: Example The following example includes a dynamic query statement with one bind variable (:g_jobname) and two select columns (ename and sal): v_stmt_str := 'SELECT ename, sal FROM emp WHERE job = :g_jobname'; This example queries for employees with the job description SALESMAN in the job column of the emp table. Table 8–2 shows sample code that accomplishes this query using the DBMS_SQL package and native dynamic SQL. 8-14 Oracle Database Application Developer’s Guide - Fundamentals Choosing Between Native Dynamic SQL and the DBMS_SQL Package Table 8–2 Querying Using the DBMS_SQL Package and Native Dynamic SQL DBMS_SQL Query Operation Native Dynamic SQL Query Operation DECLARE v_stmt_str VARCHAR2(200); v_cur_hdl INT; v_rows_processed INT; v_name VARCHAR2(10); v_salary INT; BEGIN v_cur_hdl := DBMS_SQL.OPEN_CURSOR; -- open cursor v_stmt_str := 'SELECT ename, sal FROM emp WHERE job = :g_jobname'; DBMS_SQL.PARSE(v_cur_hdl,v_stmt_str,DBMS_SQL.NATIVE); -- Supply binds (bind by name) DBMS_SQL.BIND_VARIABLE(v_cur_hdl, 'g_jobname', 'SALESMAN'); -- Describe defines DBMS_SQL.DEFINE_COLUMN(v_cur_hdl, 1, v_name, 200); DBMS_SQL.DEFINE_COLUMN(v_cur_hdl, 2, v_salary); -- Execute v_rows_processed := DBMS_SQL.EXECUTE(v_cur_hdl); LOOP -- Fetch a row IF DBMS_SQL.FETCH_ROWS(v_cur_hdl) > 0 THEN -- Fetch columns from the row DBMS_SQL.COLUMN_VALUE(v_cur_hdl, 1, v_name); DBMS_SQL.COLUMN_VALUE(v_cur_hdl, 2, v_salary); -- Process ELSE EXIT; END IF; END LOOP; DBMS_SQL.CLOSE_CURSOR(v_cur_hdl); -- close cursor END; / DECLARE TYPE EmpCurTyp IS REF CURSOR; v_emp_cursor EmpCurTyp; v_stmt_str VARCHAR2(200); v_name VARCHAR2(20); v_salary NUMBER; BEGIN v_stmt_str := 'SELECT ename, sal FROM emp WHERE job = :1'; OPEN v_emp_cursor FOR v_stmt_str USING 'SALESMAN'; LOOP FETCH v_emp_cursor INTO v_name, v_salary; EXIT WHEN v_emp_cursor%NOTFOUND; -- Process data END LOOP; CLOSE v_emp_cursor; END; / Performing DML with Dynamic SQL: Example The following example includes a dynamic INSERT statement for a table with three columns. v_stmt_str := 'INSERT INTO dept_new VALUES (:g_deptno, :g_dname, :g_loc)'; This example inserts a new row for which the column values are in the PL/SQL variables deptnumber, deptname, and location. Table 8–3 shows sample code that accomplishes this task with the DBMS_SQL package and native dynamic SQL. Coding Dynamic SQL 8-15 Choosing Between Native Dynamic SQL and the DBMS_SQL Package Table 8–3 DML Operation Using the DBMS_SQL Package and Native Dynamic SQL DBMS_SQL DML Operation Native Dynamic SQL DML Operation DECLARE v_stmt_str VARCHAR2(200); v_cur_hdl NUMBER; v_deptnumber NUMBER := 99; v_deptname VARCHAR2(20); v_location VARCHAR2(10); v_rows_processed NUMBER; BEGIN v_stmt_str := 'INSERT INTO dept VALUES (:g_deptno, :g_dname, :g_loc)'; v_cur_hdl := DBMS_SQL.OPEN_CURSOR; DBMS_SQL.PARSE(v_cur_hdl, v_stmt_str, DBMS_SQL.NATIVE); -- Supply binds DBMS_SQL.BIND_VARIABLE (v_cur_hdl, ':g_deptno', v_deptnumber); DBMS_SQL.BIND_VARIABLE (v_cur_hdl, ':g_dname', v_deptname); DBMS_SQL.BIND_VARIABLE (v_cur_hdl, ':g_loc', v_location); v_rows_processed := DBMS_SQL.EXECUTE(v_cur_hdl); DBMS_SQL.CLOSE_CURSOR(v_cur_hdl); END; / DECLARE v_stmt_str v_deptnumber v_deptname v_location VARCHAR2(200); NUMBER := 99; VARCHAR2(20); VARCHAR2(10); BEGIN v_stmt_str := 'INSERT INTO dept VALUES (:g_deptno, :g_dname, :g_loc)'; EXECUTE IMMEDIATE v_stmt_str USING v_deptnumber, v_deptname, v_location; END; / Performing DML with RETURNING Clause Using Dynamic SQL: Example The following example uses a dynamic UPDATE statement to update the location of a department, then returns the name of the department: v_stmt_str := 'UPDATE dept_new SET loc = :g_newloc WHERE deptno = :g_deptno RETURNING dname INTO :g_dname'; Table 8–4 shows sample code that accomplishes this operation using both the DBMS_SQL package and native dynamic SQL. 8-16 Oracle Database Application Developer’s Guide - Fundamentals Avoiding SQL Injection in PL/SQL Table 8–4 DML Returning Operation Using the DBMS_SQL Package and Native Dynamic SQL DBMS_SQL DML Returning Operation Native Dynamic SQL DML Returning Operation DECLARE deptname_array DBMS_SQL.VARCHAR2_TABLE; v_cur_hdl INT; v_stmt_str VARCHAR2(200); v_location VARCHAR2(20); v_deptnumber NUMBER := 10; v_rows_procsd NUMBER; BEGIN v_stmt_str := 'UPDATE dept SET loc = :g_newloc WHERE deptno = :g_deptno RETURNING dname INTO :g_dname'; v_cur_hdl := DBMS_SQL.OPEN_CURSOR; DBMS_SQL.PARSE (v_cur_hdl, v_stmt_str, DBMS_SQL.NATIVE); -- Supply binds DBMS_SQL.BIND_VARIABLE (v_cur_hdl, ':g_newloc', v_location); DBMS_SQL.BIND_VARIABLE (v_cur_hdl, ':g_deptno', v_deptnumber); DBMS_SQL.BIND_ARRAY (v_cur_hdl, ':g_dname', deptname_array); -- Execute cursor v_rows_procsd := DBMS_SQL.EXECUTE(v_cur_hdl); -- Get RETURNING column into OUT bind array DBMS_SQL.VARIABLE_VALUE (v_cur_hdl, ':g_dname', deptname_array); DBMS_SQL.CLOSE_CURSOR(v_cur_hdl); END; / DECLARE deptname_array v_stmt_str v_location v_deptnumber v_deptname DBMS_SQL.VARCHAR2_TABLE; VARCHAR2(200); VARCHAR2(20); NUMBER := 10; VARCHAR2(20); BEGIN v_stmt_str := 'UPDATE dept SET loc = :g_newloc WHERE deptno = :g_deptno RETURNING dname INTO :g_dname'; EXECUTE IMMEDIATE v_stmt_str USING v_location, v_deptnumber, OUT v_deptname; END; / Avoiding SQL Injection in PL/SQL SQL injection is a technique for maliciously exploiting applications that use client-supplied data in SQL statements. The purpose of this technique is to gain unauthorized access to a database in order to query or manipulate restricted data. This section describes SQL injection vulnerabilities in PL/SQL and explains how you can guard against them. This section contains the following topics: ■ Overview of SQL Injection Techniques ■ Guarding Against SQL Injection Overview of SQL Injection Techniques Although SQL injection techniques differ, they all exploit a single vulnerability: string input is not correctly validated and is concatenated into a dynamic SQL statement. For the purposes of this discussion, SQL injection attacks can be classified as follows: ■ Statement Modification ■ Statement Injection The following sections describe these techniques. To try the examples in your sample database, run the script in Example 8–14. Coding Dynamic SQL 8-17 Avoiding SQL Injection in PL/SQL Example 8–14 Setup for Injection Examples CONNECT hr/hr SET SERVEROUTPUT ON SET LINESIZE 150 SET ECHO OFF DROP TABLE user_pwd; CREATE TABLE user_pwd( username VARCHAR2(100), password VARCHAR2(100) ); INSERT INTO user_pwd VALUES('whitehat', 'secret'); DROP TABLE delemp; CREATE TABLE delemp AS SELECT * FROM employees; COMMIT; Statement Modification SQL modification involves deliberately altering a dynamic SQL statement so that it executes in a way unintended by the application developer. Typically, the user retrieves unauthorized data by changing the WHERE clause of a query or by inserting a UNION ALL clause. The classic example of this technique is bypassing password authentication by making a WHERE clause always TRUE. Suppose a Web form prompts a user to enter a username and password. When the user clicks Submit, the form invokes a PL/SQL stored procedure that concatenates the username and password entered in the form to build a dynamic SQL statement. The procedure executes the query to authenticate the user. Example 8–15 illustrates an authentication procedure that you can test in SQL*Plus. The ckpwd procedure uses concatenation to build a SQL query of the user_pwd table. If the user enters a username and password stored in the table, then the execution of the query retrieves a single row and the user is authenticated. The ckpwd procedure also displays the concatenated query so that you can see which query is executed. Example 8–15 ckpwd Procedure CREATE OR REPLACE PROCEDURE ckpwd (p_user IN VARCHAR2, p_pass IN VARCHAR2) IS v_query VARCHAR2(100); v_output NUMBER; BEGIN v_query := q'{SELECT COUNT(*) FROM user_pwd }' || q'{WHERE username = '}' || p_user || q'{' AND password = '}' || p_pass || q'{'}'; DBMS_OUTPUT.PUT_LINE(CHR(10)||'Built the following query:'||CHR(10)||v_query); EXECUTE IMMEDIATE v_query INTO v_output; IF v_output = 1 THEN DBMS_OUTPUT.PUT_LINE(CHR(10)||p_user||' is authenticated'); ELSE DBMS_OUTPUT.PUT_LINE(CHR(10)||'access denied'); END IF; END; / Suppose that the user whitehat enters the password secret in a Web form. You can simulate this scenario by invoking the code shown in Example 8–16 in SQL*Plus to authenticate whitehat (sample output included). 8-18 Oracle Database Application Developer’s Guide - Fundamentals Avoiding SQL Injection in PL/SQL Example 8–16 Authenticating a User with the ckpwd Procedure BEGIN ckpwd ( p_user => q'{whitehat}', p_pass => q'{secret}' ); END; / Built the following query: SELECT COUNT(*) FROM user_pwd WHERE username = 'whitehat' AND password = 'secret' whitehat is authenticated A malicious user could exploit the concatenation vulnerability and enter the username x in the Web-based form and the text shown in Example 8–17 as a password. Example 8–17 Password String Entered in Form x' OR 'x' = 'x You can simulate this scenario by executing the code in Example 8–18 in SQL*Plus (sample output included). Example 8–18 Performing Statement Modification with the ckpwd Procedure BEGIN ckpwd ( p_user => q'{x}', p_pass => q'{x' OR 'x' = 'x}' ); END; / Built the following query: SELECT COUNT(*) FROM user_pwd WHERE username = 'x' AND password = 'x' OR 'x' = 'x' x is authenticated By using the cleverly constructed password in Example 8–17, you alter the concatenated SQL statement so that the OR condition always returns TRUE. Thus, the query of the user_pwd table always succeeds no matter which username is entered. Statement Injection Statement injection occurs when a user appends one or more new SQL statements to a dynamically generated SQL statement. Anonymous PL/SQL blocks are vulnerable to this technique. Suppose a Web form prompts a user to enter a username and password. When the user clicks Submit, the form invokes a PL/SQL stored procedure that concatenates the username and password entered in the form into an anonymous PL/SQL block. The procedure then executes the anonymous block to authenticate the user. Example 8–19 illustrates an authentication procedure that you can test in SQL*Plus. The call_ckpwd procedure uses concatenation to build an anonymous block that invokes the ckpwd procedure from Example 8–15. If the user enters a username and password stored in the user_pwd table, then the execution of the block retrieves a Coding Dynamic SQL 8-19 Avoiding SQL Injection in PL/SQL single row and the user is authenticated. The call_ckpwd procedure also prints the concatenated text so that you can see which block is executed. Example 8–19 call_ckpwd Procedure CREATE OR REPLACE PROCEDURE call_ckpwd (p_user IN VARCHAR2, p_pass IN VARCHAR2) IS v_block VARCHAR2(100); BEGIN v_block := q'{BEGIN ckpwd( '}' || p_user || q'{' , '}' || p_pass || q'{' ); END; }'; DBMS_OUTPUT.PUT_LINE(CHR(10)|| 'Built the following anonymous block:'||CHR(10)||v_block); EXECUTE IMMEDIATE v_block; END; / Suppose that the user whitehat enters the password secret in a Web-based form. You can simulate this scenario by invoking the call_ckpwd procedure shown in Example 8–20 in SQL*Plus (sample output included). Example 8–20 Authenticating a User with the call_ckpwd Procedure BEGIN call_ckpwd ( p_user => q'{whitehat}', p_pass => q'{secret}' ); END; / Built the following anonymous block: BEGIN ckpwd( 'whitehat' , 'secret' ); END; Built the following query: SELECT COUNT(*) FROM user_pwd WHERE username = 'whitehat' AND password = 'secret' whitehat is authenticated If whitehat turns bad, then he could enter the string shown in Example 8–21 as the password in a Web form. Example 8–21 Bogus Password Entered in Form secret'); DELETE FROM hr.delemp WHERE UPPER('x') = UPPER('x You can simulate this technique by invoking the call_ckpwd procedure shown in Example 8–22 in SQL*Plus (sample output included). Example 8–22 Performing Statement Injection with the call_ckpwd Procedure BEGIN call_ckpwd ( p_user => q'{whitehat}', 8-20 Oracle Database Application Developer’s Guide - Fundamentals Avoiding SQL Injection in PL/SQL p_pass => q'{secret'); DELETE FROM hr.delemp WHERE UPPER('x') = UPPER('x}' ); END; / Built the following anonymous block: BEGIN ckpwd( 'whitehat' , 'secret' ); DELETE FROM hr.delemp WHERE UPPER('x') = UPPER('x'); END; Built the following query: SELECT COUNT(*) FROM user_pwd WHERE username = 'whitehat' AND password = 'secret' whitehat is authenticated The bogus password in Example 8–21 causes the system to authenticate whitehat and silently execute the injected DELETE statement. A query of the delemp table shows that the injected statement silently removed all rows from the table: SELECT * FROM delemp; no rows selected Guarding Against SQL Injection If you use dynamic SQL in your PL/SQL applications, then you must check the input text to ensure that it is exactly and only what is expected. You have the following useful techniques at your disposal: ■ Using Bind Variables to Guard Against SQL Injection ■ Using Validation Checks to Guard Against SQL Injection Using Bind Variables to Guard Against SQL Injection "Improving Performance Through Bind Variables" on page 8-11 shows how you can use bind variables to improve performance in dynamic SQL. Besides improving performance, binding placeholders renders your PL/SQL code immune to SQL injection attacks. The ckpwd procedure shown in Example 8–15 used concatenation instead of bind variables. Example 8–23 shows how you could rewrite the procedure to use bind variables instead of concatenation. Example 8–23 ckpwd_bind Procedure CREATE OR REPLACE PROCEDURE ckpwd_bind (p_user IN VARCHAR2, p_pass IN VARCHAR2) IS v_query VARCHAR2(100); v_output NUMBER; BEGIN v_query := q'{SELECT COUNT(*) FROM user_pwd WHERE username = :1 AND password = :2}'; DBMS_OUTPUT.PUT_LINE(CHR(10)||'Built the following query:'||CHR(10)||v_query); EXECUTE IMMEDIATE v_query INTO v_output USING p_user, p_pass; IF v_output = 1 THEN DBMS_OUTPUT.PUT_LINE(CHR(10)||p_user||' is authenticated'); ELSE DBMS_OUTPUT.PUT_LINE(CHR(10)||'access denied'); END IF; END; Coding Dynamic SQL 8-21 Avoiding SQL Injection in PL/SQL / If the user tries to pass the bogus password shown in Example 8–17 to the ckpwd_bind procedure, then the technique fails to authenticate the user. You can execute the block shown in Example 8–24 in SQL*Plus to test the revised version of the code (sample output included). Example 8–24 Preventing Statement Modification with the ckpwd_bind Procedure BEGIN ckpwd_bind ( p_user => q'{x}', p_pass => q'{x' OR 'x' = 'x}' ); END; / Built the following query: SELECT COUNT(*) FROM user_pwd WHERE username = :1 AND password = :2 access denied The same binding technique fixes the vulnerable call_ckpwd procedure shown in Example 8–19. By using bind variables exclusively in your code, you avoid concatenating SQL statements and thereby prevent malicious users from altering or injecting additional statements. Oracle database uses the value of the bind variable exclusively and does not interpret its contents in any way. This technique is the most effective way to prevent SQL injection in PL/SQL programs. Using Validation Checks to Guard Against SQL Injection A program should always validate user input to ensure that it is what is intended. For example, if the user is passing in a department number for a DELETE statement, then check the validity of this department number by querying the departments table. Similarly, if a user enters the name of a table to be deleted, check that this table exists by querying the ALL_TABLES view. 8-22 Oracle Database Application Developer’s Guide - Fundamentals 9 Coding Triggers Triggers are procedures that are stored in the database and are implicitly run, or fired, when something happens. Traditionally, triggers supported the execution of a PL/SQL block when an INSERT, UPDATE, or DELETE occurred on a table or view. Triggers support system and other data events on DATABASE and SCHEMA. Oracle Database also supports the execution of PL/SQL or Java procedures. This chapter discusses DML triggers, INSTEAD OF triggers, and system triggers (triggers on DATABASE and SCHEMA). Topics include: ■ Designing Triggers ■ Creating Triggers ■ Coding the Trigger Body ■ Compiling Triggers ■ Modifying Triggers ■ Enabling and Disabling Triggers ■ Viewing Information About Triggers ■ Examples of Trigger Applications ■ Responding to System Events through Triggers Designing Triggers Use the following guidelines when designing your triggers: ■ ■ ■ ■ Use triggers to guarantee that when a specific operation is performed, related actions are performed. Do not define triggers that duplicate features already built into Oracle Database. For example, do not define triggers to reject bad data if you can do the same checking through declarative integrity constraints. Limit the size of triggers. If the logic for your trigger requires much more than 60 lines of PL/SQL code, it is better to include most of the code in a stored procedure and call the procedure from the trigger. Use triggers only for centralized, global operations that should be fired for the triggering statement, regardless of which user or database application issues the statement. Coding Triggers 9-1 Creating Triggers ■ ■ Do not create recursive triggers. For example, creating an AFTER UPDATE statement trigger on the Emp_tab table that itself issues an UPDATE statement on Emp_tab, causes the trigger to fire recursively until it has run out of memory. Use triggers on DATABASE judiciously. They are executed for every user every time the event occurs on which the trigger is created. Creating Triggers Triggers are created using the CREATE TRIGGER statement. This statement can be used with any interactive tool, such as SQL*Plus or Enterprise Manager. When using an interactive tool, a single slash (/) on the last line is necessary to activate the CREATE TRIGGER statement. The following statement creates a trigger for the Emp_tab table. CREATE OR REPLACE TRIGGER Print_salary_changes BEFORE DELETE OR INSERT OR UPDATE ON Emp_tab FOR EACH ROW WHEN (new.Empno > 0) DECLARE sal_diff number; BEGIN sal_diff := :new.sal - :old.sal; dbms_output.put('Old salary: ' || :old.sal); dbms_output.put(' New salary: ' || :new.sal); dbms_output.put_line(' Difference ' || sal_diff); END; / The trigger is fired when DML operations (INSERT, UPDATE, and DELETE statements) are performed on the table. You can choose what combination of operations should fire the trigger. Because the trigger uses the BEFORE keyword, it can access the new values before they go into the table, and can change the values if there is an easily-corrected error by assigning to :NEW.column_name. You might use the AFTER keyword if you want the trigger to query or change the same table, because triggers can only do that after the initial changes are applied and the table is back in a consistent state. Because the trigger uses the FOR EACH ROW clause, it might be executed multiple times, such as when updating or deleting multiple rows. You might omit this clause if you just want to record the fact that the operation occurred, but not examine the data for each row. Once the trigger is created, entering the following SQL statement: UPDATE Emp_tab SET sal = sal + 500.00 WHERE deptno = 10; fires the trigger once for each row that is updated, in each case printing the new salary, old salary, and the difference. The CREATE (or CREATE OR REPLACE) statement fails if any errors exist in the PL/SQL block. Note: The size of the trigger cannot be more than 32K. The following sections use this example to illustrate the way that parts of a trigger are specified. 9-2 Oracle Database Application Developer’s Guide - Fundamentals Creating Triggers See Also: "Examples of Trigger Applications" on page 9-23 for more realistic examples of CREATE TRIGGER statements Types of Triggers A trigger is either a stored PL/SQL block or a PL/SQL, C, or Java procedure associated with a table, view, schema, or the database itself. Oracle Database automatically executes a trigger when a specified event takes place, which may be in the form of a system event or a DML statement being issued against the table. Triggers can be: ■ DML triggers on tables. ■ INSTEAD OF triggers on views. ■ System triggers on DATABASE or SCHEMA: With DATABASE, triggers fire for each event for all users; with SCHEMA, triggers fire for each event for that specific user. Oracle Database SQL Reference for information on trigger creation syntax See Also: Overview of System Events You can create triggers to be fired on any of the following: ■ DML statements (DELETE, INSERT, UPDATE) ■ DDL statements (CREATE, ALTER, DROP) ■ Database operations (SERVERERROR, LOGON, LOGOFF, STARTUP, SHUTDOWN) Getting the Attributes of System Events You can get certain event-specific attributes when the trigger is fired. Creating a trigger on DATABASE implies that the triggering event is outside the scope of a user (for example, database STARTUP and SHUTDOWN), and it applies to all users (for example, a trigger created on LOGON event by the DBA). Creating a trigger on SCHEMA implies that the trigger is created in the current user's schema and is fired only for that user. For each trigger, publication can be specified on DML and system events. See Also: "Responding to System Events through Triggers" on page 9-37 Naming Triggers Trigger names must be unique with respect to other triggers in the same schema. Trigger names do not need to be unique with respect to other schema objects, such as tables, views, and procedures. For example, a table and a trigger can have the same name (however, to avoid confusion, this is not recommended). When Is the Trigger Fired? A trigger is fired based on a triggering statement, which specifies: ■ The SQL statement or the system event, database event, or DDL event that fires the trigger body. The options include DELETE, INSERT, and UPDATE. One, two, or all three of these options can be included in the triggering statement specification. Coding Triggers 9-3 Creating Triggers ■ The table, view, DATABASE, or SCHEMA associated with the trigger. Exactly one table or view can be specified in the triggering statement. If the INSTEAD OF option is used, then the triggering statement may only specify a view; conversely, if a view is specified in the triggering statement, then only the INSTEAD OF option may be used. Note: For example, the PRINT_SALARY_CHANGES trigger fires after any DELETE, INSERT, or UPDATE on the Emp_tab table. Any of the following statements trigger the PRINT_SALARY_CHANGES trigger given in the previous example: DELETE INSERT INSERT UPDATE FROM Emp_tab; INTO Emp_tab VALUES ( ... ); INTO Emp_tab SELECT ... FROM ... ; Emp_tab SET ... ; Do Import and SQL*Loader Fire Triggers? INSERT triggers fire during SQL*Loader conventional loads. (For direct loads, triggers are disabled before the load.) The IGNORE parameter of the IMP command determines whether triggers fire during import operations: ■ ■ ■ If IGNORE=N (default) and the table already exists, then import does not change the table and no existing triggers fire. If the table does not exist, then import creates and loads it before any triggers are defined, so again no triggers fire. If IGNORE=Y, then import loads rows into existing tables. Any existing triggers fire, and indexes are updated to account for the imported data. How Column Lists Affect UPDATE Triggers An UPDATE statement might include a list of columns. If a triggering statement includes a column list, the trigger is fired only when one of the specified columns is updated. If a triggering statement omits a column list, the trigger is fired when any column of the associated table is updated. A column list cannot be specified for INSERT or DELETE triggering statements. The previous example of the PRINT_SALARY_CHANGES trigger could include a column list in the triggering statement. For example: ... BEFORE DELETE OR INSERT OR UPDATE OF ename ON Emp_tab ... Notes: ■ ■ ■ You cannot specify a column list for UPDATE with INSTEAD OF triggers. If the column specified in the UPDATE OF clause is an object column, then the trigger is also fired if any of the attributes of the object are modified. You cannot specify UPDATE OF clauses on collection columns. Controlling When a Trigger Is Fired (BEFORE and AFTER Options) The BEFORE or AFTER option in the CREATE TRIGGER statement specifies exactly when to fire the trigger body in relation to the triggering statement that is being run. In 9-4 Oracle Database Application Developer’s Guide - Fundamentals Creating Triggers a CREATE TRIGGER statement, the BEFORE or AFTER option is specified just before the triggering statement. For example, the PRINT_SALARY_CHANGES trigger in the previous example is a BEFORE trigger. In general, you use BEFORE or AFTER triggers to achieve the following results: ■ Use BEFORE row triggers to modify the row before the row data is written to disk. ■ Use AFTER row triggers to obtain, and perform operations, using the row ID. BEFORE row triggers are slightly more efficient than AFTER row triggers. With AFTER row triggers, affected data blocks must be read (logical read, not physical read) once for the trigger and then again for the triggering statement. Alternatively, with BEFORE row triggers, the data blocks must be read only once for both the triggering statement and the trigger. Note: BEFORE Triggers Fired Multiple Times If an UPDATE or DELETE statement detects a conflict with a concurrent UPDATE, then Oracle Database performs a transparent ROLLBACK to SAVEPOINT and restarts the update. This can occur many times before the statement completes successfully. Each time the statement is restarted, the BEFORE statement trigger is fired again. The rollback to savepoint does not undo changes to any package variables referenced in the trigger. Your package should include a counter variable to detect this situation. Ordering of Triggers A relational database does not guarantee the order of rows processed by a SQL statement. Therefore, do not create triggers that depend on the order in which rows are processed. For example, do not assign a value to a global package variable in a row trigger if the current value of the global variable is dependent on the row being processed by the row trigger. Also, if global package variables are updated within a trigger, then it is best to initialize those variables in a BEFORE statement trigger. When a statement in a trigger body causes another trigger to be fired, the triggers are said to be cascading. Oracle Database allows up to 32 triggers to cascade at any one time. However, you can effectively limit the number of trigger cascades using the initialization parameter OPEN_CURSORS, because a cursor must be opened for every execution of a trigger. Trigger Evaluation Order Although any trigger can run a sequence of operations either in-line or by calling procedures, using multiple triggers of the same type enhances database administration by permitting the modular installation of applications that have triggers on the same tables. Oracle Database executes all triggers of the same type before executing triggers of a different type. If you have multiple triggers of the same type on a single table, then Oracle Database chooses an arbitrary order to execute these triggers. Oracle Database Concepts for more information on the firing order of triggers See Also: Each subsequent trigger sees the changes made by the previously fired triggers. Each trigger can see the old and new values. The old values are the original values, and the Coding Triggers 9-5 Creating Triggers new values are the current values, as set by the most recently fired UPDATE or INSERT trigger. To ensure that multiple triggered actions occur in a specific order, you must consolidate these actions into a single trigger (for example, by having the trigger call a series of procedures). Modifying Complex Views (INSTEAD OF Triggers) An updatable view is one that lets you perform DML on the underlying table. Some views are inherently updatable, but others are not because they were created with one or more of the constructs listed in"Views that Require INSTEAD OF Triggers". Any view that contains one of those constructs can be made updatable by using an INSTEAD OF trigger. INSTEAD OF triggers provide a transparent way of modifying views that cannot be modified directly through UPDATE, INSERT, and DELETE statements. These triggers are called INSTEAD OF triggers because, unlike other types of triggers, Oracle Database fires the trigger instead of executing the triggering statement. The trigger must determine what operation was intended and perform UPDATE, INSERT, or DELETE operations directly on the underlying tables. With an INSTEAD OF trigger, you can write normal UPDATE, INSERT, and DELETE statements against the view, and the INSTEAD OF trigger works invisibly in the background to make the right actions take place. INSTEAD OF triggers can only be activated for each row. See Also: "Firing Triggers One or Many Times (FOR EACH ROW Option)" on page 9-9 Note: ■ ■ ■ The INSTEAD OF option can only be used for triggers created over views. The BEFORE and AFTER options cannot be used for triggers created over views. The CHECK option for views is not enforced when inserts or updates to the view are done using INSTEAD OF triggers. The INSTEAD OF trigger body must enforce the check. Views that Require INSTEAD OF Triggers A view cannot be modified by UPDATE, INSERT, or DELETE statements if the view query contains any of the following constructs: ■ A set operator ■ A DISTINCT operator ■ An aggregate or analytic function ■ A GROUP BY, ORDER BY, MODEL, CONNECT BY, or START WITH clause ■ A collection expression in a SELECT list ■ A subquery in a SELECT list ■ A subquery designated WITH READ ONLY 9-6 Oracle Database Application Developer’s Guide - Fundamentals Creating Triggers ■ Joins, with some exceptions, as documented in Oracle Database Administrator's Guide If a view contains pseudocolumns or expressions, then you can only update the view with an UPDATE statement that does not refer to any of the pseudocolumns or expressions. INSTEAD OF Trigger Example You may need to set up the following data structures for this example to work: Note: CREATE TABLE Prj_level Projno Resp_dept CREATE TABLE Empno Ename Job Mgr Hiredate Sal Comm Deptno Project_tab ( NUMBER, NUMBER, NUMBER); Emp_tab ( NUMBER NOT NULL, VARCHAR2(10), VARCHAR2(9), NUMBER(4), DATE, NUMBER(7,2), NUMBER(7,2), NUMBER(2) NOT NULL); CREATE TABLE Deptno Dname Loc Mgr_no Dept_type Dept_tab ( NUMBER(2) NOT NULL, VARCHAR2(14), VARCHAR2(13), NUMBER, NUMBER); The following example shows an INSTEAD OF trigger for inserting rows into the MANAGER_INFO view. CREATE OR REPLACE VIEW manager_info AS SELECT e.ename, e.empno, d.dept_type, d.deptno, p.prj_level, p.projno FROM Emp_tab e, Dept_tab d, Project_tab p WHERE e.empno = d.mgr_no AND d.deptno = p.resp_dept; CREATE OR REPLACE TRIGGER manager_info_insert INSTEAD OF INSERT ON manager_info REFERENCING NEW AS n -- new manager information FOR EACH ROW DECLARE rowcnt number; BEGIN SELECT COUNT(*) INTO rowcnt FROM Emp_tab WHERE empno = :n.empno; IF rowcnt = 0 THEN INSERT INTO Emp_tab (empno,ename) VALUES (:n.empno, :n.ename); ELSE UPDATE Emp_tab SET Emp_tab.ename = :n.ename WHERE Emp_tab.empno = :n.empno; END IF; SELECT COUNT(*) INTO rowcnt FROM Dept_tab WHERE deptno = :n.deptno; Coding Triggers 9-7 Creating Triggers IF rowcnt = 0 THEN INSERT INTO Dept_tab (deptno, dept_type) VALUES(:n.deptno, :n.dept_type); ELSE UPDATE Dept_tab SET Dept_tab.dept_type = :n.dept_type WHERE Dept_tab.deptno = :n.deptno; END IF; SELECT COUNT(*) INTO rowcnt FROM Project_tab WHERE Project_tab.projno = :n.projno; IF rowcnt = 0 THEN INSERT INTO Project_tab (projno, prj_level) VALUES(:n.projno, :n.prj_level); ELSE UPDATE Project_tab SET Project_tab.prj_level = :n.prj_level WHERE Project_tab.projno = :n.projno; END IF; END; The actions shown for rows being inserted into the MANAGER_INFO view first test to see if appropriate rows already exist in the base tables from which MANAGER_INFO is derived. The actions then insert new rows or update existing rows, as appropriate. Similar triggers can specify appropriate actions for UPDATE and DELETE. Object Views and INSTEAD OF Triggers INSTEAD OF triggers provide the means to modify object view instances on the client-side through OCI calls. See Also: Oracle Call Interface Programmer's Guide To modify an object materialized by an object view in the client-side object cache and flush it back to the persistent store, you must specify INSTEAD OF triggers, unless the object view is modifiable. If the object is read only, then it is not necessary to define triggers to pin it. Triggers on Nested Table View Columns INSTEAD OF triggers can also be created over nested table view columns. These triggers provide a way of updating elements of the nested table. They fire for each nested table element being modified. The row correlation variables inside the trigger correspond to the nested table element. This type of trigger also provides an additional correlation name for accessing the parent row that contains the nested table being modified. Note: ■ ■ These triggers: Can only be defined over nested table columns in views. Fire only when the nested table elements are modified using the THE() or TABLE() clauses. They do not fire when a DML statement is performed on the view. For example, consider a department view that contains a nested table of employees. CREATE OR REPLACE VIEW Dept_view AS SELECT d.Deptno, d.Dept_type, d.Dept_name, CAST (MULTISET ( SELECT e.Empno, e.Empname, e.Salary) FROM Emp_tab e 9-8 Oracle Database Application Developer’s Guide - Fundamentals Creating Triggers WHERE e.Deptno = d.Deptno) AS Amp_list_ Emplist FROM Dept_tab d; The CAST (MULTISET..) operator creates a multi-set of employees for each department. If you want to modify the emplist column, which is the nested table of employees, then you can define an INSTEAD OF trigger over the column to handle the operation. The following example shows how an insert trigger might be written: CREATE OR REPLACE TRIGGER Dept_emplist_tr INSTEAD OF INSERT ON NESTED TABLE Emplist OF Dept_view REFERENCING NEW AS Employee PARENT AS Department FOR EACH ROW BEGIN -- The insert on the nested table is translated to an insert on the base table: INSERT INTO Emp_tab VALUES ( :Employee.Empno, :Employee.Empname,:Employee.Salary, :Department.Deptno); END; Any INSERT into the nested table fires the trigger, and the Emp_tab table is filled with the correct values. For example: INSERT INTO TABLE (SELECT d.Emplist FROM Dept_view d WHERE Deptno = 10) VALUES (1001, 'John Glenn', 10000); The :department.deptno correlation variable in this example would have a value of 10. Firing Triggers One or Many Times (FOR EACH ROW Option) The FOR EACH ROW option determines whether the trigger is a row trigger or a statement trigger. If you specify FOR EACH ROW, then the trigger fires once for each row of the table that is affected by the triggering statement. The absence of the FOR EACH ROW option indicates that the trigger fires only once for each applicable statement, but not separately for each row affected by the statement. For example, you define the following trigger: You may need to set up the following data structures for certain examples to work: Note: CREATE TABLE Emp_log ( Emp_id NUMBER, Log_date DATE, New_salary NUMBER, Action VARCHAR2(20)); CREATE OR REPLACE TRIGGER Log_salary_increase AFTER UPDATE ON Emp_tab FOR EACH ROW WHEN (new.Sal > 1000) BEGIN INSERT INTO Emp_log (Emp_id, Log_date, New_salary, Action) VALUES (:new.Empno, SYSDATE, :new.SAL, 'NEW SAL'); END; Then, you enter the following SQL statement: Coding Triggers 9-9 Coding the Trigger Body UPDATE Emp_tab SET Sal = Sal + 1000.0 WHERE Deptno = 20; If there are five employees in department 20, then the trigger fires five times when this statement is entered, because five rows are affected. The following trigger fires only once for each UPDATE of the Emp_tab table: CREATE OR REPLACE TRIGGER Log_emp_update AFTER UPDATE ON Emp_tab BEGIN INSERT INTO Emp_log (Log_date, Action) VALUES (SYSDATE, 'Emp_tab COMMISSIONS CHANGED'); END; See Also: Oracle Database Concepts for the order of trigger firing The statement level triggers are useful for performing validation checks for the entire statement. Firing Triggers Based on Conditions (WHEN Clause) Optionally, a trigger restriction can be included in the definition of a row trigger by specifying a Boolean SQL expression in a WHEN clause. Note: A WHEN clause cannot be included in the definition of a statement trigger. If included, then the expression in the WHEN clause is evaluated for each row that the trigger affects. If the expression evaluates to TRUE for a row, then the trigger body is fired on behalf of that row. However, if the expression evaluates to FALSE or NOT TRUE for a row (unknown, as with nulls), then the trigger body is not fired for that row. The evaluation of the WHEN clause does not have an effect on the execution of the triggering SQL statement (in other words, the triggering statement is not rolled back if the expression in a WHEN clause evaluates to FALSE). For example, in the PRINT_SALARY_CHANGES trigger, the trigger body is not run if the new value of Empno is zero, NULL, or negative. In more realistic examples, you might test if one column value is less than another. The expression in a WHEN clause of a row trigger can include correlation names, which are explained later. The expression in a WHEN clause must be a SQL expression, and it cannot include a subquery. You cannot use a PL/SQL expression (including user-defined functions) in the WHEN clause. Note: You cannot specify the WHEN clause for INSTEAD OF triggers. Coding the Trigger Body The trigger body is a CALL procedure or a PL/SQL block that can include SQL and PL/SQL statements. The CALL procedure can be either a PL/SQL or a Java procedure that is encapsulated in a PL/SQL wrapper. These statements are run if the triggering statement is entered and if the trigger restriction (if included) evaluates to TRUE. 9-10 Oracle Database Application Developer’s Guide - Fundamentals Coding the Trigger Body The trigger body for row triggers has some special constructs that can be included in the code of the PL/SQL block: correlation names and the REFERENCEING option, and the conditional predicates INSERTING, DELETING, and UPDATING. The INSERTING, DELETING, and UPDATING conditional predicates cannot be used for the CALL procedures; they can only be used in a PL/SQL block. Note: Example: Monitoring Logons with a Trigger You may need to set up data structures similar to the following for certain examples to work: Note: CONNECT system/manager GRANT ADMINISTER DATABASE TRIGGER TO scott; CONNECT scott/tiger CREATE TABLE audit_table ( seq number, user_at VARCHAR2(10), time_now DATE, term VARCHAR2(10), job VARCHAR2(10), proc VARCHAR2(10), enum NUMBER); CREATE OR REPLACE PROCEDURE foo (c VARCHAR2) AS BEGIN INSERT INTO Audit_table (user_at) VALUES(c); END; CREATE OR REPLACE TRIGGER logontrig AFTER LOGON ON DATABASE -- Just call an existing procedure. The ORA_LOGIN_USER is a function -- that returns information about the event that fired the trigger. CALL foo (ora_login_user) / Example: Calling a Java Procedure from a Trigger Although triggers are declared using PL/SQL, they can call procedures in other languages, such as Java: CREATE OR REPLACE PROCEDURE Before_delete (Id IN NUMBER, Ename VARCHAR2) IS language Java name 'thjvTriggers.beforeDelete (oracle.sql.NUMBER, oracle.sql.CHAR)'; CREATE OR REPLACE TRIGGER Pre_del_trigger BEFORE DELETE ON Tab FOR EACH ROW CALL Before_delete (:old.Id, :old.Ename) / The corresponding Java file is thjvTriggers.java: import import import import public { java.sql.* java.io.* oracle.sql.* oracle.oracore.* class thjvTriggers Coding Triggers 9-11 Coding the Trigger Body public state void beforeDelete (NUMBER old_id, CHAR old_name) Throws SQLException, CoreException { Connection conn = JDBCConnection.defaultConnection(); Statement stmt = conn.CreateStatement(); String sql = "insert into logtab values ("+ old_id.intValue() +", '"+ old_ename.toString() + ", BEFORE DELETE'); stmt.executeUpdate (sql); stmt.close(); return; } } Accessing Column Values in Row Triggers Within a trigger body of a row trigger, the PL/SQL code and SQL statements have access to the old and new column values of the current row affected by the triggering statement. Two correlation names exist for every column of the table being modified: one for the old column value, and one for the new column value. Depending on the type of triggering statement, certain correlation names might not have any meaning. ■ ■ ■ A trigger fired by an INSERT statement has meaningful access to new column values only. Because the row is being created by the INSERT, the old values are null. A trigger fired by an UPDATE statement has access to both old and new column values for both BEFORE and AFTER row triggers. A trigger fired by a DELETE statement has meaningful access to :old column values only. Because the row no longer exists after the row is deleted, the :new values are NULL. However, you cannot modify :new values because ORA-4084 is raised if you try to modify :new values. The new column values are referenced using the new qualifier before the column name, while the old column values are referenced using the old qualifier before the column name. For example, if the triggering statement is associated with the Emp_tab table (with the columns SAL, COMM, and so on), then you can include statements in the trigger body. For example: IF :new.Sal > 10000 ... IF :new.Sal < :old.Sal ... Old and new values are available in both BEFORE and AFTER row triggers. A new column value can be assigned in a BEFORE row trigger, but not in an AFTER row trigger (because the triggering statement takes effect before an AFTER row trigger is fired). If a BEFORE row trigger changes the value of new.column, then an AFTER row trigger fired by the same statement sees the change assigned by the BEFORE row trigger. Correlation names can also be used in the Boolean expression of a WHEN clause. A colon (:) must precede the old and new qualifiers when they are used in a trigger body, but a colon is not allowed when using the qualifiers in the WHEN clause or the REFERENCING option. Example: Modifying LOB Columns with a Trigger You can treat LOB columns the same as other columns, using regular SQL and PL/SQL functions with CLOB columns, and calls to the DBMS_LOB package with BLOB columns: 9-12 Oracle Database Application Developer’s Guide - Fundamentals Coding the Trigger Body drop table tab1; create table tab1 (c1 clob); insert into tab1 values (' HTML Document Fragment
Some text.'); create or replace trigger trg1 before update on tab1 for each row begin dbms_output.put_line('Old value of CLOB column: '||:OLD.c1); dbms_output.put_line('Proposed new value of CLOB column: '||:NEW.c1); -- Previously, we couldn't change the new value for a LOB. -- Now, we can replace it, or construct a new value using SUBSTR, INSTR... -- operations for a CLOB, or DBMS_LOB calls for a BLOB. :NEW.c1 := :NEW.c1 || to_clob('
Standard footer paragraph.'); dbms_output.put_line('Final value of CLOB column: '||:NEW.c1); end; / set serveroutput on; update tab1 set c1 = '
Different Document Fragment
Different text.'; select * from tab1; INSTEAD OF Triggers on Nested Table View Columns In the case of INSTEAD OF triggers on nested table view columns, the new and old qualifiers correspond to the new and old nested table elements. The parent row corresponding to this nested table element can be accessed using the parent qualifier. The parent correlation name is meaningful and valid only inside a nested table trigger. Avoiding Name Conflicts with Triggers (REFERENCING Option) The REFERENCING option can be specified in a trigger body of a row trigger to avoid name conflicts among the correlation names and tables that might be named old or new. Because this is rare, this option is infrequently used. For example, assume you have a table named new with columns field1 (number) and field2 (character). The following CREATE TRIGGER example shows a trigger associated with the new table that can use correlation names and avoid naming conflicts between the correlation names and the table name: You may need to set up the following data structures for certain examples to work: Note: CREATE TABLE new ( field1 NUMBER, field2 VARCHAR2(20)); CREATE OR REPLACE TRIGGER Print_salary_changes BEFORE UPDATE ON new REFERENCING new AS Newest FOR EACH ROW BEGIN :Newest.Field2 := TO_CHAR (:newest.field1); END; Coding Triggers 9-13 Coding the Trigger Body Notice that the new qualifier is renamed to newest using the REFERENCING option, and it is then used in the trigger body. Detecting the DML Operation That Fired a Trigger If more than one type of DML operation can fire a trigger (for example, ON INSERT OR DELETE OR UPDATE OF Emp_tab), the trigger body can use the conditional predicates INSERTING, DELETING, and UPDATING to check which type of statement fire the trigger. Within the code of the trigger body, you can execute blocks of code depending on the kind of DML operation fired the trigger: IF INSERTING THEN ... END IF; IF UPDATING THEN ... END IF; The first condition evaluates to TRUE only if the statement that fired the trigger is an INSERT statement; the second condition evaluates to TRUE only if the statement that fired the trigger is an UPDATE statement. In an UPDATE trigger, a column name can be specified with an UPDATING conditional predicate to determine if the named column is being updated. For example, assume a trigger is defined as the following: CREATE OR REPLACE TRIGGER ... ... UPDATE OF Sal, Comm ON Emp_tab ... BEGIN ... IF UPDATING ('SAL') THEN ... END IF; END; The code in the THEN clause runs only if the triggering UPDATE statement updates the SAL column. This way, the trigger can minimize its overhead when the column of interest is not being changed. Error Conditions and Exceptions in the Trigger Body If a predefined or user-defined error condition or exception is raised during the execution of a trigger body, then all effects of the trigger body, as well as the triggering statement, are rolled back (unless the error is trapped by an exception handler). Therefore, a trigger body can prevent the execution of the triggering statement by raising an exception. User-defined exceptions are commonly used in triggers that enforce complex security authorizations or integrity constraints. The only exception to this is when the event under consideration is database STARTUP, SHUTDOWN, or LOGIN when the user logging in is SYSTEM. In these scenarios, only the trigger action is rolled back. Triggers on Object Tables You can use the OBJECT_VALUE pseudocolumn in a trigger on an object table since 10g Release 1 (10.1). OBJECT_VALUE means the object as a whole. This is one example of its use. You can also call a PL/SQL function with OBJECT_VALUE as the datatype of an IN formal parameter. Here is an example of the use of OBJECT_VALUE in a trigger. To keep track of updates to values in an object table tbl, a history table, tbl_history, is also created in the following example. For tbl, the values 1 through 5 are inserted into n, while m is kept at 0. The trigger is a row-level trigger that executes once for each row affected by a 9-14 Oracle Database Application Developer’s Guide - Fundamentals Coding the Trigger Body DML statement. The trigger causes the old and new values of the object t in tbl to be written in tbl_history when tbl is updated. These old and new values are :OLD.OBJECT_VALUE and :NEW.OBJECT_VALUE. An update of the table tbl is done (each value of n is increased by 1). A select from the history table to check that the trigger works is then shown at the end of the example: CREATE OR REPLACE TYPE t AS OBJECT (n NUMBER, m NUMBER) / CREATE TABLE tbl OF t / BEGIN FOR j IN 1..5 LOOP INSERT INTO tbl VALUES (t(j, 0)); END LOOP; END; / CREATE TABLE tbl_history ( d DATE, old_obj t, new_obj t) / CREATE OR REPLACE TRIGGER Tbl_Trg AFTER UPDATE ON tbl FOR EACH ROW BEGIN INSERT INTO tbl_history (d, old_obj, new_obj) VALUES (SYSDATE, :OLD.OBJECT_VALUE, :NEW.OBJECT_VALUE); END Tbl_Trg; / -------------------------------------------------------------------------------UPDATE tbl SET tbl.n = tbl.n+1 / BEGIN FOR j IN (SELECT d, old_obj, new_obj FROM tbl_history) LOOP Dbms_Output.Put_Line ( j.d|| ' -- old: '||j.old_obj.n||' '||j.old_obj.m|| ' -- new: '||j.new_obj.n||' '||j.new_obj.m); END LOOP; END; / The result of the select shows that the values of the column n have been all increased by 1. The value of m remains 0. The output of the select is: 23-MAY-05 23-MAY-05 23-MAY-05 23-MAY-05 23-MAY-05 ------ old: old: old: old: old: 1 2 3 4 5 0 0 0 0 0 ------ new: new: new: new: new: 2 3 4 5 6 0 0 0 0 0 Triggers and Handling Remote Exceptions A trigger that accesses a remote site cannot do remote exception handling if the network link is unavailable. For example: CREATE OR REPLACE TRIGGER Example AFTER INSERT ON Emp_tab FOR EACH ROW BEGIN INSERT INTO Emp_tab@Remote -- <- compilation fails here VALUES ('x'); -when dblink is inaccessible EXCEPTION WHEN OTHERS THEN Coding Triggers 9-15 Coding the Trigger Body INSERT INTO Emp_log VALUES ('x'); END; A trigger is compiled when it is created. Thus, if a remote site is unavailable when the trigger must compile, then Oracle Database cannot validate the statement accessing the remote database, and the compilation fails. The previous example exception statement cannot run, because the trigger does not complete compilation. Because stored procedures are stored in a compiled form, the work-around for the previous example is as follows: CREATE OR REPLACE TRIGGER Example AFTER INSERT ON Emp_tab FOR EACH ROW BEGIN Insert_row_proc; END; CREATE OR REPLACE PROCEDURE Insert_row_proc AS BEGIN INSERT INTO Emp_tab@Remote VALUES ('x'); EXCEPTION WHEN OTHERS THEN INSERT INTO Emp_log VALUES ('x'); END; The trigger in this example compiles successfully and calls the stored procedure, which already has a validated statement for accessing the remote database; thus, when the remote INSERT statement fails because the link is down, the exception is caught. Restrictions on Creating Triggers Coding triggers requires some restrictions that are not required for standard PL/SQL blocks. The following sections discuss these restrictions. Maximum Trigger Size The size of a trigger cannot be more than 32K. SQL Statements Allowed in Trigger Bodies The body of a trigger can contain DML SQL statements. It can also contain SELECT statements, but they must be SELECT... INTO... statements or the SELECT statement in the definition of a cursor. DDL statements are not allowed in the body of a trigger. Also, no transaction control statements are allowed in a trigger. ROLLBACK, COMMIT, and SAVEPOINT cannot be used.For system triggers, {CREATE/ALTER/DROP} TABLE statements and ALTER...COMPILE are allowed. A procedure called by a trigger cannot run the previous transaction control statements, because the procedure runs within the context of the trigger body. Note: 9-16 Oracle Database Application Developer’s Guide - Fundamentals Coding the Trigger Body Statements inside a trigger can reference remote schema objects. However, pay special attention when calling remote procedures from within a local trigger. If a timestamp or signature mismatch is found during execution of the trigger, then the remote procedure is not run, and the trigger is invalidated. Trigger Restrictions on LONG and LONG RAW Datatypes LONG and LONG RAW datatypes in triggers are subject to the following restrictions: ■ ■ A SQL statement within a trigger can insert data into a column of LONG or LONG RAW datatype. If data from a LONG or LONG RAW column can be converted to a constrained datatype (such as CHAR and VARCHAR2), then a LONG or LONG RAW column can be referenced in a SQL statement within a trigger. The maximum length for these datatypes is 32000 bytes. ■ Variables cannot be declared using the LONG or LONG RAW datatypes. ■ :NEW and :PARENT cannot be used with LONG or LONG RAW columns. Trigger Restrictions on Mutating Tables A mutating table is a table that is being modified by an UPDATE, DELETE, or INSERT statement, or a table that might be updated by the effects of a DELETE CASCADE constraint. The session that issued the triggering statement cannot query or modify a mutating table. This restriction prevents a trigger from seeing an inconsistent set of data. This restriction applies to all triggers that use the FOR EACH ROW clause. Views being modified in INSTEAD OF triggers are not considered mutating. When a trigger encounters a mutating table, a runtime error occurs, the effects of the trigger body and triggering statement are rolled back, and control is returned to the user or application. Consider the following trigger: CREATE OR REPLACE TRIGGER Emp_count AFTER DELETE ON Emp_tab FOR EACH ROW DECLARE n INTEGER; BEGIN SELECT COUNT(*) INTO n FROM Emp_tab; DBMS_OUTPUT.PUT_LINE(' There are now ' || n || ' employees.'); END; If the following SQL statement is entered: DELETE FROM Emp_tab WHERE Empno = 7499; An error is returned because the table is mutating when the row is deleted: ORA-04091: table SCOTT.Emp_tab is mutating, trigger/function may not see it If you delete the line "FOR EACH ROW" from the trigger, it becomes a statement trigger which is not subject to this restriction, and the trigger. If you need to update a mutating table, you could bypass these restrictions by using a temporary table, a PL/SQL table, or a package variable. For example, in place of a single AFTER row trigger that updates the original table, resulting in a mutating table Coding Triggers 9-17 Coding the Trigger Body error, you might use two triggers—an AFTER row trigger that updates a temporary table, and an AFTER statement trigger that updates the original table with the values from the temporary table. Declarative integrity constraints are checked at various times with respect to row triggers. Oracle Database Concepts for information about the interaction of triggers and integrity constraints See Also: Because declarative referential integrity constraints are not supported between tables on different nodes of a distributed database, the mutating table restrictions do not apply to triggers that access remote nodes. These restrictions are also not enforced among tables in the same database that are connected by loop-back database links. A loop-back database link makes a local table appear remote by defining an Oracle Net path back to the database that contains the link. Restrictions on Mutating Tables Relaxed The mutating error, discussed earlier in this section, still prevents the trigger from reading or modifying the table that the parent statement is modifying. However, starting in Oracle Database Release 8.1, a delete against the parent table causes before/after statement triggers to be fired once. That way, you can create triggers (just not row triggers) to read and modify the parent and child tables. This allows most foreign key constraint actions to be implemented through their obvious after-row trigger, providing the constraint is not self-referential. Update cascade, update set null, update set default, delete set default, inserting a missing parent, and maintaining a count of children can all be implemented easily. For example, this is an implementation of update cascade: create table p create table f create trigger update f set end; / (p1 number constraint ppk primary key); (f1 number constraint ffk references p); pt after update on p for each row begin f1 = :new.p1 where f1 = :old.p1; This implementation requires care for multirow updates. For example, if a table p has three rows with the values (1), (2), (3), and table f also has three rows with the values (1), (2), (3), then the following statement updates p correctly but causes problems when the trigger updates f: update p set p1 = p1+1; The statement first updates (1) to (2) in p, and the trigger updates (1) to (2) in f, leaving two rows of value (2) in f. Then the statement updates (2) to (3) in p, and the trigger updates both rows of value (2) to (3) in f. Finally, the statement updates (3) to (4) in p, and the trigger updates all three rows in f from (3) to (4). The relationship of the data in p and f is lost. To avoid this problem, you must forbid multirow updates to p that change the primary key and reuse existing primary key values. It could also be solved by tracking which foreign key values have already been updated, then modifying the trigger so that no row is updated twice. That is the only problem with this technique for foreign key updates. The trigger cannot miss rows that have been changed but not committed by another transaction, because the foreign key constraint guarantees that no matching foreign key rows are locked before the after-row trigger is called. 9-18 Oracle Database Application Developer’s Guide - Fundamentals Coding the Trigger Body System Trigger Restrictions Depending on the event, different event attribute functions are available. For example, certain DDL operations may not be allowed on DDL events. Check "Event Attribute Functions" on page 9-38 before using an event attribute function, because its effects might be undefined rather than producing an error condition. Only committed triggers are fired. For example, if you create a trigger that should be fired after all CREATE events, then the trigger itself does not fire after the creation, because the correct information about this trigger was not committed at the time when the trigger on CREATE events was fired. For example, if you execute the following SQL statement: CREATE OR REPLACE TRIGGER my_trigger AFTER CREATE ON DATABASE BEGIN null; END; Then, trigger my_trigger is not fired after the creation of my_trigger. Oracle Database does not fire a trigger that is not committed. Foreign Function Callouts All restrictions on foreign function callouts also apply. Who Is the Trigger User? The following statement, inside a trigger, returns the owner of the trigger, not the name of user who is updating the table: SELECT Username FROM USER_USERS; Privileges Needed to Work with Triggers To create a trigger in your schema, you must have the CREATE TRIGGER system privilege, and either: ■ Own the table specified in the triggering statement, or ■ Have the ALTER privilege for the table in the triggering statement, or ■ Have the ALTER ANY TABLE system privilege To create a trigger in another user's schema, or to reference a table in another schema from a trigger in your schema, you must have the CREATE ANY TRIGGER system privilege. With this privilege, the trigger can be created in any schema and can be associated with any user's table. In addition, the user creating the trigger must also have EXECUTE privilege on the referenced procedures, functions, or packages. To create a trigger on DATABASE, you must have the ADMINISTER DATABASE TRIGGER privilege. If this privilege is later revoked, then you can drop the trigger, but not alter it. The object privileges to the schema objects referenced in the trigger body must be granted to the trigger owner explicitly (not through a role). The statements in the trigger body operate under the privilege domain of the trigger owner, not the privilege domain of the user issuing the triggering statement. This is similar to the privilege model for stored procedures. Coding Triggers 9-19 Compiling Triggers Compiling Triggers Triggers are similar to PL/SQL anonymous blocks with the addition of the :new and :old capabilities, but their compilation is different. A PL/SQL anonymous block is compiled each time it is loaded into memory. Compilation involves three stages: 1. Syntax checking: PL/SQL syntax is checked, and a parse tree is generated. 2. Semantic checking: Type checking and further processing on the parse tree. 3. Code generation: The pcode is generated. Triggers, in contrast, are fully compiled when the CREATE TRIGGER statement is entered, and the pcode is stored in the data dictionary. Hence, firing the trigger no longer requires the opening of a shared cursor to run the trigger action. Instead, the trigger is executed directly. If errors occur during the compilation of a trigger, then the trigger is still created. If a DML statement fires this trigger, then the DML statement fails. (Runtime that trigger errors always cause the DML statement to fail.) You can use the SHOW ERRORS statement in SQL*Plus or Enterprise Manager to see any compilation errors when you create a trigger, or you can SELECT the errors from the USER_ERRORS view. Dependencies for Triggers Compiled triggers have dependencies. They become invalid if a depended-on object, such as a stored procedure or function called from the trigger body, is modified. Triggers that are invalidated for dependency reasons are recompiled when next invoked. You can examine the ALL_DEPENDENCIES view to see the dependencies for a trigger. For example, the following statement shows the dependencies for the triggers in the SCOTT schema: SELECT NAME, REFERENCED_OWNER, REFERENCED_NAME, REFERENCED_TYPE FROM ALL_DEPENDENCIES WHERE OWNER = 'SCOTT' and TYPE = 'TRIGGER'; Triggers may depend on other functions or packages. If the function or package specified in the trigger is dropped, then the trigger is marked invalid. An attempt is made to validate the trigger on occurrence of the event. If the trigger cannot be validated successfully, then it is marked VALID WITH ERRORS, and the event fails. Note: ■ ■ There is an exception for STARTUP events: STARTUP events succeed even if the trigger fails. There are also exceptions for SHUTDOWN events and for LOGON events if you login as SYSTEM. Because the DBMS_AQ package is used to enqueue a message, dependency between triggers and queues cannot be maintained. Recompiling Triggers Use the ALTER TRIGGER statement to recompile a trigger manually. For example, the following statement recompiles the PRINT_SALARY_CHANGES trigger: ALTER TRIGGER Print_salary_changes COMPILE; 9-20 Oracle Database Application Developer’s Guide - Fundamentals Enabling and Disabling Triggers To recompile a trigger, you must own the trigger or have the ALTER ANY TRIGGER system privilege. Modifying Triggers Like a stored procedure, a trigger cannot be explicitly altered: It must be replaced with a new definition. (The ALTER TRIGGER statement is used only to recompile, enable, or disable a trigger.) When replacing a trigger, you must include the OR REPLACE option in the CREATE TRIGGER statement. The OR REPLACE option is provided to allow a new version of an existing trigger to replace the older version, without affecting any grants made for the original version of the trigger. Alternatively, the trigger can be dropped using the DROP TRIGGER statement, and you can rerun the CREATE TRIGGER statement. To drop a trigger, the trigger must be in your schema, or you must have the DROP ANY TRIGGER system privilege. Debugging Triggers You can debug a trigger using the same facilities available for stored procedures. See Also: "Debugging Stored Procedures" on page 7-29 Enabling and Disabling Triggers A trigger can be in one of two distinct modes: Enabled. An enabled trigger executes its trigger body if a triggering statement is entered and the trigger restriction (if any) evaluates to TRUE. Disabled. A disabled trigger does not execute its trigger body, even if a triggering statement is entered and the trigger restriction (if any) evaluates to TRUE. Enabling Triggers By default, a trigger is automatically enabled when it is created; however, it can later be disabled. After you have completed the task that required the trigger to be disabled, re-enable the trigger, so that it fires when appropriate. Enable a disabled trigger using the ALTER TRIGGER statement with the ENABLE option. To enable the disabled trigger named REORDER of the INVENTORY table, enter the following statement: ALTER TRIGGER Reorder ENABLE; All triggers defined for a specific table can be enabled with one statement using the ALTER TABLE statement with the ENABLE clause with the ALL TRIGGERS option. For example, to enable all triggers defined for the INVENTORY table, enter the following statement: ALTER TABLE Inventory ENABLE ALL TRIGGERS; Disabling Triggers You might temporarily disable a trigger if: Coding Triggers 9-21 Viewing Information About Triggers ■ ■ ■ An object it references is not available. You need to perform a large data load, and you want it to proceed quickly without firing triggers. You are reloading data. By default, triggers are enabled when first created. Disable a trigger using the ALTER TRIGGER statement with the DISABLE option. For example, to disable the trigger named REORDER of the INVENTORY table, enter the following statement: ALTER TRIGGER Reorder DISABLE; All triggers associated with a table can be disabled with one statement using the ALTER TABLE statement with the DISABLE clause and the ALL TRIGGERS option. For example, to disable all triggers defined for the INVENTORY table, enter the following statement: ALTER TABLE Inventory DISABLE ALL TRIGGERS; Viewing Information About Triggers The following data dictionary views reveal information about triggers: ■ USER_TRIGGERS ■ ALL_TRIGGERS ■ DBA_TRIGGERS The new column, BASE_OBJECT_TYPE, specifies whether the trigger is based on DATABASE, SCHEMA, table, or view. The old column, TABLE_NAME, is null if the base object is not table or view. The column ACTION_TYPE specifies whether the trigger is a call type trigger or a PL/SQL trigger. The column TRIGGER_TYPE includes two additional values: BEFORE EVENT and AFTER EVENT, applicable only to system events. The column TRIGGERING_EVENT includes all system and DML events. Oracle Database Reference for a complete description of these data dictionary views See Also: For example, assume the following statement was used to create the REORDER trigger: Caution: You may need to set up data structures for certain examples to work: CREATE OR REPLACE TRIGGER Reorder AFTER UPDATE OF Parts_on_hand ON Inventory FOR EACH ROW WHEN(new.Parts_on_hand < new.Reorder_point) DECLARE x NUMBER; BEGIN SELECT COUNT(*) INTO x 9-22 Oracle Database Application Developer’s Guide - Fundamentals Examples of Trigger Applications FROM Pending_orders WHERE Part_no = :new.Part_no; IF x = 0 THEN INSERT INTO Pending_orders VALUES (:new.Part_no, :new.Reorder_quantity, sysdate); END IF; END; The following two queries return information about the REORDER trigger: SELECT Trigger_type, Triggering_event, Table_name FROM USER_TRIGGERS WHERE Trigger_name = 'REORDER'; TYPE TRIGGERING_STATEMENT TABLE_NAME ---------------- -------------------------- -----------AFTER EACH ROW UPDATE INVENTORY SELECT Trigger_body FROM USER_TRIGGERS WHERE Trigger_name = 'REORDER'; TRIGGER_BODY -------------------------------------------DECLARE x NUMBER; BEGIN SELECT COUNT(*) INTO x FROM Pending_orders WHERE Part_no = :new.Part_no; IF x = 0 THEN INSERT INTO Pending_orders VALUES (:new.Part_no, :new.Reorder_quantity, sysdate); END IF; END; Examples of Trigger Applications You can use triggers in a number of ways to customize information management in Oracle Database. For example, triggers are commonly used to: ■ Provide sophisticated auditing ■ Prevent invalid transactions ■ Enforce referential integrity (either those actions not supported by declarative integrity constraints or across nodes in a distributed database) ■ Enforce complex business rules ■ Enforce complex security authorizations ■ Provide transparent event logging ■ Automatically generate derived column values ■ Enable building complex views that are updatable ■ Track system events Coding Triggers 9-23 Examples of Trigger Applications This section provides an example of each of these trigger applications. These examples are not meant to be used exactly as written: They are provided to assist you in designing your own triggers. Auditing with Triggers: Example Triggers are commonly used to supplement the built-in auditing features of Oracle Database. Although triggers can be written to record information similar to that recorded by the AUDIT statement, triggers should be used only when more detailed audit information is required. For example, use triggers to provide value-based auditing for each row. Sometimes, the AUDIT statement is considered a security audit facility, while triggers can provide financial audit facility. When deciding whether to create a trigger to audit database activity, consider what Oracle Database's auditing features provide, compared to auditing defined by triggers, as shown in Table 9–1. Table 9–1 Comparison of Built-in Auditing and Trigger-Based Auditing Audit Feature Description DML and DDL Auditing Standard auditing options permit auditing of DML and DDL statements regarding all types of schema objects and structures. Comparatively, triggers permit auditing of DML statements entered against tables, and DDL auditing at SCHEMA or DATABASE level. Centralized Audit Trail All database audit information is recorded centrally and automatically using the auditing features of Oracle Database. Declarative Method Auditing features enabled using the standard Oracle Database features are easier to declare and maintain, and less prone to errors, when compared to auditing functions defined by triggers. Auditing Options can be Audited Any changes to existing auditing options can also be audited to guard against malicious database activity. Session and Execution time Auditing Using the database auditing features, records can be generated once every time an audited statement is entered (BY ACCESS) or once for every session that enters an audited statement (BY SESSION). Triggers cannot audit by session; an audit record is generated each time a trigger-audited table is referenced. Auditing of Unsuccessful Data Access Database auditing can be set to audit when unsuccessful data access occurs. However, unless autonomous transactions are used, any audit information generated by a trigger is rolled back if the triggering statement is rolled back. For more information on autonomous transactions, refer to Oracle Database Concepts. Sessions can be Audited Connections and disconnections, as well as session activity (physical I/Os, logical I/Os, deadlocks, and so on), can be recorded using standard database auditing. When using triggers to provide sophisticated auditing, AFTER triggers are normally used. By using AFTER triggers, auditing information is recorded after the triggering statement is subjected to any applicable integrity constraints, preventing cases where the audit processing is carried out unnecessarily for statements that generate exceptions to integrity constraints. Choosing between AFTER row and AFTER statement triggers depends on the information being audited. For example, row triggers provide value-based auditing for each table row. Triggers can also require the user to supply a "reason code" for 9-24 Oracle Database Application Developer’s Guide - Fundamentals Examples of Trigger Applications issuing the audited SQL statement, which can be useful in both row and statement-level auditing situations. The following example demonstrates a trigger that audits modifications to the Emp_tab table for each row. It requires that a "reason code" be stored in a global package variable before the update. This shows how triggers can be used to provide value-based auditing and how to use public package variables. You may need to set up the following data structures for the examples to work: Note: CREATE OR REPLACE PACKAGE Auditpackage AS Reason VARCHAR2(10); PROCEDURE Set_reason(Reason VARCHAR2); END; CREATE TABLE Emp99 ( Empno NOT NULL NUMBER(4), Ename VARCHAR2(10), Job VARCHAR2(9), Mgr NUMBER(4), Hiredate DATE, Sal NUMBER(7,2), Comm NUMBER(7,2), Deptno NUMBER(2), Bonus NUMBER, Ssn NUMBER, Job_classification NUMBER); CREATE TABLE Audit_employee ( Oldssn NUMBER, Oldname VARCHAR2(10), Oldjob VARCHAR2(2), Oldsal NUMBER, Newssn NUMBER, Newname VARCHAR2(10), Newjob VARCHAR2(2), Newsal NUMBER, Reason VARCHAR2(10), User1 VARCHAR2(10), Systemdate DATE); CREATE OR REPLACE TRIGGER Audit_employee AFTER INSERT OR DELETE OR UPDATE ON Emp99 FOR EACH ROW BEGIN /* AUDITPACKAGE is a package with a public package variable REASON. REASON could be set by the application by a command such as EXECUTE AUDITPACKAGE.SET_REASON(reason_string). Note that a package variable has state for the duration of a session and that each session has a separate copy of all package variables. */ IF Auditpackage.Reason IS NULL THEN Raise_application_error(-20201, 'Must specify reason' || ' with AUDITPACKAGE.SET_REASON(Reason_string)'); END IF; /* If the preceding conditional evaluates to TRUE, the Coding Triggers 9-25 Examples of Trigger Applications user-specified error number and message is raised, the trigger stops execution, and the effects of the triggering statement are rolled back. Otherwise, a new row is inserted into the predefined auditing table named AUDIT_EMPLOYEE containing the existing and new values of the Emp_tab table and the reason code defined by the REASON variable of AUDITPACKAGE. Note that the "old" values are NULL if triggering statement is an INSERT and the "new" values are NULL if the triggering statement is a DELETE. */ INSERT INTO Audit_employee VALUES (:old.Ssn, :old.Ename, :old.Job_classification, :old.Sal, :new.Ssn, :new.Ename, :new.Job_classification, :new.Sal, auditpackage.Reason, User, Sysdate ); END; Optionally, you can also set the reason code back to NULL if you wanted to force the reason code to be set for every update. The following simple AFTER statement trigger sets the reason code back to NULL after the triggering statement is run: CREATE OR REPLACE TRIGGER Audit_employee_reset AFTER INSERT OR DELETE OR UPDATE ON Emp_tab BEGIN auditpackage.set_reason(NULL); END; Notice that the previous two triggers are both fired by the same type of SQL statement. However, the AFTER row trigger is fired once for each row of the table affected by the triggering statement, while the AFTER statement trigger is fired only once after the triggering statement execution is completed. This next trigger also uses triggers to do auditing. It tracks changes made to the Emp_tab table and stores this information in AUDIT_TABLE and AUDIT_TABLE_VALUES. You may need to set up the following data structures for the example to work: Note: CREATE TABLE Audit_table ( Seq NUMBER, User_at VARCHAR2(10), Time_now DATE, Term VARCHAR2(10), Job VARCHAR2(10), Proc VARCHAR2(10), enum NUMBER); CREATE SEQUENCE Audit_seq; CREATE TABLE Audit_table_values ( Seq NUMBER, Dept NUMBER, Dept1 NUMBER, Dept2 NUMBER); CREATE OR REPLACE TRIGGER Audit_emp AFTER INSERT OR UPDATE OR DELETE ON Emp_tab FOR EACH ROW DECLARE Time_now DATE; 9-26 Oracle Database Application Developer’s Guide - Fundamentals Examples of Trigger Applications Terminal CHAR(10); BEGIN -- get current time, and the terminal of the user: Time_now := SYSDATE; Terminal := USERENV('TERMINAL'); -- record new employee primary key IF INSERTING THEN INSERT INTO Audit_table VALUES (Audit_seq.NEXTVAL, User, Time_now, Terminal, 'Emp_tab', 'INSERT', :new.Empno); -- record primary key of the deleted row: ELSIF DELETING THEN INSERT INTO Audit_table VALUES (Audit_seq.NEXTVAL, User, Time_now, Terminal, 'Emp_tab', 'DELETE', :old.Empno); -- for updates, record the primary key -- of the row being updated: ELSE INSERT INTO Audit_table VALUES (audit_seq.NEXTVAL, User, Time_now, Terminal, 'Emp_tab', 'UPDATE', :old.Empno); -- and for SAL and DEPTNO, record old and new values: IF UPDATING ('SAL') THEN INSERT INTO Audit_table_values VALUES (Audit_seq.CURRVAL, 'SAL', :old.Sal, :new.Sal); ELSIF UPDATING ('DEPTNO') THEN INSERT INTO Audit_table_values VALUES (Audit_seq.CURRVAL, 'DEPTNO', :old.Deptno, :new.DEPTNO); END IF; END IF; END; Integrity Constraints and Triggers: Examples Triggers and declarative integrity constraints can both be used to constrain data input. However, triggers and integrity constraints have significant differences. Declarative integrity constraints are statements about the database that are always true. A constraint applies to existing data in the table and any statement that manipulates the table. See Also: Chapter 6, "Maintaining Data Integrity in Application Development" Triggers constrain what a transaction can do. A trigger does not apply to data loaded before the definition of the trigger; therefore, it is not known if all data in a table conforms to the rules established by an associated trigger. Although triggers can be written to enforce many of the same rules supported by Oracle Database's declarative integrity constraint features, triggers should only be used to enforce complex business rules that cannot be defined using standard integrity constraints. The declarative integrity constraint features provided with Oracle Database offer the following advantages when compared to constraints defined by triggers: Centralized integrity checks. All points of data access must adhere to the global set of rules defined by the integrity constraints corresponding to each schema object. Coding Triggers 9-27 Examples of Trigger Applications Declarative method. Constraints defined using the standard integrity constraint features are much easier to write and are less prone to errors, when compared with comparable constraints defined by triggers. While most aspects of data integrity can be defined and enforced using declarative integrity constraints, triggers can be used to enforce complex business constraints not definable using declarative integrity constraints. For example, triggers can be used to enforce: ■ ■ ■ UPDATE SET NULL, and UPDATE and DELETE SET DEFAULT referential actions. Referential integrity when the parent and child tables are on different nodes of a distributed database. Complex check constraints not definable using the expressions allowed in a CHECK constraint. Referential Integrity Using Triggers There are many cases where referential integrity can be enforced using triggers. Note, however, you should only use triggers when there is no declarative support for the action you are performing. When using triggers to maintain referential integrity, declare the PRIMARY (or UNIQUE) KEY constraint in the parent table. If referential integrity is being maintained between a parent and child table in the same database, then you can also declare the foreign key in the child table, but disable it; this prevents the corresponding PRIMARY KEY constraint from being dropped (unless the PRIMARY KEY constraint is explicitly dropped with the CASCADE option). To maintain referential integrity using triggers: ■ ■ A trigger must be defined for the child table that guarantees values inserted or updated in the foreign key correspond to values in the parent key. One or more triggers must be defined for the parent table. These triggers guarantee the desired referential action (RESTRICT, CASCADE, or SET NULL) for values in the foreign key when values are updated or deleted in the parent key. No action is required for inserts into the parent table (no dependent foreign keys exist). The following sections provide examples of the triggers necessary to enforce referential integrity. The Emp_tab and Dept_tab table relationship is used in these examples. Several of the triggers include statements that lock rows (SELECT... FOR UPDATE). This operation is necessary to maintain concurrency as the rows are being processed. Foreign Key Trigger for Child Table The following trigger guarantees that before an INSERT or UPDATE statement affects a foreign key value, the corresponding value exists in the parent key. The mutating table exception included in the following example allows this trigger to be used with the UPDATE_SET_DEFAULT and UPDATE_CASCADE triggers. This exception can be removed if this trigger is used alone. CREATE OR REPLACE TRIGGER Emp_dept_check BEFORE INSERT OR UPDATE OF Deptno ON Emp_tab FOR EACH ROW WHEN (new.Deptno IS NOT NULL) -- Before a row is inserted, or DEPTNO is updated in the Emp_tab -- table, fire this trigger to verify that the new foreign -- key value (DEPTNO) is present in the Dept_tab table. 9-28 Oracle Database Application Developer’s Guide - Fundamentals Examples of Trigger Applications DECLARE Dummy INTEGER; -- to be used for cursor fetch Invalid_department EXCEPTION; Valid_department EXCEPTION; Mutating_table EXCEPTION; PRAGMA EXCEPTION_INIT (Mutating_table, -4091); -- Cursor used to verify parent key value exists. -- present, lock parent key's row so it can't be -- deleted by another transaction until this -- transaction is committed or rolled back. CURSOR Dummy_cursor (Dn NUMBER) IS SELECT Deptno FROM Dept_tab WHERE Deptno = Dn FOR UPDATE OF Deptno; BEGIN OPEN Dummy_cursor (:new.Deptno); FETCH Dummy_cursor INTO Dummy; If ---IF Verify parent key. If not found, raise user-specified error number and message. If found, close cursor before allowing triggering statement to complete: Dummy_cursor%NOTFOUND THEN RAISE Invalid_department; ELSE RAISE valid_department; END IF; CLOSE Dummy_cursor; EXCEPTION WHEN Invalid_department THEN CLOSE Dummy_cursor; Raise_application_error(-20000, 'Invalid Department' || ' Number' || TO_CHAR(:new.deptno)); WHEN Valid_department THEN CLOSE Dummy_cursor; WHEN Mutating_table THEN NULL; END; The following trigger is defined on the DEPT_TAB table to enforce the UPDATE and DELETE RESTRICT referential action on the primary key of the DEPT_TAB table: UPDATE and DELETE RESTRICT Trigger for Parent Table CREATE OR REPLACE TRIGGER Dept_restrict BEFORE DELETE OR UPDATE OF Deptno ON Dept_tab FOR EACH ROW -- Before a row is deleted from Dept_tab or the primary key -- (DEPTNO) of Dept_tab is updated, check for dependent -- foreign key values in Emp_tab; rollback if any are found. DECLARE Dummy INTEGER; -- to be used for cursor fetch Employees_present EXCEPTION; employees_not_present EXCEPTION; -- Cursor used to check for dependent foreign key values. CURSOR Dummy_cursor (Dn NUMBER) IS SELECT Deptno FROM Emp_tab WHERE Deptno = Dn; BEGIN Coding Triggers 9-29 Examples of Trigger Applications OPEN Dummy_cursor (:old.Deptno); FETCH Dummy_cursor INTO Dummy; -- If dependent foreign key is found, raise user-specified -- error number and message. If not found, close cursor -- before allowing triggering statement to complete. IF Dummy_cursor%FOUND THEN RAISE Employees_present; -- dependent rows exist ELSE RAISE Employees_not_present; -- no dependent rows END IF; CLOSE Dummy_cursor; EXCEPTION WHEN Employees_present THEN CLOSE Dummy_cursor; Raise_application_error(-20001, 'Employees Present in' || ' Department ' || TO_CHAR(:old.DEPTNO)); WHEN Employees_not_present THEN CLOSE Dummy_cursor; END; This trigger does not work with self-referential tables (tables with both the primary/unique key and the foreign key). Also, this trigger does not allow triggers to cycle (such as, A fires B fires A). Caution: UPDATE and DELETE SET NULL Triggers for Parent Table: Example The following trigger is defined on the DEPT_TAB table to enforce the UPDATE and DELETE SET NULL referential action on the primary key of the DEPT_TAB table: CREATE OR REPLACE TRIGGER Dept_set_null AFTER DELETE OR UPDATE OF Deptno ON Dept_tab FOR EACH ROW -- Before a row is deleted from Dept_tab or the primary key -- (DEPTNO) of Dept_tab is updated, set all corresponding -- dependent foreign key values in Emp_tab to NULL: BEGIN IF UPDATING AND :OLD.Deptno != :NEW.Deptno OR DELETING THEN UPDATE Emp_tab SET Emp_tab.Deptno = NULL WHERE Emp_tab.Deptno = :old.Deptno; END IF; END; DELETE Cascade Trigger for Parent Table: Example The following trigger on the DEPT_TAB table enforces the DELETE CASCADE referential action on the primary key of the DEPT_TAB table: CREATE OR REPLACE TRIGGER Dept_del_cascade AFTER DELETE ON Dept_tab FOR EACH ROW -- Before a row is deleted from Dept_tab, delete all -- rows from the Emp_tab table whose DEPTNO is the same as -- the DEPTNO being deleted from the Dept_tab table: BEGIN DELETE FROM Emp_tab WHERE Emp_tab.Deptno = :old.Deptno; END; 9-30 Oracle Database Application Developer’s Guide - Fundamentals Examples of Trigger Applications Typically, the code for DELETE CASCADE is combined with the code for UPDATE SET NULL or UPDATE SET DEFAULT to account for both updates and deletes. Note: The following trigger ensures that if a department number is updated in the Dept_tab table, then this change is propagated to dependent foreign keys in the Emp_tab table: UPDATE Cascade Trigger for Parent Table: Example -- Generate a sequence number to be used as a flag for -- determining if an update has occurred on a column: CREATE SEQUENCE Update_sequence INCREMENT BY 1 MAXVALUE 5000 CYCLE; CREATE OR REPLACE PACKAGE Integritypackage AS Updateseq NUMBER; END Integritypackage; CREATE OR REPLACE PACKAGE BODY Integritypackage AS END Integritypackage; -- create flag col: ALTER TABLE Emp_tab ADD Update_id NUMBER; CREATE OR REPLACE TRIGGER Dept_cascade1 BEFORE UPDATE OF Deptno ON Dept_tab DECLARE Dummy NUMBER; -- Before updating the Dept_tab table (this is a statement -- trigger), generate a new sequence number and assign -- it to the public variable UPDATESEQ of a user-defined -- package named INTEGRITYPACKAGE: BEGIN SELECT Update_sequence.NEXTVAL INTO Dummy FROM dual; Integritypackage.Updateseq := Dummy; END; CREATE OR REPLACE TRIGGER Dept_cascade2 AFTER DELETE OR UPDATE OF Deptno ON Dept_tab FOR EACH ROW -- For each department number in Dept_tab that is updated, -- cascade the update to dependent foreign keys in the -- Emp_tab table. Only cascade the update if the child row -- has not already been updated by this trigger: BEGIN IF UPDATING THEN UPDATE Emp_tab SET Deptno = :new.Deptno, Update_id = Integritypackage.Updateseq --from 1st WHERE Emp_tab.Deptno = :old.Deptno AND Update_id IS NULL; /* only NULL if not updated by the 3rd trigger fired by this same triggering statement */ END IF; IF DELETING THEN -- Before a row is deleted from Dept_tab, delete all Coding Triggers 9-31 Examples of Trigger Applications -- rows from the Emp_tab table whose DEPTNO is the same as -- the DEPTNO being deleted from the Dept_tab table: DELETE FROM Emp_tab WHERE Emp_tab.Deptno = :old.Deptno; END IF; END; CREATE OR REPLACE TRIGGER Dept_cascade3 AFTER UPDATE OF Deptno ON Dept_tab BEGIN UPDATE Emp_tab SET Update_id = NULL WHERE Update_id = Integritypackage.Updateseq; END; Because this trigger updates the Emp_tab table, the Emp_dept_check trigger, if enabled, is also fired. The resulting mutating table error is trapped by the Emp_dept_check trigger. You should carefully test any triggers that require error trapping to succeed to ensure that they always work properly in your environment. Note: Trigger for Complex Check Constraints: Example Triggers can enforce integrity rules other than referential integrity. For example, this trigger performs a complex check before allowing the triggering statement to run. You may need to set up the following data structures for the example to work: Note: CREATE TABLE Salgrade ( Grade Losal Hisal Job_classification NUMBER, NUMBER, NUMBER, NUMBER) CREATE OR REPLACE TRIGGER Salary_check BEFORE INSERT OR UPDATE OF Sal, Job ON Emp99 FOR EACH ROW DECLARE Minsal NUMBER; Maxsal NUMBER; Salary_out_of_range EXCEPTION; BEGIN /* Retrieve the minimum and maximum salary for the employee's new job classification from the SALGRADE table into MINSAL and MAXSAL: */ SELECT Minsal, Maxsal INTO Minsal, Maxsal FROM Salgrade WHERE Job_classification = :new.Job; /* If the employee's new salary is less than or greater than the job classification's limits, the exception is raised. The exception message is returned and the pending INSERT or UPDATE statement that fired the trigger is rolled back:*/ IF (:new.Sal < Minsal OR :new.Sal > Maxsal) THEN 9-32 Oracle Database Application Developer’s Guide - Fundamentals Examples of Trigger Applications RAISE Salary_out_of_range; END IF; EXCEPTION WHEN Salary_out_of_range THEN Raise_application_error (-20300, 'Salary '||TO_CHAR(:new.Sal)||' out of range for ' ||'job classification '||:new.Job ||' for employee '||:new.Ename); WHEN NO_DATA_FOUND THEN Raise_application_error(-20322, 'Invalid Job Classification ' ||:new.Job_classification); END; Complex Security Authorizations and Triggers: Example Triggers are commonly used to enforce complex security authorizations for table data. Only use triggers to enforce complex security authorizations that cannot be defined using the database security features provided with Oracle Database. For example, a trigger can prohibit updates to salary data of the Emp_tab table during weekends, holidays, and non-working hours. When using a trigger to enforce a complex security authorization, it is best to use a BEFORE statement trigger. Using a BEFORE statement trigger has these benefits: ■ ■ The security check is done before the triggering statement is allowed to run, so that no wasted work is done by an unauthorized statement. The security check is performed only once for the triggering statement, not for each row affected by the triggering statement. This example shows a trigger used to enforce security. You may need to set up the following data structures for the example to work: Note: CREATE TABLE Company_holidays (Day DATE); CREATE OR REPLACE TRIGGER Emp_permit_changes BEFORE INSERT OR DELETE OR UPDATE ON Emp99 DECLARE Dummy INTEGER; Not_on_weekends EXCEPTION; Not_on_holidays EXCEPTION; Non_working_hours EXCEPTION; BEGIN /* check for weekends: */ IF (TO_CHAR(Sysdate, 'DY') = 'SAT' OR TO_CHAR(Sysdate, 'DY') = 'SUN') THEN RAISE Not_on_weekends; END IF; /* check for company holidays:*/ SELECT COUNT(*) INTO Dummy FROM Company_holidays WHERE TRUNC(Day) = TRUNC(Sysdate); /* TRUNC gets rid of time parts of dates: */ IF dummy > 0 THEN RAISE Not_on_holidays; END IF; /* Check for work hours (8am to 6pm): */ IF (TO_CHAR(Sysdate, 'HH24') < 8 OR Coding Triggers 9-33 Examples of Trigger Applications TO_CHAR(Sysdate, 'HH24') > 18) THEN RAISE Non_working_hours; END IF; EXCEPTION WHEN Not_on_weekends THEN Raise_application_error(-20324,'May not change ' ||'employee table during the weekend'); WHEN Not_on_holidays THEN Raise_application_error(-20325,'May not change ' ||'employee table during a holiday'); WHEN Non_working_hours THEN Raise_application_error(-20326,'May not change ' ||'Emp_tab table during non-working hours'); END; See Also: Oracle Database Security Guide for details on database security features Transparent Event Logging and Triggers Triggers are very useful when you want to transparently perform a related change in the database following certain events. The REORDER trigger example shows a trigger that reorders parts as necessary when certain conditions are met. (In other words, a triggering statement is entered, and the PARTS_ON_HAND value is less than the REORDER_POINT value.) Derived Column Values and Triggers: Example Triggers can derive column values automatically, based upon a value provided by an INSERT or UPDATE statement. This type of trigger is useful to force values in specific columns that depend on the values of other columns in the same row. BEFORE row triggers are necessary to complete this type of operation for the following reasons: ■ ■ The dependent values must be derived before the INSERT or UPDATE occurs, so that the triggering statement can use the derived values. The trigger must fire for each row affected by the triggering INSERT or UPDATE statement. The following example illustrates how a trigger can be used to derive new column values for a table whenever a row is inserted or updated. You may need to set up the following data structures for the example to work: Note: ALTER TABLE Emp99 ADD( Uppername VARCHAR2(20), Soundexname VARCHAR2(20)); CREATE OR REPLACE TRIGGER Derived BEFORE INSERT OR UPDATE OF Ename ON Emp99 /* Before updating the ENAME field, derive the values for the UPPERNAME and SOUNDEXNAME fields. Users should be restricted from updating these fields directly: */ FOR EACH ROW BEGIN :new.Uppername := UPPER(:new.Ename); 9-34 Oracle Database Application Developer’s Guide - Fundamentals Examples of Trigger Applications :new.Soundexname := SOUNDEX(:new.Ename); END; Building Complex Updatable Views Using Triggers: Example Views are an excellent mechanism to provide logical windows over table data. However, when the view query gets complex, the system implicitly cannot translate the DML on the view into those on the underlying tables. INSTEAD OF triggers help solve this problem. These triggers can be defined over views, and they fire instead of the actual DML. Consider a library system where books are arranged under their respective titles. The library consists of a collection of book type objects. The following example explains the schema. CREATE OR REPLACE TYPE Book_t AS OBJECT ( Booknum NUMBER, Title VARCHAR2(20), Author VARCHAR2(20), Available CHAR(1) ); CREATE OR REPLACE TYPE Book_list_t AS TABLE OF Book_t; Assume that the following tables exist in the relational schema: Table Book_table (Booknum, Section, Title, Author, Available) Booknum Section Title Author Available 121001 Classic Iliad Homer Y 121002 Novel Gone With the Wind Mitchell M N Library consists of library_table(section). Section Geography Classic You can define a complex view over these tables to create a logical view of the library with sections and a collection of books in each section. CREATE OR REPLACE VIEW Library_view AS SELECT i.Section, CAST (MULTISET ( SELECT b.Booknum, b.Title, b.Author, b.Available FROM Book_table b WHERE b.Section = i.Section) AS Book_list_t) BOOKLIST FROM Library_table i; Make this view updatable by defining an INSTEAD OF trigger over the view. CREATE OR REPLACE TRIGGER Library_trigger INSTEAD OF INSERT ON Library_view FOR EACH ROW Bookvar BOOK_T; i INTEGER; BEGIN INSERT INTO Library_table VALUES (:NEW.Section); FOR i IN 1..:NEW.Booklist.COUNT LOOP Coding Triggers 9-35 Examples of Trigger Applications Bookvar := Booklist(i); INSERT INTO book_table VALUES ( Bookvar.booknum, :NEW.Section, Bookvar.Title, Bookvar.Author, bookvar.Available); END LOOP; END; / The library_view is an updatable view, and any INSERTs on the view are handled by the trigger that gets fired automatically. For example: INSERT INTO Library_view VALUES ('History', book_list_t(book_t(121330, 'Alexander', 'Mirth', 'Y'); Similarly, you can also define triggers on the nested table booklist to handle modification of the nested table element. Tracking System Events Using Triggers Fine-Grained Access Control Using Triggers: Example System triggers can be used to set application context. Application context is a relatively new feature that enhances your ability to implement fine-grained access control. Application context is a secure session cache, and it can be used to store session-specific attributes. In the example that follows, procedure set_ctx sets the application context based on the user profile. The trigger setexpensectx ensures that the context is set for every user. CONNECT secdemo/secdemo CREATE OR REPLACE CONTEXT Expenses_reporting USING Secdemo.Exprep_ctx; REM ================================================================= REM Creation of the package which implements the context: REM ================================================================= CREATE OR REPLACE PACKAGE Exprep_ctx AS PROCEDURE Set_ctx; END; SHOW ERRORS CREATE OR REPLACE PACKAGE BODY Exprep_ctx IS PROCEDURE Set_ctx IS Empnum NUMBER; Countrec NUMBER; Cc NUMBER; Role VARCHAR2(20); BEGIN -- SET emp_number: SELECT Employee_id INTO Empnum FROM Employee WHERE Last_name = SYS_CONTEXT('userenv', 'session_user'); DBMS_SESSION.SET_CONTEXT('expenses_reporting','emp_number', Empnum); -- SET ROLE: SELECT COUNT (*) INTO Countrec FROM Cost_center WHERE Manager_id=Empnum; IF (countrec > 0) THEN DBMS_SESSION.SET_CONTEXT('expenses_reporting','exp_role','MANAGER'); 9-36 Oracle Database Application Developer’s Guide - Fundamentals Responding to System Events through Triggers ELSE DBMS_SESSION.SET_CONTEXT('expenses_reporting','exp_role','EMPLOYEE'); END IF; -- SET cc_number: SELECT Cost_center_id INTO Cc FROM Employee WHERE Last_name = SYS_CONTEXT('userenv','session_user'); DBMS_SESSION.SET_CONTEXT(expenses_reporting','cc_number',Cc); END; END; CALL Syntax CREATE OR REPLACE TRIGGER Secdemo.Setexpseetx AFTER LOGON ON DATABASE CALL Secdemo.Exprep_etx.Set_otx Responding to System Events through Triggers System event publication lets applications subscribe to database events, just like they subscribe to messages from other applications. The system events publication framework includes the following features: ■ ■ ■ Infrastructure for publish/subscribe, by making the database an active publisher of events. Integration of data cartridges in the server. The system events publication can be used to notify cartridges of state changes in the server. Integration of fine-grained access control in the server. By creating a trigger, you can specify a procedure that runs when an event occurs. DML events are supported on tables, and system events are supported on DATABASE and SCHEMA. You can turn notification on and off by enabling and disabling the trigger using the ALTER TRIGGER statement. This feature is integrated with the Advanced Queueing engine. Publish/subscribe applications use the DBMS_AQ.ENQUEUE() procedure, and other applications such as cartridges use callouts. See Also: ■ ■ Oracle Database SQL Reference Oracle Streams Advanced Queuing User's Guide and Reference for details on how to subscribe to published events How Events Are Published Through Triggers When events are detected by the database, the trigger mechanism executes the action specified in the trigger. As part of this action, you can use the DBMS_AQ package to publish the event to a queue so that subscribers receive notifications. Note: Only system-defined database events can be detected this way. You cannot define your own event conditions. When an event occurs, the database fires all triggers that are enabled on that event, with some exceptions: Coding Triggers 9-37 Responding to System Events through Triggers ■ ■ If the trigger is actually the target of the triggering event, it is not fired. For example, a trigger for all DROP events is not fired when it is dropped itself. If a trigger has been modified but not committed within the same transaction as the firing event. For example, recursive DDL within a system trigger might modify a trigger, which prevents the modified trigger from being fired by events within the same transaction. You can create more than one trigger on an object. When an event fires more than one trigger, the firing order is not defined and so you should not rely on the triggers being fired in a particular order. Publication Context When an event is published, certain runtime context and attributes, as specified in the parameter list, are passed to the callout procedure. A set of functions called event attribute functions are provided. See Also: "Event Attribute Functions" on page 9-38 for information on event-specific attributes For each supported system event, you can identify and predefine event-specific attributes for the event. You can choose the parameter list to be any of these attributes, along with other simple expressions. For callouts, these are passed as IN arguments. Error Handling Return status from publication callout functions for all events are ignored. For example, with SHUTDOWN events, the database cannot do anything with the return status. See Also: "List of Database Events" on page 9-41 for details on return status Execution Model Traditionally, triggers execute as the definer of the trigger. The trigger action of an event is executed as the definer of the action (as the definer of the package or function in callouts, or as owner of the trigger in queues). Because the owner of the trigger must have EXECUTE privileges on the underlying queues, packages, or procedure, this behavior is consistent. Event Attribute Functions When the database fires a trigger, you can retrieve certain attributes about the event that fired the trigger. You can retrieve each attribute with a function call. Table 9–2 describes the system-defined event attributes. 9-38 Oracle Database Application Developer’s Guide - Fundamentals Responding to System Events through Triggers Note: ■ ■ ■ Table 9–2 To make these attributes available, you must first run the CATPROC.SQL script. The trigger dictionary object maintains metadata about events that will be published and their corresponding attributes. In earlier releases, these functions were accessed through the SYS package. We recommend you use these public synonyms whose names begin with ora_. System-Defined Event Attributes Attribute Type Description Example ora_client_ip_address VARCHAR2 Returns IP address of the client in a LOGON event when the underlying protocol is TCP/IP DECLARE v_addr VARCHAR2(11); IF (ora_sysevent = 'LOGON') THEN v_addr := ora_client_ip_address; END IF; END; ora_database_name VARCHAR2(50) Database name. DECLARE v_db_name VARCHAR2(50); BEGIN v_db_name := ora_database_name; END; ora_des_encrypted_password VARCHAR2 The DES-encrypted password of the user being created or altered. IF (ora_dict_obj_type = 'USER') THEN INSERT INTO event_table VALUES (ora_des_encrypted_password); END IF; ora_dict_obj_name VARCHAR(30) Name of the dictionary object on which the DDL operation occurred. INSERT INTO event_table VALUES ('Changed object is ' || ora_dict_obj_name); ora_dict_obj_name_list (name_list OUT ora_name_list_t) BINARY_INTEGER Return the list of object names of objects being modified in the event. IF (ora_sysevent='ASSOCIATE STATISTICS') THEN number_modified := ora_dict_obj_name_list(name_list); END IF; ora_dict_obj_owner VARCHAR(30) Owner of the dictionary object on which the DDL operation occurred. INSERT INTO event_table VALUES ('object owner is' || ora_dict_obj_owner); ora_dict_obj_owner_list (owner_list OUT ora_name_list_t) BINARY_INTEGER Returns the list of IF (ora_sysevent='ASSOCIATE STATISTICS') object owners of objects THEN number_of_modified_objects := being modified in the ora_dict_obj_owner_list(owner_list); event. END IF; ora_dict_obj_type VARCHAR(20) Type of the dictionary object on which the DDL operation occurred. INSERT INTO event_table VALUES ('This object is a ' || ora_dict_obj_type); ora_grantee (user_list OUT ora_name_list_t) BINARY_INTEGER Returns the grantees of a grant event in the OUT parameter; returns the number of grantees in the return value. IF (ora_sysevent = 'GRANT') THEN number_of_users=ora_grantee(user_list); END IF; ora_instance_num NUMBER Instance number. IF (ora_instance_num = 1) THEN INSERT INTO event_table VALUES ('1'); END IF; Coding Triggers 9-39 Responding to System Events through Triggers Table 9–2 (Cont.) System-Defined Event Attributes Attribute Type Description Example ora_is_alter_column (column_name IN VARCHAR2) BOOLEAN Returns true if the specified column is altered. IF (ora_sysevent = 'ALTER' AND ora_dict_obj_type = 'TABLE') THEN alter_column := ora_is_alter_column('C'); END IF; ora_is_creating_nested_table BOOLEAN Returns true if the current event is creating a nested table IF (ora_sysevent = 'CREATE' and ora_dict_obj_type = 'TABLE' and ora_is_creating_nested_table) THEN INSERT INTO event_table VALUES ('A nested table is created'); END IF; ora_is_drop_column (column_name IN VARCHAR2) BOOLEAN Returns true if the specified column is dropped. IF (ora_sysevent = 'ALTER' AND ora_dict_obj_type = 'TABLE') THEN drop_column := ora_is_drop_column('C'); END IF; ora_is_servererror BOOLEAN Returns TRUE if given error is on error stack, FALSE otherwise. IF (ora_is_servererror(error_number)) THEN INSERT INTO event_table VALUES ('Server error!!'); END IF; ora_login_user VARCHAR2(30) Login user name. SELECT ora_login_user FROM dual; ora_partition_pos BINARY_INTEGER In an INSTEAD OF trigger for CREATE TABLE, the position within the SQL text where you could insert a PARTITION clause. -- Retrieve ora_sql_txt into -- sql_text variable first. v_n := ora_partition_pos; v_new_stmt := SUBSTR(sql_text,1,v_n - 1) || ' ' || my_partition_clause || ' ' || SUBSTR(sql_text, v_n)); ora_privilege_list (privilege_list OUT ora_name_list_t) BINARY_INTEGER Returns the list of IF (ora_sysevent = 'GRANT' OR privileges being ora_sysevent = 'REVOKE') THEN granted by the grantee number_of_privileges := or the list of privileges ora_privilege_list(priv_list); revoked from the END IF; revokees in the OUT parameter; returns the number of privileges in the return value. ora_revokee (user_list OUT ora_name_list_t) BINARY_INTEGER Returns the revokees of IF (ora_sysevent = 'REVOKE') THEN a revoke event in the number_of_users := ora_revokee(user_list); OUT parameter; returns the number of revokees in the return value. ora_server_error NUMBER Given a position (1 for top of stack), it returns the error number at that position on error stack INSERT INTO event_table VALUES ('top stack error ' || ora_server_error(1)); ora_server_error_depth BINARY_INTEGER Returns the total number of error messages on the error stack. n := ora_server_error_depth; -- This value is used with other functions -- such as ora_server_error ora_server_error_msg (position in binary_integer) VARCHAR2 Given a position (1 for top of stack), it returns the error message at that position on error stack INSERT INTO event_table VALUES ('top stack error message' || ora_server_error_msg(1)); 9-40 Oracle Database Application Developer’s Guide - Fundamentals Responding to System Events through Triggers Table 9–2 (Cont.) System-Defined Event Attributes Attribute Type Description Example ora_server_error_num_params (position in binary_integer) BINARY_INTEGER Given a position (1 for top of stack), it returns the number of strings that have been substituted into the error message using a format like %s. n := ora_server_error_num_params(1); ora_server_error_param (position in binary_integer, param in binary_integer) VARCHAR2 Given a position (1 for top of stack) and a parameter number, returns the matching substitution value (%s, %d, and so on) in the error message. -- For example, the second %s in a -- message: "Expected %s, found %s" param := ora_server_error_param(1,2); ora_sql_txt (sql_text out ora_name_list_t) BINARY_INTEGER Returns the SQL text of the triggering statement in the OUT parameter. If the statement is long, it is broken into multiple PL/SQL table elements. The function return value shows the number of elements are in the PL/SQL table. sql_text ora_name_list_t; v_stmt VARCHAR2(2000); ... n := ora_sql_txt(sql_text); FOR i IN 1..n LOOP v_stmt := v_stmt || sql_text(i); END LOOP; INSERT INTO event_table VALUES ('text of triggering statement: ' || v_stmt); ora_sysevent VARCHAR2(20) System event firing the trigger: Event name is same as that in the syntax. INSERT INTO event_table VALUES (ora_sysevent); ora_with_grant_option BOOLEAN Returns true if the privileges are granted with grant option. IF (ora_sysevent = 'GRANT' and ora_with_grant_option = TRUE) THEN INSERT INTO event_table VALUES ('with grant option'); END IF; space_error_info (error_number OUT NUMBER, error_type OUT VARCHAR2, object_owner OUT VARCHAR2, table_space_name OUT VARCHAR2, object_name OUT VARCHAR2, sub_object_name OUT VARCHAR2) BOOLEAN Returns true if the error IF (space_error_info(eno,typ,owner,ts,obj, is related to an subobj) = TRUE) THEN out-of-space condition, DBMS_OUTPUT.PUT_LINE('The object '|| obj and fills in the OUT || ' owned by ' || owner || parameters with ' has run out of space.'); information about the END IF; object that caused the error. List of Database Events This section describes important system events and client events. System Events System events are related to entire instances or schemas, not individual tables or rows. Triggers created on startup and shutdown events must be associated with the database instance. Triggers created on error and suspend events can be associated with either the database instance or a particular schema. Table 9–3 contains a list of system manager events. Coding Triggers 9-41 Responding to System Events through Triggers Table 9–3 System Manager Events Event When Fired? Conditions STARTUP When the database is opened. None allowed Restrictions Transaction Attribute Functions No database operations allowed in the trigger. Starts a separate transaction and commits it after firing the triggers. ora_sysevent ora_login_user ora_instance_num ora_database_name Starts a separate transaction and commits it after firing the triggers. ora_sysevent ora_login_user ora_instance_num ora_database_name Starts a separate transaction and commits it after firing the triggers. ora_sysevent ora_login_user ora_instance_num ora_database_name Starts a separate transaction and commits it after firing the triggers. ora_sysevent ora_login_user ora_instance_num ora_database_name ora_server_error ora_is_servererror space_error_info Return status ignored. SHUTDOWN Just before the server starts the shutdown of an instance. None allowed This lets the cartridge shutdown completely. For abnormal instance shutdown, this event may not be fired. DB_ROLE_CHANGE When the database is opened for the first time after a role change. SERVERERROR When the error eno occurs. If no condition is given, then this event fires whenever an error occurs. No database operations allowed in the trigger. Return status ignored. None allowed Return status ignored. ERRNO = eno Depends on the error. Return status ignored. The trigger does not fire on ORA-1034, ORA-1403, ORA-1422, ORA-1423, and ORA-4030 because they are not true errors or are too serious to continue processing. It also fails to fire on ORA-18 and ORA-20 because a process is not available to connect to the database to record the error. Client Events Client events are the events related to user logon/logoff, DML, and DDL operations. For example: CREATE OR REPLACE TRIGGER On_Logon AFTER LOGON ON The_user.Schema BEGIN Do_Something; END; The LOGON and LOGOFF events allow simple conditions on UID and USER. All other events allow simple conditions on the type and name of the object, as well as functions like UID and USER. The LOGON event starts a separate transaction and commits it after firing the triggers. All other events fire the triggers in the existing user transaction. The LOGON and LOGOFF events can operate on any objects. For all other events, the corresponding trigger cannot perform any DDL operations, such as DROP and ALTER, on the object that caused the event to be generated. The DDL allowed inside these triggers is altering, creating, or dropping a table, creating a trigger, and compile operations. If an event trigger becomes the target of a DDL operation (such as CREATE TRIGGER), it cannot be fired later during the same transaction Table 9–4 contains a list of client events. 9-42 Oracle Database Application Developer’s Guide - Fundamentals Responding to System Events through Triggers Table 9–4 Client Events Event When Fired? Attribute Functions BEFORE ALTER When a catalog object is altered. ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_type ora_dict_obj_name ora_dict_obj_owner ora_des_encrypted_password (for ALTER USER events) ora_is_alter_column (for ALTER TABLE events) ora_is_drop_column (for ALTER TABLE events) When a catalog object is dropped. ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_type ora_dict_obj_name ora_dict_obj_owner When an analyze statement is issued ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_name ora_dict_obj_type ora_dict_obj_owner When an associate statistics statement is issued ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_name ora_dict_obj_type ora_dict_obj_owner ora_dict_obj_name_list ora_dict_obj_owner_list When an audit or noaudit statement is issued ora_sysevent ora_login_user ora_instance_num ora_database_name When an object is commented ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_name ora_dict_obj_type ora_dict_obj_owner When a catalog object is created. ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_type ora_dict_obj_name ora_dict_obj_owner ora_is_creating_nested_table (for CREATE TABLE events) AFTER ALTER BEFORE DROP AFTER DROP BEFORE ANALYZE AFTER ANALYZE BEFORE ASSOCIATE STATISTICS AFTER ASSOCIATE STATISTICS BEFORE AUDIT AFTER AUDIT BEFORE NOAUDIT AFTER NOAUDIT BEFORE COMMENT AFTER COMMENT BEFORE CREATE AFTER CREATE Coding Triggers 9-43 Responding to System Events through Triggers Table 9–4 (Cont.) Client Events Event When Fired? Attribute Functions BEFORE DDL AFTER DDL When most SQL DDL statements are issued. Not fired for ALTER DATABASE, CREATE CONTROLFILE, CREATE DATABASE, and DDL issued through the PL/SQL procedure interface, such as creating an advanced queue. ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_name ora_dict_obj_type ora_dict_obj_owner BEFORE DISASSOCIATE STATISTICS When a disassociate statistics statement is issued ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_name ora_dict_obj_type ora_dict_obj_owner ora_dict_obj_name_list ora_dict_obj_owner_list When a grant statement is issued ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_name ora_dict_obj_type ora_dict_obj_owner ora_grantee ora_with_grant_option ora_privileges BEFORE LOGOFF At the start of a user logoff ora_sysevent ora_login_user ora_instance_num ora_database_name AFTER LOGON After a successful logon of a user. ora_sysevent ora_login_user ora_instance_num ora_database_name ora_client_ip_address BEFORE RENAME When a rename statement is issued. ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_name ora_dict_obj_owner ora_dict_obj_type AFTER DISASSOCIATE STATISTICS BEFORE GRANT AFTER GRANT AFTER RENAME 9-44 Oracle Database Application Developer’s Guide - Fundamentals Responding to System Events through Triggers Table 9–4 (Cont.) Client Events Event When Fired? Attribute Functions BEFORE REVOKE When a revoke statement is issued ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_name ora_dict_obj_type ora_dict_obj_owner ora_revokee ora_privileges AFTER SUSPEND After a SQL statement is suspended because of an out-of-space condition. The trigger should correct the condition so the statement can be resumed. ora_sysevent ora_login_user ora_instance_num ora_database_name ora_server_error ora_is_servererror space_error_info BEFORE TRUNCATE When an object is truncated ora_sysevent ora_login_user ora_instance_num ora_database_name ora_dict_obj_name ora_dict_obj_type ora_dict_obj_owner AFTER REVOKE AFTER TRUNCATE Coding Triggers 9-45 Responding to System Events through Triggers 9-46 Oracle Database Application Developer’s Guide - Fundamentals 10 Developing Flashback Applications This chapter discusses the following flashback topics: ■ Overview of Flashback Features ■ Database Administration Tasks Before Using Flashback Features ■ Using Flashback Query (SELECT ... AS OF) ■ Using the DBMS_FLASHBACK Package ■ Using ORA_ROWSCN ■ Using Flashback Version Query ■ Using Flashback Transaction Query ■ Flashback Tips See Also: ■ ■ Oracle Database Backup and Recovery Advanced User's Guide and Oracle Database Administrator's Guide for information on flashback features designed for database administration tasks such as Oracle Flashback Database and Oracle Flashback Table Oracle Database SQL Reference for the syntax of SQL extensions for flashback features Overview of Flashback Features Oracle Database has a group of features, known collectively as flashback, that provide ways to view past states of database objects or to return database objects to a previous state without using point-in-time media recovery. You can use flashback features of the database to do the following: ■ ■ ■ Perform queries that return past data. Perform queries that return metadata that shows a detailed history of changes to the database. Recover tables or rows to a previous point in time. Flashback features use the Automatic Undo Management system to obtain metadata and historical data for transactions. They rely on undo data, which are records of the effects of individual transactions. For example, if a user executes an UPDATE statement to change a salary from 1000 to 1100, then Oracle would store the value 1000 in the undo data. Developing Flashback Applications 10-1 Overview of Flashback Features Undo data is persistent and survives a database shutdown. By using flashback features, you can employ undo data to query past data or recover from logical corruptions. Besides using it in flashback operations, Oracle Database uses undo data to perform the following actions: ■ Roll back active transactions ■ Recover terminated transactions by using database or process recovery ■ Provide read consistency for SQL queries Oracle Database Concepts for more information about flashback features and automatic undo management See Also: Application Development Features In application development, you can use flashback features to report on historical data or undo erroneous changes. Flashback features include the following: ■ Oracle Flashback Query You can use this feature to retrieve data for a time in the past that you specify using the AS OF clause of the SELECT statement. ■ Oracle Flashback Version Query You can use this feature to retrieve metadata and historical data for a specific time interval. For example, you can view all the rows of a table that ever existed during a given time interval. Metadata about the different versions of rows includes start and end time, type of change operation, and identity of the transaction that created the row version. You use the VERSIONS BETWEEN clause of the SELECT statement to create a Flashback Version Query. ■ Oracle Flashback Transaction Query You can use this feature to retrieve metadata and historical data for a given transaction or for all transactions within a given time interval. You can also obtain the SQL code to undo the changes to particular rows affected by a transaction. Typically, you use Flashback Transaction Query in conjunction with a Flashback Version Query that provides the transaction IDs for the rows of interest. To perform a Flashback Transaction Query, select from the FLASHBACK_TRANSACTION_QUERY view. ■ DBMS_FLASHBACK package You can use this feature to set the internal Oracle clock to a time in the past so that you can examine data current at that time. Database Administration Features You can use the following features for application development or interactively as a database user or administrator: ■ DBMS_FLASHBACK package ■ Flashback Query ■ Flashback Version Query ■ Flashback Transaction Query Typically, you use the following flashback features only in database administration: ■ Oracle Flashback Table 10-2 Oracle Database Application Developer’s Guide - Fundamentals Database Administration Tasks Before Using Flashback Features You can use this feature to recover a table to its state at a previous point in time. You can restore table data while the database is on line, undoing changes to only the specified table. ■ Oracle Flashback Drop You can use this feature to recover a dropped table. This reverses the effects of a DROP TABLE statement. ■ Oracle Flashback Database You can use this feature to quickly return the database to an earlier point in time, by undoing all of the changes that have taken place since then. This is fast, because you do not have to restore database backups. Flashback Database, Flashback Table, and Flashback Drop are primarily data recovery mechanisms and are therefore documented elsewhere. The other flashback features, while valuable in data recovery scenarios, are useful for application development. They are the focus of this chapter. See Also: ■ ■ ■ ■ Oracle Database Backup and Recovery Advanced User's Guide Oracle Database Administrator's Guide to learn about the Flashback Drop feature Oracle Database Administrator's Guide to learn about the Flashback Table feature Oracle Database Administrator's Guide to learn about Automatic Undo Management Database Administration Tasks Before Using Flashback Features Before you can use flashback features in your application, you must perform the following administrative tasks to configure your database. Consult with your database administrator to perform these tasks: ■ ■ Create an undo tablespace with enough space to keep the required data for flashback operations. The more often users update the data, the more space is required. Calculating the space requirements is usually performed by a database administrator. You can find the calculation formula in the Oracle Database Administrator's Guide. Enable Automatic Undo Management, as explained in Oracle Database Administrator's Guide. In particular, you must set the following database initialization parameters: – UNDO_MANAGEMENT – UNDO_TABLESPACE Note that for an undo tablespace with a fixed size, Oracle Database automatically performs the following actions: – Tunes the system to give the best possible undo retention for the undo tablespace. For an automatically extensible undo tablespace, Oracle Database retains undo data longer than the longest query duration as well as the low threshold of undo retention specified by the UNDO_RETENTION parameter. Developing Flashback Applications 10-3 Using Flashback Query (SELECT ... AS OF) Note: You can query V$UNDOSTAT.TUNED_UNDORETENTION to determine the amount of time for which undo is retained for the current undo tablespace. ■ ■ ■ Specify the RETENTION GUARANTEE clause for the undo tablespace to ensure that unexpired undo is not discarded. Setting UNDO_RETENTION is not, by itself, a strict guarantee. If the system is under space pressure, then Oracle can overwrite unexpired undo with freshly generated undo. Specifying RETENTION GUARANTEE prevents this behavior. Grant flashback privileges to users, roles, or applications that need to use flashback features as follows: – For the DBMS_FLASHBACK package, grant the EXECUTE privilege on DBMS_FLASHBACK to provide access to the features in this package. – For Flashback Query and Flashback Version Query, grant FLASHBACK and SELECT privileges on specific objects to be accessed during queries or grant the FLASHBACK ANY TABLE privilege to allow queries on all tables. – For Flashback Transaction Query, grant the SELECT ANY TRANSACTION privilege. – For Execution of undo SQL code, grant SELECT, UPDATE, DELETE, and INSERT privileges for specific tables, as appropriate, to permit execution of undo SQL code retrieved by a Flashback Transaction Query. To use the Flashback Transaction Query feature in Oracle Database 10g, the database must be running with version 10.0 compatibility, and must have supplemental logging turned on with the following SQL statement: ALTER DATABASE ADD SUPPLEMENTAL LOG DATA; ■ To enable flashback operations on specific LOB columns of a table, use the ALTER TABLE command with the RETENTION option. Because undo data for LOB columns can be voluminous, you must define which LOB columns to use with flashback operations. See Also: ■ ■ Oracle Database Backup and Recovery Advanced User's Guide and Oracle Database Administrator's Guide to learn about DBA tasks such as setting up automatic undo management and granting privileges Oracle Database Application Developer's Guide - Large Objects to learn about LOB storage and the RETENTION parameter Using Flashback Query (SELECT ... AS OF) You perform a Flashback Query by using a SELECT statement with an AS OF clause. You use a Flashback Query to retrieve data as it existed at some time in the past. The query explicitly references a past time by means of a timestamp or SCN. It returns committed data that was current at that point in time. Potential uses of Flashback Query include: ■ Recovering lost data or undoing incorrect, committed changes. For example, if you mistakenly delete or update rows, and then commit them, you can immediately undo the mistake. 10-4 Oracle Database Application Developer’s Guide - Fundamentals Using Flashback Query (SELECT ... AS OF) ■ ■ ■ ■ ■ Comparing current data with the corresponding data at some time in the past. For example, you might run a daily report that shows the change in data from yesterday. You can compare individual rows of table data or find intersections or unions of sets of rows. Checking the state of transactional data at a particular time. For example, you could verify the account balance of a certain day. Simplifying application design, by removing the need to store some kinds of temporal data. By using a Flashback Query, you can retrieve past data directly from the database. Applying packaged applications such as report generation tools to past data. Providing self-service error correction for an application, thereby enabling users to undo and correct their errors. Oracle Database SQL Reference for details on the syntax of the SELECT... AS OF statement See Also: Examining Past Data: Example This example uses a Flashback Query to examine the state of a table at a previous time. Suppose that a DBA discovers at 12:30 PM that the row for employee Chung had been deleted from the employees table. The DBA also knows that at 9:30AM the data for Chung was correctly stored in the database. The DBA can use a Flashback Query to examine the contents of the table at 9:30 to find out what data had been lost. If appropriate, the DBA can then re-insert the lost data. Example 10–1 retrieves the state of the record for Chung at 9:30AM, April 4, 2004: Example 10–1 Retrieving a Row with Flashback Query SELECT * FROM employees AS OF TIMESTAMP TO_TIMESTAMP('2004-04-04 09:30:00', 'YYYY-MM-DD HH:MI:SS') WHERE last_name = 'Chung'; The update in Example 10–2 restores Chung's information to the employees table: Example 10–2 Reinserting a Row After a Flashback Query INSERT INTO employees (SELECT * FROM employees AS OF TIMESTAMP TO_TIMESTAMP('2004-04-04 09:30:00', 'YYYY-MM-DD HH:MI:SS') WHERE last_name = 'Chung'); Tips for Using Flashback Query Keep the following in mind when using a Flashback Query (SELECT ... AS OF): ■ ■ ■ You can specify or omit the AS OF clause for each table and specify different times for different tables. Use an AS OF clause in a query to perform DDL operations (such as creating and truncating tables) or DML operations (such as inserting and deleting) in the same session as the query. To use the results of a Flashback Query in a DDL or DML statement that affects the current state of the database, use an AS OF clause inside an INSERT or CREATE TABLE AS SELECT statement. When choosing whether to use a timestamp or an SCN in Flashback Query, remember that Oracle Database uses SCNs internally and maps these to Developing Flashback Applications 10-5 Using the DBMS_FLASHBACK Package timestamps at a granularity of 3 seconds. If a possible 3-second error (maximum) is important to a Flashback Query in your application, then use an SCN instead of a timestamp. Refer to "Flashback Tips – General". ■ You can create a view that refers to past data by using the AS OF clause in the SELECT statement that defines the view. If you specify a relative time by subtracting from the current time on the database host, then the past time is recalculated for each query. For example: CREATE VIEW hour_ago AS SELECT * FROM employees AS OF TIMESTAMP (SYSTIMESTAMP - INTERVAL '60' MINUTE); -- SYSTIMESTAMP refers to the time zone of the database host environment ■ You can use the AS OF clause in self-joins, or in set operations such as INTERSECT and MINUS, to extract or compare data from two different times. You can store the results by preceding a Flashback Query with a CREATE TABLE AS SELECT or INSERT INTO TABLE SELECT statement. For example, the following query reinserts into table employees the rows that existed an hour ago: INSERT INTO employees (SELECT * FROM employees AS OF TIMESTAMP (SYSTIMESTAMP - INTERVAL '60' MINUTE)) -- SYSTIMESTAMP refers to the time zone of the database host environment MINUS SELECT * FROM employees); Using the DBMS_FLASHBACK Package In general, the DBMS_FLASHBACK package provides the same functionality as Flashback Query, but Flashback Query is sometimes more convenient. The DBMS_FLASHBACK package acts as a time machine: you can turn back the clock, carry out normal queries as if you were at that time in the past, and then return to the present. Because you can use the DBMS_FLASHBACK package to perform queries on past data without special clauses such as AS OF or VERSIONS BETWEEN, you can reuse existing PL/SQL code to query the database at times in the past. You must have the EXECUTE privilege on the DBMS_FLASHBACK package. To use the DBMS_FLASHBACK package in your PL/SQL code: 1. Call DBMS_FLASHBACK.ENABLE_AT_TIME or DBMS_FLASHBACK.ENABLE_AT_SYSTEM_CHANGE_NUMBER to turn back the clock to a specified time in the past. Afterwards all queries retrieve data that was current at the specified time. 2. Perform normal queries, that is, without any special flashback-feature syntax such as AS OF. The database is automatically queried at the specified past time. Perform only queries; do not try to perform DDL or DML operations. 3. Call DBMS_FLASHBACK.DISABLE to return to the present. You must call DISABLE before calling ENABLE again for a different time. You cannot nest ENABLE /DISABLE pairs. You can use a cursor to store the results of queries. To do this, open the cursor before calling DBMS_FLASHBACK.DISABLE. After storing the results and then calling DISABLE, you can do the following: ■ Perform INSERT or UPDATE operations to modify the current database state by using the stored results from the past. 10-6 Oracle Database Application Developer’s Guide - Fundamentals Using ORA_ROWSCN ■ Compare current data with the past data. After calling DISABLE, open a second cursor. Fetch from the first cursor to retrieve past data; fetch from the second cursor to retrieve current data. You can store the past data in a temporary table, and then use set operators such as MINUS or UNION to contrast or combine the past and current data. You can call DBMS_FLASHBACK.GET_SYSTEM_CHANGE_NUMBER at any time to obtain the current System Change Number (SCN). Note that the current SCN is always returned; this takes no account of previous calls to DBMS_FLASHBACK.ENABLE*. See Also: ■ ■ Oracle Database PL/SQL Packages and Types Reference for details about the DBMS_FLASHBACK package Oracle Database Reference and Oracle Database Backup and Recovery Reference for information about SCNs Using ORA_ROWSCN ORA_ROWSCN is a pseudocolumn of any table that is not fixed or external. It represents the SCN of the most recent change to a given row, that is, the latest COMMIT operation for the row. For example: SELECT ora_rowscn, last_name, salary FROM employees WHERE employee_id = 7788; ORA_ROWSCN ---------202553 NAME ---Fudd SALARY -----3000 The latest COMMIT operation for the row took place at approximately SCN 202553. You can use function SCN_TO_TIMESTAMP to convert an SCN, like ORA_ROWSCN, to the corresponding TIMESTAMP value. ORA_SCN is in fact a conservative upper bound of the latest commit time: the actual commit SCN can be somewhat earlier. ORA_SCN is more precise (closer to the actual commit SCN) for a row-dependent table (created using CREATE TABLE with the ROWDEPENDENCIES clause). Noteworthy uses of ORA_ROWSCN in application development include concurrency control and client cache invalidation. To see how you might use it in concurrency control, consider the following scenario. Your application examines a row of data, and records the corresponding ORA_ROWSCN as 202553. Later, the application needs to update the row, but only if its record of the data is still accurate. That is, this particular update operation depends, logically, on the row not having been changed. The operation is therefore made conditional on the ORA_ROWSCN being still 202553. Here is an equivalent interactive command: UPDATE employees SET salary = salary + 100 WHERE employee_id = 7788 AND ora_rowscn = 202553; 0 rows updated. Developing Flashback Applications 10-7 Using Flashback Version Query The conditional update fails in this case, because the ORA_ROWSCN is no longer 202553. This means that some user or another application changed the row and performed a COMMIT more recently than the recorded ORA_ROWSCN. Your application queries again to obtain the new row data and ORA_ROWSCN. Suppose that the ORA_ROWSCN is now 415639. The application tries the conditional update again, using the new ORA_ROWSCN. This time, the update succeeds, and it is committed. Here is an interactive equivalent: SQL> UPDATE employees SET salary = salary + 100 WHERE empno = 7788 AND ora_rowscn = 415639; 1 row updated. SQL> COMMIT; Commit complete. SQL> SELECT ora_rowscn, name, salary FROM employees WHERE empno = 7788; ORA_ROWSCN ---------465461 NAME ---Fudd SALARY -----3100 The SCN corresponding to the new COMMIT is 465461. Besides using ORA_ROWSCN in an UPDATE statement WHERE clause, you can use it in a DELETE statement WHERE clause or the AS OF clause of a Flashback Query. See Also: ■ Oracle Database SQL Reference Using Flashback Version Query You use a Flashback Version Query to retrieve the different versions of specific rows that existed during a given time interval. A new row version is created whenever a COMMIT statement is executed. You specify a Flashback Version Query using the VERSIONS BETWEEN clause of the SELECT statement. Here is the syntax: VERSIONS {BETWEEN {SCN | TIMESTAMP} start AND end} where start and end are expressions representing the start and end of the time interval to be queried, respectively. The interval is closed at both ends: the upper and lower limits specified (start and end) are both included in the time interval. The Flashback Version Query returns a table with a row for each version of the row that existed at any time during the time interval you specify. Each row in the table includes pseudocolumns of metadata about the row version, described in Table 10–1. This information can reveal when and how a particular change (perhaps erroneous) occurred to your database. 10-8 Oracle Database Application Developer’s Guide - Fundamentals Using Flashback Version Query Table 10–1 Flashback Version Query Row Data Pseudocolumns Pseudocolumn Name Description Starting System Change Number (SCN) or TIMESTAMP when the row version was created. This identifies the time when the data first VERSIONS_STARTTIME took on the values reflected in the row version. You can use this to identify the past target time for a Flashback Table or Flashback Query operation. VERSIONS_STARTSCN If this is NULL, then the row version was created before the lower time bound of the query BETWEEN clause. VERSIONS_ENDSCN VERSIONS_ENDTIME VERSIONS_XID SCN or TIMESTAMP when the row version expired. This identifies the row expiration time. If this is NULL, then either the row version was still current at the time of the query or the row corresponds to a DELETE operation. Identifier of the transaction that created the row version. VERSIONS_OPERATION Operation performed by the transaction: I for insertion, D for deletion, or U for update. The version is that of the row that was inserted, deleted, or updated; that is, the row after an INSERT operation, the row before a DELETE operation, or the row affected by an UPDATE operation. Note: For user updates of an index key, a Flashback Version Query may treat an UPDATE operation as two operations, DELETE plus INSERT, represented as two version rows with a D followed by an I VERSIONS_OPERATION. A given row version is valid starting at its time VERSIONS_START* up to, but not including, its time VERSIONS_END*. That is, it is valid for any time t such that VERSIONS_START* <= t < VERSIONS_END*. For example, the following output indicates that the salary was 10243 from September 9, 2002, included, to November 25, 2003, not included. VERSIONS_START_TIME ------------------09-SEP-2003 VERSIONS_END_TIME ----------------25-NOV-2003 SALARY -----10243 Here is a typical Flashback Version Query: SELECT versions_startscn, versions_starttime, versions_endscn, versions_endtime, versions_xid, versions_operation, name, salary FROM employees VERSIONS BETWEEN TIMESTAMP TO_TIMESTAMP('2003-07-18 14:00:00', 'YYYY-MM-DD HH24:MI:SS') AND TO_TIMESTAMP('2003-07-18 17:00:00', 'YYYY-MM-DD HH24:MI:SS') WHERE name = 'JOE'; Pseudocolumn VERSIONS_XID provides a unique identifier for the transaction that put the data in that state. You can use this value in connection with a Flashback Transaction Query to locate metadata about this transaction in the FLASHBACK_TRANSACTION_QUERY view, including the SQL required to undo the row change and the user responsible for the change – see "Using Flashback Transaction Query" on page 10-10. Oracle Database SQL Reference for information on the Flashback Version Query pseudocolumns and the syntax of the VERSIONS clause See Also: Developing Flashback Applications 10-9 Using Flashback Transaction Query Using Flashback Transaction Query A Flashback Transaction Query is a query on the view FLASHBACK_TRANSACTION_QUERY. You use a Flashback Transaction Query to obtain transaction information, including SQL code that you can use to undo each of the changes made by the transaction. See Also: Oracle Database Backup and Recovery Advanced User's Guide. and Oracle Database Administrator's Guide for information on how a DBA can use the Flashback Table feature to restore an entire table, rather than individual rows As an example, the following statement queries the FLASHBACK_TRANSACTION_QUERY view for transaction information, including the transaction ID, the operation, the operation start and end SCNs, the user responsible for the operation, and the SQL code to undo the operation: SELECT xid, operation, start_scn,commit_scn, logon_user, undo_sql FROM flashback_transaction_query WHERE xid = HEXTORAW('000200030000002D'); As another example, the following query uses a Flashback Version Query as a subquery to associate each row version with the LOGON_USER responsible for the row data change. SELECT xid, logon_user FROM flashback_transaction_query WHERE xid IN (SELECT versions_xid FROM employees VERSIONS BETWEEN TIMESTAMP TO_TIMESTAMP('2003-07-18 14:00:00', 'YYYY-MM-DD HH24:MI:SS') AND TO_TIMESTAMP('2003-07-18 17:00:00', 'YYYY-MM-DD HH24:MI:SS')); Flashback Transaction Query and Flashback Version Query: Example This example demonstrates the use of a Flashback Transaction Query in conjunction with a Flashback Version Query. The example assumes simple variations of the employees and departments tables in the sample hr schema. In this example, a DBA carries out the following series of actions in SQL*Plus: connect hr/hr CREATE TABLE emp (empno NUMBER PRIMARY KEY, empname VARCHAR2(16), salary NUMBER); INSERT INTO emp VALUES (111, 'Mike', 555); COMMIT; CREATE TABLE dept (deptno NUMBER, deptname VARCHAR2(32)); INSERT INTO dept VALUES (10, 'Accounting'); COMMIT; At this point, emp and dept have one row each. In terms of row versions, each table has one version of one row. Next, suppose that an erroneous transaction deletes employee id 111 from table emp: UPDATE emp SET salary = salary + 100 WHERE empno = 111; INSERT INTO dept VALUES (20, 'Finance'); DELETE FROM emp WHERE empno = 111; COMMIT; 10-10 Oracle Database Application Developer’s Guide - Fundamentals Using Flashback Transaction Query Subsequently, a new transaction reinserts employee id 111 with a new employee name into the emp table. INSERT INTO emp VALUES (111, 'Tom', 777); UPDATE emp SET salary = salary + 100 WHERE empno = 111; UPDATE emp SET salary = salary + 50 WHERE empno = 111; COMMIT; At this point, the DBA detects the application error and needs to diagnose the problem. The DBA issues the following query to retrieve versions of the rows in the emp table that correspond to empno 111. The query uses Flashback Version Query pseudocolumns. connect dba_name/password SELECT versions_xid XID, versions_startscn START_SCN, versions_endscn END_SCN, versions_operation OPERATION, empname, salary FROM hr.emp VERSIONS BETWEEN SCN MINVALUE AND MAXVALUE where empno = 111; XID ---------------0004000700000058 000200030000002D 000200030000002E 3 rows selected START_SCN END_SCN ---------- --------113855 113564 112670 113564 OPERATION ---------I D I EMPNAME ---------Tom Mike Mike SALARY ---------927 555 555 The results table reads chronologically, from bottom to top. The third row corresponds to the version of the row in emp that was originally inserted in the table when the table was created. The second row corresponds to the row in emp that was deleted by the erroneous transaction. The first row corresponds to the version of the row in emp that was reinserted with a new employee name. The DBA identifies transaction 000200030000002D as the erroneous transaction and issues the following Flashback Transaction Query to audit all changes made by this transaction: SELECT xid, start_scn START, commit_scn COMMIT, operation OP, logon_user USER, undo_sql FROM flashback_transaction_query WHERE xid = HEXTORAW('000200030000002D'); XID START COMMIT OP USER UNDO_SQL ---------------- ---------- ------------------------------000200030000002D 195243 195244 DELETE HR insert into "HR"."EMP" ("EMPNO","EMPNAME","SALARY") values ('111','Mike','655'); 000200030000002D 195243 195244 INSERT where ROWID = 'AAAKD4AABAAAJ3BAAB'; HR delete from "HR"."DEPT" 000200030000002D 195243 195244 UPDATE HR update "HR"."EMP" set "SALARY" = '555' where ROWID = 'AAAKD2AABAAAJ29AAA'; 000200030000002D 195243 113565 BEGIN HR 4 rows selected The rightmost column (undo_sql) contains the SQL code that will undo the corresponding change operation. The DBA can execute this code to undo the changes Developing Flashback Applications 10-11 Flashback Tips made by that transaction. The USER column (logon_user) shows the user responsible for the transaction. A DBA might also be interested in knowing all changes made in a certain time window. In our scenario, the DBA performs the following query to view the details of all transactions that executed since the erroneous transaction identified earlier (including the erroneous transaction itself): SELECT xid, start_scn, commit_scn, operation, table_name, table_owner FROM flashback_transaction_query WHERE table_owner = 'HR' AND start_timestamp >= TO_TIMESTAMP ('2002-04-16 11:00:00','YYYY-MM-DD HH:MI:SS'); XID ---------------0004000700000058 0004000700000058 0004000700000058 000200030000002D 000200030000002D 000200030000002D START_SCN --------195245 195245 195245 195243 195243 195243 COMMIT_SCN ---------195246 195246 195246 195244 195244 195244 OPERATION --------UPDATE UPDATE INSERT DELETE INSERT UPDATE TABLE_NAME ---------EMP EMP EMP EMP DEPT EMP TABLE_OWNER ----------HR HR HR HR HR HR 6 rows selected Flashback Tips The following tips and restrictions apply to using flashback features. Flashback Tips – Performance ■ ■ ■ ■ ■ ■ For better performance, generate statistics on all tables involved in a Flashback Query by using the DBMS_STATS package, and keep the statistics current. Flashback Query always uses the cost-based optimizer, which relies on these statistics. The performance of a query into the past depends on how much undo data must be accessed. For better performance, use queries to select small sets of past data using indexes, not to scan entire tables. If you must do a full table scan, consider adding a parallel hint to the query. The performance cost in I/O is the cost of paging in data and undo blocks that are not already in the buffer cache. The performance cost in CPU use is the cost of applying undo information to affected data blocks. When operating on changes in the recent past, flashback features essentially CPU bound. Use index structures for Flashback Version Query: the database keeps undo data for index changes as well as data changes. Performance of index lookup-based Flashback Version Query is an order of magnitude faster than the full table scans that are otherwise needed. In a Flashback Transaction Query, the type of the xid column is RAW(8). To take advantage of the index built on the xid column, use the HEXTORAW conversion function: HEXTORAW(xid). Flashback Query against a materialized view does not take advantage of query rewrite optimizations. See Also: Oracle Database Performance Tuning Guide 10-12 Oracle Database Application Developer’s Guide - Fundamentals Flashback Tips Flashback Tips – General ■ ■ ■ ■ Use the DBMS_FLASHBACK package or other flashback features? Use ENABLE/DISABLE calls to the DBMS_FLASHBACK package around SQL code that you do not control, or when you want to use the same past time for several consecutive queries. Use Flashback Query, Flashback Version Query, or Flashback Transaction Query for SQL that you write, for convenience. A Flashback Query, for example, is flexible enough to do comparisons and store results in a single query. To obtain an SCN to use later with a flashback feature, use DBMS_FLASHBACK.GET_SYSTEM_CHANGE_NUMBER. You can compute or retrieve a past time to use in a query by using a function return value as a timestamp or SCN argument. For example, you can perform date and time calculations by adding or subtracting an INTERVAL value to the value of the SYSTIMESTAMP function. You can query locally or remotely (Flashback Query, Flashback Version Query, or Flashback Transaction Query). for example here is a remote Flashback Query: (SELECT * FROM employees@some_remote_host AS OF TIMESTAMP (SYSTIMESTAMP - INTERVAL '60' MINUTE); ■ ■ ■ ■ To ensure database consistency, always perform a COMMIT or ROLLBACK operation before querying past data. Remember that all flashback processing is done using the current session settings, such as national language and character set, not the settings that were in effect at the time being queried. Some DDLs that alter the structure of a table, such as drop/modify column, move table, drop partition, and truncate table/partition, invalidate any existing undo data for the table. It is not possible to retrieve data from a point earlier than the time such DDLs were executed. Trying such a query results in error ORA-1466. This restriction does not apply to DDL operations that alter the storage attributes of a table, such as PCTFREE, INITRANS, and MAXTRANS. Use an SCN to query past data at a precise time. If you use a timestamp, the actual time queried might be up to 3 seconds earlier than the time you specify. Internally, Oracle Database uses SCNs; these are mapped to timestamps at a granularity of every 3 seconds. For example, assume that the SCN values 1000 and 1005 are mapped to the times 8:41 and 8:46 AM respectively. A query for a time between 8:41:00 and 8:45:59 AM is mapped to SCN 1000; a Flashback Query for 8:46 AM is mapped to SCN 1005. Due to this time-to-SCN mapping, if you specify a time that is slightly after a DDL operation (such as a table creation) the database might actually use an SCN that is just before the DDL operation. This can result in error ORA-1466. ■ You cannot retrieve past data from a V$ view in the data dictionary. Performing a query on such a view always returns the current data. You can, however, perform queries on past data in other views of the data dictionary, such as USER_TABLES. Developing Flashback Applications 10-13 Flashback Tips 10-14 Oracle Database Application Developer’s Guide - Fundamentals 11 Developing Applications with the PL/SQL Web Toolkit Java is not the only language that can do network operations and produce dynamic Web content. PL/SQL has a number of features that you can use to make your database available on the Web and make back-office data accessible on the intranet. This chapter discusses the following topics: ■ Developing PL/SQL Web Applications: Overview ■ Using the mod_plsql Gateway ■ Generating HTML Output with PL/SQL ■ Passing Parameters to a PL/SQL Web Application ■ Performing Network Operations within PL/SQL Stored Procedures Developing PL/SQL Web Applications: Overview This section contains the following topics: ■ Invoking a PL/SQL Web Application ■ Implementing a PL/SQL Web Application Invoking a PL/SQL Web Application Typically, a Web application written in PL/SQL is a set of stored procedures that interact with Web browsers through HTTP. A set of interlinked, dynamically generated HTML pages forms the user interface of a web application. The program flow of a PL/SQL Web application is similar to that in a CGI Perl script. Developers often use CGI scripts to produce Web pages dynamically, but such scripts are often not optimal for accessing Oracle Database. Delivering Web content with PL/SQL stored procedures provides the power and flexibility of database processing. For example, you can use DML, dynamic SQL, and cursors. You also eliminate the process overhead of forking a new CGI process to handle each HTTP request. Figure 11–1 illustrates the generic process for a PL/SQL Web application. Developing Applications with the PL/SQL Web Toolkit 11-1 Developing PL/SQL Web Applications: Overview Figure 11–1 PL/SQL Web Application Stored Procedure 3 4 2 Web Browser PL/SQL Web Toolkit 5 1 Web Server The process includes the following steps: 1. A user visits a Web page, follows a hypertext link, or submits data in a form, which causes the browser to send a HTTP request for a URL to an HTTP server. 2. The HTTP server invokes a stored procedure on an Oracle database according to the data encoded in the URL. The data in the URL takes the form of parameters to be passed to the stored procedure. 3. The stored procedure calls subprograms in the PL/SQL Web Toolkit. Typically, subprograms such as HTP.Print generate Web pages dynamically. A generated Web page varies depending on the database contents and the input parameters. 4. The subprograms pass the dynamically generated page to the Web server. 5. The Web server delivers the page to the client. Implementing a PL/SQL Web Application You can implement a Web browser-based application entirely in PL/SQL with the Oracle Database components described in this section. PL/SQL Web Toolkit This set of PL/SQL packages is a generic interface that enables you to use stored procedures called by mod_plsql at runtime. In response to a browser request, a PL/SQL procedure updates or retrieves data from Oracle Database according to the user input. It then generates an HTTP response to the browser, typically in the form of a file download or HTML to be displayed. The Web Toolkit API enables stored procedures to perform actions such as the following: ■ Obtain information about an HTTP request ■ Generate HTTP headers such as content-type and mime-type ■ Set browser cookies ■ Generate HTML pages Table 11–1 describes commonly used PL/SQL Web Toolkit packages. 11-2 Oracle Database Application Developer’s Guide - Fundamentals Using the mod_plsql Gateway Table 11–1 Commonly Used Packages in the PL/SQL Web Toolkit Package Description of Contents HTF Function versions of the procedures in the htp package. The function versions do not directly generate output in a Web page. Instead, they pass their output as return values to the statements that invoke them. Use these functions when you need to nest function calls. HTP Procedures that generate HTML tags. For instance, the procedure htp.anchor generates the HTML anchor tag, . OWA_CACHE Functions and procedures that enable the PL/SQL gateway cache feature to improve performance of your PL/SQL Web application. You can use this package to enable expires-based and validation-based caching with the PL/SQL gateway file system. OWA_COOKIE Subprograms that send and retrieve HTTP cookies to and from a client Web browser. Cookies are strings a browser uses to maintain state between HTTP calls. State can be maintained throughout a client session or longer if a cookie expiration date is included. OWA_CUSTOM The authorize function used by cookies. OWA_IMAGE Subprograms that obtain the coordinates where a user clicked an image. Use this package when you have an image map whose destination links invoke a PL/SQL gateway. OWA_OPT_LOCK Subprograms that impose database optimistic locking strategies to prevent lost updates. Lost updates can otherwise occur if a user selects, and then attempts to update, a row whose values have been changed in the meantime by another user. OWA_PATTERN Subprograms that perform string matching and string manipulation with regular expressions. OWA_SEC Subprograms used by the PL/SQL gateway for authenticating requests. OWA_TEXT Subprograms used by package OWA_PATTERN for manipulating strings. You can also use them directly. OWA_UTIL The following types of utility subprograms: ■ ■ ■ WPG_DOCLOAD Dynamic SQL utilities to produce pages with dynamically generated SQL code. HTML utilities to retrieve the values of CGI environment variables and perform URL redirects. Date utilities for correct date-handling. Date values are simple strings in HTML, but must be properly treated as an Oracle Database datatype. Subprograms that download documents from a document repository that you define using the DAD configuration. See Also: Oracle Database PL/SQL Packages and Types Reference for syntax, descriptions, and examples for the PL/SQL Web Toolkit packages Using the mod_plsql Gateway As explained in detail in the Oracle Application Server mod_plsql User's Guide, mod_plsql maps Web client requests to PL/SQL stored procedures over HTTP. Refer to this documentation for instructions. Developing Applications with the PL/SQL Web Toolkit 11-3 Generating HTML Output with PL/SQL See Also: ■ ■ Oracle Application Server mod_plsql User's Guide to learn how to configure and use mod_plsql Oracle HTTP Server Administrator's Guide to obtain mod_plsql reference material Generating HTML Output with PL/SQL Traditionally, PL/SQL Web applications use function calls to generate each HTML tag for output. These functions are part of the PL/SQL Web Toolkit packages that come with Oracle Database. Example 11–1 illustrates how to generate a simple HTML page by calling the HTP functions that correspond to each HTML tag. Example 11–1 Displaying HTML Tags with HTP Functions CREATE OR REPLACE PROCEDURE html_page IS BEGIN HTP.HTMLOPEN; HTP.HEADOPEN; HTP.TITLE('Title'); HTP.HEADCLOSE; ----- generates generates generates generates
Hello -- generates HTP.BODYOPEN( cattributes => 'TEXT="#000000" BGCOLOR="#FFFFFF"'); -- generatesHeading in the HTML File
HTP.HEADER(1, 'Heading in the HTML File'); HTP.PARA; -- generatesHTP.PRINT('Some text in the HTML file.'); HTP.BODYCLOSE; -- generates HTP.HTMLCLOSE; -- generates END; An alternative to making function calls that correspond to each tag is to use the HTP.PRINT function to print the text and tags together. Example 11–2 illustrates this technique. Example 11–2 Displaying HTML Tags with HTP.PRINT CREATE OR REPLACE PROCEDURE html_page2 IS BEGIN HTP.PRINT(''); HTP.PRINT('
'); HTP.PRINT(''); HTP.PRINT('Title of the HTML File '); HTP.PRINT(''); HTP.PRINT(''); HTP.PRINT('Heading in the HTML File
'); HTP.PRINT('Some text in the HTML file.'); HTP.PRINT(''); HTP.PRINT(''); END; 11-4 Oracle Database Application Developer’s Guide - Fundamentals Passing Parameters to a PL/SQL Web Application Chapter 12, "Developing PL/SQL Server Pages" describes an additional method for delivering using PL/SQL to generate HTML content. PL/SQL server pages enables you to build on your knowledge of HTML tags and avoid learning a new set of function calls. In an application written as a set of PL/SQL server pages, you can still use functions from the PL/SQL Web toolkit to do the following: ■ Simplify the processing involved in displaying tables ■ Store persistent data (cookies) ■ Work with CGI protocol internals Passing Parameters to a PL/SQL Web Application To be useful in a wide variety of situations, a Web application must be interactive enough to allow user choices. To keep the attention of impatient Web surfers, you should streamline the interaction so that users can specify these choices very simply, without excessive decision-making or data entry. The main methods of passing parameters to PL/SQL Web applications are: ■ ■ Using HTML form tags. The user fills in a form on one Web page, and all the data and choices are transmitted to a stored procedure when the user clicks the Submit button on the page. Hard-coded in the URL. The user clicks on a link, and a set of predefined parameters are transmitted to a stored procedure. Typically, you would include separate links on your Web page for all the choices that the user might want. This section contains the following topics: ■ Passing List and Dropdown List Parameters from an HTML Form ■ Passing Radio Button and Checkbox Parameters from an HTML Form ■ Passing Entry Field Parameters from an HTML Form ■ Passing Hidden Parameters from an HTML Form ■ Uploading a File from an HTML Form ■ Submitting a Completed HTML Form ■ Handling Missing Input from an HTML Form ■ Maintaining State Information Between Web Pages Passing List and Dropdown List Parameters from an HTML Form List boxes and drop-down lists are implemented with the HTML tag
Source Exif Data:
File Type : PDF File Type Extension : pdf MIME Type : application/pdf PDF Version : 1.3 Linearized : Yes Marked : True Subject : Oracle Database Modify Date : 2005:11:10 15:03:13Z Create Date : 2005:11:10 15:03:13Z Page Count : 440 Creation Date : 2005:11:10 15:03:13Z Mod Date : 2005:11:10 15:03:13Z Producer : Acrobat Distiller 5.0.5 (Windows) Author : Oracle Corporation Metadata Date : 2005:11:10 15:03:13Z Creator : Oracle Corporation Description : Oracle Database Title : Oracle Database Application Developer’s Guide - Fundamentals Page Mode : UseOutlinesEXIF Metadata provided by EXIF.tools