Make a Windows application using .net framework

Developing Windows-Based and Web-Enabled

Information Systems

Boca Raton London New York

CRC Press is an imprint of the Taylor & Francis Group, an informa business

Nong Ye Teresa Wu

Developing Windows-Based and Web-Enabled

Information Systems

CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742

© 2015 by Taylor & Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works Version Date: 20140716

International Standard Book Number-13: 978-1-4822-5387-0 (eBook - PDF)

This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the valid- ity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or uti- lized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopy- ing, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com (http:// www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.

Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com

and the CRC Press Web site at http://www.crcpress.com

v

Contents

Preface .............................................................................................................................................xv Acknowledgments ..................................................................................................................... xvii Authors ......................................................................................................................................... xix Overview ...................................................................................................................................... xxi

Section I Foundation of Information Systems

1. Boolean Algebra and Digital Logic Circuits for Computer Hardware ........................ 3 1.1 Boolean Logic ................................................................................................................ 3

1.1.1 Binary Values and Boolean Operators .......................................................... 3 1.1.2 Basic Properties of Boolean Algebra ............................................................. 4 1.1.3 Conversion of a Truth Table for a Function into a Boolean Expression ... 5

1.2 Digital Logic Circuits ................................................................................................... 8 1.2.1 Digital Logic Gates .......................................................................................... 8 1.2.2 Combinational Circuits ................................................................................. 10 1.2.3 Sequential Circuits ......................................................................................... 13

1.3 Summary ...................................................................................................................... 17

2. Digital Data Representation ............................................................................................... 19 2.1 Representation of Numbers ....................................................................................... 19

2.1.1 Conversion between Unsigned Binary and Decimal Numbers ............. 19 2.1.2 Representation of Signed Integers ............................................................... 22

2.1.2.1 Signed Magnitude Method ........................................................... 23 2.1.2.2 One’s Complement Method .......................................................... 23 2.1.2.3 Two’s Complement Method .......................................................... 25

2.1.3 Representation of Signed Floating Point Values ....................................... 27 2.2 Representation of Alphabet and Control Characters ............................................. 27 2.3 Error Detection and Correction ................................................................................ 28 2.4 Summary ...................................................................................................................... 32

3. Computer and Network System Software....................................................................... 33 3.1 The Operating System ................................................................................................ 33

3.1.1 Process Management ..................................................................................... 33 3.1.2 Storage Management ..................................................................................... 34 3.1.3 I/O Management ............................................................................................ 35

3.2 Networking and Communication Software ........................................................... 35 3.2.1 OSI Reference Model ..................................................................................... 35 3.2.2 Transmission Control Protocol/Internet Protocol .................................... 36

3.3 Summary ...................................................................................................................... 41

4. Overview of Information Systems .................................................................................... 43 4.1 Information System Concepts ................................................................................... 43

vi Contents

4.2 The Role of Information System in Business .......................................................... 44 4.2.1 Transaction Processing System .................................................................... 47 4.2.2 Management Information System ............................................................... 47 4.2.3 Decision Support System .............................................................................. 47 4.2.4 Executive Information System and Strategic Information System ......... 47 4.2.5 Electronic Business and Electronic Commerce ......................................... 48 4.2.6 Mobile Commerce .......................................................................................... 48

4.3 Post-PC Information Age: Five Trends in the Future Information System Applications ................................................................................................................. 48 4.3.1 Mobile .............................................................................................................. 48 4.3.2 Social Media ................................................................................................... 49 4.3.3 Big Data ........................................................................................................... 49 4.3.4 Cloud Computing .......................................................................................... 50 4.3.5 Consumerization of IT .................................................................................. 50

4.4 Summary ...................................................................................................................... 50

Section II Database Design and Development

5. Conceptual Data Modeling: Entity-Relationship Modeling ....................................... 55 5.1 Types, Attributes, and Instances of Entities ............................................................ 55 5.2 Types, Attributes, Instances, and Degrees of Relationships between Entities ..... 59 5.3 Maximum and Minimum Cardinalities of a Relationship ................................... 60 5.4 Associative Entities ..................................................................................................... 66 5.5 Weak Entities ............................................................................................................... 71 5.6 Superclass and Subclass of Entities .......................................................................... 71 5.7 Summary ...................................................................................................................... 76

6. Logical Database Design: Relational Modeling and Normalization ........................ 79 6.1 Relational Model of Database and Data Integrity Constraints ............................ 79 6.2 Transformation of an E-R Model to a Relational Model ........................................ 83

6.2.1 Transformation of Entities and Their Attributes....................................... 83 6.2.2 Transformation of Superclass and Subclass Entities and Their

Attributes ........................................................................................................ 86 6.2.3 Transformation of Relationships and Their Attributes ............................ 86 6.2.4 Transformation of Association Entities ...................................................... 93 6.2.5 Transformation of Weak Entities ................................................................. 93

6.3 Normalization ............................................................................................................. 96 6.3.1 Data Redundancy and Data Anomalies ..................................................... 96 6.3.2 Functional Dependency ................................................................................ 98 6.3.3 Normalization and Normal Forms ........................................................... 100

6.4 Summary .................................................................................................................... 104

7. Database Implementation in Microsoft Access ............................................................ 109 7.1 Tables for Data Storage ............................................................................................. 109

7.1.1 Setting the Primary Key for a Table .......................................................... 109 7.1.2 Data Types of a Data Field .......................................................................... 110 7.1.3 Field Size Property of a Data Field ............................................................ 111

viiContents

7.1.4 Format Property of a Data Field ................................................................ 113 7.1.5 Input Mask Property of a Data Field ......................................................... 113 7.1.6 Default Value Property of a Data Field ..................................................... 114 7.1.7 Validation Rule and Validation Text Properties of a Data Field............ 115 7.1.8 Required Property of a Data Field ............................................................ 115 7.1.9 Indexed Property of a Data Field ............................................................... 115 7.1.10 Adding and Deleting Records of a Table.................................................. 115

7.2 Relationships of Tables ............................................................................................. 117 7.3 Queries for Data Retrieval ....................................................................................... 120

7.3.1 Select Queries Using One Table ................................................................. 120 7.3.2 Select Queries with Joins of Multiple Tables ............................................ 123 7.3.3 Select Queries with Parameters and Calculated Fields .......................... 126 7.3.4 Select Queries with Groupings of Records .............................................. 128 7.3.5 Crosstab Queries .......................................................................................... 132 7.3.6 Embedded Select Queries ........................................................................... 134

7.4 Action Queries ........................................................................................................... 140 7.5 Summary .................................................................................................................... 141

8. Structured Query Language ............................................................................................. 151 8.1 Introduction to SQL .................................................................................................. 151 8.2 Backus–Naur Form ................................................................................................... 152 8.3 SQL Syntax ................................................................................................................. 152

8.3.1 SQL Data Definition Language and Data Types ..................................... 153 8.3.1.1 CREATE SCHEMA ...................................................................... 153 8.3.1.2 CREATE VIEW ............................................................................. 154 8.3.1.3 CREATE DOMAIN ...................................................................... 154 8.3.1.4 CREATE TABLE............................................................................ 155

8.3.2 SQL Data Manipulation Laguage: Data Queries ..................................... 157 8.3.2.1 Basic Structure of the SELECT Command ............................... 157 8.3.2.2 Aggregate Functions .................................................................... 160 8.3.2.3 Nested Subqueries........................................................................ 162 8.3.2.4 JOIN Query ................................................................................... 163

8.3.3 SQL Data Manipulation Language: Data Modification ......................... 165 8.3.3.1 INSERT ........................................................................................... 165 8.3.3.2 UPDATE ......................................................................................... 167 8.3.3.3 DELETE .......................................................................................... 167

8.3.4 SQL Data Manipulation Language: Relation Modification ................... 168 8.3.4.1 ALTER TABLE .............................................................................. 168 8.3.4.2 DROP TABLE ................................................................................ 169

8.4 Summary .................................................................................................................... 169

9. MySQL .................................................................................................................................. 177 9.1 Introduction ............................................................................................................... 177 9.2 Get Ready to Work with MySQL ............................................................................ 178

9.2.1 MySQL Installation on Windows .............................................................. 179 9.2.2 MySQL Installation on Mac OS X .............................................................. 183

9.3 Working with MySQL Command Line Client ...................................................... 184 9.3.1 Data Types..................................................................................................... 184

9.3.1.1 Numeric Data Types .................................................................... 184

viii Contents

9.3.1.2 String (Character) Data Type ...................................................... 185 9.3.1.3 Date and Time (Temporal) Data Types ..................................... 185 9.3.1.4 NULL Value .................................................................................. 186

9.3.2 Practice Data Definition Language in MySQL ........................................ 186 9.3.2.1 Statements for Database Operations ......................................... 186 9.3.2.2 Statements for Relation Operations ........................................... 188

9.3.3 Practice Data Manipulation Language in MySQL .................................. 189 9.3.3.1 Adding Records to a Table .......................................................... 189 9.3.3.2 Querying Tables ........................................................................... 190 9.3.3.3 Updating Tables ............................................................................ 190 9.3.3.4 Deleting Records .......................................................................... 191

9.3.4 MySQL Transaction Control Language .................................................... 191 9.3.5 MySQL Data Control Language ................................................................ 192 9.3.6 MySQL Utilities ............................................................................................ 193

9.3.6.1 SHOW Statement .......................................................................... 193 9.3.6.2 DESCRIBE Statement ................................................................... 194 9.3.6.3 HELP Statement ............................................................................ 194

9.4 Working with MySQL Workbench ......................................................................... 194 9.4.1 Data Modeling .............................................................................................. 195 9.4.2 SQL Development ........................................................................................ 196 9.4.3 Server Administration ................................................................................ 197

9.5 Summary .................................................................................................................... 198

10. Object-Based Database Systems ...................................................................................... 201 10.1 Object-Oriented Database Management System .................................................. 201

10.1.1 Object-Oriented Database Concepts ......................................................... 201 10.1.1.1 Objects and Identities .................................................................. 202 10.1.1.2 Complex Objects ........................................................................... 202 10.1.1.3 Encapsulation ................................................................................ 202 10.1.1.4 Classes and Inheritance............................................................... 203 10.1.1.5 Overloading, Overriding, and Late Binding ............................ 204

10.1.2 Object-Oriented Database Design and Modeling ................................... 204 10.1.2.1 Object Modeling ........................................................................... 204 10.1.2.2 Object Definition Language ........................................................ 206 10.1.2.3 Object Query Language .............................................................. 207

10.1.3 Summary of OODBMS ................................................................................ 208 10.2 Object RDBMS ........................................................................................................... 209

10.2.1 Object-Relational Model ............................................................................. 209 10.2.2 Object-Relational Query Language ........................................................... 210 10.2.3 Summary of ORDBMS ................................................................................ 211

10.3 Summary .................................................................................................................... 212

Section III Windows Application Development

11. Windows Forms and Controls in Microsoft Visual Studio ....................................... 217 11.1 Visual Basic Development Environment ............................................................... 217 11.2 Windows Forms ........................................................................................................ 220 11.3 GUI Controls .............................................................................................................. 223

ixContents

11.4 GUI of the Healthcare Information System Application ..................................... 227 11.5 Execution of a Windows-Based Application Outside Visual Studio ................. 227 11.6 Summary .................................................................................................................... 229

12. Visual Basic Programming in Microsoft Visual Studio ............................................. 235 12.1 Variables and Data Types ......................................................................................... 235 12.2 Constants .................................................................................................................... 238 12.3 Operators and Mathematical Functions in Expressions ..................................... 238 12.4 Control Structures ..................................................................................................... 238 12.5 Summary .................................................................................................................... 243

13. Database Connection in Microsoft Visual Studio ....................................................... 245 13.1 Adding an Access Database as a Data Source ...................................................... 245 13.2 Presenting Data from Tables and Simple Select Queries .................................... 251 13.3 Architecture of Database Connection .................................................................... 251 13.4 Displaying Related Data in Parent and Child Tables .......................................... 257 13.5 Displaying Data Fields Using Controls Other Than DataGridView ................. 259 13.6 Handling Crosstab and Select Queries with Parameters .................................... 261 13.7 Updating Data to an Access Database ................................................................... 271 13.8 Visual Basic Code to Retrieve, Add, and Delete Data of an Access Database ...... 272 13.9 Summary .................................................................................................................... 276

14. Working with VBA in Excel ............................................................................................. 277 14.1 Introduction to VBA ................................................................................................. 277 14.2 Excel Object Models .................................................................................................. 281 14.3 VBA for Excel ............................................................................................................. 282

14.3.1 Visual Basic Editor ....................................................................................... 282 14.3.1.1 Project Explorer Window ............................................................ 283 14.3.1.2 Code Window ............................................................................... 283 14.3.1.3 Property Window ......................................................................... 283 14.3.1.4 Immediate Window ..................................................................... 283 14.3.1.5 Object Browser Window ............................................................. 284

14.3.2 Graphical User Interface: Forms and Controls ........................................ 284 14.3.2.1 Controls in Excel ........................................................................... 284 14.3.2.2 Controls with Forms .................................................................... 285

14.3.3 Modules: Sub and Function Procedure .................................................... 286 14.3.4 Modules: Variables ....................................................................................... 287

14.3.4.1 Procedure-Only Variables ........................................................... 288 14.3.4.2 Module-Only Variables ............................................................... 289 14.3.4.3 Public Variables ............................................................................ 289 14.3.4.4 Static Variables .............................................................................. 289

14.4 Programming with VBA and Excel ........................................................................ 289 14.4.1 Excel: Range Object ..................................................................................... 289

14.4.1.1 Range.............................................................................................. 290 14.4.1.2 Cells ................................................................................................ 290 14.4.1.3 Offset .............................................................................................. 290 14.4.1.4 Some Useful Properties and Methods for the Range Object ... 291

14.4.2 VBA Functions and Excel Functions ......................................................... 291

x Contents

14.4.2.1 VBA Functions .............................................................................. 292 14.4.2.2 Excel Functions ............................................................................. 293

14.4.3 Some Important Events in Excel ................................................................ 294 14.5 Mini Project ................................................................................................................ 295

14.5.1 Introduction .................................................................................................. 295 14.5.2 Spreadsheet Specifications ......................................................................... 296

14.6 Summary .................................................................................................................... 297

15. Database Connectivity with VBA ................................................................................... 301 15.1 ActiveX Data Object .................................................................................................. 301 15.2 ADO.NET Objects in Data Provider Libraries ...................................................... 301

15.2.1 Connection Object ....................................................................................... 301 15.2.2 Command Object ......................................................................................... 302 15.2.3 Recordset Object .......................................................................................... 303

15.3 Illustrative Example 1: Connecting to MySQL ..................................................... 303 15.4 Illustrative Example 2: Connecting to an Access Database ................................ 307 15.5 Illustrative Example 3: Connecting to XML .......................................................... 308 15.6 Summary .................................................................................................................... 310

Section IV Web Application Development

16. Web Applications in Microsoft Visual Studio ............................................................. 315 16.1 Starting a Web Application in Microsoft Visual Studio ...................................... 315 16.2 Hypertext Markup Language ................................................................................. 319 16.3 Multiple Web Forms ................................................................................................. 320 16.4 Database Connection ................................................................................................ 324 16.5 Summary .................................................................................................................... 334

17. Working with XML (I) ....................................................................................................... 337 17.1 What Is Markup Language? .................................................................................... 337

17.1.1 Standard Generalized Markup Language ............................................... 337 17.1.2 Hypertext Markup Language .................................................................... 338 17.1.3 Extensible Markup Language .................................................................... 339

17.2 XML Basics: Elements and Attributes .................................................................... 339 17.2.1 Elements ........................................................................................................ 339 17.2.2 Attributes ...................................................................................................... 340 17.2.3 Text ................................................................................................................. 341 17.2.4 Empty Elements ........................................................................................... 341 17.2.5 Well-Formed XML ....................................................................................... 341

17.3 XML Namespaces ..................................................................................................... 343 17.4 XML Validation ......................................................................................................... 345

17.4.1 DTD and XML Validation ........................................................................... 345 17.4.1.1 DTD Structure .............................................................................. 346 17.4.1.2 Internal versus External DTD .................................................... 350

17.4.2 XML Schema and XML Validation ........................................................... 352 17.4.2.1 XML Schema Data Types ............................................................ 352 17.4.2.2 XML Schema Namespaces .......................................................... 352

xiContents

17.4.2.3 XML Schema Elements and Attributes ..................................... 353 17.4.3 Referencing XML Schema for Validation ................................................. 355

17.5 XML Editor Tools ...................................................................................................... 357 17.6 Summary .................................................................................................................... 357

18. Working with XML (II) ..................................................................................................... 365 18.1 XML Parsers ............................................................................................................... 365 18.2 XML Query ................................................................................................................ 370 18.3 XML Transformation ................................................................................................ 373 18.4 Interfacing XML with Database Applications ...................................................... 376

18.4.1 Relational Database versus XML ............................................................... 376 18.4.2 XML and Microsoft Access ........................................................................ 380

18.4.2.1 Exporting Access Database Tables to XML .............................. 380 18.4.2.2 Importing XML to an Access Database .................................... 381

18.4.3 XML and MySQL ......................................................................................... 385 18.4.3.1 Exporting MySQL to XML .......................................................... 385 18.4.3.2 Importing XML to MySQL ......................................................... 386

18.4.4 XML and Microsoft Excel ........................................................................... 387 18.5 Summary .................................................................................................................... 387

19. Web Services ........................................................................................................................ 393 19.1 Introduction to Web Services .................................................................................. 393 19.2 Web Service Protocols .............................................................................................. 394

19.2.1 Hypertext Transfer Protocol ....................................................................... 394 19.2.1.1 HTTP-GET ..................................................................................... 394 19.2.1.2 HTTP-POST ................................................................................... 395

19.2.2 Simple Object Access Protocol ................................................................... 395 19.2.3 Web Services Definition Language ........................................................... 395 19.2.4 UDDI Language ........................................................................................... 395

19.3 Web Service Example ............................................................................................... 396 19.3.1 Creating Web Service .................................................................................. 396

19.3.1.1 Web Services Definition Language ............................................ 397 19.3.1.2 SOAP Message .............................................................................. 399

19.3.2 Using the Web Service ................................................................................ 401 19.4 Summary .................................................................................................................... 404

Section V Design of Information Systems

20. Computing Efficiency of Algorithms ............................................................................. 409 20.1 Analysis of the Running Time of a Non-Recursive Algorithm .......................... 409 20.2 Asymptotic Notations for the Computing Efficiency of Algorithms ................ 413 20.3 Summary .................................................................................................................... 418

21. User Interface Design and Usability .............................................................................. 421 21.1 Usability Criteria ....................................................................................................... 421 21.2 Principles and Guidelines of User Interface Design ............................................ 422 21.3 Summary .................................................................................................................... 426

xii Contents

22. Computer and Network Security .................................................................................... 427 22.1 Security Risks Involving Assets, Vulnerabilities, and Threats .......................... 427 22.2 Various Types of Cyber Attacks .............................................................................. 429

22.2.1 Brute Force Attack ....................................................................................... 430 22.2.2 Bypassing Attack ......................................................................................... 430 22.2.3 Code Attachment ......................................................................................... 432 22.2.4 Mobile Code .................................................................................................. 432 22.2.5 Denial of Service .......................................................................................... 432 22.2.6 Tampering ..................................................................................................... 433 22.2.7 Man in the Middle ....................................................................................... 433 22.2.8 Probing and Scanning ................................................................................. 434 22.2.9 Spoofing ........................................................................................................ 434 22.2.10 Adding ........................................................................................................... 434 22.2.11 Insider Threat ............................................................................................... 434

22.3 Cyber Security Protection ........................................................................................ 435 22.3.1 Cyber Attack Prevention ............................................................................. 435

22.3.1.1 Data Encryption ........................................................................... 435 22.3.1.2 Two Types of Firewalls: Screening Routers and

Application Gateways .................................................................. 440 22.3.1.3 Authentication and Authorization............................................. 441

22.3.2 Cyber Attack Detection ............................................................................... 442 22.3.3 Cyber Attack Response ............................................................................... 443

22.4 Summary .................................................................................................................... 444

23. Data Mining ......................................................................................................................... 447 23.1 Learning a Binary Decision Tree and Classifying Data Using a Decision

Tree .............................................................................................................................. 447 23.1.1 Description of a Data Set ............................................................................ 447 23.1.2 Elements of a Decision Tree........................................................................ 449 23.1.3 Decision Tree with the Minimum Description Length .......................... 450 23.1.4 Split Selection Methods ............................................................................... 450 23.1.5 Algorithm for the Top-Down Construction of a Binary Decision Tree ... 456 23.1.6 Classifying Data Using a Decision Tree ................................................... 458

23.2 Learning a Non-Binary Decision Tree ................................................................... 460 23.3 Handling Numeric and Missing Values of Attribute Variables ......................... 467 23.4 Advantages and Shortcomings of the Decision Tree Algorithm ....................... 468 23.5 Software ...................................................................................................................... 469 23.6 Summary .................................................................................................................... 469

24. Expert Systems .................................................................................................................... 471 24.1 Structure of the Knowledge Base ........................................................................... 471 24.2 Forward Chaining and Backward Chaining by the Inference Engine ............. 472 24.3 Conflict Resolution .................................................................................................... 476 24.4 Variable Binding ........................................................................................................ 476 24.5 Certainty Factor ......................................................................................................... 477 24.6 Software and Advantages of Expert Systems to Algorithms of Computer

Programs .................................................................................................................... 478 24.7 Summary .................................................................................................................... 479

xiiiContents

25. Decision Support Systems ................................................................................................ 481 25.1 What Is a Decision? ................................................................................................... 481

25.1.1 Components of a Decision .......................................................................... 481 25.1.1.1 Decision Problem ......................................................................... 481 25.1.1.2 Decision Alternatives ................................................................... 482 25.1.1.3 Decision Constraints .................................................................... 482 25.1.1.4 Decision Criteria ........................................................................... 482

25.1.2 Types of Decisions ....................................................................................... 482 25.1.2.1 Decisions According to Their Nature ....................................... 482 25.1.2.2 Decisions According to Their Scope .......................................... 483

25.2 Decision-Making Process......................................................................................... 484 25.3 Decision Support Systems ....................................................................................... 485

25.3.1 Data Sources ................................................................................................. 485 25.3.2 Extraction, Transformation, and Loading ................................................ 486 25.3.3 Data Warehouse ........................................................................................... 486 25.3.4 Data Marts .................................................................................................... 486 25.3.5 Decision Models ........................................................................................... 486

25.3.5.1 Simple Weighted Average Methodology .................................. 486 25.3.5.2 Compromise Programming ........................................................ 487 25.3.5.3 Genetic Algorithms ...................................................................... 488 25.3.5.4 Analytic Hierarchy Process ........................................................ 488

25.3.6 Interfaces ....................................................................................................... 489 25.4 Implementing Decision Support Systems ............................................................. 489

25.4.1 Identifying Problem and Objectives ......................................................... 491 25.4.2 Identifying Data Sources ............................................................................ 491 25.4.3 Data Analysis and Preparation .................................................................. 491 25.4.4 Designing Decision Models ....................................................................... 492 25.4.5 Evaluation of Results ................................................................................... 492 25.4.6 Implementing and Monitoring the Decision Support System .............. 492

25.5 Summary .................................................................................................................... 493

Section VI Case Studies

26. Development of a Healthcare Information System Using Microsoft Access and Visual Studio ............................................................................................................... 499 26.1 Enterprise Modeling ................................................................................................. 499 26.2 Conceptual Data Modeling...................................................................................... 501 26.3 Relational Modeling ................................................................................................. 505 26.4 Database Implementation Using Microsoft Access.............................................. 505 26.5 Windows-Based Application and Web-Enabled Application Using

Microsoft Visual Studio ........................................................................................... 505 26.6 Summary .................................................................................................................... 537

27. Development of an Online System for Imaging Device Productivity Evaluation in Radiology Practices ................................................................................... 539 27.1 Introduction ............................................................................................................... 539 27.2 Imaging Equipment Utilization .............................................................................. 539

xiv Contents

27.3 REES Design .............................................................................................................. 540 27.3.1 Imaging API Design: DICOM Receiver and Parser ................................ 541 27.3.2 Database Design ........................................................................................... 541

27.3.2.1 DICOM Knowledge Database .................................................... 541 27.3.2.2 Patient-Centric Imaging Exam Database .................................. 542

27.3.3 Application Layer ......................................................................................... 543 27.3.3.1 Benchmark Metric for Imaging Exam Efficiency

Assessment .................................................................................... 543 27.3.3.2 Web Interface ................................................................................ 546

27.4 Use Case ..................................................................................................................... 546 27.5 Conclusion .................................................................................................................. 548 27.6 Summary .................................................................................................................... 548

28. Development of a Radiology Skin Dose Simulation Tool Using VBA and Database ............................................................................................................................... 551 28.1 Quality Assurance in Radiology Exam ................................................................. 551 28.2 Peak Skin Exposure .................................................................................................. 552 28.3 Simulation Tool for Peak Skin Dose Monitoring .................................................. 552

28.3.1 Patient Phantom ........................................................................................... 552 28.3.2 Virtual Equipment Configuration and Geometry .................................. 555 28.3.3 Imaging Exam Database ............................................................................. 556 28.3.4 GUI for Physicist Validation ....................................................................... 556 28.3.5 GUI for Technologist Validation ................................................................ 557 28.3.6 Peak Skin Dose Calculation Engine .......................................................... 558

28.4 Discussion .................................................................................................................. 558 28.5 Summary .................................................................................................................... 558

References ................................................................................................................................... 561 Index ............................................................................................................................................. 565

xv

Preface

An information system is an integrated system that contains databases, programs, and graphical user interface to generate, send, receive, store, and process data into meaningful information and knowledge for supporting decision making. Two types of information systems are common: those for Windows applications used by individual users and those for Web applications accessible simultaneously by multiple users through the Internet. This book covers concepts, methods, programming languages, and software tools for designing and developing both Window-based and Web-enabled information systems. Specifically, data modeling and design methods, including entity–relationship modeling, relational modeling and normalization, and object-oriented data concepts are covered. The use of Windows application development tools and languages, including Microsoft Access, MySQL, structured query language (SQL), Visual Studio, Visual Basic for Applications (VBA), and extensible markup language (XML), is introduced and explained. Concepts, methods, and tools will be illustrated in a self-contained manner with easy-to-understand small examples and large-scale case studies. This book will enable readers in engineering, business, and science fields to gain knowledge and skills for developing Windows-based and Web-enabled information systems.

xvii

Acknowledgments

Nong Ye would like to thank her family, Baijun and Alice, for their love, understanding, and unconditional support. Teresa Wu wishes to express her love and gratitude to her beloved family, Aimin and Tristan, who support and encourage her in spite of all the time it took her away from them. Teresa Wu would also like to thank Eugene Rex Jalao, who provided support, offered comments, and assisted in the editing and proofreading of some book materials. We would like to thank Cindy Carelli, a senior editor at CRC Press. This book would not have been possible without her being responsive, helpful, understanding, and supportive for a few years since we started preparing this book. It has been a great pleasure working with her. We also thank other people at CRC Press who helped us in publishing this book.

xix

Authors

Nong Ye is a professor at the School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, Tempe, Arizona. She holds a PhD in Industrial Engineering from Purdue University, West Lafayette, Indiana, an M.S. degree in Computer Science from the Chinese Academy of Sciences, Beijing, People’s Republic of China, and a B.S. degree in Computer Science from Peking University, Beijing, People’s Republic of China. Her book publications include this book, Data Mining: Theories, Algorithms, and Examples, The Handbook of Data Mining, and Secure Computer and Network Systems: Modeling, Analysis, and Design. Her publications also include over 80 journal papers in the fields of data mining, statistical data analysis and modeling, computer and network security, quality-of-service optimization, quality control, human–computer interaction, and human factors.

Teresa Wu is an associate professor at the School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, Tempe, Arizona. She holds a PhD in Industrial Engineering from the University of Iowa. Her book publications include this book and Managing Supply Chain Risk and Vulnerability: Tools and Methods for Supply Chain Decision Makers. Her publications also include over 60 journal papers in the fields of infor- mation systems, decision algorithms, data mining, and health informatics.

xxi

Overview

Topics for developing information system applications are organized into six sections as follows:

I Foundation of Information Systems II Database Design and Development III Windows Application Development IV Web Application Development V Design of Information Systems VI Case Studies

Section I of this book describes computer hardware and software that provide the foun- dation for developing information system applications. Section I covers the following:

• Boolean algebra and digital logic circuits for computer hardware in Chapter 1 • Digital data representation in Chapter 2 • Computer and network system software in Chapter 3 • Computer and network applications for information systems in Chapter 4

Section II describes concepts, models, and tools related to database design and devel- opment. A database is an organized collection of data and is the core of an information system. Data in a database are organized using a model, such as a relational model, an object-oriented model, or an object-relational model. Data are then retrieved via data que- ries. Section II covers the following:

• Conceptual database design: entity–relational modeling in Chapter 5 • Logical database design: Relational modeling and normalization in Chapter 6 • Database implementation in Microsoft Access in Chapter 7 • Structured query language in Chapter 8 • MySQL: A relational database management system in Chapter 9 • Object-oriented database systems in Chapter 10

With the development of a database, applications are then developed to enable users to access and manipulate data in a database through a graphical user interface (GUI). Section III of this book covers the development of Windows-based applications with a database embedded. Section III includes the following:

• Windows forms and controls in Microsoft Visual Studio in Chapter 11 • Visual Basic programming in Microsoft Visual Studio in Chapter 12 • Database connection in Microsoft Visual Studio in Chapter 13

xxii Overview

• Windows forms and controls with Microsoft VBA in Chapter 14 • Data connectivity with VBA embedded within Microsoft Excel in Chapter 15

Section IV of this book describes the development of Web-enabled applications. Some fundamental concepts and standards (including XML) are explained. The development of Web applications and Web services is illustrated. Section IV includes the following:

• Web applications in Microsoft Visual Studio in Chapter 16 • Working with XML (I) in Chapter 17 • Working with XML (II) in Chapter 18 • Web services in Chapter 19

With Section II to Section IV covering the database and GUI aspects of information sys- tems, Section V addresses some other aspects of information systems: computing efficiency of algorithms, usability, security, extraction of useful information and knowledge from data through data mining, and knowledge-based reasoning to support decision making. Section V includes the following:

• Computing efficiency of algorithms in Chapter 20 • User interface design and usability in Chapter 21 • Computer and network security in Chapter 22 • Data mining in Chapter 23 • Expert systems in Chapter 24 • Decision support systems in Chapter 25

Section VI of this book demonstrates the development of information systems through three case studies. Section VI includes the following:

• Development of a healthcare information system using Microsoft Access and Visual Studio in Chapter 26

• Development of an online system for imaging device productivity evaluation in radiology practices in Chapter 27

• Development of radiology skin dose simulation tools using VBA and database in Chapter 28

Distinctive Features

Many existing textbooks are written for readers having a background in computer sci- ence and knowledge, experience, and skills in programming and working on developing large-scale computer and network software. However, many professionals in engineering, business, and science fields have needs to develop information systems for managing and using data to support decision making in their fields, although they are not professional programmers or software developers. This book aims to help students and professionals

xxiiiOverview

in engineering, business, and science fields, who are not professional programmers or soft- ware developers, learn concepts, methods, and software tools for developing Windows- based and Web-enabled information systems. This book takes readers through the entire process of developing a computer and network application for an information system, including the following:

1. Design and development of a database 2. Development of a Windows-based or Web-enabled application with a database

embedded and with GUI to allow users to access and manipulate data in the database

3. Design of the information system for computing efficiency, usability, security, and capabilities of turning data into useful information and knowledge and perform- ing knowledge-based reasoning for decision support

This book provides both concepts and operational details of developing a Windows- based or a Web-enabled application for an information system using Microsoft software development environments and tools. In each chapter, small data examples are used to manually walk through concepts and operational details. Hence, this book will enable readers in non–computer science fields to obtain the conceptual understanding and the practical skills of independently developing Windows-based or Web-enabled applications for their specialized information systems.

Teaching Support

Topics covered in this book involve different levels of difficulty. An instructor, who uses this book as a textbook for a course on information systems, may select book materials to meet their instructional needs based on the level of the course and the difficulty levels of book materials. The book materials in Chapters 5 to 7, 11 to 16, and 26, which cover database design and Windows-based and Web-enabled applications, are appropriate for both undergraduate-level and graduate-level courses. The case studies in Chapters 26 to 28 can be selectively used by the instructor to illustrate topics covered in Chapters 5 to 7, 11 to 16. The book materials in Chapters 17 to 19, which cover XML standards and Web services, may be appropriate for a graduate-level course only. The book materials in Section V, including Chapters 20 to 25, take the development of information systems from the basic components of a database and GUI to more advanced design issues. Hence, the book materials in Section V are appropriate for a graduate-level course. Concepts in Section I of this book, which explain the foundation of hardware and system software for computers and networks, are required to understand the materials on computer and network security in Chapter 22. Although the book materials in Section I of this book are not required for learning the concepts and skills of developing information systems in Section II to Section IV, readers who are interested in learning how Windows and Web applications are enabled by hardware and system software for computers and networks are encouraged to read the book materials in Section I to unveil the mystery of the digital revolution enabled by computing technologies and to obtain a complete picture of com- puter and network applications.

xxiv Overview

Exercises are provided at the end of each chapter. Additional teaching support materials can be obtained from the publisher and include the following:

• Solution manual • Lecture notes, which include the outline of topics, figures, tables, and equations

from this book to support teaching

xxv

Overview of the Book

Topics for developing information system applications are organized into six sections as follows:

I Foundation of Information Systems II Database Design and Development III Windows Application Development IV Web Application Development V Design of Information Systems VI Case Studies

Section I of this book describes computer hardware and software that provide the foun- dation for developing information system applications. Section I covers the following:

• Boolean algebra and digital logic circuits for computer hardware in Chapter 1 • Digital data representation in Chapter 2 • Computer and network system software in Chapter 3 • Computer and network applications for information systems in Chapter 4

Section II describes concepts, models, and tools related to database design and devel- opment. A database is an organized collection of data and is the core of an information system. Data in a database are organized using a model, such as a relational model, an object-oriented model, or an object-relational model. Data are then retrieved via data que- ries. Section II covers the following:

• Conceptual database design: entity–relational modeling in Chapter 5 • Logical database design: Relational modeling and normalization in Chapter 6 • Database implementation in Microsoft Access in Chapter 7 • Structured query language in Chapter 8 • MySQL: A relational database management system in Chapter 9 • Object-oriented database systems in Chapter 10

With the development of a database, applications are then developed to enable users to access and manipulate data in a database through a graphical user interface (GUI). Section III of this book covers the development of Windows-based applications with a database embedded. Section III includes the following:

• Windows forms and controls in Microsoft Visual Studio in Chapter 11 • Visual Basic programming in Microsoft Visual Studio in Chapter 12 • Database connection in Microsoft Visual Studio in Chapter 13

xxvi Overview of the Book

• Windows forms and controls with Microsoft VBA in Chapter 14 • Data connectivity with VBA embedded within Microsoft Excel in Chapter 15

Section IV of this book describes the development of Web-enabled applications. Some fundamental concepts and standards (including XML) are explained. The development of Web applications and Web services is illustrated. Section IV includes the following:

• Web applications in Microsoft Visual Studio in Chapter 16 • Working with XML (I) in Chapter 17 • Working with XML (II) in Chapter 18 • Web services in Chapter 19

With Section II to Section IV covering the database and GUI aspects of information sys- tems, Section V addresses some other aspects of information systems: computing efficiency of algorithms, usability, security, extraction of useful information and knowledge from data through data mining, and knowledge-based reasoning to support decision making. Section V includes the following:

• Computing efficiency of algorithms in Chapter 20 • User interface design and usability in Chapter 21 • Computer and network security in Chapter 22 • Data mining in Chapter 23 • Expert systems in Chapter 24 • Decision support systems in Chapter 25

Section VI of this book demonstrates the development of information systems through three case studies. Section VI includes the following:

• Development of a healthcare information system using Microsoft Access and Visual Studio in Chapter 26

• Development of an online system for imaging device productivity evaluation in radiology practices in Chapter 27

• Development of radiology skin dose simulation tools using VBA and database in Chapter 28

Distinctive Features of This Book

Many existing textbooks are written for readers having a background in computer sci- ence and knowledge, experience, and skills in programming and working on developing large-scale computer and network software. However, many professionals in engineering, business, and science fields have needs to develop information systems for managing and using data to support decision making in their fields, although they are not professional programmers or software developers. This book aims to help students and professionals

xxviiOverview of the Book

in engineering, business, and science fields, who are not professional programmers or soft- ware developers, learn concepts, methods, and software tools for developing Windows- based and Web-enabled information systems. This book takes readers through the entire process of developing a computer and network application for an information system, including the following:

1. Design and development of a database 2. Development of a Windows-based or Web-enabled application with a database

embedded and with GUI to allow users to access and manipulate data in the database

3. Design of the information system for computing efficiency, usability, security, and capabilities of turning data into useful information and knowledge and perform- ing knowledge-based reasoning for decision support

This book provides both concepts and operational details of developing a Windows- based or a Web-enabled application for an information system using Microsoft software development environments and tools. In each chapter, small data examples are used to manually walk through concepts and operational details. Hence, this book will enable readers in non–computer science fields to obtain the conceptual understanding and the practical skills of independently developing Windows-based or Web-enabled applications for their specialized information systems.

Teaching Support

Topics covered in this book involve different levels of difficulty. An instructor, who uses this book as a textbook for a course on information systems, may select book materials to meet their instructional needs based on the level of the course and the difficulty levels of book materials. The book materials in Chapters 5 to 7, 11 to 16, and 26, which cover database design and Windows-based and Web-enabled applications, are appropriate for both undergraduate-level and graduate-level courses. The case studies in Chapters 26 to 28 can be selectively used by the instructor to illustrate topics covered in Chapters 5 to 7, 11 to 16. The book materials in Chapters 17 to 19, which cover XML standards and Web services, may be appropriate for a graduate-level course only. The book materials in Section V, including Chapters 20 to 25, take the development of information systems from the basic components of a database and GUI to more advanced design issues. Hence, the book materials in Section V are appropriate for a graduate-level course. Concepts in Section I of this book, which explain the foundation of hardware and system software for computers and networks, are required to understand the materials on computer and network security in Chapter 22. Although the book materials in Section I of this book are not required for learning the concepts and skills of developing information systems in Section II to Section IV, readers who are interested in learning how Windows and Web applications are enabled by hardware and system software for computers and networks are encouraged to read the book materials in Section I to unveil the mystery of the digital revolution enabled by computing technologies and to obtain a complete picture of com- puter and network applications.

xxviii Overview of the Book

Exercises are provided at the end of each chapter. Additional teaching support materials are available at the book website, which include the following and can be obtained from the publisher:

• Solution manual • Lecture notes, which include the outline of topics, figures, tables, and equations

from this book to support teaching

Section I

Foundation of Information Systems

Information systems are computer and network applications running on networks of com- puters. Figure I.1 shows the computer architecture in a hierarchy from the hardware level of digital logic circuits to the software levels of machine instruction code, operating sys- tem, system software, and applications. There are many types of computer and network applications for scientific computing, data visualization, data management, decision sup- port, and so on. Information systems are computer and network applications for managing data and information to provide decision support.

Section I of this book covers computer hardware and system software, which lay the foundation for the development of information systems that is covered after Section I. Computer hardware is made up of digital logic circuits that can be designed using Boolean logic. Boolean logic and digital logic circuits are covered in Chapter 1. Digital logic circuits directly support the representation and manipulation of binary data. Chapter 2 explains how non-binary data and information are represented using binary data. System software, including the operating system and network software, is described in Chapter 3. Chapter 4 gives an overview of various information systems.

Machine instruction code

Operating system and system software

Digital logic circuits

Computer and network applications

FIGURE I.1 Hierarchy of Computer Hardware and Software.

3

1 Boolean Algebra and Digital Logic Circuits for Computer Hardware

Computer hardware is made of electrical circuits. High and low voltages of electrical cir- cuits are considered as true and false or 1 and 0, respectively, of binary data in Boolean logic, which is also known as “symbolic logic,” and deals with true and false values in logical reasoning. The implementation of a digital circuit can be expressed as a Boolean expression. Hence, Boolean logic is used to design digital circuits (called “digital logic circuits”). This chapter first introduces Boolean logic and then gives the design of digi- tal logic circuits for such computer components as the arithmetic logic unit (ALU) of the central processing unit (CPU) and memory to illustrate how computer hardware works to process and store binary data.

1.1 Boolean Logic

This section introduces binary values and Boolean operators, gives properties of Boolean algebra, and describes the method of converting a truth table for a function into a Boolean expression, which has one-to-one correspondence to the implementation of a digital logic circuit for the function.

1.1.1 Binary Values and Boolean Operators

Boolean logic represents two values: false, with 0 as the binary value representation, and true, with 1 as the binary value representation. There are three basic Boolean operators: AND, OR, and NOT, which are used to construct Boolean expressions. Tables 1.1, 1.2, and 1.3 give the truth table for the three Boolean operators. A truth table for a Boolean expres- sion lists all the combinations of values for the input variables in the Boolean expression and gives the resulting values of the Boolean expression for input values. In Table 1.1, for the AND operator, there are two input variables, x and y, which take binary values, 0 and 1, and one binary output for the result of the AND operation, x AND y, which is represented by xy. For the AND operator, the output result is true (1) only when x takes the true value (1) and y takes the true value (1). In Table 1.2, the output result of the OR operator, x OR y, which is represented by x + y, is true (1) when either x or y takes the true value (1). In Table 1.3, the output of the NOT operator, NOT x, which is represented by x , takes the opposite value of x.

4 Developing Windows-Based and Web-Enabled Information Systems

1.1.2 Basic Properties of Boolean Algebra

Table 1.4 gives the basic properties of Boolean algebra, showing a basic set of identical Boolean expressions. These properties can be proven using the truth table. For example, Table 1.5 gives the truth table to prove the OR form of de Morgan’s law. In Table 1.5, fourth column gives the values for the left side of the equation for the de Morgan’s law, and the last column gives the values for the right side of the equation for the de Morgan’s law. The

TABLE 1.1

Truth Table for the Boolean Operator, AND

x y xy

0 0 0 0 1 0 1 0 0 1 1 1

TABLE 1.2

Truth Table for the Boolean Operator, OR

x y x + y

0 0 0 0 1 1 1 0 1 1 1 1

TABLE 1.3

Truth Table for the Boolean Operator, NOT

x x

0 1 1 0

TABLE 1.4

Properties with Identical Expressions in Boolean Algebra

Name of Property AND Form OR Form

Identity properties 1x = x1 = x xx = x

0 + x = x + 0 = x x + x = x

Null property 0x = x0 = 0 1 + x = x + 1 = 1 Inverse property xx = 0 x x+ = 1 Commutative property xy = yx x + y = y + x Associative property (xy)z = x(yz) (x + y) + z = x + (y + z) Distributive property x + yz = (x + y) (x + z) x (y + z) = xy + xz de Morgan’s law xy x y= + x y xy+ =

5Boolean Algebra and Digital Logic Circuits for Computer Hardware

basic properties of Boolean algebra can be used to prove more complex identical Boolean expressions and simplify a Boolean expression. Example 1.1 shows using the basic proper- ties of Boolean algebra to simplify a Boolean expression. Example 1.2 proves two identical Boolean expressions.

Example 1.1

Simplify the following Boolean expression:

( )( )

( )( ) (

x y x y

x y x y x yy

+ +

+ + = + AND form of distributiive property

AND form of inverse property

)

(= +x 0 ))

( )= x OR form of identity property

Example 1.2

Use the basic properties of Boolean algebra to prove the following Boolean equation:

x x y x

x x y xx xy

( )

( ) (

+ =

+ = + OR form of distributive prooperty

AND form of identity property

)

( )= +

=

x xy

x1++

= +

xy

x y

( )

( ) (

AND form of identity property

OR f1 oorm of distributive property

OR form of nu

)

(= x1 lll property

AND form of identity property

)

( )= x

1.1.3 Conversion of a Truth Table for a Function into a Boolean Expression

To implement a function using a digital logic circuit, we often use a truth table to express the function and then covert the truth table into the corresponding Boolean expression, which is used as the design of the digital logic circuit implementing the function. Table 1.6 gives the truth table for a parity generator, which produces a parity bit for data containing 3 bits of x, y, z. If the 3 bits of x, y, z have an odd number of value 1, the parity generator produces the parity bit of 1. If the input variables x, y, z have an even number of value 1, the parity generator produces the parity bit of 0. When the 3-bit data are transmitted, the

TABLE 1.5

Truth Table to Prove the OR Form of de Morgan’s Law

x y x + y x y++ x y xy

0 0 0 1 1 1 1 0 1 1 0 1 0 0 1 0 1 0 0 1 0 1 1 1 0 0 0 0

6 Developing Windows-Based and Web-Enabled Information Systems

parity bit is transmitted along with the data. If the received data and the parity bit are not consistent, there is an error in transmitting the data and the parity bit. For example, if the data of 000 and its parity bit of 0 are transmitted but the received data are 100 and the received parity bit is 0, we know that a transmission error occurs because the data of 100 should have the parity bit of 1. The parity bit can be used for detecting only an odd number of bit transmission errors. For example, if the data of 000 and the corresponding parity bit of 0 are transmitted and there are 2-bit transmission errors to produce the received data of 101 and the received parity bit of 0, we cannot detect the transmission errors because the received parity bit of 0 is the correct parity for the received data of 101.

Table 1.7 gives the truth table for a 1-bit full adder that takes carry-in. A function expressed in a truth table can be represented as a Boolean expression in the

sum-of-product form in the following steps:

Step 1. Express the input variables in a product term for each row of the truth table that has the output variable taking the value of 1, by expressing an input variable x with the value of 0 as x and an input variable x with the value of 1 as x.

Step 2. Add all the product terms from Step 1 to produce the sum-of-product form of a Boolean expression.

Step 3. Simplify the Boolean expression in Step 2.

TABLE 1.6

Truth Table for the Parity Generator

x y z Parity Bit

0 0 0 0 0 0 1 1 0 1 0 1 0 1 1 0 1 0 0 1 1 0 1 0 1 1 0 0 1 1 1 1

TABLE 1.7

Truth Table for the 1-Bit Adder

Inputs Outputs

x y z (Carry-In) Sum Carry-Out

0 0 0 0 0 0 0 1 1 0 0 1 0 1 0 0 1 1 0 1 1 0 0 1 0 1 0 1 0 1 1 1 0 0 1 1 1 1 1 1

7Boolean Algebra and Digital Logic Circuits for Computer Hardware

These steps are illustrated in Examples 1.3 and 1.4.

Example 1.3

Produce a Boolean expression in the sum-of-product form for the truth table in Table 1.6. In Step 1, Rows 2, 3, 5, and 8 have the output value of 1 among the eight rows of data.

The product terms for these rows with the output value of 1 are

Row 2: xyz Row 3: xyz Row 5: xyz Row 8: xyz

In Row 2, we have the input variables x = 0, y = 0, and z = 1. The only product term of the input variables producing the output value of 1 is xyz for Row 2. Similarly, xyz, xyz, and xyz are the only product form of the input variables producing the output value of 1 for Rows 3, 5, and 8, respectively.

In Step 2, the product terms for Rows 2, 3, 5, and 8 are added together through the OR operator into the following Boolean expression in the sum-of-product form:

xyz xyz xyz xyz+ + + .

This Boolean expression cannot be further simplified in Step 3. In this example, the input values in Rows 2, 3, 5, or 8 make one of the product terms in

the Boolean expression of the parity generator produce the output value of 1 and, thus, make the entire Boolean expression produce the output value of 1. Any other combina- tions of input values (those in the other rows of the truth table) are different from those that are represented by the product terms in the Boolean expression make each of the product terms produce the output value of 0 and thus make the entire Boolean expres- sion produce the output value of 0. Hence, the Boolean expression constructed in the above three steps produces the desired output value for each row of input values.

Example 1.4

Produce a Boolean expression in the sum-of-product form for the output variable, carry- out, in the 1-bit full adder in Table 1.7.

In Step 1, Rows 4, 6, 7, and 8 have the output value of 1 among the eight rows of data. The product terms for these rows with the output value of 1 are

Row 4: xyz Row 6: xyz Row 7: xyz Row 8: xyz

In Step 2, the product terms for Rows 4, 6, 7, and 8 are added together through the OR operator into the following Boolean expression in the sum-of-product form, which is simplified in Step 3:

xyz xyz xy z z xyz xyz xy+ + + = + +( ) .

Note that the output for the sum bit in Table 1.7 is the same as the output for the parity bit in Table 1.6. Hence, the Boolean expression from Example 1.3, xyz xyz xyz xyz+ + + , can be used for the sum bit in Table 1.7, where z is carry-in.

8 Developing Windows-Based and Web-Enabled Information Systems

1.2 Digital Logic Circuits

This section introduces digital logic gates for a set of basic Boolean operators and describes combinational circuits and sequential circuits that are used to build computer hardware components such as the ALU in the CPU and memory.

1.2.1 Digital Logic Gates

Digital logic gates are digital circuits built for a set of basic Boolean operators, including AND, OR, NOT, and other operators such as exclusive-OR (XOR) and NOT AND (NAND). Table 1.8 gives the truth table for XOR, which produces the output value of 1 if only one of the two input variables x and y has the value of 1, that is, if the input variables have the value of 1 exclusively. Table 1.9 gives the truth table for x NAND y, whose Boolean expres- sion is xy. Figure 1.1 shows the symbols used to represent the logic gates for AND, OR, NOT, XOR, and NAND.

TABLE 1.9

Truth Table for NAND

x y x NAND y

0 0 1 0 1 1 1 0 1 1 1 0

x y

x AND y x y

x OR y y NOT y

NOT gateAND gate OR gate

x y

x XOR y

XOR gate

x y

x NAND y

NAND gate

FIGURE 1.1 Symbols for logic gates.

TABLE 1.8

Truth Table for XOR

x y x XOR y

0 0 0 0 1 1 1 0 1 1 1 0

9Boolean Algebra and Digital Logic Circuits for Computer Hardware

XOR is useful for many functions. For example, XOR can be used to clear the content of x stored at a memory location on a computer by writing the result of x XOR x = 0 to the memory location. That is, x = x XOR x = 0. XOR can also be used to exchange the values of x and y stored at two memory locations by performing the following operations in sequence:

x = x XOR y

y = x XOR y

x = x XOR y.

Table 1.10 shows the change of values in x and y step by step in the above operations. The values of y in the second last column in Table 1.10 are the same as the original values of x in the first column. The values of x in the last column in Table 1.10 are the same as the original values of y in the second column.

The NAND gate can be used to construct a digital logic circuit for any function. For example, the AND, NOT, and OR gates can be constructed using only the NAND gates as follows:

xy xy xy xy

xx x

xx yy x y x y x y

= = =

= = + = +

Figure 1.2 gives the logic diagrams showing the implementation of digital logic circuits based on the Boolean expressions of xy xy, xx, and xx yy for the AND, NOT, and OR gates, respectively.

TABLE 1.10

Illustration of Operations for Using XOR to Exchange Values of x and y

x y x = x XOR y y = x XOR y x = x XOR y

0 0 0 0 0 0 1 1 0 1 1 0 1 1 0 1 1 0 1 1

x y xy

AND gate NOT gate

x x

OR gate

x

y x + y

FIGURE 1.2 Construction of the AND, NOT, and OR gates, using only the NAND gates.

10 Developing Windows-Based and Web-Enabled Information Systems

Example 1.5

Use the logic diagram to show the implementation of the digital logic circuit for the Boolean expression yz xyz xyz+ + from Example 1.4.

Figure 1.3 gives the logic diagram for the Boolean expression.

1.2.2 Combinational Circuits

A combinational circuit produces an output based on entirely the current inputs without memory or storage of the past inputs. Figure 1.4 gives the combinational circuit that imple- ments the 1-bit half adder without the carry-in bit. Figure 1.5 shows a combinational circuit of the 1-bit full adder in Table 1.7.

y

z

x

FIGURE 1.3 Logic diagram for yz xyz xyz+ + .

x y

Carry-out

Sum

FIGURE 1.4 Logic diagram of the digital logic circuit for the 1-bit half adder.

x y

z (Carry-in) Carry-out

Sum

FIGURE 1.5 Logic diagram of the digital logic circuit for the 1-bit full adder.

11Boolean Algebra and Digital Logic Circuits for Computer Hardware

The Boolean expression for the sum bit in the combinational circuit of the full adder in Figure 1.5 is Sum = z XOR (x XOR y). As shown below, this Boolean expression is equal to the Boolean expression for the sum bit in the sum-of-product form from Example 1.4.

Sum XOR XOR XOR= = + = + +z x y z xy xy z xy xy z xy   ( )  ( ) ( ) ( ++

= + + ( )( ) = + + + xy

xyz xyz z xy xy xyz xyz z x y

)

( )(xx y

xyz xyz xxz xyz xyz yyz xyz xyz x

+

= + + + + + = + +

)

yyz xyz+ .

As shown below, the Boolean expression for the carry-out bit in the combinational cir- cuit in Figure 1.5 is equal to the Boolean expression for the carry-out bit from Example 1.4.

Carry-out XOR= + = + + = + +xy x y z xy xy xy z xy xyz xyz( ) ( )

The digital logic circuit for the 1-bit full adder in Figure 1.5 can be used to construct a ripple-carry adder for any bit length (e.g., 128 bits) by feeding the carry-out bit of the adder for 0 bit to the carry-in bit of the adder for the next bit.

Figure 1.6 gives the logic diagram of the digital logic circuit for the 2- to 4-bit decoder, which has a 2-bit input representing a memory address. Two bits can represent four dif- ferent memory addresses since 22 = 4. The decoder has four output bits, each of which corresponds to one of the four memory locations. When a memory address is given to the decoder as the input, the decoder produces the output value of 1 for the memory location represented by the input and produces the output value of 0 for the other memory loca- tions. Hence, the decoder is to select a memory location specified by the memory address. For example, for the input of 00 (x = 0 and y = 0), the output line for xy in Figure 1.6 has the value of 1, and the three other output lines have the value of 0. Hence, the memory address of 00 selects the output line for xy in Figure 1.6. The n-to-2n decoder with n input bits for a memory address and 2n output lines for the memory locations can be designed similarly.

Figure 1.7 gives the logic diagram of the digital logic circuit for a multiplexer. The mul- tiplexer in Figure 1.7 lets two control bits C1 and C0 determine which of the four input lines, I3, I2, I1, and I0 become the output line. Specifically, if the control bits C1C0 are 00, the multiplexer makes I0 the output; if the control bits C1C0 are 01, the multiplexer makes I1 the

y x

xy

xy

xy

xy

FIGURE 1.6 Logic diagram of the digital logic circuit for a 2- to 4-bit decoder.

12 Developing Windows-Based and Web-Enabled Information Systems

output; if the control bits C1C0 are 10, the multiplexer makes I2 the output; and if the control bits C1C0 are 11, the multiplexer makes I3 the output.

CPU is an important part of computer hardware. It consists of the ALU, registers including the program counter, and the control unit. Figure 1.8 shows the digital logic circuit for a simple 2-bit ALU, which performs four basic operations: AND, OR, NOT, and addition. The digital logic circuit for the ALU uses the digital logic circuits of the half

C1 C0

I3

I2

I1

I0

C1C0I3

C1C0I2

C1C0I1

C1C0I0

FIGURE 1.7 Logic diagram of the digital logic circuit for a multiplexer.

C0C1

C1 C0C1 C0C1

C0C1

y1y0x1x0

O1O0 Carry-out

C0

FIGURE 1.8 Logic diagram of the digital logic circuit for a simple ALU.

13Boolean Algebra and Digital Logic Circuits for Computer Hardware

adder in Figure 1.4, the full adder in Figure 1.5, and the decoder in Figure 1.6. The two control bits C1 and C0 determine which operation is performed on two inputs, x1x0 and y1y0. If C0C1 are 00, x1y1 and x0y0 for AND is performed. If C0C1 are 01, x1 + y1 and x0 + y0 for OR is performed. If C0C1 are 10, x x1 0 for NOT is performed. If C0C1 are 11, an addition of x1x0 and y1y0 is performed.

1.2.3 Sequential Circuits

The outputs of a sequential circuit depend not only on the current inputs, but also on the effect of the past inputs on the state of the sequential circuit. Table 1.11 gives the truth table of a sequential circuit called the set/reset (SR) flip-flop. Figure 1.9 shows the logic diagram of the SR flip-flop. When S ≠ 1 or R ≠ 1, the two outputs of the SR flip-flop take opposite val- ues, Q and Q. However, when S = 1 and R = 1, Q t Q t( ) ( )+ = + =1 1 0 and Q t Q t( ) ( )+ = + =1 1 0, regardless of what value Q(t) takes. The same output value of 0 for Q and Q contradicts the expected opposite values for the two outputs. Hence, the outputs for S = 1 and R = 1 are considered undefined.

The SR flip-flop in Figure 1.9 is called an “asynchronous sequential circuit,” since it updates its outputs when any input value changes. A synchronous sequential circuit uses a clock to synchronize the update of the outputs. A clock can be implemented by a circuit that emits a series of pulses with high and low levels, as shown in Figure 1.10. A level- triggered circuit updates its outputs when the clock signal changes its level to high or low. Figure 1.11 shows the digital logic circuit for the SR flip-flop that has the clock signal as an input and changes the output value when the level of the clock signal is high with the value of 1.

S

R

Q

Q

FIGURE 1.9 SR flip-flop without clock.

TABLE 1.11

Truth Table of the SR Flip-Flop

S R Present State Q(t) Next State Q(t + 1)

0 0 0 0 0 0 1 1 0 1 0 0 0 1 1 0 1 0 0 1 1 0 1 1 1 1 0 Undefined 1 1 1 Undefined

14 Developing Windows-Based and Web-Enabled Information Systems

The JK flip-flop, which is named after its inventor, Jack Kilby, addresses the problem of the undefined outputs of the SR flip-flop for S = 1 and R = 1 using the digital logic circuit shown in Figure 1.12. Table 1.12 gives the truth table of the JK flip-flop. By having Q and Q join R and S inputs of the SR flip-flop, respectively, the JK flip-flop eliminates the input combination of S = 1 and R = 1 to the SR flip-flop. When J = 1 and K = 1, Q and Q change their state to the opposite value (Figure 1.12).

The data (D) flip-flop addresses the problem of undefined outputs in the SR flip-flop by eliminating the input combination of S = 1 and R = 1, and the input combination of S = 0 and R = 0 as shown in Figure 1.13. Table 1.13 gives the truth table for the D flip-flop. As seen in Table 1.13, the D flip-flop can be used as a memory unit since it sets the state and the output of the memory unit to the input value. Figure 1.14 gives the logic diagram symbols of the SR, JK, and D flip-flops.

Figure 1.15 shows a 4-bit register that is made of four D flip-flops. Figure 1.16 shows a memory that can store four words, with 2 bits in each word. In Figure 1.16, A1A0 are the inputs for a 2- to 4-bit decoder and give the memory address to select one of the four

S

R Q

Q

C (clock)

FIGURE 1.11 SR flip-flop with clock.

Q

QJ

K

C

FIGURE 1.12 JK flip-flop with clock.

High

Low

FIGURE 1.10 Clock signal with a series of pulses.

15Boolean Algebra and Digital Logic Circuits for Computer Hardware

TABLE 1.12

Truth Table of the JK Flip-Flop

J K Present State Q(t) Next State Q(t + 1)

0 0 0 0 0 0 1 1 0 1 0 0 0 1 1 0 1 0 0 1 1 0 1 1 1 1 0 1 1 1 1 0

D

Q

Q

C

FIGURE 1.13 D flip-clop with clock.

S

R

Q

Q

C

SR flip-flop

J

K

Q

Q

C

JK flip-flop

D Q

QC

D flip-flop

FIGURE 1.14 Logic diagram symbols for the SR, JK, and D flip-flops.

TABLE 1.13

Truth Table for the D Flip-Flop

D Q(t) Q(t + 1)

0 0 0 0 1 0 1 0 1 1 1 1

16 Developing Windows-Based and Web-Enabled Information Systems

D Q

C

D Q

C

D Q

C

D Q

CClock

I3

I2

I1

I0

O3

O2

O1

O0

FIGURE 1.15 Logic diagram of the sequential circuit for a 4-bit register.

D Q C

D Q C

D Q C

D Q C

O1

O0D Q C

D Q C

D Q C

D Q C

Word3 Word2 Word1 Word0

A0 A1

C lo

ck W

ri te

-e na

bl e

I1

I0

FIGURE 1.16 Logic diagram of the sequential circuit for a 2-bit memory to store four words.

17Boolean Algebra and Digital Logic Circuits for Computer Hardware

words. If A1A0 are 00, Word0 is selected. If A1A0 are 01, Word1 is selected; if A1A0 are 10, Word2 is selected; and if A1A0 are 11, Word3 is selected. When the bit for Write-Enable in Figure 1.16 is set to 1, the D flip-flops for the selected word are allowed to update its out- puts from its inputs, that is, to write the input data to the memory at the selected word location. When the bit for Write-Enable in Figure 1.16 is set to 0, the memory content at the selected word location is read since the outputs O1O0 present what are stored in the memory.

1.3 Summary

This chapter gives the basics of how computer hardware, including key components such as the ALU in the CPU and memory, is made of digital logic circuits. Boolean algebra, which is useful in designing digital logic circuits, is introduced. The method of designing a digital logic circuit for a function is described to include first using a truth table to express the function and then converting the truth table into a Boolean expression, which is directly implemented by a digital logic circuit. The combinational or sequential circuits for some key computer hardware components such as the ALU and memory are provided in this chapter. With an understanding of how electronic devices are used to construct computer hardware and how computer hardware stores and operates on binary data in this chapter, we explain how binary data are used to represent non-binary numbers, characters, and other forms of information on comput- ers in the next chapter.

EXERCISES

1. Use a truth table to verify the AND and OR forms of the distributive property in Table 1.4.

2. Use a truth table to verify the AND form of de Morgan’s law in Table 1.4. 3. Given three input variables with binary values, x, y, and z, a majority voting func-

tion produces the output of 1 if the majority of the input variables take the value of 1, and the output of 0, otherwise. Express the majority voting function using a truth table, and express this function using a Boolean expression in the sum-of- product form.

4. Use a truth table to express the 2-bit adder, which performs the addition of two input bits with sum and carry-out but no carry-in. Express this function in a Boolean expression in the sum-of-product form.

5. Table 1.14 expresses the function of the parity checker for examining whether the received parity bit is consistent with the received data to detect transmis- sion errors. Express the parity checker using a Boolean expression in the sum-of- product form.

6. Develop a method of converting a truth table into a Boolean expression in the product-of-sum form.

7. Use only the NAND gates to construct a digital logic circuit for the XOR gate, and use the logic diagram to show the implementation of the digital logic circuit.

18 Developing Windows-Based and Web-Enabled Information Systems

8. Design a combinational circuit for a 3- to 8-bit decoder. 9. Design a sequential circuit for a memory that can store four words with 4 bits in

each word. 10. Design a synchronous sequential circuit for a 4-bit synchronous counter, which

goes through a predetermined sequence of states that reflect the binary number sequence 0000, 0001, 0010, 0011, …, 1111, and then back to 0000, as the clock pulses. The counter uses an input, Count-Enable, to enable counting when Count-Enable is set to 1. The sequential circuit does not change the state if Count-Enable is set to 0.

TABLE 1.14

Truth Table for the Parity Checker

x y z Parity Bit Parity Checker (Error)

0 0 0 0 0 0 0 0 1 1 0 0 1 0 1 0 0 1 1 0 0 1 0 0 1 0 1 0 1 0 0 1 1 0 0 0 1 1 1 1 1 0 0 0 1 1 0 0 1 0 1 0 1 0 0 1 0 1 1 1 1 1 0 0 0 1 1 0 1 1 1 1 1 0 1 1 1 1 1 0

19

2 Digital Data Representation

This chapter describes how numbers, characters, and other forms of information are repre- sented on computers using binary data. Some methods of detecting and correcting errors in data storage and transmission are also given in this chapter.

2.1 Representation of Numbers

In this section, we first introduce the conversion between decimal numbers and binary numbers for unsigned numbers. We then give the representation of signed numbers. An unsigned number is a number without the negative sign. A memory address is an unsigned integer on computers. A signed number has a positive or a negative sign. Data used in computation are typically declared as signed numbers on computers.

2.1.1 Conversion between Unsigned Binary and Decimal Numbers

A decimal number has a base of 10 and digit values of 0, 1, 2, …, 9. A binary number has a base of 2 and digit values of 0 and 1. The representation of a decimal number and a binary number using their base is given as follows:

Decimal number: 123.4510 = 1 × 102 + 2 × 101 + 3 × 100 + 4 × 10−1 + 5 × 10−2

Binary number: 101.012 = 1 × 22 + 0 × 21 + 1 × 20 + 0 × 2−1 + 1 × 2−2

The conversion of an unsigned binary number to a decimal number is carried out by computing the sum using the base representation of the binary number, as shown:

101.012 = 1 × 22 + 0 × 21 + 1 × 20 + 0 × 2−1 + 1 × 2−2 = 410 + 010 + 110 + 010 + 0.2510 = 5.2510

We use the decimal number system in our daily activities. An unsigned decimal number must be represented on computers using a binary number that is supported by computer hardware. A binary integer with N binary digits (called “bits”) can represent an unsigned decimal integer from 0 to 2N−1. For example, the decimal integers from 0 to 15 can be rep- resented by 4-bit binary numbers, as shown in Table 2.1.

One method of converting an unsigned decimal number to a binary number is the repeated subtraction method, which has the following steps:

1. Look for the highest power of 2 that is smaller than the decimal number. 2. Subtract that power of 2 from the decimal number. 3. Take the result of the subtraction in Step 2 as the decimal number.

20 Developing Windows-Based and Web-Enabled Information Systems

4. If the decimal number is 0 or meets another stopping criterion, go to Step 5; other- wise, go back to Step 1.

5. Construct the binary number by filling 1 in each bit position, where the power of 2 is used in the subtraction in Step 2 and 0 in each bit position, where the power of 2 is not used in the subtraction in Step 2.

A stopping criterion in Step 4 can be defined using the value after the decimal point, for example, the value after the decimal point <0.01, or the number of binary digits after the deci- mal point, for example, two binary digits after the decimal point. The repeated subtraction method requires the knowledge of decimal numbers for powers of 2, as shown in Table 2.2.

Example 2.1 illustrates the repeated subtraction method.

Example 2.1

Use the repeated subtraction method to convert the decimal number 123.45 into a binary number, with four binary digits after the decimal point.

In Step 1, we determine that the highest power of 2 smaller than 123.45 is 26 = 64. In Step 2, we perform the subtraction:

123.45 − 26 = 59.45 or 123.45 = 1 × 26 + 59.45

59.45 is taken as the next decimal number in Step 3. Since the decimal number is not 0, we go back to Step 1. The remaining steps are given as follows:

59.45 − 25 = 27.45 or 123.45 = 1 × 26 + 1 × 25 + 27.45 27.45 − 24 = 11.45 or 123.45 = 1 × 26 + 1 × 25 + 1 × 24 + 11.45 11.45 − 23 = 3.45 or 123.45 = 1 × 26 + 1 × 25 + 1 × 24 + 1 × 23 + 3.45 3.45 − 21 = 1.45 or 123.45 = 1 × 26 + 1 × 25 + 1 × 24 + 1 × 23 + 0 × 22 + 1 × 21 + 1.45 1.45 − 20 = 0.45 or 123.45 = 1 × 26 + 1 × 25 + 1 × 24 + 1 × 23 + 0 × 22 + 1 × 21 + 1 × 20 + 0.45

TABLE 2.1

Decimal Integers Represented by 4-Bit Binary Numbers

Decimal Numbers Binary Numbers

0 0000 1 0001 2 0010 3 0011 4 0100 5 0101 6 0110 7 0111 8 1000 9 1001 10 1010 11 1011 12 1100 13 1101 14 1110 15 1111

21Digital Data Representation

0.45 − 2−2 = 0.20 or 123.45 = 1 × 26 + 1 × 25 + 1 × 24 + 1 × 23 + 0 × 22 + 1 × 21 + 1 × 20 + 0 × 2−1 + 1 × 2−2 + 0.20

0.20 − 2−3 = 0.075 or 123.45 = 1 × 26 + 1 × 25 + 1 × 24 + 1 × 23 + 0 × 22 + 1 × 21 + 1 × 20 + 0 × 2−1 + 1 × 2−2 + 1 × 2−3 + 0.075

0.075 − 2−4 = 0.0125 or 123.45 = 1 × 26 + 1 × 25 + 1 × 24 + 1 × 23 + 0 × 22 + 1 × 21 + 1 × 20 + 0 × 2−1 + 1 × 2−2 + 1 × 2−3 + 1 × 2−4 + 0.0125

We stop with four binary digits after the decimal point. In Step 5, we construct the binary number: 1111011.0111.

Another method of converting a decimal number into a binary number is the division remainder method, which has the following steps for the integer part of the decimal number:

1. Divide the decimal number by 2, obtain the remainder of the division, and take the quotient of the division as the next decimal number.

2. If the decimal number is 0, go to Step 3; otherwise, go back to Step 1. 3. Construct the binary number by reading the division remainders from the last step

to the first step of Step 1 performed.

The division remainder method has the following steps for the fractional part of the decimal number:

1. Multiply the decimal number by 2, obtain the unit digit of the multiplication product, and take the fractional part of the multiplication product as the next decimal number.

2. If the decimal number is 0 or a required number of digits after the decimal point is obtained, go to Step 3; otherwise, go back to Step 1.

3. Construct the binary number by reading the unit digit of the multiplication prod- uct from the first step to the last step of Step 1 performed.

TABLE 2.2

Decimal Numbers for Some Powers of 2

Power of 2 Decimal Number

2−4 1 2

0 0625 4 = .

2−3 1 2

0 125 3 = .

2−2 1 2

0 25 2 = .

2−1 1 2

0 5 1 = .

20 1 21 2 22 4 23 8 24 16 25 32 26 64 27 128 29 256 210 512

22 Developing Windows-Based and Web-Enabled Information Systems

Example 2.2

Use the division remainder method to convert the decimal number 123.45 into a binary number, with four binary digits after the decimal point.

We first convert the integer part of the decimal number: 123. In Step 1, we divide 123 by 2, obtain 1 as the reminder of the division, and take the quotient, 61, as the next deci- mal number. In Step 2, because the decimal number is not 0, we go back to Step 1. The complete procedure is shown as follows.

123 ÷ 2 = 61 with the remainder of 1 or 123 = 61 × 21 + 1 × 20 (note that the remain- der is highlighted in bold)

61 ÷ 2 = 30 with the remainder of 1 or 123 = (30 × 21 + 1) × 21 + 1 × 20 = 30 × 22 + 1 × 21 + 1 × 20

30 ÷ 2 = 15 with the remainder of 0 or 123 = 15 × 23 + 0 × 22 + 1 × 21 + 1 × 20 15 ÷ 2 = 7 with the remainder of 1 or 123 = 7 × 24 + 1 × 23 + 0 × 22 + 1 × 21 + 1 × 20 7 ÷ 2 = 3 with the remainder of 1 or 123 = 3 × 25 + 1 × 24 + 1 × 23 + 0 × 22 + 1 × 21 + 1 × 20 3 ÷ 2 = 1 with the remainder of 1 or 123 = 1 × 26 + 1 × 25 + 1 × 24 + 1 × 23 + 0 ×

22 + 1 × 21 + 1 × 20 1 ÷ 2 = 0 with the remainder of 1 or 123 = (0 × 2 + 1) × 26 + 1 × 25 + 1 × 24 + 1 ×

23 + 0 × 22 + 1 × 21 + 1 × 20 = 1 × 26 + 1 × 25 + 1 × 24 + 1 × 23 + 0 × 22 + 1 × 21 + 1 × 20

As illustrated above, the last remainder is associated with the highest power of 2. Hence, we obtain the binary integer of 1111011 by reading the remainder from the last step to the first step of Step 1 performed.

We then convert the fractional part of the decimal value: 0.45. In Step 1, we multiply 0.45 by 2, obtain 0 as the unit digit of the multiplication product, and take the fractional part of the multiplication product, 0.90, as the next decimal value. In Step 2, because the decimal value is not 0, we go back to Step 1. The complete procedure is shown as follows:

0.45 × 2 = 0.90 or 0.45 = (0 + 0.90) × 2−1 = 0 × 2−1 + 0.90 × 2−1 (note that the unit digit is highlighted in bold)

0.90 × 2 = 1.80 or 0.45 = 0 × 2−1 + 0.90 × 2−1 = 0 × 2−1 + (1 + 0.80) × 2−1 × 2−1 = 0 × 2−1 + 1 × 2−2 + 0.80 × 2−2

0.80 × 2 = 1.60 or 0.45 = 0 × 2−1 + 1 × 2−2 + 0.80 × 2−2 = 0 × 2−1 + 1 × 2−2 + (1 + 0.60) × 2−1 × 2−2 = 0 × 2−1 + 1 × 2−2 + 1 × 2−3 + 0.60 × 2−3

0.60 × 2 = 1.20 or 0.45 = 0 × 2−1 + 1 × 2−2 + 1 × 2−3 + 0.60 × 2−2 = 0 × 2−1 + 1 × 2−2 + 1 × 2−3 + (1 + 0.20) × 2−1 × 2−3 = 0 × 2−1 + 1 × 2−2 + 1 × 2−3 + 1 × 2−4 + 0.20 × 2−4

As illustrated above, the unit digit from the first multiplication product is associated with the highest power of 2. Hence, we obtain the binary number of 0.0111 by reading the unit digit from the first step to the last step of Step 1 performed. Hence, the division remainder method produces the binary number 1111011.0111, which is the same as the result in Example 2.1.

2.1.2 Representation of Signed Integers

Using the methods in Section 2.1.1, we can convert an unsigned decimal number into an unsigned binary number that is directly stored and manipulated on comput- ers. This section introduces the signed magnitude method, one’s complement method and two’s complement method of representing a signed integer. ALU described in Chapter 1 includes the addition operation. The addition of signed integers represented by the signed magnitude method, one’s complement method and two’s complement method is described in this section.

23Digital Data Representation

2.1.2.1 Signed Magnitude Method

The signed magnitude method uses the left-most bit of a binary integer to represent the sign, specifically, 0 for the positive sign and 1 for the negative sign. An N-bit binary inte- ger can represent any value in the range [−(2N–1 − 1), (2N–1 − 1)]. For example, using 8 bits, we have the left-most bit representing the sign of the binary integer and the remaining 7 bits representing the magnitude (absolute value) of the binary integer. Hence, an 8-bit binary integer can represent any value in the range [−(27 − 1), (27 − 1)] or [−127, 127]. Table 2.3 gives examples of using 8 bits and the signed magnitude method to represent signed numbers −123, 123, 0, −127, and 127. To perform the addition of two signed integers using the signed magnitude method, we need to process the sign bit separately from the num- ber bits. For example, if we perform the addition of one positive integer and one negative integer, we need to determine which integer has a smaller magnitude (a smaller absolute value), perform the subtraction of the smaller integer from the larger integer using the number bits, and assign the sign of the larger integer as the sign of the subtraction result. Hence, it is not straightforward to perform the addition of signed integers in digital circuits.

2.1.2.2 One’s Complement Method

One’s complement method uses the diminished complement of an integer. Given an inte- ger a with base b and N digits, the diminished complement of a is defined by (bN − 1) − a. For example, given a decimal integer 43 with a base of 10 and 3 digits, the diminished complement of 43 is (103 − 1) − 43 = 999 − 43 = 956. To see how the diminished comple- ment of an integer is used in the addition, we look at an example of performing 156 − 43 as follows:

156 43 156 43 999 999 156 956

− = − + = + +

x x

the diminished( complement of 43) = + = + − +

x x

999 1112 999 1112 1000 1 ((end carry around: the carry-out beyond three ddigits is moved as the

unit digit and added) = +x 9999 1000 1 112 1 113

− + + = =

x x,

TABLE 2.3

Examples of Representing Signed Numbers Using the Signed Magnitude Method

Signed Decimal Numbers 8-Bit Binary Numbers

−123 11111011 123 01111011 0 00000000 −127 11111111 127 01111111

24 Developing Windows-Based and Web-Enabled Information Systems

which is the result of 156 − 43 = 113. As illustrated above, the addition with one negative integer can be performed by representing the negative integer (−43) using the diminished complement of that integer without the negative sign, adding the positive integer and the diminished complement, and performing the end carry-around.

Given a binary integer 0101 with a base of 2 and 4 digits, the diminished complement of 0101 is (24 − 1) − 0101 = 1111 − 0101 = 1010. Hence, obtaining the diminished complement of a binary integer is easy to implement in digital circuits by simply flipping each bit.

One’s complement method uses the diminished complement to represent a negative binary integer. A binary integer with N bits can represent a signed integer in the range of [−(2N−1 − 1), (2N−1  − 1)], with the left-most bit indicating the sign. For example, a binary integer with 4 bits can present an integer in the range of [−(24−1 − 1), (24−1 − 1)] or [−7, 7], as shown  in Table 2.4. Using one’s complement method to represent a signed integer, we have the left- most bit of 0 for a positive integer and the left-most bit of 1 for a negative integer. Using one’s complement method, we convert and represent a negative integer using its dimin- ished complement and leave a positive integer unchanged. Note that both 0000 and 1111 represent 0 in one’s complement method.

The addition of two binary integers is carried out in the following steps:

1. Represent the binary integers using one’s complement method. 2. Perform the addition of the binary integers represented by one’s complement

method. 3. Perform the end carry-around to obtain the result.

Example 2.3

Perform the addition of one positive 4-bit binary integer and one negative 4-bit binary integer, 0100 − 0011 (4 − 3), using one’s complement method.

TABLE 2.4

Representation of Signed Binary Integers with 4 Bits Using One’s Complement Method

Signed Integer Binary Integer

–7 1000 −6 1001 −5 1010 −4 1011 −3 1100 −2 1101 −1 1110 0 0000 or 1111 1 0001 2 0010 3 0011 4 0100 5 0101 6 0110 7 0111

25Digital Data Representation

In Step 1, we represent −0011 using its diminished complement 1100. In Step 2, we perform the addition:

0100

1100

1

+ (the deminished complement of 0011)

= 00000.

In Step 3, we perform the end carry-around by removing the carry-out bit 1 beyond the 4 bits and adding that bit at the unit position:

0000

1

0001

+

= ,

which is 1.

Example 2.4

Perform the addition of one positive 4-bit binary integer and one negative 4-bit binary integer, 0011 − 0100 (3 − 4), using one’s complement method.

In Step 1, we represent −0100 using its diminished complement 1011. In Step 2, we perform the addition:

0011

1011

1

+

=

(the diminished complement of 0100)

1110.

In Step 3, we perform the end carry-around:

1110

0

1110

+

=

which is −1.

2.1.2.3 Two’s Complement Method

Two’s complement method is the most common method used on computers. In Table 2.4, for one’s complement method, we have 0000 and 1111 representing 0. Two’s complement method addresses this problem by using the complement that is defined by bN − a for an integer a with a base of b and N digits. That is, two’s complement is one’s complement plus 1. For example, the complement of 0101 is 24 − 0101 = 10000 − 0101 = 1011. It is easy to obtain the complement of a binary integer in digital circuits by flipping each bit and then adding 1. Table 2.5 shows the representation of signed binary integers using two’s comple- ment method and 4 bits. Using N bits, two’s complement method can represent signed integers in the range [−2N–1, 2N–1 − 1].

26 Developing Windows-Based and Web-Enabled Information Systems

The addition of two signed integers with N bits represented by two’s complement method is performed in the following steps:

1. Represent the binary integers using two’s complement method. 2. Perform the addition of the binary integers represented by two’s complement

method. 3. Discard the carry-out beyond N bits.

Example 2.5

Perform the addition of one positive 4-bit integer and one negative 4-bit integer, 0100 − 0011 (4 − 3), using two’s complement method.

In Step 1, we represent −0011 by its complement 1101. In Step 2, we perform the addition:

0100

1101

10001

+

=

(the complement of 0011)

.

In Step 3, we drop the carry-out bit 1 beyond 4 bits and obtain 0001 as the result.

Example 2.6

Perform the addition of one positive 4-bit integer and one negative 4-bit integer, 0011 − 0100 (3 − 4), using two’s complement method.

TABLE 2.5

Representation of Signed Binary Integers with 4 Bits Using Two’s Complement Method

Signed Integer Binary Integer

−8 1000 −7 1001 −6 1010 −5 1011 −4 1100 −3 1101 −2 1110 −1 1111 0 0000 1 0001 2 0010 3 0011 4 0100 5 0101 6 0110 7 0111

27Digital Data Representation

In Step 1, we represent −0100 by its complement 1100. In Step 2, we perform the addition:

0011

1100

1111

+

=

(the complement of 0100)

.

Since there is no carry-out, Step 3 produces the result 1111, which is −1.

2.1.3 Representation of Signed Floating Point Values

A floating point value is represented using the sign bit (0 for the positive sign and 1 for the negative sign), bits for the exponent, and bits for the fractional part of the floating point value, which is called “significand.” For example, a floating point value, 100.01, is equal to 0.10001 × 23, whose representation is shown in Table 2.6 using 16 bits, with 1 bit for the sign, 5 bits for the exponent, and 10 bits for the significand. The number of bits used for the exponent determines the range of values that can be represented. The number of bits used for the significand determines the precision of the value that is represented.

An exponent may be a positive integer or a negative integer. To represent a signed expo- nent, a bias value is used. For example, 5 bits for an exponent can represent a value in the range [0, 31]. A value near the middle of the range, for example, 16, is used as the bias value for a 5-bit exponent. The biased exponent is obtained by adding the bias value to an expo- nent. For example, the exponent value of 3 is represented by the biased exponent, which is 16 + 3 = 19 or 10000 + 11 = 10011. The exponent value of −1 is represented by the biased exponent, which is 16 − 1 = 15 or 10000 − 1 = 1111. Hence, in the biased exponent represen- tation, a biased exponent larger than the bias value represents a positive exponent, and a biased exponent smaller than the bias value represents a negative exponent. The exponent is obtained by subtracting the bias value from the biased exponent.

A floating point number, 100.01, is equal to 0.10001 × 23, 0.010001 × 24, 0.0010001 × 25, and so on, which correspond to different representations of the same floating point number. To produce a unique representation of a floating point number, normalization is necessary to require that the left-most bit of the significand must be 1. With normalization, 100.01 can be represented only in the form in Table 2.6 for 0.10001 × 23.

2.2 Representation of Alphabet and Control Characters

Text and control information is represented on computers by using codes for alpha- bet, number, control, and other characters. Table 2.7 gives American Standard Code for Information Interchange (ASCII) character codes that use 8 bits and have code values from 0 to 127 (27 – 1). A code itself uses 7 bits, and the left-most bit (the eighth bit) is used as the

TABLE 2.6

Representation of a Floating Point Number

Sign Exponent Significand

0 00011 1000100000

28 Developing Windows-Based and Web-Enabled Information Systems

parity bit, as described in Chapter 1. For example, the character code for letter A is 65 or 1000001. If the parity bit is 0, the complete code for letter A is 01000001.

ASCII includes 128 characters. Unicode and Universal Character Set covers more char- acters. For example, Unicode uses 16 bits and covers mathematical symbols, characters in languages other than English, and so on.

2.3 Error Detection and Correction

As described in Section 2.2, the code for a character includes a parity bit for error detec- tion. When data are stored on computers or transmitted over networks, errors may occur.

TABLE 2.7

ASCII Character Codes

Character Code Character Code Character Code Character Code

Null 0 Space 32 @ 64 ` 96 Start of heading 1 ! 33 A 65 a 97 Start of text 2 “ 34 B 66 b 98 End of text 3 # 35 C 67 c 99 End of transmission 4 $ 36 D 68 d 100 Enquiry 5 % 37 E 69 e 101 Acknowledge 6 & 38 F 70 f 102 Bell (beep) 7 ‘ 39 G 71 g 103 Backspace 8 ( 40 H 72 h 104 Horizontal tab 9 ) 41 I 73 i 105 Line feed, new line 10 * 42 J 74 j 106 Vertical tab 11 + 43 K 75 k 107 Form feed, new page 12 , 44 L 76 l 108 Carriage return 13 - 45 M 77 m 109 Shift out 14 . 46 N 78 n 110 Shift in 15 / 47 O 79 o 111 Data link escape 16 0 48 P 80 p 112 Device control 1 17 1 49 Q 81 q 113 Device control 2 18 2 50 R 82 r 114 Device control 3 19 3 51 S 83 s 115 Device control 4 20 4 52 T 84 t 116 Negative acknowledge 21 5 53 U 85 u 117 Synchronous idle 22 6 54 V 86 v 118 End of transmission block 23 7 55 W 87 w 119 Cancel 24 8 56 X 88 x 120 End of medium 25 9 57 Y 89 y 121 Substitute 26 : 58 Z 90 z 122 Escape 27 ; 59 [ 91 { 123 File separator 28 < 60 \ 92 | 124 Group separator 29 = 61 ] 93 } 125 Record separator 30 > 62 ^ 94 ~ 126 Unit separator 31 ? 63 _ 95 Delete 127

29Digital Data Representation

Detecting and correcting errors are required for reliable data storage and transmission. The parity bit method can detect an odd number of errors but cannot detect an even num- ber of errors or identify where errors occur to correct errors. This section describes the Hamming algorithm of detecting and correcting errors.

An ASCII code uses 7 bits representing a character and 1 parity bit. If k check bits are used to detect errors in m data bits, there are, totally, n = m + k code bits. The Hamming distance of two codes is the number of bits in which two codes are different. For example, 1000001 (65), along with its parity bit 0, produces the ASCII code 01000001 for letter A, and 1000011 (67), along with its parity bit 1, produces the ASCII code 11000011 for letter C. The two codes, 01000001 and 11000011 differ in two bit positions and have a Hamming distance of 2. For a code system such as ASCII, the minimum Hamming distance, Dmin, is the small- est distance among all pairs of codes in the code system. If the number of bit errors that occur to a code is no more than Dmin, the errors can be detected because Dmin is the smallest Hamming distance among all pairs of valid codes, and thus, the code with no more than Dmin bits of errors is different from any valid codes in the system. We detect the presence of error(s) in a given code if the given code is different from any valid code in the system.

If the presence of error(s) in a given code is detected, we correct error(s) in a given code to the valid code that is closest to the given code with error(s) in the Hamming distance. Considering that two valid codes, x and y, have the smallest Hamming distance Dmin in the code system and p bits of errors occur to x, producing x′, correcting x′ to its original valid code x requires that x′ has a smaller distance to x than y. Because the Hamming distance of x′ and x is p, the Hamming distance of x′ and y must be greater than p, with the small- est value being p + 1. Hence, the Hamming distance between x and y, Dmin ≥ p + p + 1 or Dmin ≥ 2p + 1, is required to correct p bits of errors such that the correction to the original valid code is guaranteed. That is, Dmin of a code system must be 2p + 1 to correct p bits of errors.

Given m data bits, we need to determine how many check bits are required to detect and correct errors. Let us consider the simple case of detecting and correcting 1 bit of error. Let k denote the number of check bits and n denote the total numbers of bits in a code: n = m + k. Using m data bits, we can create 2m valid codes with Dmin = 1. For each valid code including m data bits and k check bits, there are n possible positions where 1 bit of error can occur, and thus, there are n possible invalid codes with 1 bit of error. Hence, for each code, there are n + 1 bit patterns (including 1 valid code and n invalid codes). For 2m valid codes, there are (n + 1)2m bit patterns. For n bits, there are totally 2n possible bit patterns. The following inequality should hold:

(n + 1)2m ≤ 2n

(m + k + 1)2m ≤ 2m+k

m + k + 1 ≤ 2k

This inequality can be used to determine k, the number of check bits to detect and correct 1 bit of error.

Example 2.7

Determine the number of check bits needed for a code with 4 data bits to detect and correct 1 bit of error.

30 Developing Windows-Based and Web-Enabled Information Systems

For m = 4, we have:

4 + k + 1 ≤ 2k

5 + k ≤ 2k

Comparing the values of 5 + k and 2k, we obtain k ≥ 3. Hence, we can let k = 3.

After determining the number of check bits for detecting and correcting 1 bit of error, the Hamming algorithm is used to construct Hamming codes. The Hamming algorithm has the following steps:

1. Number n bit positions from right to left, starting with 1. 2. Let each bit position, which is a power of 2, be a check bit; let other bits be data bits,

and put in the values of data bits. 3. Write each bit position as the sum of the numbers that are powers of 2, using the

highest power of 2 first, and then determine which bit positions each check bit contributes to the sum of the numbers for these bit positions.

4. Use the parity bit to determine the value of each check bit by considering the val- ues at bit positions where the check bit contributes to the sum of the numbers.

Example 2.8 illustrates these steps.

Example 2.8

Construct the Hamming code with data bits 0100, using the Hamming algorithm and the even parity bit such as one in Table 1.6.

Using the result from Example 2.7, we know that we need 3 check bits for 4 data bits, a total of 7 bits, for the code. In Step 1 of the Hamming algorithm, we number 7 bit posi- tions from right to left, starting from 1, as follows:

Bit position 7 6 5 4 3 2 1 Value

In Step 2, we let bit positions 1, 2, and 4 be the check bits because these bit positions are powers of 2, and put in the values of data bits 0100 as follows:

Bit position 7 6 5 4 3 2 1 Value 0 1 0 0

In Step 3, we write each bit position as the sum of the numbers that are powers of 2 as follows:

1 = 1 2 = 2 3 = 1 + 2 4 = 4 5 = 1 + 4 6 = 2 + 4 7 = 1 + 2 + 4

Hence, the check bit at position 1 contributes to the sum of the numbers for bit posi- tions 1, 3, 5, and 7. The check bit at position 2 contributes to the sum of the numbers for

31Digital Data Representation

bit positions 2, 3, 6, and 7. The check bit at position 4 contributes to bit positions 4, 5, 6, and 7. In Step 4, we compute the value of each check bit using the even parity. Because the check bit at position 1 contributes to the sum of numbers for bit positions 1, 3, 5, and 7:

Bit position 7 5 3 1 Value 0 0 0 ?

the even parity bit at bit position 1 is 0. Because the check bit at position 2 contributes to the sum of numbers for bit positions 2, 3, 6, and 7:

Bit position 7 6 3 2 Value 0 1 0 ?

the even parity bit at bit position 2 is 1. Because the check bit at position 4 contributes to the sum of numbers for bit positions 4, 5, 6, and 7:

Bit position 7 6 5 4 Value 0 1 0 ?

the even parity bit at bit position 4 is 1. Putting all the check bits and data bits together, we have the Hamming code 0101010 as follows:

Bit position 7 6 5 4 3 2 1 Value 0 1 0 1 0 1 0

After constructing the code, we can detect a 1-bit error using the check bits, as shown in Example 2.9.

Example 2.9

Detect and correct a 1-bit error in a code 0101110. Note that the code 0101110 is produced by introducing a 1-bit error to the Hamming code 0101010 from Example 2.8 at bit position 3.

The parity bit at bit position 1 checks bit positions 1, 3, 5, and 7, which have values of 0, 1, 0, and 0, respectively, and violate the even parity. Hence, a 1-bit error occurs at bit position 1, 3, 5, or 7. The parity bit at bit position 2 checks bit positions 2, 3, 6, and 7, which have values of 1, 1, 1, and 0, respectively, and violate the even parity. Hence, a 1-bit error occurs at bit position 2, 3, 6, or 7. The parity bit at bit position 4 checks bit positions 4, 5, 6, and 7, which have values of 1, 0, 1, and 0, respectively, and conform to the even parity. Hence, there is no 1-bit error at bit positions 4, 5, 6, and 7. Taking all the results of the parity checks:

• A 1-bit error at bit position 1, 3, 5, or 7, using the check bit at position 1 • A 1-bit error at bit position 2, 3, 6, or 7, using the check bit at position 2 • No 1-bit error at bit positions 4, 5, 6, and 7, using the check bit at position 4

A 1-bit error at position 3 is detected because there is no 1-bit error at positions 4, 5, 6, and 7, and both check bits 2 and 3 indicate the possibility of the 1-bit error at position 3. A straightforward method of detecting a 1-bit error is using the sum of the bit positions of the check bits that indicate an error. Because check bits 1 and 2 indicates an error, 1 + 2 = 3. Hence, the 1-bit error occurs at bit position 3. To correct the 1-bit error, we sim- ply flip the bit at position 3 from 1 to 0.

32 Developing Windows-Based and Web-Enabled Information Systems

2.4 Summary

This chapter illustrates how objects (e.g., numbers and characters) used by end users are supported on computer hardware using binary data. Because the data representation of objects uses check bits to detect and correct errors that may occur to stored or transmitted data, error detection and correction methods such as the Hamming algorithm are also introduced in this chapter. Based on the understanding of materials in this chapter, the next chapter gives an overview of how computer system software, especially the operating system and network software, manages computer and network hardware resources and allows application software to provide services to end users without directly handling the implementation details of computer and network hardware.

EXERCISES

1. What decimal integers can be represented by 3-bit binary integers? Show the rep- resentation of these decimal integers by using 3-bit binary integers.

2. Convert 111011.01111 into a decimal number. 3. Use the repeated subtraction method to convert 12.34 into a binary number, with

two digits after the decimal point. 4. Use the division remainder method to convert 12.34 into a binary number, with

two digits after the decimal point. 5. Use the signed magnitude method and an 8-bit binary number to represent −12

and 12. 6. Use one’s complement method and an 8-bit binary number to represent −12 and 12. 7. Use two’s complement method and an 8-bit binary number to represent −12 and 12. 8. Perform the addition of one 8-bit positive number and one 8-bit negative number,

00010111 − 00001001, using one’s complement method. 9. Perform the addition of one 8-bit positive number and one 8-bit negative number,

00001001 − 00010111, using one’s complement method. 10. Perform the addition of one 8-bit positive number and one 8-bit negative number,

00010111 − 00001001, using two’s complement method. 11. Perform the addition of one 8-bit positive number and one 8-bit negative number,

00001001 − 00010111, using two’s complement method. 12. Determine the number of check bits that are needed for a code system with 8 data

bits to detect and correct 1 bit of error. 13. Construct the Hamming code with 8 data bits 01001011, using the Hamming

algorithm. 14. Detect and correct a 1-bit error in the Hamming code 010111010110.

33

3 Computer and Network System Software

System software manages computer hardware resources and provides libraries of system functions that application software can use to create a variety of services to end users with- out handling computer hardware directly. This chapter introduces two important kinds of system software: the operating system that manages data processing and storage on a computer and network software that manages data communication between computers.

3.1 The Operating System

Figure 3.1 gives the von Neumann model of computer architecture (Null and Lobur, 2006), which puts computer hardware devices into three categories: CPU (which consists of the ALU, registers, and control unit), memory, and input/output (I/O) devices. The hardware implementation of the ALU and memory is described in Chapter 1. I/O devices include the hard drive, network interface card (NIC), keyboard, mouse, USB, printer, etc. The CPU gets program instructions and data from memory, executes program instructions, places the results in registers or memory, and responds to events such as requests from I/O devices. Interactions of the CPU, memory, and I/O devices are managed by the operating system. This section describes three main functions of the operating system:

• Process management • Storage management • I/O management

The operating system also carries out other functions for managing networking and com- munication, security, quality of service, etc. Networking and communication management is described in Section 3.2.

3.1.1 Process Management

A process is a program in execution. A process has its execution environment to keep track of the process status and data, which include the running state (e.g., ready or waiting), the memory address of the next program instruction, the values of variables, the contents of CPU registers, page tables, resources being used, etc. A process may create multiple threads to perform multiple tasks simultaneously. A thread is the smallest unit that can be scheduled for the CPU to execute. Threads share the same execution environment as their parent process.

Since the CPU executes only one process/thread at a given time, the operating system schedules processes and threads to determine which process/thread uses the CPU at a given time. Examples of process/thread scheduling methods are round robin; first come,

34 Developing Windows-Based and Web-Enabled Information Systems

first serve; shortest job first; shortest remaining time first; and priority-based scheduling. The round robin method is the most popular method, which lets all processes/threads share the CPU time by taking turns in using the CPU for a given amount of time. In the round robin method, more processes/threads will result in more waiting time of each process/thread before its turn of using the CPU. In the first come, first serve method, a process/thread that comes to request for the CPU first will be executed and completed by the CPU first, and thus, processes/threads use the CPU one by one instead of sharing CPU time. In the shortest job first method, a process/thread that requires the shortest CPU pro- cessing time is scheduled to use the CPU first. In the shortest remaining time first method, a process/thread that requires the least amount of CPU time to complete the process is scheduled to use the CPU first. If it is difficult to estimate the CPU processing time or the remaining CPU processing time of a process/thread, the shortest job first method or the shortest remaining time first method should not be employed. The priority-based schedul- ing method schedules processes/threads based on the priority of each process/thread that may be assigned to reflect the importance or another aspect of the process/thread.

The round robin method is a preemptive scheduling method because a process/thread is interrupted, pushed out of running on the CPU, and placed into the waiting state from the running state when the process/thread uses all its allocated amount of CPU time. The first come, first serve method; the shortest job first method; the shortest remaining time first method; and the priority-based method are non-preemptive scheduling methods because a process/thread is not interrupted before it completes its execution on the CPU. When a process/thread is interrupted before its completion of execution, the operating system performs the context switch to store the execution environment of the interrupted process/thread and bring in the execution environment of the next process/thread. When it is the turn for a process/thread to use the CPU again, the operating system restores the execution environment of the process/thread to resume the execution of the process/ thread. When a process completes its execution, the operating system deletes the process and cleans up after the process termination.

3.1.2 Storage Management

The operating system manages the storage of data in memory, the hard drive, the USB drive, and other data storage devices. The operating system breaks a data file into pages of standard size for data storage efficiency. If data file is stored entirely on a single block of memory or storage space, the block of memory or storage space left after a data file is deleted may not be usable for another data file of a different size, resulting in unused memory or storage space. Breaking a data file into pages of standard size allows a page of

CPU Memory Input/output devices

Data bus

Address bus

Control bus

FIGURE 3.1 The von Neumann model of computer architecture.

35Computer and Network System Software

any data file to fill in a page space left by a deleted page, maximizing the use of memory or storage space. Hence, pages of a data file may be stored at different locations of memory or permanent storage. Pages, which are currently used by a process, are stored in memory. If the pages needed are not in memory, they need to be brought into memory from a perma- nent storage such as the hard drive, the USB drive, etc.

The operating system creates a virtual memory system to manage pages and performs the translation of a virtual memory address and a physical memory address. The memory/ storage management performed by the operating system includes the following:

• Breaking files into pages • Translating virtual memory addresses to physical memory addresses • Transferring pages between memory and a permanent storage device • Maintaining page tables to locate pages • Allocating memory/storage space to pages • Releasing memory/storage space of deleted pages • Keeping track of free memory/storage space

The operating system also performs file management, including file creation, file deletion, directory creation, and directory deletion.

3.1.3 I/O Management

I/O devices usually have device drivers, which are software packages implementing func- tions for particular I/O devices to allow interactions with I/O devices. For example, a particu- lar printer has its special driver that may come with the operating system to allow application software (e.g., Microsoft Word) to send data to the printer for printing. The operating sys- tem provides system calls that application software can use to interact with I/O devices.

3.2 Networking and Communication Software

Figure 3.2 shows the open systems interconnect (OSI) reference model, which defines the functions required to implement networking and data communication between comput- ers. The OSI reference model organizes networking functions in seven layers: application, presentation, session, transport, network, data link, and physical. Transmission control protocol/internet protocol (TCP/IP) is the most common software suite that implements networking functions in the OSI reference model and has four layers: application, trans- port, network, and link. Figure 3.2 shows the mapping of OSI and TCP/IP layers. The OSI reference model and TCP/IP are described in the following sections.

3.2.1 OSI Reference Model

At the application layer, a network application (e.g., Web application) on a computer at the source of data communication creates data packets containing application data (e.g., data containing a uniform resource locator (URL) entered by a user of the Web applica- tion to search for a website) and sends data packets to the same network application on a

36 Developing Windows-Based and Web-Enabled Information Systems

computer at the destination of the data communication. The client–server architecture is commonly used by a network application to have the client software and the server soft- ware of the application work together. For example, Internet Explorer is the client software of the Web application in the Windows operating system and works together with Internet Information Server, which is the server software of the Web application.

The presentation layer manages the presentation of communicated data by defining the data presentation format (e.g., the format of floating point numbers) that is under- stood by computers of different kinds on the network. The session layer is responsible for establishing synchronized dialogs through data communication sessions between two computers on the network. Many types of data communication on the network involve dialogs between a client and its server, and there are many data packets being transmitted between the client and the server. Because all client–server dialogs share the same net- work, data packets for different client–server dialogs are mixed in the network bandwidth. A data communication session is established to identify all data packets that belong to a synchronized dialog between the client and the server.

The transport layer takes care of the reliability of data communication by providing acknowledgment, error checking, and retransmission. The network layer is responsible for finding a routing path to relay data packets from the source computer to the destination computer. The network layer also breaks down a data packet into smaller datagrams that are suitable to be transmitted over a particular physical transmission medium. The data link layer detects and corrects errors of physical data transmission that occur in the physi- cal layer. The physical layer transmits binary bits across the physical transmission medium.

3.2.2 Transmission Control Protocol/Internet Protocol

TCP/IP covers networking functions in the OSI reference model in four layers: application, transport, network, and link. The application layer of the TCP/IP contains network applica- tion program suites (e.g., hypertext transfer protocol (HTTP) for the Web application, simple mail transfer protocol (SMTP) for the email application, and a domain name server for trans- lating a network domain name to an IP address), each of which implements networking func- tions at the application and presentation layers of the OSI reference model. The transport

Application

Presentation

Session

Transport

Network

Data link

Physical

Application, e.g., HTTP, SMTP, etc.

Transport

Network

Link

OSI: TCP/IP:

FIGURE 3.2 The OSI reference model and the TCP/IP.

37Computer and Network System Software

layer of TCP/IP contains program suites such as TCP, user datagram protocol (UDP) and Internet Control Message Protocol (ICMP), which implement the networking functions at the session and transport layers of the OSI reference model. The term “protocol” is used to refer to a program suite. TCP is used for connection-oriented reliable data delivery through acknowl- edgment, error checking, and retransmission. Network applications such as HTTP and SMTP need to use TCP for connection-oriented reliable data delivery. UDP is used to provide con- nectionless, efficient data transmission without guarantee for reliable data delivery. Network applications such as domain name server (DNS) do not require reliable data delivery and, thus, use UDP for data transmission. ICMP is used to exchange command and control infor- mation between two computers on a network. The network layer of TCP/IP contains mainly IP, which implements the networking functions at the network layer of the OSI reference model to handle the routing of network data and break down network data to fit into the physical transmission medium. The link layer of TCP/IP contains network drivers that imple- ment the networking functions at the data link and physical layers of the OSI reference model.

Figure 3.3 illustrates how data generated by the client program of the HTTP Web appli- cation on a computer at the data source are transmitted over the Ethernet to the server program of the HTTP Web application on a computer at the data destination through the TCP/IP. On the source computer, the HTTP client at the application layer of the TCP/IP generates the HTTP data packet and passes it down to the TCP at the transport layer of the TCP/IP. The TCP generates the TCP segment by adding the TCP header to the HTTP data packet. The following are the important fields in the TCP header to assist connection- oriented reliable data delivery:

• Source port: indicates which application sends the data • Destination port: indicates which application should receive the data • Sequence number: works with the acknowledgement number to identify the data

stream in a network session between the application programs on the source com- puter and the destination computer

Source computer:

HTTP Client

TCP

IP

Ethernet driver

TCP/IP layer:

Application

Transport

Network

Link

HTTP data packet

TCP segment

IP datagram

Router:

IP

Ethernet driver Ethernet frame

Destination computer:

HTTP Server

TCP

IP

Ethernet driver

HTTP data packet

TCP segment

IP datagram

Actual data path Virtual data path

Ethernet frame IP datagram

FIGURE 3.3 Data transmission through TCP/IP layers.

38 Developing Windows-Based and Web-Enabled Information Systems

• Acknowledgement number: works with the sequence number to identify the data stream in a network session

• Data offset: indicates the header length • Flags: contain six control bits—URG, ACK, PSH, RST, SYN, and FIN • Window: indicates the number of bytes allowed in a segment • Header checksum: applies to the TCP header fields to ensure the correct transmis-

sion of the TCP header fields

TCP passes the TCP segment down to the IP at the network layer of TCP/IP. The IP gen- erates the IP datagram by adding the IP header to the TCP segment. The following are the important fields in the IP header:

• Packet ID: a serial number assigned to each IP datagram • Header length: indicates the length of the IP header • Total length of the datagram: indicates the length of the IP datagram • Flags: indicate whether the datagram may be fragmented by intermediate network

nodes between the source computer and the destination computer • Fragment offset: indicates the location of a fragment within a datagram • Time to live: specifies the maximum number of hops over the network path to

reach the destination computer • Protocol number: indicates the protocol (TCP, UDP, etc.) at the transport layer that

calls the IP • Source IP address: contains the IP address of the source computer • Destination IP address: contains the IP address of the destination computer • Header checksum: applies to the IP header fields to ensure the correct transmis-

sion of the IP header fields

The IP passes the IP datagram down to the Ethernet driver on the source computer. The Ethernet driver generates the Ethernet frame by adding the Ethernet header to the IP datagram and sends the Ethernet frame to the next router on the network path to the destination computer. On the router, the Ethernet driver rips off the Ethernet header of the Ethernet frame and passes the IP datagram up to the IP at the network layer. The IP obtains the destination IP address from the IP header fields, determines the next router to relay the IP datagram to the destination computer, and passes the IP datagram down to the Ethernet driver of the router. The Ethernet driver generates the Ethernet frame and sends it to the next router. When the Ethernet frame reaches the destination computer, the Ethernet driver on the destination computer rips off the Ethernet header of the Ethernet frame and passes the IP datagram up to the IP at the network layer. The IP rips off the IP header and passes the TCP segment up to the TCP at the transport layer. The TCP obtains the destination port to identify the server program of the HTTP Web application, rips off the TCP header, and passes the HTTP data packet to the HTTP server at the application layer. Although network programs at the higher layers of the TCP/IP (i.e., the application, trans- port, and network layers) on two computers do not communicate with each other directly but through the link layer along the actual data path, the network program at each layer responds to data only at that layer after data are passed up from the link layer, making the

39Computer and Network System Software

virtual path at each layer between two computers. Figure 3.4 shows the packaging of data by TCP/IP layers, each of which adds its header to data.

When a user requests a website to start a session of interactive data stream with the website, a network session needs to be established for identifying the data stream in this network session between the client program of the Web application on the source com- puter and the server program of the Web application on the destination computer. The net- work session is established through a three-way hand-shaking procedure, which includes the exchange of three TCP segments between the source computer and the destination computer, as shown in Figure 3.5. At first, the source computer sends to the destination computer a TCP segment with the control bit SYN being set to the TRUE value and a given sequence number m to request opening a network session. The destination computer responds to this request by sending a TCP segment with SYN being set to TRUE, a sequence number n, ACK being set to TRUE, and the acknowledgement number generated using the sequence number m (e.g., m + 1). Because the purpose of ACK along with the acknowledge- ment number is to acknowledge receiving the sequence number m from the source com- puter, the acknowledgement number must be related to the sequence number m. In Figure 3.5, the method of adding 1 to the sequence number m is used to create the acknowledge- ment number. A more complex method of creating an acknowledgement number related to a sequence number is used in TCP/IP software. When the source computer receives the TCP segment from the destination computer, the source computer knows that its request for opening a network session is received and sends a TCP segment to the destination computer to acknowledge receiving the sequence number n. When the destination com- puter receives this TCP segment for acknowledging the receipt of the sequence number n, a network session is established between the client program on the source computer and

TCP/IP layer:

Application

Transport

Network

Link

Packaging of data:

DataHTTP data packet:

DataTCP segment:

DataIP datagram:

DataEthernet frame:

TCP header

TCP headerIP header

TCP headerIP headerEthernet header

FIGURE 3.4 Data packaging through TCP/IP layers.

1. a TCP segment with SYN = TRUE, and sequence number = m

Source computer Destination computer

2. a TCP segment with SYN = TRUE, sequence number = n, ACK = TRUE, and acknowledge number = m + 1

3. a TCP segment with ACK = TRUE, and acknowledge number = n + 1

FIGURE 3.5 TCP three-way hand-shaking procedure.

40 Developing Windows-Based and Web-Enabled Information Systems

the server program on the destination computer, with two sequence numbers m and n to identify data in this network session. Data from the source computer to the destination computer are identified using a sequence number based on m, and data from the destina- tion computer to the source computer are identified using a sequence number based on n. Although the network bandwidth is a shared resource and is used to transmit data belong- ing to many different network sessions, data in a particular network session are identified by a unique pair of sequence numbers for this network session. The purpose of three steps in the TCP three-way hand-shaking procedure is to exchange the two sequence numbers between the source computer and the destination computer. Hence, a network session is not a time slot reserved on the network bandwidth but a virtual session identified by two sequence numbers.

After a network session is established, the sequence number and the acknowledgement number in a TCP header also take part in reliable data delivery. Suppose that application data are sent from the source computer to the destination computer through a sequence of TCP segments that have the related sequence numbers, m, m + 100, m + 200, m + 300, and so on. When the destination computer receives the first TCP segment from the source com- puter with the sequence number m, the destination computer sends to the source computer a TCP segment with the acknowledgement number m + 100 to acknowledge receiving the TCP segment with the sequence number m and expecting to receive the next TCP segment with the sequence number m + 100. If the TCP segment with the sequence number m is lost on the network and does not reach the destination computer, the destination computer will not send the TCP segment for acknowledgement. If the source computer sends out the TCP segment with the sequence number m but does not receive the acknowledgement within a time period, the source computer knows that the TCP segment is lost and will retransmit the TCP segment with the sequence number m again. Hence, the sequence number and the acknowledgement number work together for reliable data delivery through acknowledge- ment and retransmission.

UDP provides connectionless data transmissions and supports network applications that do not require reliable data delivery. For example, DNS lets domain name servers peri- odically send updates of domain name mapping information to each other. DNS does not require reliable data delivery of those updates. DNS uses UDP to send updates. The UDP header includes fields such as source and destination ports, header length, and checksum but does not include the sequence number and the acknowledgement number.

The IP header fields support routing and breaking a TCP segment into IP datagrams whose size fits to the transmission size allowed by the physical transmission medium. The flags and fragment offset fields of the IP header are used to break up a TCP segment into smaller IP datagrams that fit to the transmission size of the physical transmission medium on the source computer and reassemble IP datagrams into the original TCP segment on the destination computer.

Routing is based on the source IP address and the destination IP address in the IP header. The IP address (e.g., 191.167.2.0) of a computer indicates which local area network, regional area network, and Internet backbone network the computer belongs to. Data from the source computer are sent to the router for the local area network of the computer. Each router on the Internet keeps a routing table with entries of destination, next hop (i.e., another router), and a certain routing metric (e.g., the hop count to reach the destination through the next hop). When a router receives an IP datagram, the router obtains the destination IP address from the IP header, looks up the routing table using the destination IP address, selects the next hop to the destination, and sends data to the next hop. In general, data are routed from the source computer to the router of the local area network for the source

41Computer and Network System Software

computer, to the router of the regional area network for the local area network, to the Internet backbone router for the regional area network, to the regional area network for the local area network of the destination computer, to the local area network for the destination computer, and finally, to the destination computer. However, some routers on the Internet may contain incorrect routing entries, which may cause a data packet to fall into an end- less routing loop on the Internet instead of reaching the destination computer. To prevent a data packet from staying on the Internet forever in an endless routing loop, the time to live (TTL) field of the IP header sets the maximum number of hops that a data packet can travel before reaching the destination computer. When a data packet reaches a router, the router decreases the TTL value by 1. When the TTL value of a data packet becomes 0 and the data packet has not reached the destination, the data packet is dropped by a router.

The link layer of the TCP/IP contains the software driver for the NIC on a computer to handle the physical transmission of signals representing binary data across a physical transmission medium. The link layer of the TCP/IP includes the address resolution proto- col to map an IP address to a medium address control address, which is built into each NIC and uniquely identifies each individual computer on the network.

3.3 Summary

This chapter describes how system software, especially the operating system and network software, manages computer and network resources and enables application software to provide end-user services of data processing, data storage, and data communication with- out directly dealing with the implementation details of the computer and network hard- ware. The operating system manages the scheduling of multiple processes/threads on the CPU, the storage and retrieval of data in memory and other storage devices, and interac- tions with I/O devices. The network software, especially the TCP/IP, allows heterogeneous computers on the Internet to locate each other and transmit data over the shared network bandwidth. The next chapter introduces a variety of computer and network applications that are built on system software, including information system applications.

EXERCISES

1. What scheduling method is commonly used on a computer to schedule the execu- tion of processes/threads on the CPU? Explain the pros and cons of using this CPU scheduling method in comparison with other CPU scheduling methods described in this chapter.

2. Discuss the advantages of breaking a data file into pages and storing pages in memory or other storages such as the hard drive.

3. Describe how the sequence number and the acknowledgement number in the TCP header are used in the three-way hand-shaking procedure to establish a network session.

4. Describe how the sequence number and the acknowledgement number in the TCP header are used for reliable data delivery through acknowledgement and retransmission.

5. Describe why the time to live field in the IP header is needed for the proper rout- ing of data on the Internet.

43

4 Overview of Information Systems

4.1 Information System Concepts

An information system is an organized combination of people, hardware, software, com- munication networks, and data resources that collects and transforms data into meaning- ful information and knowledge to assist the decision-making process in an organization. Figure 4.1 is a schematic view of an information system structure that shows that peo- ple, technology, and organization are interconnected within a societal and business environment.

In an organization, information is one of the most valuable resources for decision mak- ing at different levels (e.g., operational decisions, tactical decisions, and strategic decisions). However, people often get confused between information and data. Taking a hospital as an example, raw data may include the patient’s name; the patient’s ID; the patient’s address; and the patient’s medical history, medical exam results, and treatments, just to name a few. When the raw facts are organized in a meaningful manner, they become information that adds additional value beyond the aggregated individual facts. For instance, the medical physicians may find the information about the patient’s history and medical exam results to be more critical for treatment planning. The financial executives may find the informa- tion about the types of medical exams serve more purposes for billing and reimburse- ment. Transforming the raw data into meaningful information for different purposes thus becomes a critical procedure that requires specific knowledge related to the task or deci- sion making (Ravishankar et al., 2011). As shown in Figure 4.2, from data to information to knowledge is an iterative process, where the focus lies in understanding the relationships among the data items to gather the information and discovering the underlying pattern in the information to obtain the knowledge.

During the transformation process, people sometimes organize or process data mentally or manually, while in most cases, a computer is utilized to automate such processes. No matter how the data are handled, the key is to ensure the resulting information is useful and valuable. Stair and Reynolds (2013) comprehensively review the value of informa- tion and summarize the basic characteristics that valuable information requires to support decisions (see Table 4.1).

The valuable information has been used in a massive number of applications across all levels of activities in a business setting. These will be reviewed in the next section.

44 Developing Windows-Based and Web-Enabled Information Systems

4.2 The Role of Information System in Business

In general, information systems used in an organization can be classified into three levels (Figure 4.3):

• Operational information system: It delivers information to the point of business where information is used as part of an operational process. For example, most retail stores now use a computer-based system for point-of-sale transactions. This operational information system can be used to help record customer purchases and keep track of product inventory levels.

Environment

Technology People

Organizations

Information systems

FIGURE 4.1 Information system components.

Data

Information

Knowledge

Understanding relations

Understanding patterns

FIGURE 4.2 Data, information, and knowledge.

45Overview of Information Systems

Strategic information

systems

Tactical information systems

Operational information systems

FIGURE 4.3 Information systems in business.

TABLE 4.1

Characteristics of Valuable Information

Characteristics Description

Accessible Information should be easily accessible to authorized users so that they can obtain it in the right format and at the right time to meet their needs.

Accurate Accurate information is error free. Errors could occur due to different reasons. If there are some problems in the knowledge required for the process, the information rendered may have errors. In other cases, inaccurate information is generated because the data contain errors. This is commonly known as garbage in, garbage out.

Complete Complete information contains all the important facts for decision making. For example, a patient report that does not include patient demographic information (e.g., age, gender) is not complete.

Economical Information should be relatively economical to produce. There is a trade-off between the value of information and the costs of producing it. For example, if collecting the data takes lots of resources and time, it is not economical.

Flexible Information should be flexible to be used for different purposes. For example, patient information can be used by physicians for treatment planning. Such information can also be used by the quality assurance committee for workflow improvements and the patient safety committee to manage patient safety events.

Relevant Relevant information is important to the decision maker. It is determined based on the usefulness of information with respect to the decision making process.

Reliable Reliable information refers to the correctness of the information. If the reliability of data collection is poor, it will directly impact the information produced. Thus, it is questionable to trust such information.

Secure The value of information could be lost due to issues such as unauthorized access or intentionally damaging its existence. Information should be secured from access by unauthorized users.

Simple Information should be simple, not complex. Sophisticated and detailed information might not be needed. In fact, too much information can cause information overload, whereby a decision maker has too much information and is unable to determine what is really important.

Timeliness Decisions should be made at the right time to achieve effectiveness. Verifiable Information should be verifiable. This means that the user can check it to make sure that it is

correct, perhaps by checking many sources for the same information.

Source: Stair, R., Reynolds, G., Fundamentals of Information Systems, Course Technology, 7th edn., 2013.

46 Developing Windows-Based and Web-Enabled Information Systems

• Tactical information system: It is the application to analyze business trends, fre- quently compare a specific metric (such as sales) to the same metric from a pre- vious month or year, and generate a report to support the decision making of employees and managers. For example, a retail store manager uses the tactical information system to make decisions such as which merchandise needs to be added or discontinued.

• Strategic (executive) information system: It supports strategic decisions for com- petitive advantages. For example, some retail stores (e.g., Kohl’s) decide to install touch-screen kiosks to help customers locate the merchandise. In a similar man- ner, some healthcare providers decide to install kiosks to increase efficiency and patient satisfaction in emergency departments, waiting rooms, and ambulatory settings. Patients can easily check in, perform payment transactions, and get help in locating the service unit through complicated facility layouts.

In a typical organization, these three systems often interact with each other. The infor- mation system that supports operational decisions may also provide data to, or may be accepting data from, tactical and strategic information systems (SIS), and vice versa.

Under these three categories, the business applications of information systems have expanded significantly over the years, from traditional transactional processing systems (TPS), management information systems (MIS), and decision support systems (DSS), to executive information system/strategic information system (EIS/SIS), to the more recently emerged mobile business systems (M-business). The general description on each milestone of the information evolution is provided in Table 4.2.

TABLE 4.2

Evolution of Information System Applications

Milestone Application System Description

1950s–1960s TPS TPS functions for information storage and retrieval. The system is used for transaction processing, record keeping, accounting, etc.

1960s–1970s MIS MIS can process the data into useful information reports to help the manager make decisions.

1970s–1980s DSS DSS can provide special supports to handle different management challenges for the manager. The report in DSS must be interactive and informative. Compared to MIS, DSS usually has a decision model (e.g., analytical models) component to realize the support.

1980s–1990s EIS/SIS EIS/SIS provides corporate executives with simple information gathered from multiple sources across the organization, which are opportune to the strategic goal of the company.

1990s–2000s E-business/E-commerce The rapid growth of Internet, Intranet, Extranet and other interconnected global networks in the 1990s dramatically changed the capabilities of information systems in business at the beginning of the 21st century. Internet-based and Web-enabled enterprise and global E-business and E-commerce systems are becoming common-place in the operations and management of today’s business enterprises.

2000s–present M-business Mobile commerce connects with mobile customers, builds insights through more powerful analytics, delivers more convenient and relevant engagements, and improves management and customer service with a seamless integration of front-end functionality and back-end data.

47Overview of Information Systems

4.2.1 Transaction Processing System

Since the 1950s, computers have been more commonly used for business applications. Many of these early systems were designed to reduce costs by automating routine, labor- intensive business transactions. A business transaction initiates the accounting cycle of the business activities. They are recorded to maintain accurate records to ensure account- ability, to establish historical business activity data, and to provide information to decision makers for determining business strategies. TPS is an information system that involves the collection, modification, and retrieval of all these transaction data.

The major functions of a TPS usually include paying employees and suppliers, controlling inventory, sending invoices, and ordering supplies. Today, a majority of business organiza- tions use enterprise resource planning (ERP) systems for these tasks that are, indeed, a set of integrated TPS. An ERP system facilitates the flow of information between the functional areas of a business and manages the important business operations for an entire multi-site, global organization. Using smartphones and mobile devices, a timely report can be generated from an ERP system these days. According to Forbes 2013 data (http://www.forbes.com), the worldwide ERP software market size is about $24.5 billion, and the top two vendors are SAP and Oracle, with the market shares being 25% and 13%, respectively.

4.2.2 Management Information System

MIS is a system that is used to design and implement procedures, processes, and rou- tines, providing detailed reports in an accurate, consistent, and timely manner. In an MIS, modern, computerized systems continuously collect relevant data which, in turn, are pro- cessed, integrated, and stored. Valuable information transformed from the data is made available to users who have the authority to access it, in the form that fits their needs. An MIS typically provides standard reports generated with data from the ERP/TPS and is used to improve operational efficiency (Munoz et al., 2011). For example, Dell used manu- facturing MIS software to develop a variety of reports in its manufacturing processes and to account for costs (http://www.dell.com). As a result, Dell was able to double its product variety while saving about $1 million annually in manufacturing costs.

4.2.3 Decision Support System

DSS is a collection of integrated software applications that form the backbone of an orga- nization’s decision-making process. It is a data-driven application that focuses on the col- lection and availability of data to analyze. Whereas an MIS helps an organization to “do things right,” a DSS helps a manager to “do the right things.” In general, a DSS consists of a collection of models used to support decision makers or users (models), a collection of facts and information to assist in decision making (database), and systems and procedures (user interface) that help decision makers and other users interact with the system. In this book, some basic DSS models will be discussed in Chapter 25.

4.2.4 Executive Information System and Strategic Information System

To some extent, the EIS and the SIS serve the same purpose for an organization. An EIS/ SIS is an infrastructure (a combination of software and hardware) that supplies an enter- prise’s executives up-to-the-minute operational data gathered and sifted from various data sources. The typical information mix presented to the executives may include financial information, work in process, inventory figures, sales figures, market trends, industry

48 Developing Windows-Based and Web-Enabled Information Systems

statistics and market price of the firm’s shares, etc. It differs from a DSS as it targets top- level executives and not managers.

4.2.5 Electronic Business and Electronic Commerce

Electronic business (E-business) and electronic commerce (E-commerce) involve any business transaction executed electronically between companies (business-to-business), companies and customers (business-to-customer), customers and other customers (customer- to-customer), a business and the public sector, and customers and the public sector. This term was first introduced by IBM in 1997, and now, the business model built around it has been widely adopted. E-business and E-commerce have offered tremendous oppor- tunities for businesses of all sizes to enter the global market with competitive advan- tages. With the security built into today’s browsers and with digital certificates now available for individuals and companies, much of the early concern about the security of business transaction on the Web has been addressed, and E-business/E-commerce is accelerating.

4.2.6 Mobile Commerce

Mobile commerce (M-commerce) is a newly emerging business model that uses mobile devices to browse, buy, and sell products and services. With the boom of smartphone and tablet ownership, M-commerce has followed the wave. In 2012, M-commerce sales in the United States hit $24.66 billion, which was an 81% increase from $13.63 billion in 2011. More impressively, looking at the 2012 holiday season alone, mobile visitors accounted for one-third of the holiday E-commerce traffic. BMW, the German car company, has invested about $100 million in developing mobile applications for its cars and other prod- ucts (Murphy, 2011). Today, M-commerce has exploded in popularity with advances in smartphones and tablets.

As various information applications used in business are reviewed, an emerging trend is observed: the development of advanced applications from PC-based applications to Web applications to mobile applications has fundamentally shifted the types of informa- tion technologies being used. This unveils the new information age: post-PC. Apparently, understanding the basic characteristic of this new information age offers the opportunity to develop and use information systems for the future.

4.3 Post-PC Information Age: Five Trends in the Future Information System Applications

There are five intertwined trends that have been observed in the new information age (Hinchcliffe, 2011): mobile, social media, big data, cloud computing, and consumerization of IT (Figure 4.4). Each is explained in the following section.

4.3.1 Mobile

In the post-PC era, one of the biggest trends being seen today is the move toward mobile devices. In most developed countries, the vast majority of adults have a mobile phone, and

49Overview of Information Systems

typically, people have their mobile phones within their reach 24/7, as compared to having an access to the laptop or PC. Another mobile device gaining popularity is the tablet. For organizations, this increase in mobility has a wide range of implications, from increased collaboration to the ability to manage a business in real-time to a change in the way that new (or existing) customers can be reached.

4.3.2 Social Media

Another notable trend is social media. We may be one of the over 1.15 billion (and still growing) Facebook users who share status updates or pictures with friends and family or we may be using a social network such as Google+ to stay tuned with the activities of our social “circles.” A recent survey of over 8000 faculty members done by Pearson Learning Solutions found that 41% of college professors use social media as a teaching tool, up from around 34% in 2012. Similarly, companies are using social media to get people (employees, customers) to participate in innovation and other activities.

4.3.3 Big Data

The transformation of social and work interactions enabled by this 24/7 connectivity has given rise to a third trend—big data. A study by the research firm International Data Corporation (IDC) estimates that, in 2011, 1.8 zettabytes of data were generated and con- sumed. The number is forecast to be 50 times more by the end of the decade. For many orga- nizations in this new information age, value and business opportunities are created from the big data. In the “old economy,” the largest/most valuable organizations (such as GE or Ford) have 100,000 to 200,000 employees and the largest organizations in the “new economy” (such as Microsoft, HP, or Oracle) have 50,000 to 100,000 employees, whereas in the “new infor- mation age economy,” thriving companies such as Facebook and Twitter have a mere 5000 to 20,000 employees (Hofmann, 2011). However, it should be noted that the big data poses tremendous challenges in terms of data storage, data analysis, and transforming the data to useful information and knowledge, which leads to the fourth trend—cloud computing.

Mobile

Social media

Big data

Cloud computing

Consumerization of IT

Changes in organization and

society

FIGURE 4.4 Five trends of the new information age.

50 Developing Windows-Based and Web-Enabled Information Systems

4.3.4 Cloud Computing

Whereas, traditionally, each user would install applications for various tasks—from creat- ing, storing, and managing documents on the computer—much of the functionality previ- ously offered by the applications installed on each individual computer is now offered by the applications “in the cloud,” accessed via a Web browser (e.g., Internet Explorer, Google Chrome, and Safari). One example is the various services offered by Google, such as Gmail (e-mail), Google Docs (word processing), or Google Calendar, all of which are accessed via a Web browser, freeing users from the task of installing or updating traditional desk- top applications. Today, cloud computing is becoming the mainstream technology tool for business. The report from Parallels, a software company (http://www.parallels.com/), projects that the cloud service business is expected to grow annually at 28% to be a $95 billion industry.

4.3.5 Consumerization of IT

Lastly, the most significant trend according to the research firm Gartner is the consumer- ization of IT. Driven by the advances in consumer-oriented mobile devices and the ability to access data and applications in the cloud, today’s employees are increasingly using their own devices for work-related purposes and/or are using software that they are used to (e.g., social media) in the workplace. Although this trend may raise concerns related to security or compliance or may increase the need to support the employees’ own devices, it may also provide opportunities, such as increased productivity, higher retention rates of talented employees, or higher customer satisfaction (TrendMicro, 2011).

4.4 Summary

This chapter gives an overview of the information system concept, including data, infor- mation, and knowledge. The basic characteristics of valuable information are reviewed. Various types of information systems using valuable information are discussed. Observing the emergence of the new information age, five trends for future information system applica- tions are introduced. We note that the core of any information system is data. Throughout the remainder of this book, the focus is on the design and implementation of information system applications to store, retrieve and manage data (database), and use the data for dif- ferent end users in various forms (VB.NET and VBA for Windows applications, ASP.NET for Web applications) to support decisions.

EXERCISES

1. An information system is an organized combination of the following except: a. Environment b. Data c. Technology d. Organizations e. All of the above are valid components of information systems

51Overview of Information Systems

2. Information, data, and knowledge mean the same thing. True or false? 3. Collecting data and analyzing its patterns produces knowledge. True or false? 4. An information system should be flexible enough and economical during imple-

mentation. True or false? 5. Which of the following statements are true? a. A tactical information system delivers information from point-of-sale transac-

tions to provide tactical reports for middle management. b. An SIS supports top management, which affects a wide range of business

departments, such as finance, inventory management, and warehousing. c. An operational information system compares a performance metric trend for

a couple of months to develop operational advantages over competitors. d. All of the above are true. e. None of the above is true. 6. M-business is preceded by E-business in the evolution of information systems.

True or false? 7. A set of integrated transaction processing systems is called “material require-

ments planning” or MRP. True or false? 8. What are the two typical sources of information for MIS? 9. What is the main difference between a SIS and a DSS? 10. E-business is a business model that offers business to a market and sells their

products online through a mobile device. 11. The following trend examples are observed during the post-PC information age

except: a. Facebook b. Smartphones c. Big data algorithms d. Gmail e. All of the above are valid examples 12. The purpose of big data is to analyze huge amounts of data for the purpose of

improving business value proposition to customers. True or false? 13. An information system hosted entirely online is an example of an application of

cloud computing. True or false? 14. An application that lets you navigate the streets of New York on a smartphone is

an example of consumerization of IT. True or false?

Section II

Database Design and Development

A database is a major component of information systems. A database contains an organized col- lection of related data. The relational data model is widely used to organize data in a database. There are many database software packages, such as Microsoft Access, Microsoft Structured Query Language (SQL) Server, MySQL, and Oracle, which implement database management systems (DBMSs) to support the definition, creation, access, and management of a relational database by end users. This book introduces the use of Microsoft Access and MySQL for data- base development. The development of a relational database includes four phases:

• Enterprise modeling: Specifies data needed for a database and data queries for using data in an enterprise environment

• Conceptual data modeling: Uses the entity–relationship (E-R) modeling to define the conceptual organization of related data, which can easily be understood by end users in the enterprise environment without reference to the specific rela- tional data structure

• Logical database design: Uses the relational modeling to transform the ER model of a database into the relational database design, which can be directly imple- mented using a relational DBMS

• Physical database implementation: Implements the relational data model to create a relational database with data storage and query capacities

Section II of this book covers the entire process of developing a database. Enterprise modeling is illustrated using a case study for a health care information system. Chapter 5 describes the E-R modeling for conceptual data modeling. Chapter 6 presents relational modeling along with database normalization for logical database design. Chapter 7 shows the use of Microsoft Access to create a relational database and data queries. Chapter 8 introduces the SQL, which is a standard database language supported by many different DBMS software packages. Because SQL is interoperable among different DBMS, it allows us to define, create, access, and manage a new database, or transport an existing database, using even a DBMS that we are not familiar with. Chapter 9 shows the use of MySQL to create a relational database and data queries. Chapter 10 introduces the concepts of object- oriented database.

55

5 Conceptual Data Modeling: Entity-Relationship Modeling

An E-R model organizes data using entities, relationships, and their attributes. Such an organi- zation of data is natural and easy to understand but is not the way in which data are stored on computers. An E-R model will need to be transformed into a relational model, which organizes data using the data structure called “relation,” which is similar to a table with columns and rows. Because concepts for a relational model involve the implementation of data storage on computers and may not be natural to our understanding, we first use an E-R model to describe a conceptual organization of data and then transform the E-R model into a relational model.

This chapter describes the elements and concepts of E-R modeling, including types, attributes, and instances of entities; types, attributes, instances, and degrees of relation- ships between entities; maximum and minimum cardinalities of relationships; associative entities and weak entities; and superclass and subclass entities. A case study of a database for a healthcare information system is used to illustrate the development of an E-R model. Chapter 26 gives the requirements for the healthcare database in Section 26.1.

5.1 Types, Attributes, and Instances of Entities

An entity can represent a person, a place, an object, an event, a concept, or others. The fol- lowing are some examples of entities that may be included in an E-R model:

• Person: student, employee, client • Object: restaurant, movie theater, machine • Place: city, park, warehouse, room • Event: marriage, lease, war • Concept: project, account, course

An entity is represented by a rectangular box in an E-R diagram, which is used to define an E-R model. An attribute of an entity describes a property or characteristic of the entity and holds data about the entity. An attribute is represented by an ellipse in an E-R model. An entity may have one or more attributes.

Example 5.1

Use an E-R diagram to represent the Patient and the InsurancePlan entities in the healthcare database. The Patient entity has the attributes PatientID, SSN, PatientName, Address, Phone, Gender, DateOfBirth, Employer, Position, E-Name, and E-Phone. The InsurancePlan entity has the attributes PlanID, PlanName, Description, MemberPremium, and EmployerPremium.

56 Developing Windows-Based and Web-Enabled Information Systems

Figure 5.1 gives the E-R diagram containing the Patient and the InsurancePlan entities, and their attributes.

The set of attributes for an entity can also be represented in the text format. For example, the InsurancePlan entity and its attributes in Example 5.1 can be expressed as follows:

InsurancePlan = {PlanID, PlanName, Description, MemberPremium, and EmployerPremium}

In an E-R model, an entity box represents a type or collection of entities that have the same attributes. For example, in the E-R model of the healthcare database, the Patient entity represents a collection of all patients in the database. Each individual patient is an instance of the entity type Patient. All individual patients have the same attributes. For example, the InsurancePlan entity in the E-R model of the healthcare database represents a collection of all insurance plans in the healthcare database. A particular insurance plan is an instance of the entity type InsurancePlan.

Example 5.2

Give two instances of the Patient and the InsurancePlan entities described in Example 5.1. Table 5.1 gives two instances of the Patient entity. Table 5.2 gives two instances of the

InsurancePlan entity.

Hence, an entity and its attributes in an E-R model of a database describe the data orga- nization for the entity type (i.e., what type of attributes for what entity type are kept in the database), while instances of the entity type and attribute values for instances are the data

InsurancePlan

PlanID

PlanName

Description

MemberPremium

EmployerPremium

Patient

SSN

PatientName

Address

Phone

Gender

DateOfBirth

Employer

Position

E-Name

E-Phone

PatientID

FIGURE 5.1 An E-R model containing the Patient entity, the InsurancePlan entity, and their attributes in the healthcare database.

57Conceptual Data Modeling

stored in the database. When the value of an attribute for an instance is not known or not available, a NULL value can be used for the attribute of the instance. For example, a particular patient (e.g., PA2), which is an instance of the Patient entity in Example 5.2, may not have an employer. This instance of the Patient entity can have NULL for the Employer attribute.

Each entity type should have the key attribute(s) whose value is unique for each instance of the entity and, thus, can be used to uniquely identify each instance. A key attribute is underlined in an E-R model. For example, PatientID is the key attribute of the Patient entity and is underlined in the E-R model shown in Figure 5.1. Each patient has a unique value of PatientID so that each patient can be uniquely identified by the value of PatientID. Consider the City entity in Figure 5.2 that has the three attributes: CityName, StateName, and Population. Because two different cities may have the same city name (e.g., Miami in Arizona and Miami in Florida), using CityName alone cannot uniquely identity each of the two cities that have the same city name. Hence, both CityName and StateName can be used together as the key attributes to uniquely identify each instance of the City entity. The key attribute of an entity cannot have the NULL value because the NULL value of the key attribute cannot identify a particular instance of the entity.

There are various types of attributes: simple versus composite, single-valued versus multi-valued, and stored versus derived. Because different types of attributes are imple- mented in a database on computers differently, each attribute type must be identified in an E-R model.

A simple or atomic attribute does not have any other attributes as components. A composite attribute has some other attributes as components. For example, we can have EmergencyContact, which contains E-Name and E-Phone, as a composite attribute of the Patient entity in the E-R model of the healthcare database, as shown in Figure 5.3.

City

CityName StateName Population

FIGURE 5.2 An E-R model containing the City entity and its attributes.

TABLE 5.1

Two Instances of the Patient Entity

PatientID SSN Patient Name Address Phone Gender DateOfBirth Employer Position E-Name E-Phone

PA1 111-11-1111 John Smith

321 Mill Avenue

480-965-1111 M 1/1/1983 ASU Staff Helen Smith

480-965-1112

PA2 222-22-2222 Mary Davis

432 Mill Avenue

480-965-2222 F 2/2/1995 NULL Student Jack Davis

480-965-2223

TABLE 5.2

Two Instances of the InsurancePlan Entity

PlanID PlanName Description MemberPremium EmployerPremium

PL1 Care HMO A HMO plan $100 $800 PL2 Care PPO A PPO plan $300 $600

58 Developing Windows-Based and Web-Enabled Information Systems

Pa tie

nt

SS N

Pa tie

nt N am

e

A dd

re ss

Ph on

e

G en

de r

D at eO

fB ir th

Em pl oy er

Po si tio

n

Em er ge nc

yC on

ta ct

E- N am

e

E- Ph

on e

Pa tie

nt ID

In su ra nc eC

om pa ny

In su ra nc eP la n

En ro ll

O  |

O ffe

r >  |

| |

M em

be rI D

C om

pa ny

N am

e

A dd

re ss

Ph on

e

C ar eP

ro vi de r

M ed ic in e

V isi t

D at eO

fV isi t

R ec ov er yT

im e

Sy m pt om

s

D ise

as e

M ed ic in eN

am e

D es cr ip tio

n

Pr ov id er ID

Pr ov id er N am

e

A dd

re ss

Ph on

e

G en de r

Sp ec ia lty

A ffi lia te dH

os pi ta l

Ed uc

at io n

D eg re e

Y ea r

La ng

ua ge s

Sc he du

le St ar tT im

e

Pa tie

nt N am

eC om

pa ny

ID Pl an

ID

Pl an

N am

e

D es cr ip tio

n

M em

be rP re m iu m

Em pl oy er Pr em

iu m

En ro llm

en tD

at e

G ro up

ID D ep en

dO n

C la im

Pr oc es s

C la im

ID

Se rv ic eN

am e

Pa tie

nt Pa

ym en t

Se rv ic e

In su ra nc eP

ay m en

t

Ba la nc

e

Ph ar m ac y

Pr es cr ip tio

n

Fi ll

K ee p

Ph ar m ac yI D

Ph ar m ac yN

am e

A dd

re ss

M ed ic in eI D

U ni tP ri ce

C ur re nt In ve nt or y

M in im

um In ve nt or yC on

ta in

Q ua nt ity

R efi

llF re qu

en cy

La st Pi ck up

D at e

Pr es cr ip tio

nI D

Ph on

e

Se rv ic eC

od e

In cl ud

e

T re at m en

t

G en

er at e

O rd er

T re at m en

tI D

A ge

>  |

Pa tie

nt ID

Pr ov id er ID

C la im

ID

FI G

U R

E 5 .3

T h

e E

-R m

o d

el o

f th

e h

ea lt

h ca

re d

at ab

as e.

59Conceptual Data Modeling

That is, the EmergencyContact attribute consists of the E-Name and E-Contact attributes. In an E-R model, a line is used to connect a composite attribute to its component attri- bute. All the other attributes of the Patient entity are the simple attributes. In the E-R model of the healthcare database (see Figure 5.3), the following attributes are the com- posite attributes: the EmergencyContact attribute of the Patient entity, the Education and Schedule attributes of  the CareProvider entity, and the Service attributes of the Claim entity. All the other attributes in the E-R model of the healthcare database are the simple attributes.

A single-valued attribute of an entity has only one value for an instance of the entity. A multi-valued attribute of an entity may have more than one value for an instance of the entity. For example, in the E-R model of the healthcare database, the Language attribute of the CareProvider entity is a multi-valued attribute because an instance of the CareProvider entity, for example, a physician, may speak both English and Spanish and, thus, have two values for the Language attribute. A multi-valued attribute is denoted by a double-lined ellipse (see Figure 5.3). In the E-R model of the healthcare database, the Language and Schedule attributes of the CareProvider entity and the Service attribute of the Claim entity are the multi-valued attributes, and all the other attributes in the E-R model are the single- valued attributes.

The value of a derived attribute can be derived from values of some stored attribute(s) and, thus, does not need to be stored in a database. For example, in the E-R model of the healthcare database, we may add and consider the Age attribute of the Patient entity as a derived attribute because the value of Age can be computed from the DateOfBirth attri- bute of the Patient entity. A derived attribute is denoted by a dash-lined ellipse (see Figure 5.3). The value of a stored attribute cannot be derived from the values of some other attri- butes and must be stored in a database.

In summary, there are three pairs of different attribute types: simple versus compos- ite, single-valued versus multi-valued, and stored versus derived. An entity attribute falls into one of the two types in each pair. For example, in the E-R model of the healthcare database in Figure 5.3, the Schedule attribute of the CareProvider entity is a composite, multi-valued, stored attribute, and the Gender attribute of the Patient entity is a simple, single-valued, stored attribute.

5.2 Types, Attributes, Instances, and Degrees of Relationships between Entities

A relationship represents an association between entities. For example, in the E-R model of  the healthcare database (see Figure 5.3), there is the Enroll relationship between the Patient and the InsurancePlan entities because a patient is enrolled in an insurance plan and the Enroll relationship links a patient to an insurance plan. A relationship is repre- sented by a diamond-shaped box with lines connecting entities involved in the relation- ship (see Figure 5.3). A relationship may have no attributes or may have a set of attributes. Unlike an entity type, a relationship type does not need a key attribute. An attribute of a relationship stores information about the relationship. For example, the Enroll relation- ship between the Patient and the InsurancePlan entities has the attributes EnrollmentDate, GroupID, and MemberID. Each of the three attributes is not an attribute of either the Patient or the InsurancePlan entity because the value of EnrollmentDate, GroupID, and MemberID

60 Developing Windows-Based and Web-Enabled Information Systems

depends on both a specific instance of the Patient entity and a specific instance of the InsurancePlan entity involved in the Enroll relationship. For example, MemberID cannot be an attribute of the Patient entity because the MemberID of a given patient depends on which insurance plan this patient is enrolled in. MemberID cannot be an attribute of the InsurancePlan entity because an insurance plan has many members and a particular MemberID is linked to a particular patient.

A relationship in an E-R model represents a relationship type or a collection of relation- ship instances. An instance of a relationship type involves an instance of each entity type involved in the relationship. To identify an instance of a relationship, we should identify the instance of each entity type involved in the instance of the relationship, which can be done by using the value of the key attribute(s) of each entity type. For example, when a patient is enrolled in an insurance plan, an instance of the Enroll relationship is created. This instance of the Enroll relationship can be identified by the value of PatientID and the value of PlanID involved in the instance of the relationship.

Example 5.3

Give two instances of the Enroll relationship. Table 5.3 gives two instances of the Enroll relationship.

The number of entity types involved in a relationship is called the “degree” of a rela- tionship. In the E-R model of the healthcare database (see Figure 5.3), the DependOn rela- tionship represents the relationship between a patient and his/her dependents. Because the DependOn relationship involves only the Patient entity, the DependOn relationship is a degree-1 or unary relationship. The Enroll relationship between the Patient and the InsurancePlan entities is a degree-2 or binary relationship. The Visit relationship associ- ates the Patient, the CareProvider, and the Claim entities, and is a degree-3 or ternary relationship, which is converted into an associative entity, as shown in Figure 5.3 (see the description of associative entity in Section 5.4).

5.3 Maximum and Minimum Cardinalities of a Relationship

Given entity A, entity B, and their relationship R in Figure 5.4, the maximum cardinality of relationship R describes the maximum number of entity B’s instances that can be asso- ciated with any instance of entity A (see Figure 5.4a and b) and the maximum number of entity A’s instances that can be associated with any instance of entity B (see Figure 5.4c and d). There are only two values for the maximum cardinality: one or many (more than one).

TABLE 5.3

Two Instances of the Enroll Relationship

PatientID PlanID EnrollmentDate GroupID MemberID

PA1 PL1 1/1/2013 10000 1234567 PA2 PL1 1/1/2013 10000 2345678

61Conceptual Data Modeling

Figure 5.5 gives examples of one-to-one, one-to-many, many-to-one, and many-to-many relationships for the maximum cardinality. Suppose that all instances of entity A, entity B, and relationship R are shown in each example in Figure 5.5. In Figure 5.5a, for the one- to-one relationship between entities A and B, any instance of entity A is associated with, at maximum, one instance of entity B (a1 with b1, a2 with b3, a3 with b5, a5 with b4, and no association of a4 and a6 with any instance of entity B), and any instance of entity B is associated with, at maximum, one instance of entity A (b1 with a1, b3 with a2, b4 with a5, b5 with a3, and no association of b2 with any instance of entity A). In Figure 5.5b, for the one-to-many relationship, there is an instance of entity A that is associated with more than one instance of entity B (e.g., a2 with b2 and b3), and any instance of entity B is associated with, at maximum, one instance of entity A (b1 with a1, b2 with a2, b3 with a2, b4 with a3, and b5 with a6). In Figure 5.5c, for the many-to-one relationship, any instance of entity A is associated with, at maximum, one instance of entity B (a1 with b1, a2 with b3, a3 with b3, a4 with b1, a5 with b5, and a6 with b1), and there is an instance of entity B that is associated with more than one instance of entity A (e.g., b1 with a1, a4 and a6). In Figure 5.5d, for the many-to-many relationship, there is an instance of entity A that is associated with more than one instance of entity B (e.g., a1 with b1 and b2), and there is an instance of entity B that is associated with more than one instance of entity A (e.g., b3 with a2 and a3).

The minimum cardinality of relationship R between entities A and B describes the minimum number of entity B’s instances that can be associated with any instance of entity A (see Figure 5.6a and b) and the minimum number of entity A’s instances that can be associated with any instance of entity B (see Figure 5.6c and d). There are only two values for the minimum cardinality: mandatory or optional. The mandatory value means that the minimum number of instances is at least 1. The optional value means that the minimum number of instances is 0. Figure 5.6a shows that the association of any instance of entity A with an instance of entity B is optional, that is, there is an instance of entity A that is not associated with any instance of entity B. Figure 5.6b shows that the association of any instance of entity A with an instance of entity B is mandatory, that is, any instance of entity A is associated with at least one instance of entity B. Figure 5.6c shows that the

A BR

(a)

(b) (d)

(c)

|

A BR <

A BR|

A BR>

FIGURE 5.4 Representation of the maximum cardinality of a relationship. (a) Given any instance of entity A, the maxi- mum number of entity B’s instances that can be associated with the instance of entity A is one (denoted by |). (b) Given any instance of entity A, the maximum number of entity B’s instances that can be associated with the instance of entity A is many (denoted by <). (c) Given any instance of entity B, the maximum number of entity A’s instances that can be associated with the instance of entity B is one (denoted by |). (d) Given any instance of entity B, the maximum number of entity A’s instances that can be associated with the instance of entity B is many (denoted by <).

62 Developing Windows-Based and Web-Enabled Information Systems

association of any instance of entity B with an instance of entity A is optional, that is, there is an instance of entity B that is not associated with any instance of entity A. Figure 5.6d shows that the association of any instance of entity B with an instance of entity A is mandatory, that is, any instance of entity B is associated with at least one instance of entity A.

A BR| O O |

(a) a1

a2

a3

a4

a5

a6

b1

b2

b3

b4

b5

A BR| | O <

(b) a1

a2

a3

a4

a5

a6

b1

b2

b3

b4

b5

A BR> O | |

(c) a1

a2

a3

a4

a5

a6

b1

b2

b3

b4

b5

A BR> | | <

(d) a1

a2

a3

a4

a5

a6

b1

b2

b3

b4

b5

FIGURE 5.5 Examples of one-to-one, one-to-many, many-to-one, and many-to-many relationships for the maximum cardinality, and optional-to-optional, mandatory-to-optional, optional-to-mandatory, and mandatory-to-mandatory rela- tionships for the minimum cardinality. (a) One-to-one and optional-to-optional. (b) One-to-many and mandatory- to-optional. (c) Many-to-one and optional-to-mandatory. (d) Many-to-many and mandatory-to-mandatory.

63Conceptual Data Modeling

Figure 5.5 shows examples of optional-to-optional, mandatory-to-optional, optional-to- mandatory, and mandatory-to-mandatory relationships for the minimum cardinality. As shown in Figure 5.5, the pair of symbols for the minimum cardinality are placed close to the diamond-shaped relationship, and the pair of symbols for the maximum cardinality are placed close to the entity boxes. In Figure 5.5a, for the optional-to-optional relation- ship, there is an instance of entity A (e.g., a4) that is not associated with any instance of entity B, and there is an instance of entity B (e.g., b2) that is not associated with any instance of entity A. In Figure 5.5b, for the mandatory-to-optional relationship, every instance of entity B is associated with an instance of entity A, indicating that the asso- ciation of entity B with entity A is mandatory, and there is an instance of entity A (e.g., a4) that is not associated with any instance of entity B. In Figure 5.5c, for the optional- to-mandatory relationship, every instance of entity A is associated with an instance of entity B, indicating that the association of entity A with entity B is mandatory, and there is an instance of entity B (e.g., b4) that is not associated with any instance of entity A. In Figure 5.5d, for the mandatory- to-mandatory relationship, every instance of entity A is associated with an instance of entity B, and every instance of entity B is associated with an instance of entity A, indicating that the association between entities A and B is mandatory.

The maximum and minimum cardinalities of a relationship determine how instances of a relationship are stored in a database on computers (see Chapter 6 for detailed discus- sions). Hence, the maximum and minimum cardinalities of a relationship must be identi- fied in an E-R model.

Example 5.4

Figure 5.7 gives five instances of the Enroll relationship between the Patient and the InsurancePlan entities. Based on these instances, determine the maximum and mini- mum cardinalities of the Enroll relationship.

For the maximum cardinality, an insurance plan may have many patients enrolled, for example, IP1 has PA1, PA2, and PA3 enrolled, and a patient is enrolled in one insur- ance at maximum. Hence, the maximum cardinality of the Enroll relationship between the Patient and the InsurancePlan entities is many-to-one, as shown in Figure 5.3. For

A BR O

(a)

A BR |

(b)

A BRO

(c)

A BR|

(d)

FIGURE 5.6 Representation of the minimum cardinality of a relationship. (a) Given any instance of entity A, the minimum number of entity B’s instances that can be associated with the instance of entity A is 0 (denoted by a circle). (b) Given any instance of entity A, the minimum number of entity B’s instances that can be associated with the instance of entity A is at least one (denoted by |). (c) Given any instance of entity B, the minimum number of entity A’s instances that can be associated with the instance of entity B is 0 (denoted by a circle). (d) Given any instance of entity B, the minimum number of entity A’s instances that can be associated with the instance of entity B is at least one (denoted by |).

64 Developing Windows-Based and Web-Enabled Information Systems

the minimum cardinality, each insurance plan has at least one patient enrolled, and a patient (e.g., PA4) may not enroll in any insurance plan. Hence, the minimum cardinal- ity of the Enroll relationship between the Patient and the InsurancePlan entities is man- datory to optional, as shown in Figure 5.3.

Example 5.5

The DependOn relationship for the Patient entity in the E-R model of the healthcare database is a unary relationship representing which patient is a dependent of another patient. Figure 5.8 gives some instances of the DependOn relationship for the Patient entity. In Figure 5.8, we use a copy of the Patient entity and its instances (PA1, PA2, PA3, PA4, PA5, and PA6) to represent the unary relationship as a binary relationship so that we can clearly see which patient depends on which patient. Figure 5.8 shows that PA1 and PA2 depend on PA3, PA4 depends on PA5, and PA6 has no dependent and does not depend on anyone. Based on these instances, determine the maximum and minimum cardinalities of the DependOn relationship.

For the maximum cardinality, one patient may have many dependents (e.g., PA3 has PA1 and PA2 as dependents), and one patient depends on only one other patient. Hence, the DependOn relationship is a many-to-one relationship, as shown in Figure 5.9. For the minimum cardinality, there is a patient (e.g., PA6) who has no dependent and does not depend on another patient. Hence, the DependOn relationship is an optional-to-optional

Patient

PA1

PA2

PA3

PA4

PA5

PA6

IP1

IP2

IP3

InsurancePlanEnroll

FIGURE 5.7 Examples of instances for the Enroll relationship between the Patient and the InsurancePlan entities.

Patient

PA1

PA2

PA3

PA4

PA5

PA6

DependOn Patient

PA1

PA2

PA3

PA4

PA5

PA6

FIGURE 5.8 Examples of instances for the DependOn relationship for the Patient entity.

65Conceptual Data Modeling

relationship, as shown in Figure 5.9. Flipping the copy of the Patient entity back to the Patient entity produces the representation of the maximum and minimum cardinalities for the DependOn relationship in Figure 5.3.

We cannot show the maximum and minimum cardinalities on a relationship of degree-3 or higher, such as the Visit relationship in Figure 5.3 and the degree-3 relationship shown in Figure 5.10. For example, if we place the Many value (<) for the maximum cardinality next to entity A in Figure 5.10, it is not clear whether the Many value is for the relationship with entity B or with entity C. Furthermore, the degree-3 relationship of entities A, B, and C is not equivalent to the binary relationship of entities A and B, the binary relationship of entities A and C, and the binary relationship of entities B and C all together. Before we can mark the maximum and minimum cardinalities of a relationship of degree-3 or more, we need to first convert a relationship of degree-3 or higher to an associative entity, which is described in Section 5.4.

Example 5.6

Determine the maximum and minimum cardinalities for each relationship in the E-R model of the healthcare database.

Figure 5.3 shows the maximum and minimum cardinalities of each relationship in the E-R model of the healthcare database. The DependOn relationship of the Patient entity is a many-to-one and optional-to-optional relationship, as described in Example 5.5. As described in Example 5.4, the Enroll relationship of the Patient and the InsurancePlan entities is a many-to-one and a mandatory-to-optional relationship because an insur- ance plan must have one patient as a member and may have many patients as members. We assume that a patient has, at maximum, one insurance plan and may not have any insurance plan. The Offer relationship of the InsurancePlan and the InsuranceCompany entities is a many-to-one and a mandatory-to-mandatory relationship because an insur- ance plan must be offered by only one insurance company, and an insurance company must offer an insurance plan and may offer many insurance plans. The Include rela- tionship of the InsurancePlan and the CareProvider entities is a many-to-many and a mandatory-to-mandatory relationship because an insurance plan includes at least one healthcare provider and may include many healthcare providers, and a healthcare

Patient PatientDependOn> O |O |

FIGURE 5.9 Maximum and minimum cardinalities of the DependOn relationship for the Patient entity.

A BR>

C

FIGURE 5.10 Example of a degree-3 relationship.

66 Developing Windows-Based and Web-Enabled Information Systems

provider is included in at least one insurance plan and may be included in many insur- ance plans. The Visit relationship of the Patient, the Claim, and the CareProvider enti- ties is converted into an associative entity, which is described in Section 5.4. The Process relationship between the InsuranceCompany and the Claim entities is a one-to-many and a mandatory-to-mandatory relationship because an insurance company must pro- cess at least one claim and may process many claims, and a claim must be processed by one insurance company. The Generate relationship of the Visit associative entity and the Treatment entity is a one-to-many and a mandatory-to-optional relationship because a visit to a healthcare provider may generate many treatments due to many diseases diagnosed or may generate no treatment if the visit is for a physical exam, and a treat- ment must be generated by one visit. The Order relationship of the Treatment and the Prescription entities is a one-to-one and a mandatory-to-optional relationship because a treatment may or may not produce a prescription order, and a prescription must be ordered for a treatment. The Contain relationship of the Prescription and the Medicine entities is a many-to-many and an optional-to- mandatory relationship because a pre- scription must contain at least one medicine and may contain many medicines, and a medicine may be contained in many prescriptions or may not be contained in any prescription. The Fill relationship between the Prescription and the Pharmacy entities is a many-to-one and a mandatory-to-mandatory relationship because a pharmacy store fills at least one prescription and may fill many prescriptions, and a prescription must be filled by only one pharmacy store. The Keep relationship between the Pharmacy and the Medicine entities is a many-to-many and an optional-to-mandatory relation- ship because the pharmacy store keeps at least one medicine and may keep many medi- cines, and a medicine may be kept by many pharmacy stores but may not be kept by any pharmacy store.

5.4 Associative Entities

An associative entity is created by converting a many-to-many binary relationship or a relationship of degree-3 or higher into an entity. No other type of relationship than a many-to-many binary relationship and a relationship of degree-3 or higher can be con- verted into an associative entity. An associative entity is denoted by adding a rectangular box to the diamond representation of the relationship.

The associative entity keeps the attributes of the relationship as the attributes of the associative entity. The key attributes of entities involved in the relationship are used as the composite key attributes for the associative entity. Hence, any instance of an associa- tive entity must have a value for each of the composite key attribute to uniquely identify the instance of the associative entity. This also means that any instance of an associative entity must involve an instance of each entity involved in the relationship. Considering that a patient’s visit to a healthcare provider for an annual physical exam does not gener- ate a treatment, the Treatment entity in the E-R model of the healthcare database is not included in the Visit relationship, although a visit may generate a treatment (e.g., a visit for treating a disease). If the Treatment entity is included in the Visit relationship and the Visit associative entity has TreatmentID as one of the composite key attributes, a visit gen- erating no treatment will have no value for TreatmentID, which is not allowed. Hence, the Treatment entity is not included in the Visit relationship because a visit may not involve a treatment.

67Conceptual Data Modeling

In general, it is not necessary to convert a many-to-many binary relationship or a rela- tionship of degree-3 or higher to an associative entity because an E-R model with or without the conversion is transformed into the same relational model. However, we have to convert the Visit relationship in the E-R model of the healthcare database into the Visit associative entity so that we can establish a relationship between the Visit associative entity and the Treatment entity.

Example 5.7

Figure 5.11 shows the Visit relationship of the Patient, the CareProvider, and the Claim entities in the E-R model of the healthcare database. Convert the Visit relationship into an associative entity.

Figure 5.3 shows the associative entity Visit. This Visit associative entity keeps the attribute of the Visit Relationship: DateOfVisit. The key attributes of the entities involved in the Visit relationship, PatientID, ProviderID, and ClaimID, are also added as the composite key attributes of the Visit associative entity.

A new key attribute can be created and used for an associative entity without using the key attributes of the entities involved in the relationship as the composite key attributes. For example, instead of using PatientID, ProviderID, and ClaimID as the composite key attri- butes of the Visit associative entity in Example 5.7, we can create a new attribute, VisitID, and use it as the key attribute of the Visit associative entity, as shown in Figure 5.12.

After converting a relationship into an associative entity, we can describe the maximum and minimum cardinalities between the associative entity and each of the other entities.

Example 5.8

For the Visit associative entity with the composite key attributes, specify the maximum and minimum cardinalities between the Visit associative entity and the Patient entity, between the Visit associative entity and the CareProvider entity, and between the Visit associative entity and the Claim entity.

For the maximum and minimum cardinalities of the Visit associative entity with the Patient entity, each instance of the Visit associative entity must involve one instance of the Patient entity, producing the One value for the maximum cardinality and the Mandatory value for the minimum cardinality on the side of the Patient entity. Because a patient may have many visits or no visit, we have the Many value for the maximum cardinality and the Optional value for the minimum cardinality on the side of the Visit associative entity. For the maximum and minimum cardinalities of the Visit associative entity with the CareProvider entity, each instance of the Visit associative entity must involve one instance of the CareProvider entity, producing the One value for the maxi- mum cardinality and the Mandatory value for the minimum cardinality on the side of the CareProvider entity. Because a healthcare provider may be involved in many visits or no  visit, we have the Many value for the maximum cardinality and the Optional value for the minimum cardinality on the side of the Visit associative entity. For the maximum and minimum cardinalities of the Visit associative entity with the Claim entity, each instance of the Visit associative entity must involve one instance of the Claim entity, producing the One value for the maximum cardinality and the Mandatory value for the minimum cardinality on the side of the Claim entity. Because a claim must be for one visit, we have the One value for the maximum cardinality and the mandatory value for the minimum cardinality on the side of the Visit associative entity. Figure 5.13 shows all the values of the maximum and minimum cardinalities.

68 Developing Windows-Based and Web-Enabled Information Systems

Pa tie

nt

SS N

Pa tie

nt N am

e

A dd

re ss

Ph on

e

G en de r

D at eO

fB ir th

Em pl oy er

Po si tio

n

Em er ge nc

yC on

ta ct

E- N am

e

E- Ph

on e

Pa tie

nt ID

C ar eP

ro vi de r

V isi t

D at eO

fV isi t

Pr ov id er ID

Pr ov id er N am

e

A dd

re ss

Ph on

e

G en

de r

Sp ec ia lty

A ffi lia te dH

os pi ta l

Ed uc at io n

D eg re e

Y ea r

La ng

ua ge s

Sc he du

le St ar tT im

e

Pa tie

nt N am

e

C la im

C la im

ID

Se rv ic eN

am e

Pa tie

nt Pa

ym en

t

Se rv ic e

In su ra nc

eP ay m en

t

B al an

ce

Se rv ic eC

od e

A ge

FI G

U R

E 5 .1

1 E

-R m

o d

el c

o n

ta in

in g

t h

e V

is it

r el

at io

n sh

ip o

f th

e P

at ie

n t,

t h

e C

ar eP

ro v

id er

, a n

d t

h e

C la

im e

n ti

ti es

i n

t h

e h

ea lt

h ca

re d

at ab

as e.

69Conceptual Data Modeling

Pa tie

nt

SS N

Pa tie

nt N

am e

A dd

re ss

Ph on

e

G en

de r

D at

eO fB

ir th

Em pl

oy er

Po si

tio n

Em er

ge nc

yC on

ta ct

E- N

am e

E- Ph

on e

Pa tie

nt ID

C ar

eP ro

vi de

r

V isi

t

D at

eO fV

isi t

Pr ov

id er

ID

Pr ov

id er

N am

e

A dd

re ss

Ph on

e

G en

de r

Sp ec

ia lty

A ffi

lia te

dH os

pi ta

l

Ed uc

at io

n D

eg re

e

Y ea

r

La ng

ua ge

s

Sc he

du le

St ar

tT im

e

Pa tie

nt N

am e

C la

im

C la

im ID

Se rv

ic eN

am e

Pa tie

nt Pa

ym en

t

Se rv

ic e

In su

ra nc

eP ay

m en

t

Ba la

nc e

Se rv

ic eC

od e

A ge

Pr ov

id er

ID

C la

im ID

Pa tie

nt ID

V isi

tID

FI G

U R

E 5 .1

2 C

o n

v er

si o

n o

f th

e V

is it

r el

at io

n sh

ip t

o a

n a

ss o

ci at

iv e

en ti

ty i

n t

h e

h ea

lt h

ca re

d at

ab as

e.

70 Developing Windows-Based and Web-Enabled Information Systems

Pa tie

nt

SS N

Pa tie

nt N

am e

A dd

re ss

Ph on

e

G en

de r

D at

eO fB

ir th

Em pl

oy er

Po si

tio n

Em er

ge nc

yC on

ta ct

E- N

am e

E- Ph

on e

Pa tie

nt ID

C ar

eP ro

vi de

r

V isi

t

D at

eO fV

isi t

Pr ov

id er

ID

Pr ov

id er

N am

e

A dd

re ss

Ph on

e

G en

de r

Sp ec

ia lty

A ffi

lia te

dH os

pi ta

l

Ed uc

at io

n D

eg re

e

Y ea

r

La ng

ua ge

s

Sc he

du le

St ar

tT im

e

Pa tie

nt N

am e

C la

im

C la

im ID

Se rv

ic eN

am e

Pa tie

nt Pa

ym en

t

Se rv

ic e

In su

ra nc

eP ay

m en

t

Ba la

nc e

Se rv

ic eC

od e

A ge

Pr ov

id er

ID

C la

im ID

Pa tie

nt ID

FI G

U R

E 5 .1

3 M

ax im

u m

a n

d m

in im

u m

c ar

d in

al it

ie s

fo r

th e

V is

it a

ss o

ci at

iv e

en ti

ty , a

n d

t h

e P

at ie

n t,

t h

e C

ar eP

ro v

id er

, a n

d t

h e

C la

im e

n ti

ti es

i n

t h

e h

ea lt

h ca

re d

at ab

as e.

71Conceptual Data Modeling

5.5 Weak Entities

A weak entity type depends on a strong entity type to exist in a database. A strong entity type exists independently of other entity types. The examples of entity types introduced in Sections 5.1 to 5.4 are all strong entity types. A weak entity type is denoted by a box with double lines (see Figure 5.14). The diamond representing the relationship between a weak entity type and its associated strong entity type also has double lines. Suppose that a data- base for a neighborhood community has an entity type Person. If we want to add the pet(s) of a person to the database, a weak entity type, Pet, is added because no pet(s) of a person can exist in the database if the person does not exist in the database. Figure 5.14 gives the E-R model containing the Person entity, the Pet entity, and their relationship. Hence, the existence of an instance of a weak entity type depends on the existence of an instance of a strong entity type in a database.

5.6 Superclass and Subclass of Entities

In the E-R model of the healthcare database shown in Figure 5.3, the Patient and the CareProvider entities have a common set of attributes (ID, Name, Address, Phone, and Gender) and their own special attributes (SSN, DateOfBirth, Age, Employer, Position, and EmergencyContact for the Patient entity, and Specialty, AffiliatedHospital, Education, Languages, and Schedule for the CareProvider entity). As shown in Figure 5.15, a superclass, the Person entity, which has the attributes common to the Patient and the CareProvider entities, can be created and added to the E-R model of the healthcare database while mak- ing the Patient and the CareProvider entities as the subclass entities of the superclass entity Person. Figure 5.15 shows how the connection between a superclass entity and its subclass entities is represented in an E-R model.

Although only the attributes special to each subclass entity are placed as the attributes of the subclass entities, all the attributes of the superclass entity are inherited by each sub- class entity of the superclass entity. This property between a superclass entity and each of its subclass entities is called “attribute inheritance.” Because of attribute inheritance, the key attribute(s) of the superclass entity is also the key attribute(s) of each subclass entity. Hence, there is no key attribute(s) placed under each subclass entity in an E-R model. An E-R model using superclass and subclass entities is called an “enhanced E-R model.” In this book, we refer to both an E-R model with superclass and subclass entities and an E-R model without superclass and subclass entities as an E-R model.

Person PetHas| | O <

PersonID Name Address Phone PetID AgeType

FIGURE 5.14 Example of a weak entity type.

72 Developing Windows-Based and Web-Enabled Information Systems

The participation property of a superclass entity with its subclass entities describes whether an instance of a superclass entity must participate as an instance of a subclass entity. There are two types of participation: total participation and partial participation. In the total participation, each instance of a superclass entity must be an instance of at least one subclass entity. The total participation is denoted by double lines between the superclass entity and the circle connecting the superclass entity to its subclass entities, as shown in Figure 5.15. The total participation in Figure 5.15 indicates that each instance of the Person entity must be an instance of the Patient entity, the CareProvider entity, or both the Patient and the CareProvider entities. In the partial participation, an instance of a superclass entity does not have to be an instance of any subclass entity. The partial partici- pation is denoted by changing the double lines for the total participation into a single line, as shown in Figure 5.16. Figure 5.16 shows an example of the partial participation between the superclass entity Customer and the subclass entity Member in a database for an online store. The Customer entity has only one subclass entity, Member, which has two attributes for the login name and the password used to log in to an online account. Because a cus- tomer of the store may not sign up as a member, an instance of the Customer entity may not be an instance of the Member entity.

If a superclass entity has at least two subclass entities, the disjoint property of subclass entities for a superclass entity defines whether an instance of the superclass entity can be an instance of only one subclass entity or multiple subclass entities. There are two types of disjoint: disjoint and overlap. In the disjoint type (denoted by “D,” as shown in Figure 5.17, with the assumption that a patient cannot be a healthcare provider and a healthcare provider cannot be a patient), an instance of the superclass entity can be an instance of only one subclass entity. In the overlap type (denoted by “O,” as shown in Figure 5.15, with the assumption that a healthcare provider can be a patient), an instance of the superclass entity can be an instance of multiple subclass entities.

For the overlap property in Figure 5.15, if there are 100 patients (10 of 100 patients are also healthcare providers) and 20 healthcare providers, the Person superclass entity has 110

Patient

SSN

Name Address Phone Gender

DateOfBirth

Employer

Position

EmergencyContact

E-Name

E-Phone

ID

CareProvider

Specialty

AffiliatedHospital

Education Degree

Year

Languages

Schedule StartTime

PatientName

Age

Person

O

FIGURE 5.15 E-R model of the superclass entity, Person, and its subclass entities, Patient and CareProvider, with the total participation and overlap properties.

73Conceptual Data Modeling

instances, including 90 instances who are patients only, 10 instances who are healthcare providers only, and 10 instances who are both patients and healthcare providers. For each instance of the Person entity who is both a patient and a healthcare provider, we have an instance of the Patient entity and an instance of the CareProvider entity for that person.

However, if we assume that a patient cannot be a healthcare provider and a healthcare provider cannot be a patient in the healthcare database, it is now the disjoint property for the Person entity and its subclass entities, as shown in Figure 5.17. Suppose that there are 100 instances of the Patient entity and 20 instances of the CareProvider entity, the Person superclass entity has 120 instances for 100 patients and 20 healthcare providers. For each of the 120 instances for the Person entity, we can find an instance of either the Patient or the CareProvider entity for the same person. If an instance of the Person entity is a patient, we

Member

Login

Name Address PhoneID

Customer

Password

FIGURE 5.16 E-R model of the superclass entity, Customer, and its subclass entity, Member, with the partial participation.

Patient

SSN

Name Address Phone Gender

DateOfBirth

Employer

Position

EmergencyContact

E-Name

E-Phone

ID

CareProvider

Specialty

AffiliatedHospital

Education Degree

Year

Languages

Schedule StartTime

PatientName

Age

Person

D

FIGURE 5.17 E-R model of the superclass entity, Person, and its subclass entities, Patient and CareProvider, with the total participation and disjoint properties.

74 Developing Windows-Based and Web-Enabled Information Systems

have an instance of the Patient entity for that patient. If an instance of the Person entity is a healthcare provider, we have an instance of the CareProvider entity for that person.

We may add a new attribute of a superclass entity to indicate which subclass an instance of the superclass entity falls into. This new attribute plays the role of the subclass discrimi- nator. If we have the disjoint property for the superclass entity and its subclass entities, the new attribute is a single-valued attribute. If we have the overlap property for the super- class entity and its subclass entities, the new attribute is a multi-valued attribute, as shown by the Type attribute in Figure 5.18.

For customers in Figure 5.16, if we use only the Customer entity in the E-R model and keep  the special attributes of members, Login and Password, as the attributes of the Customer entity, many instances of the Customer entity who are not members have the NULL value for the Login and Password attributes, leaving much computer space reserved but unused. Creating the Member entity as a subclass entity of the Customer entity and keeping the Login and Password attributes as the attributes of the Member entity, as shown in Figure 5.16, eliminates the unused computer space of having the NULL value in the Login and Password attributes for customers who are not members.

Hence, if a subset of instances for an entity in an E-R model has their special attributes not shared by all instances of the entity, a subclass entity should be created to keep the special attributes for this subset of instances. This process of creating a subclass entity to keep special attributes for a subset of instances is called “specialization.” If two entities in an  E-R model share a number of common attributes, a superclass can be created to  keep the  common attributes. This process of creating a superclass entity to keep common attri- butes of subclass entities is called “generalization.” In the E-R model for the  healthcare database, the Patient and the CareProvider entities share many common attributes. Through generalization, the Person entity is created as the superclass entity of the Patient and the CareProvider entities. The E-R model in Figure 5.3 becomes the E-R model with superclass and subclass entities in Figure 5.19.

Patient

SSN

Name Address Phone Gender

DateOfBirth

Employer

Position

EmergencyContact

E-Name

E-Phone

ID

CareProvider

Specialty

AffiliatedHospital

Education Degree

Year

Languages

Schedule StartTime

PatientName

Age

Person

O

Type

FIGURE 5.18 E-R model of the superclass entity, Person, and its subclass entities, Patient and CareProvider, with the subclass discriminator.

75Conceptual Data Modeling

Pa tie

nt D at eO

fB ir th

Em pl oy er

Po si tio

n

Em er ge nc

yC on

ta ct

E- N am

e

E- Ph

on e

In su ra nc eC

om pa ny

In su ra nc eP la n

En ro ll

O  |

O ffe

r >  |

| |

M em

be rI D

C om

pa ny

N am

e

A dd

re ss

Ph on

e

C ar eP

ro vi de r

M ed ic in e

V isi t

D at eO

fV isi t

Sy m pt om

s

D ise

as e

M ed ic in eN

am e

D es cr ip tio

n

Sp ec ia lty

A ffi lia te dH

os pi ta l

Ed uc

at io

n D

eg re

e

Y ea

r

La ng

ua ge

s

Sc he

du le

St ar

tT im

e

Pa tie

nt N

am eC om

pa ny

ID Pl

an ID

Pl an

N am

e

D es

cr ip

tio n

M em

be rP

re m

iu m

Em pl

oy er

Pr em

iu m

En ro

llm en

tD at

e

G ro

up ID

D ep

en dO

n

C la

im

Pr oc

es s

C la

im ID

Se rv

ic eN

am e

Pa tie

nt Pa

ym en

t

Se rv

ic e

In su

ra nc

eP ay

m en

t

Ba la

nc e

Ph ar

m ac

y Pr

es cr

ip tio

n

Fi ll

K ee p

Ph ar m ac yI D

Ph ar m ac yN

am e

A dd

re ss

M ed ic in eI D

U ni tP ri ce

C ur re nt In ve nt or y

M in im

um In ve nt or yC on

ta in

Q ua nt ity

Pr es cr ip tio

nI D

Ph on

e

Pa tie

nt ID

C la im

ID

Pr ov id er ID

Se rv ic eC

od e

In cl ud

e

T re at m en t

G en

er at e

O rd er

T re at m en

tI D

A ge

>  |

Pe rs on

O

Pe rs on

N am

e A dd

re ss

Ph on

e G en

de r

ID

SS N

La st Pi ck up

D at e

R efi

llF re qu

en cy

R ec ov er yT

im e

FI G

U R

E 5 .1

9 E

-R m

o d

el o

f th

e h

ea lt

h ca

re d

at ab

as e

w it

h s

u p

er cl

as s

an d

s u

b cl

as s

en ti

ti es

.

76 Developing Windows-Based and Web-Enabled Information Systems

5.7 Summary

This chapter describes E-R modeling to define the conceptual organization of data using concepts that are natural and easy to understand. The data organization is specified using entities along with their attributes, relationships along with their attributes, and maxi- mum and minimum cardinalities of relationships, which further define how instances of entities in a relationship are associated. An E-R model may also include associative enti- ties, weak entities, superclass entities, and subclass entities. The next chapter describes the transformation of an E-R model into a relational data model.

EXERCISES

1. Give two instances of the InsuranceCompany entity in the E-R model of the healthcare database in Figure 5.19.

2. Give two instances of the CareProvider entity in the E-R model of the healthcare database in Figure 5.19.

3. Give two instances of the Order relationship in the E-R model of the healthcare database in Figure 5.19.

4. Give two instances of the Visit relationship in the E-R model of the healthcare database in Figure 5.11.

5. A list of database application projects is available at http://www.dssbooks.com/ web/ProjectsManual.html. Project 48 describes a database application for a movie theater. The movie theater database should include data about the following: • Customers with their ID, name, address, phone number, and e-mail • Employees with their ID, name, date of hire, employment history, and salary,

where employment history is a multi-valued attribute • Producers with their ID, name, contact information (address, phone number,

and e-mail), and current balance • Movies with their ID, title, year of production, rating, description, awards won,

and actors, where awards won and actors are multi-valued attributes • Showrooms with their name, location, and capacity

When a show of a movie is assigned to a showroom, the database records the date and time of the show and the total number of tickets available for the show. When a customer purchases ticket(s) for a show of a movie, the database records the ticket number, date, unit price, number of tickets purchased, and amount paid. The number of tickets purchased for a show decreases the total number of tickets available for the show. When a customer signs up as a member to buy tickets online, the member’s login name and password are recorded. When the movie theater purchases movie(s) from a producer, the database records the transaction number, purchase price, purchase date, payment due date, and amount due.

The manager of the movie theater queries the database to obtain certain infor- mation. The following is a list of data queries:

77Conceptual Data Modeling

1. Financial information a. Create a query that presents monthly revenues from ticket sales, monthly

expenses from salaries, monthly expenses from movie purchases, and monthly earnings during the current year.

b. Create a query that presents yearly revenues, salary expenses, and movie expenses during the current year.

2. Create a query that lists the five best movies of the current year based on the number of awards won.

3. Create a query that lists the five most expensive movies of the current year. 4. Create a query that displays the total number of tickets sold per movie in

decreasing order of tickets sold. 5. Create a query that presents the average room usage per showroom in the cur-

rent year. 6. Create a query that lists the 100 most preferred customers based on the amounts

that the customers spent in the movie theater. 7. Create a query to present information about the producer that the movie the-

ater did the most business during the current year. 8. Create a query that prompts for a date and a movie title and then presents the

number of tickets available per show for the movie on that date. 9. Create a query that prompts for a movie title and then presents the weekly

show schedule of the movie and the total number of tickets available per show. 10. Create a query that prompts for a customer name and presents information

about the tickets purchased by the customer during the current month.

Given the above requirements for the movie theater database, use an E-R dia- gram to describe the Customer, the Employee, the Producer, the Movie, and the Showroom entities, along with their attributes. Give two instances of each entity.

6. Given the requirements of the movie theater database in Exercise 5, use an E-R dia- gram to describe all entities and relationships in this database without the maxi- mum and minimum cardinalities of each relationship.

7. Add the maximum and minimum cardinalities to each relationship in the E-R model produced in Exercise 6.

8. Create superclass and subclass entities, if needed, in the E-R model produced in Exercise 7.

79

6 Logical Database Design: Relational Modeling and Normalization

The relational model was created by Codd (1971, 1990) for database management and has been used predominantly to design databases. In this chapter, we first define a “relation” as the data structure to store data in the relational database model and introduce several data integrity constraints. We then describe the methodology of transforming an entity- relationship (E-R) model into a relational model of database. We also describe normaliza- tion, which is used to identify and remove data anomalies in a relational model.

6.1 Relational Model of Database and Data Integrity Constraints

In a relational database model, relations are used to store data in a database. A relation is a data structure (called a “data schema”) that consists of columns and rows. A column speci- fies a data attribute, and a row specifies a data record. Figure 6.1 shows the InsurancePlan relation to represent the InsurancePlan entity in the E-R model of the healthcare database in Figure 5.19. In Figure 6.1, the InsurancePlan relation has five columns to describe the five attributes of the InsurancePlan entity and has two rows or two data records to store two instances of the InsurancePlan entity. The relational schema for InsurancePlan can also be defined in the following text form:

InsurancePlan (PlanID, PlanName, Description, MemberPremium, and InsurancePremium)

A relation must have the primary key consisting of one or more attributes as the key attribute(s). The primary key should uniquely identify each record in the relation. That is, each record in the relation has a unique value of the primary key. The primary key is underlined in a relation. The InsurancePlan relation has PlanID as the primary key.

Example 6.1

Give the InsuranceCompany relation that represents the InsuranceCompany entity and its attributes in the E-R model of the healthcare database in Figure 6.2a (see also Figure 5.19).

Figure 6.2b gives the InsuranceCompany relation with CompanyID as the primary key and two records for two instances of the InsuranceCompany entity.

When an entity in an E-R model is represented by a relation in a relational model, the key attribute(s) of the entity is typically taken as the primary key of the relation. The fol- lowing procedure can be followed to select the primary key for a relation or determine whether the selected primary key is appropriate:

80 Developing Windows-Based and Web-Enabled Information Systems

1. Identify all super keys, where a super key is a set of one or more attributes that uniquely identify each record in the relation.

2. Determine the candidate key(s) from all super keys, where a candidate key is a super key with the smallest number of attribute(s).

3. Select the primary key from the candidate key(s) to meet the following criteria: a. The primary key should be unique at all times. b. The primary key should have a value that never changes for each record. c. The primary key should never have the NULL value.

InsurancePlan

PlanID

PlanName

Description

MemberPremium

EmployerPremium

InsurancePlan

PlanID MemberPremium

PL Care PPO A PPO plan $300 $600

PL1 Care HMO An HMO plan $100 $500

DescriptionPlanName EmployerPremium

FIGURE 6.1 Representation of the InsurancePlan entity by the InsurancePlan relation in the relational model of the healthcare database.

InsuranceCompany

CompanyID

C1 ABC

InsuranceCompany

CompanyName

Address

Phone

CompanyID (a)

(b)

PhoneAddressCompanyName

480-965-1111111 Mill Avenue

C2 XYZ 480-965-2222222 Mill Avenue

FIGURE 6.2 Representation of the InsuranceCompany entity by the InsuranceCompany relation in the relational model of the healthcare database. (a) The InsuranceCompany entity in the E-R model of the healthcare database. (b) The InsuranceCompany relation representing the InsuranceCompany entity.

81Logical Database Design

Example 6.2

Determine the super key(s), the candidate key(s), and the primary key for the InsuranceCompany relation in Figure 6.2.

There are 13 super keys, each of which can uniquely identify a record of the InsuranceCompany relation:

1. CompanyID 2. Address 3. Phone 4. CompanyID, Address 5. CompanyID, Phone 6. CompanyID, CompanyName 7. Address, Phone 8. Address, CompanyName 9. Phone, CompanyName 10. CompanyID, Address, Phone 11. CompanyID, Address, CompanyName 12. CompanyID, Phone, CompanyName 13. CompanyID, CompanyName, Address, Phone

Because there may exist two companies that have the same name, the CompanyName attribute cannot uniquely identify a record of the InsuranceCompany relation. Among the 13 super keys, there are 3 candidate keys that have only one attribute:

1. CompanyID 2. Address 3. Phone

Considering that a company’s address and phone number may change, CompanyID is selected as the primary key of the InsuranceCompany relation. CompanyID is created to be unique for each company at all times, does not change, and does not have the NULL value.

A relation has the following properties:

• The relation is uniquely identified by its name. Hence, different relations must use different names.

• Each cell, which is defined by a given column and a given row, contains exactly one atomic value.

• Each record is unique. Hence, we cannot have two identical records. • Different attributes of the relation have different attribute names. However, an

attribute of one relation and an attribute of another relation may use the same attribute name, for example, ID.

• Values of an attribute are from the same domain, for example, integer. • The order of attributes is irrelevant. • The order of records is irrelevant.

An E-R model has entities and relationships of entities. As shown in Figure 6.1, an entity in an E-R model can be represented by a relation in a relational model. Because a relational model consists only of relations, a relationship of entities in an E-R model must also be represented using the data structure of a relation. A relational model

82 Developing Windows-Based and Web-Enabled Information Systems

uses the pairing of the primary key and the foreign key in a relation to represent a relationship of entities in an E-R model. As shown in Figure 6.3, the Offer relation- ship of the InsurancePlan and the InsuranceCompany entities is represented in a relational model by adding CompanyID (the primary key of the InsuranceCompany relation) to the InsurancePlan relation as the foreign key that is paired with PlanID (the primary key of the InsuranceCompany relation) for representing the Offer rela- tionship of the InsurancePlan and the InsuranceCompany entities. With the pairing of PlanID and CompanyID in the InsurancePlan relation, we can see that the company, C1, offers two insurance plans, PL1 and PL2. CompanyID in the InsurancePlan relation is called the “foreign key” because it comes from the primary key of another relation, the InsuranceCompany relation. In a relational model, a line is drawn, starting from the for- eign key and pointing back to its primary key, as shown in Figure 6.3, to indicate where the foreign key comes from.

Data in a relational database model are subject to three types of data integrity constraints: domain constraint, entity constraint, and referential constraint. The domain constraint requires that all the values of an attribute must be from the same domain. For example, in the InsurancePlan relation, all the values of MemberPremium must be currency values. The entity constraint requires that every relation in a relational database model must have the primary key and that a value of the primary key cannot be NULL. For example, if a Customer relation has Email as an attribute, we cannot use Email as the primary key of the Customer relation because some customers may not have an e-mail address, and we cannot let the primary key take the NULL value to represent an unknown or unavailable e-mail address. The referential constraint requires that a foreign key must take either one of the values that its primary key has or the NULL value. Figure 6.4 shows the violation of the referential constraint because CompanyID as a foreign key in the InsurancePlan rela- tion has a value of C3, whereas CompanyID as the primary key of the InsuranceCompany relation does not have the value of C3.

InsurancePlan

PlanID MemberPremium

InsuranceCompany

DescriptionPlanName

CompanyID

C1 ABC 111 Mill Avenue 480-965-1111

C2 XYZ 222 Mill Avenue 480-965-1111

CompanyName Address Phone

CompanyIDEmployerPremium

PL1 Care HMO An HMO plan $100 $800 C1

PL2 Care PPO A PPO plan $300 $600 C1

FIGURE 6.3 Pairing of a primary key and a foreign key and the referential integrity constraint to represent the Offer rela- tionship of the InsuranceCompany and the InsurancePlan entities.

83Logical Database Design

6.2 Transformation of an E-R Model to a Relational Model

This section describes the transformation of entities and their attributes, the transforma- tion of superclass and subclass entities and their attributes, the transformation of relation- ships and their attributes, the transformation of associative entities and their attributes, and the transformation of weak entities and their attributes.

6.2.1 Transformation of Entities and Their Attributes

A regular entity in an E-R model is represented by a relation in a relational model. The name of the entity can be used as the name of the relation. If the key attribute(s) of the entity meets the requirements of the primary key, as described in Section 6.1, the key attribute(s) of the entity becomes the primary key of the relation. Each simple, stored, and single-valued attribute of the entity becomes an attribute of the relation with the same attribute name. Figure 6.2 shows the transformation of the InsuranceCompany entity to the InsuranceCompany relation.

A derived attribute of an entity will not be included as an attribute of the relation repre- senting the entity because the value of a derived attribute can be obtained from values of stored attributes through data queries, as described in Chapter 7. If an entity has a compos- ite attribute, the relation representing the entity will include only the simple, stored, and single-valued attributes that make up the composite attribute.

Example 6.3

Transform the Patient entity in the E-R model of the healthcare database in Figure 6.5a (see also Figure 5.19) to a relation.

Figure 6.5b shows the Patient relation representing the Patient entity. Because the Patient entity is a subclass of the Person entity and inherits the ID attribute from the Person entity as the key attribute, the ID attribute is included in the Patient relation

InsurancePlan

PlanID PlanName Description MemberPremium EmployerPremium CompanyID

PL1 Care HMO A HMO plan $100 $800 C1

PL2 Care PPO A PPO plan $300 $600 C1

InsuranceCompany

CompanyID CompanyName Address Phone

C1 ABC 111 Mill Ave 480-965-1111

C2 XYZ 222 Mill Ave 480-965-2222

PL3 HealthPlan A PPO plan $350 $650 C3

FIGURE 6.4 Illustration of a violation of the referential integrity constraint.

84 Developing Windows-Based and Web-Enabled Information Systems

as the primary key. The Age attribute is a derived attribute that is not included in the Patient relation. The EmergencyContact attribute is a composite attribute that is not included in the Patient relation. Instead, the E-Name attribute and the E-Phone attribute, which make up the EmergencyContact attribute, are included in the Patient relation.

To represent a multi-valued attribute of an entity in an E-R model, we create a new rela- tion that includes the multi-valued attribute and the key attribute(s) of the entity as the composite key attributes in the primary key of the new relation.

Example 6.4

Transform the Claim entity in the E-R model of the healthcare database in Figure 6.6a (see also Figure 5.19) to a relational model.

Figure 6.6b shows the Claim and the ClaimService relations created for the multi- valued attribute of Service. The claim relation represents the Claim entity. Because the Service attribute is a multi-valued attribute, we create a new relation, the ClaimService relation, which includes ClaimID (the primary key of the Claim relation) and the ServiceCode, ServiceName, PatientPayment, and InsurancePayment attributes that make up the composite attribute of Service.

Because the ServiceCode attribute can uniquely identify each service, only ServiceCode (instead of all the attributes making up the composite Service attribute) is used together with ClaimID as the composite primary key for the ClaimService relation.

Figure 6.6c shows three records of the Claim relation to represent a claim containing three services if a new relation is not created for the multi-valued Service attribute, but instead, the Service attribute is kept in the Claim relation. The Balance attribute of this claim, which has the value of $600, repeats three times in the three records for the three services in this claim. If the Claim entity has other simple, single-valued attributes, these attributes also have to repeat the same value in the three records for the three services of the claim. Having the repetitive storage of the same data (the Balance of a

PatientDateOfBirth

Employer

Position

EmergencyContact

E-Name

E-Phone

Age

SSN

Patient

ID SSN DateOfBirth Employer Position E-Name E-Phone

(a)

(b)

FIGURE 6.5 Representation of a derived attribute and a composite attribute in a relational model. (a) The Patient entity in the E-R model of the healthcare database. (b) The Patient relation representing the Patient entity.

85Logical Database Design

claim) creates data redundancy in the database, which is undesirable because of the following:

• Redundant data waste precious storage space on the computer. • It takes more computing time to update data (e.g., updating of a claim’s balance

in multiple records). • There is a possibility of data inconsistency (e.g., the balance of a claim updated

in one record but not in other records containing the balance of the claim) and other types of data anomalies described in Section 6.3.1.

Note that, because the value of ClaimID is the same for the three records in the ClaimService relation, ServiceCode must join ClaimID as the composite primary key of the ClaimService relation in order to uniquely identify each record of the ClaimService relation.

Hence, without creating a new relation for a multi-valued attribute, all the single-valued attributes of an entity have to repeat the same value in multiple records created for storing multiple values of the multi-valued attribute. Figure 6.6b shows the representation of the same claim and its three services when a new relation is created for the multi-valued attri- bute. The representation in Figure 6.6b has no data redundancy.

Claim

ClaimID

ServiceName PatientPayment

Service

InsurancePayment

Balance

ServiceCode

Claim

ClaimID Balance

ClaimService

ClaimID ServiceCode ServiceName PatientPayment InsurancePayment

(a)

(b)

ClaimService

ClaimID Balance ServiceCode ServiceName PatientPayment InsurancePayment

CL1 $600

CL1 SC1 A $15 $135

CL1 SC1 A $15$600 $135

CL1 SC2 B $0$600 $200

CL1 SC3 C $0$600 $250

CL1 SC2 B $0 $200

CL1 SC3 C $0 $250

(c)

FIGURE 6.6 Representation of a multi-valued attribute in a relational model. (a) The Claim entity in the E-R model of the healthcare database. (b) The Claim relation and the Claim service relation representing the Claim entity. (c) The same value of the Balance attribute is repeated three times in three records for representing multiple values of the multi-valued Service attribute if a new relation is not created for the multi-valued Service attribute.

86 Developing Windows-Based and Web-Enabled Information Systems

6.2.2 Transformation of Superclass and Subclass Entities and Their Attributes

Given a superclass entity and its subclass entity, the superclass entity is transformed to a relation, and the subclass entity is transformed to another relation. Because an instance of the subclass entity has its corresponding instance in the superclass entity, the correspon- dence of an instance in a subclass entity to its superclass instance is represented by includ- ing the primary key of the relation for the superclass entity in the relation for the subclass entity as the primary key.

Example 6.5

Transform the superclass entity, Person, and its two subclass entities, Patient and CareProvider, in the E-R model of the healthcare database in Figure 6.7a (see also Figure 5.19) to a relational model.

Figure 6.7b shows the relational model that includes the Person, the Patient, and the CareProvider relations for representing the Person, the Patient, and the CareProvider entities, respectively. The primary key of the Person relation is added to the Patient and the CareProvider relations as a foreign key for representing the relationship between the superclass entity of Person with its two subclass entities of Patient and CareProvider. The ProviderLanguage relation is created to represent the multi-valued attribute of Language for the CareProvider entity. The ProviderSchedule relation is created to represent the multi-valued attribute of Schedule for the CareProvider entity. Because the Schedule attribute is made up of the StartTime attribute and the PatientName attribute and the StartTime attribute alone can uniquely identify each value of the Schedule attribute, the StartTime attribute is used together with the ID attribute in the ProviderSchedule relation as the composize primary key.

6.2.3 Transformation of Relationships and Their Attributes

In this section, we first look at transforming binary relationships and then extend the methods of transforming binary relationships to transforming unary relationships and relationships of degree-3 or higher.

As described in Section 6.1, a relationship is represented in a relational model through the pairing of a primary key and a foreign key in a relation. For a binary many-to-one or one-to-many relationship, we use the “one migrates to many” rule to add the primary key of the relation on the “one” side of the many-to-one or the one-to-many maximum cardi- nality to the relation on the “many” side as a foreign key so that the pairing of the primary key and the foreign key is included in the relation on the “many” side of the many-to-one or the one-to-many maximum cardinality. Any attribute of the relationship goes where the foreign key is. Example 6.6 further explains the “one migrates to many” rule.

Example 6.6

Transform the binary, many-to-one Offer relationship of the InsurancePlan and the InsuranceCompany entities in the E-R mode of the healthcare database (see Figures 5.19 and 6.8a) to a relational model.

At first, we transform the InsurancePlan and the InsuranceCompany entities to the InsurancePlan and the InsuranceCompany relations, respectively. To represent the Offer relationship, there are two relations to add a foreign key and, thus, create the pairing of PlanID and CompanyID: adding CompanyID as a foreign key to the CompanyPlan relation (see Figure 6.8b) or adding PlanID as a foreign key to the InsuranceCompany relation (see Figure 6.8c).

87Logical Database Design

The Offer relationship of the InsurancePlan and the InsuranceCompany entities has the many-to-one maximum cardinality. That is, an insurance company may offer many insurance plans, and one insurance plan is offered by only one insurance company. If we add PlanID as a foreign key to the InsuranceCompany relation, as shown in Figure 6.8c, this many-to-one maximum cardinality causes the record of an insurance company to repeat in the InsuranceCompany relation as many times as the number of the insurance plans that the insurance company offer, for example,

PatientDateOfBirth

Employer

Position

EmergencyContact

E-Name

E-Phone

CareProvider

Specialty

AffiliatedHospital

Education Degree

Year

Languages

Schedule StartTime

PatientName

Age

Person

O

PersonName Address Phone GenderID

SSN

Person

ID

ID

ID

PersonName

Specialty

ID Language

ID StartTime PatientName

AffiliatedHospital Degree Year

SSN DateOfBirth Employer Position E-Name E-Phone

Address Phone Gender

CareProvider

(a)

(b)

Patient

ProviderLanguage

ProviderSchedule

FIGURE 6.7 Representation of superclass and subclass entities in a relational model. (a) The Person, Patient, and CareProvider entities in the E-R model of the healthcare database. (b) The relational model representing the E-R model in (a).

88 Developing Windows-Based and Web-Enabled Information Systems

InsuranceCompanyInsurancePlan Offer> | |  |

CompanyName

Address

Phone

CompanyID

PlanID

PlanName

Description

MemberPremium

EmployerPremium

InsurancePlan

PlanID PlanName Description MemberPremium EmployerPremium CompanyID

PL1 Care HMO A HMO plan $100 $500 C1

PL2 Care PPO A PPO plan $300 $600 C1

InsuranceCompany

CompanyID CompanyName Address Phone

C1 ABC 111 Mill Ave 480-965-1111

C2 XYZ 222 Mill Ave 480-965-2222

InsurancePlan

PlanID PlanName Description MemberPremium EmployerPremium

PL1 Care HMO A HMO plan $100 $500

PL2 Care PPO A PPO plan $300 $600

InsuranceCompany

CompanyID CompanyName Address Phone PlanID

C1 ABC 111 Mill Ave 480-965-1111 PL1

C2 XYZ 222 Mill Ave 480-965-2222 PL3

C1 ABC 111 Mill Ave 480-965-1111 PL2

PL3 Med HMO A HMO plan $150 $550

PL4 Med PPO A PPO plan $350 $650

C2 XYZ 222 Mill Ave 480-965-2222 PL4

PL3 Med HMO A HMO plan $150 $550 C2

PL4 Med PPO A PPO plan $350 $650 C2

(a)

(b)

(c)

FIGURE 6.8 Representation of a one-to-many or many-to-one binary relationship in a relational model. (a) The Offer rela- tionship of the InsurancePlan entity and the InsuranceCompany entity in the E-R model of the healthcare database. (b) Adding CompanyID as a foreign key to the InsurancePlan relation. (c) Adding PlanID as a foreign key to the InsuranceCompany relation.

89Logical Database Design

two times of CI and two times of C2 in Figure 6.8c. The repetitive storage of the same data (CompanyID, CompanyName, Address, and Phone of an insurance company) in the database creates data redundancy, which is undesirable, as discussed in Section 6.2. In Figure 6.8c, because there are multiple records in the InsuranceCompany rela- tion containing the same data of an insurance company, CompanyID alone cannot uniquely identify each of these multiple records for the same insurance company. PlanID needs to be included in the primary key of the InsuranceCompany relation so that CompanyID and PlanID together can uniquely identify each record of the InsuranceCompany relation.

Various problems of adding PlanID as a foreign key to the InsuranceCompany relation in Figure 6.8c do not exist if we add CompanyID as a foreign key to the InsurancePlan relation according to the “one migrates to many” rule (see Figure 6.8b) because an insur- ance plan is offered by only one insurance company and CompanyID of the insurance company can be simply added to the existing record of the insurance plan without add- ing any new records to the InsurancePlan relation.

Hence, to transform a many-to-one or one-to-many relationship, we add the pri- mary key of the relation on the “one” side of the maximum cardinality (e.g., the InsuranceCompany relation in the Offer relationship) to the relation on the “many” side of the maximum cardinality (e.g., the InsurancePlan relation in the Offer relationship). That is, with the goal of eliminating data redundancy and associated problems, we use the “one migrates to many” rule to transform a many-to-one or one-to-many relation- ship and determine where to add a foreign key for creating the pairing of the primary key and the foreign key to represent the relationship. If a relationship has attribute(s), the attribute(s) of the relationship goes where the foreign key is and, thus, is included in the relation where the pairing of the primary key and the foreign key is because this is where the relationship is represented. The Offer relationship in Figure 6.8 does not have any attribute.

When transforming a many-to-many binary relationship, we first represent two entities involved in the binary relationship by two relations. To represent the many-to-many rela- tionship, adding the primary key of one relation to another relation as a foreign key creates data redundancy and associated problems. Hence, a new relation is created to include the primary keys of both relations in the new relation as the foreign keys whose pairing repre- sents the relationship. The two foreign keys in the new relation are used as the composite primary key for the new relation. Any attribute of the relationship goes where the foreign keys are.

Example 6.7

Transform the binary many-to-many Keep relationship of the Pharmacy and the Medicine entities in the E-R model of the healthcare database (see Figures 5.19 and 6.9a) to a relational model.

Figure 6.9b gives the relational model with the Pharmacy, the Medicine, and the Keep relations. The Keep relationship has the attributes of UnitPrice, CurrentInventory, and MinimumInventory. These three attributes go along with the foreign keys and are included in the Keep relation.

For a binary one-to-one relationship, adding a foreign key to either relation does not make any difference with regard to the maximum cardinality. If the minimum cardi- nality of the relationship is mandatory-to-mandatory or optional-to-optional, add- ing a foreign key to either relation does not make any difference with regard to the minimum cardinality. However, if the minimum cardinality of the relationship is

90 Developing Windows-Based and Web-Enabled Information Systems

mandatory-to-optional or optional-to-mandatory, we use the “mandatory migrates to optional” rule to add the primary key of the relation on the “mandatory” side of the min- imum cardinality to the relation on the “optional” side as the foreign key. Any attribute of the relationship goes into the relation containing the foreign key. This rule is further explained in Example 6.8.

Example 6.8

Transform the Order relationship of the Treatment and the Prescription entities in the E-R model of the healthcare database (see Figures 5.19 and 6.10a) to a relational model.

We first transform the Prescription and the Treatment entities to the Prescription and the Treatment relations, respectively. Figure 6.10c shows one way of representing the Order relationship in the relational model by adding PrescriptionID (the primary key of the Prescription relation) to the Treatment relation as the foreign key. This representa- tion of the relationship causes the unused computer storage space for the PrescriptionID attribute in the records of TR3 and TR4 because a prescription is optional for a treat- ment, and the two treatments, TR3 and TR4, do not produce an order of a prescrip- tion. Computer space still needs to be reserved for the PrescriptionID attribute in the Treatment relation even for treatments that do not produce a prescription.

(a)

(b) Pharmacy

PharmacyID

Medicine

MedicineName

Description

Pharmacy

Keep

PharmacyID

PharmacyName

Address

MedicineID

UnitPrice CurrentInventory MinimumInventory

Phone

Medicine

Keep

PharmacyName Address

MedicineID MedicineName Description

PharmacyID MedicineID UnitPrice CurrentInventory MinimumInventory

Phone

FIGURE 6.9 Representation of a many-to-many binary relationship in a relational model. (a) The Keep relationship of the InsurancePlan entity and the InsurancePlan entity in the E-R model of the healthcare database. (b) The transfor- mation of the Pharmacy entity, the Medicine entity, and the Keep relationship to a relational model.

91Logical Database Design

Figure 6.10b shows another way of representing the Order relationship by following the “mandatory migrates to optional” rule and adding TreatmentID (the primary key of the Treatment relation on the “mandatory” side of the minimum cardinality) to the Prescription relation on the “optional” side of the minimum cardinality as the foreign key. This representation of the relationship in Figure 6.10b does not produce any unused computer space as seen in Figure 6.10c.

DiseasePrescription

PrescriptionID Treatment

Order

TreatmentID

(a)

Prescription

PrescriptionID TreatmentID

Treatment

TreatmentID Disease

PR1 TR1

PR2 TR2

TR1 A

TR2 B

TR3 C

TR4 D

TR1 PR1

PR2

A

TR2 B

TR3 C

TR4 D

(b)

Prescription

PrescriptionID

Treatment

TreatmentID Disease PrescriptionID

PR1

PR2

(c)

FIGURE 6.10 Representation of a one-to-one, mandatory-to-optional/optional-to-mandatory relationship in a relational model. (a) The Order relationship of the Prescription entity and the Treatment entity in the E-R model of the healthcare database. (b) Adding a foreign key to the Prescription relation according to the “mandatory migrates to optional” rule. (c) Adding a foreign key to the Treatment relation.

92 Developing Windows-Based and Web-Enabled Information Systems

It can be seen that maximizing data storage efficiency or minimizing wasted data space (e.g., data redundancy in Figure 6.6c and Figure 6.8c or unused data space in Figure 6.10c) is the principle that determines the “one migrates to many” rule for transforming a binary one-to-many or many-to-one relationship, the creation of a new relation for representing a binary many-to-many relationship, and the “mandatory migrates to optional” rule for transforming a binary one-to-one relationship. Wasted data space does not only increase the database cost—wasted data space such as data redundancy is also associated with various kinds of data anomalies, which are described in Section 6.3. Hence, maximizing data storage efficiency is an important principle to follow when we implement a database on computers.

The above methods of transforming a binary relationship can be extended to transform a unary relationship. If a unary relationship has a many-to-many relationship, we create a new relation to represent the relationship. If a unary relationship has a one-to-many/many- to-one or one-to-one maximum cardinality, we apply the “one migrates to many” rule or the “mandatory migrates to optional” rule. However, because there is only one entity involved in a unary relation, either the “one migrates to many” rule or the “mandatory migrates to optional” rule puts the foreign key in the same relation containing the primary key. That is, if a unary relationship has a one-to-many/many-to-one or one-to-one maximum cardinality, the foreign key is added to the only relation that also contains the primary key.

Example 6.9

Transform the unary DependOn relationship of the Patient entity in the E-R model of the healthcare database (see Figures 5.19 and 6.11a) to a relational model.

Figure 6.11b shows the Patient relation. In the many-to-one unary relationship, a patient who is a policy holder can have many patients depending on the policy holder. Hence, the policy holder on the “one” side of the maximum cardinality migrates as a foreign key, as shown in Figure 6.11b. The primary key, ID, comes from the superclass entity of the patient entity.

PatientDateOfBirth

Employer

Position

EmergencyContact

E-Name

E-Phone

DependOn

Age

SSN

(a)

(b) Patient

ID SSN DateOfBirth Employer Position E-Name E-Phone PolicyHolder

FIGURE 6.11 Representation of a one-to-many or many-to-one unary relationship in a relational model. (a) The Patient entity and the DependOn relationship in the E-R model of the healthcare database. (b) The transformation of the Patient entity and the DependOn relationship to a relational model.

93Logical Database Design

To transform a relationship of degree-3 or higher, each entity involved in the relation- ship is represented by a relation. In addition, a new relation is created for representing the relationship. The new relation contains and uses the primary key of each relation repre- senting each entity as the composite primary key of the new relation. Example 6.10 illus- trates the transformation of a degree-3 relationship. A new relation is created to represent a relationship of degree-3 or higher for the same reason why we create a new relation to represent a binary many-to-many relationship for preventing data redundancy.

Example 6.10

Transform the degree-3 Visit relationship of the Patient, the CareProvider, and the Claim entities in the E-R model of the healthcare database (see Figure 5.11, which is copied in Figure 6.12a) to a relational model.

Figure 6.12b shows the relational model for the Visit relationship of the Patient, the CareProvider, and the Claim entities. The Visit relation is created to represent the Visit relationship. The Visit relation includes PatientID, ProviderID, and ClaimID as the for- eign keys and use them together as the composite primary key.

6.2.4 Transformation of Association Entities

As described in Chapter 5, a binary many-to-many relationship or a relationship of degree-3 or higher can be converted to an associate entity. As a binary many-to-many rela- tionship or a relationship of degree-3 or higher is represented by a new relation in a rela- tional model, an associative entity is represented by a new relation in a relational model. Example 6.11 illustrates the transformation of an associative entity to a relational model.

Example 6.11

Transform the associative entity, Visit, in the E-R model of the healthcare database (see Figures 5.19 and 6.13a) to a relational model.

Figure 6.13b shows the Visit relation that represents the associative entity and contains three foreign keys, PatientID, ProviderID, and ClaimID, as the composite primary key. Hence, the Visit relationship in Example 6.10 and the Visit associative entity in Example 6.11, which is converted from the Visit relationship, have the same representation in the relational model.

6.2.5 Transformation of Weak Entities

Like a regular entity, a weak entity is represented by a relation in a relational model. Because a weak entity depends on a strong entity, the primary key of the strong entity needs to be included in the relation for the weak entity as a foreign key, which is included in the primary key of the relation for the weak entity, so that no instance of the weak entity can be stored in the database without association with an instance of the strong entity. Example 6.12 illustrates the transformation of a weak entity to the relational model.

Example 6.12

Transform the E-R model containing the weak entity, Pet, in Figure 5.14, which is copied in Figure 6.14a to a relational model.

The E-R model is transformed to a relational model shown in Figure 6.14b. In the Pet relation, PersonID (the primary key of the strong entity) is included as a foreign key and used together with PetID as the primary key. If PersonID is not included in the pri- mary key of the Pet relation, a record of the Pet relation can exist, with PersonID having

94 Developing Windows-Based and Web-Enabled Information Systems

Pa tie

nt

SS N

Pa tie

nt N

am e

A dd

re ss

Ph on

e

G en

de r

D at

eO fB

ir th

Em pl

oy er

Po si

tio n

Em er

ge nc

yC on

ta ct

E- N

am e

E- Ph

on e

Pa tie

nt ID

C ar

eP ro

vi de

r

V is

it D

at eO

fV is

it

Pr ov

id er

ID

Pr ov

id er

N am

e

A dd

re ss

Ph on

e

G en

de r

Sp ec

ia lty

A ffi

lia te

dH os

pi ta

l

Ed uc

at io

n D

eg re

e

Y ea

r

La ng

ua ge

s

Sc he

du le

St ar

tT im

e

Pa tie

nt N

am e

C la

im

C la

im ID

Se rv

ic eN

am e

Pa tie

nt Pa

ym en

t

Se rv

ic e

In su

ra nc

eP ay

m en

t

Ba la

nc e

Se rv

ic eC

od e

A ge

(a )

FI G

U R

E 6

.1 2

R ep

re se

n ta

ti o

n o

f a

d eg

re e-

3 re

la ti

o n

sh ip

in a

r el

at io

n al

m o

d el

. ( a)

T h

e V

is it

r el

at io

n sh

ip o

f th

e P

at ie

n t e

n ti

ty , t

h e

C ar

eP ro

v id

er e

n ti

ty , a

n d

t h

e C

la im

e n

ti ty

in t

h e

E -R

m

o d

el o

f th

e h

ea lt

h ca

re d

at ab

as e.

( b)

T h

e tr

an sf

o rm

at io

n o

f th

e V

is it

r el

at io

n sh

ip o

f th

e P

at ie

n t

en ti

ty , t

h e

C ar

eP ro

v id

er e

n ti

ty , a

n d

t h

e C

la im

e n

ti ty

t o

a r

el at

io n

al

m o

d el

.

95Logical Database Design

the NULL value. This means that a pet exists without a person whom the pet depends on, which is not allowed because a pet is a weak entity and has to depend on a person. Having PersonID join PetID as the composite primary key of the Pet relation enforces that PersonID in the Pet relation cannot have the NULL value since a primary key attribute cannot have the NULL value. Hence, a record of the Pet relation must have a specific value of PetID and a specific value of PersonID, enforcing the dependence of a pet on a person.

ProviderID Language

PatientID SSN PatientName Address Phone Gender DateOfBirth Employer Position E-Name E-Phone

Patient

ProviderID ProviderName Address Phone Gender Specialty AffiliatedHospital Degree Year

ProviderLanguage

ProviderSchedule

ProviderID StartTime PatientName

Provider

ClaimID Balance

ClaimService

ClaimID ServiceCode ServiceName PatientPayment InsurancePayment

Claim

Visit

PatientID ProviderID ClaimID DateOfVisit

(b)

FIGURE 6.12 (Continued) Representation of a degree-3 relationship in a relational model. (a) The Visit relationship of the Patient entity, the CareProvider entity, and the Claim entity in the E-R model of the healthcare database. (b) The transformation of the Visit relationship of the Patient entity, the CareProvider entity, and the Claim entity to a relational model.

Visit

DateOfVisit

PatientID

ClaimID

ProviderID

(a)

(b)

Visit

PatientID ProviderID ClaimID DateOfVisit

FIGURE 6.13 Representation of an associate entity in a relational model. (a) The associative entity, Visit, in the E-R model of the healthcare database. (b) The transformation of the associative entity, Visit, to a relational model.

96 Developing Windows-Based and Web-Enabled Information Systems

Examples 6.1 to 6.11 transform many parts of the E-R model for the healthcare database to the relational model. Figure 6.15 gives the complete relational model for the entire E-R model of the healthcare database.

The transformation of an E-R model to a relational model in this section shows why we need to identify features such as types of attributes (e.g., single-valued versus multi-valued, simple versus composite, and stored versus derived), the degree of a relationship, the maxi- mum and minimum cardinalities of a relationship, superclass and subclass entities, and weak entities in an E-R model. These features identified in an E-R model determine how data are organized into relations in a relational model to maximize data storage efficiency.

6.3 Normalization

If we do a good job of developing an E-R model for a database and then transforming the E-R model to a relational model using the methods described in Section 6.2, the resulting relational model is expected to be in a desired form for maximizing data storage efficiency. However, we may have a relational model that is not in a desired form and contains data redundancy or wasted data space due to a poorly developed E-R model or an existing data- base that is poorly built. Normalization helps us determine whether a relation in a relational model is in a desired form for maximizing data storage efficiency. If a relation is not in a desired form, we use normalization to bring the relation to the desired form.

In this section, we first look into data redundancy and associated data anomalies to explain why data redundancy should be eliminated. We then introduce the concept of functional dependency and use this concept to define normal forms. We finally describe methods of normalization to bring a relation to a desired normal form.

6.3.1 Data Redundancy and Data Anomalies

Figure 6.16 gives the PrescriptionFilling relation that contains data about pharmacies fill- ing medicine prescriptions. Four records of the relation are also included in Figure 6.16.

Person PetHas||  <

PersonID

PersonID

PersonID

Name Address Phone

Name Address Phone

PetID

PetID

AgeType

AgeType

(a)

(b)Person

Pet

FIGURE 6.14 Representation of a weak entity in a relational model. (a) An E-R model containing a weak entity, Pet. (b) The transformation of the E-R model in Part (a) to a relational model.

97Logical Database Design

As indicated by the records of the relation, a pharmacy can fill multiple prescriptions (e.g., PH1 fills PR1 and PR2), a prescription can contain multiple medicines (e.g., PR1 contains M1 and M2), a medicine can be in multiple prescriptions (e.g., M1 is in PR1 and PR3), and a prescription can be filled by only one pharmacy.

Data redundancy exists in several places in the PrescriptionFilling relation. First of all, the name, address, and phone number of a pharmacy repeat as many times as the

Patient PatientID SSN DateOfBirth Employer Position E-Name E-Phone PolicyHolder PlanID MemberID GroupID EnrollmentDate

InsurancePlan PlanID PlanName Description MemberPremium EmployerPremium CompanyID

InsuranceCompany CompanyID CompanyName Address Phone

Claim ClaimID Balance CompanyID

ClaimService ClaimID ServiceCode ServiceName PatientPayment InsurancePayment

CareProvider ProviderID Speciality AffiliatedHospital Degree Year

ProviderLanguage ProviderID Language

ProviderSchedule ProviderID StartTime PatientName

Visit VisitID PatientID ProviderID ClaimID DateOfVisit

Treatment TreatmentID Disease RecoveryTime Symptoms VisitID

Prescription PrescriptionID PharmacyID TreatmentID

Medicine MedicineID MedicineName Description

Contain PrescriptionID MedicineID Quantity LastPickupDate RefillFrequency

Pharmacy PharmacyID PharmacyName Address Phone

Keep PharmacyID MedicineID UnitPrice CurrentInventory MinimumInventory

Include PlanID ProviderID

Person ID PersonName Address Phone Gender

FIGURE 6.15 Relational model of the healthcare database.

98 Developing Windows-Based and Web-Enabled Information Systems

number of medicines in the prescriptions filled by the pharmacy (e.g., the name, address, and phone number of PH1 repeat three times in the first three records in Figure 6.16). Second, the name and description of a medicine repeats as many times as the number of prescriptions that contain the medicine (e.g., the name and description of M1 repeats two times in the first and fourth records in Figure 6.16). Not only does data redundancy cause the waste of computer space, but data redundancy also causes data anomalies such as insertion anomaly, deletion anomaly, and update anomaly.

An insertion anomaly occurs when we want to add a new record to a relation but not all the data are available to allow us adding the new record. Suppose that we want to add a new pharmacy with the following values of PharmacyID, PharmacyName, Address, and Phone—PH3, PC, PCC, and P333, respectively—to the PrescriptionFilling relation. Because this is a new Pharmacy and there is no prescription filled by the Pharmacy yet, there are no data of prescriptions for this pharmacy. We can give only the NULL value for the PrescriptionID, MedicineID, MedicineName, Description, Quantity, LastPickupDate, and RefillFrequency attributes of the PrescriptionFilling relation. However, PrescriptionID and MedicineID are the primary keys and cannot have the NULL value. Hence, we cannot add a new record for this new pharmacy to the PrescriptionFilling relation because we do not have a specific value for the primary key of this new record.

A deletion anomaly occurs when the deletion of a record results in a loss of data. For example, if we delete two prescriptions, PR1 and PR2, filled by the pharmacy, PH1, by deleting the first three data records, data of the pharmacy, PH1, and two medicines, M2 and M3, are removed from the relation, causing the undesired data loss.

An update anomaly occurs when an update of an attribute value needs to be made at multiple places. For example, if the name of the pharmacy, PH1, is changed, this change has to be made at multiple places in the first three data records.

6.3.2 Functional Dependency

Normalization is performed based on the analysis of functional dependencies that are present in a relation. A functional dependency is defined in this section. The next section describes normalization, which includes a series of normal forms based on the analysis of functional dependencies.

Given that A and B are two sets of attributes, B functionally depends on A, or in other terms, A determines B, if a given value of A uniquely determines the value of B. The func- tional dependency of B on A is denoted as A → B or represented using a dependency dia- gram as shown in Figure 6.17. A is called “determinant(s),” and B is called “dependent(s).”

PrescriptionFilling PharmacyID PharmacyName Address Phone PrescriptionID MedicineID MedicineName Description Quantity LastPickupDate RefillFrequency

PH1 PA PAA P111 PR1 M1 MA MAAA 30 1/1/2013 Monthly

PH1 PA PAA P111 PR1 M2 MB MBBB 20 1/3/2013 Quarterly

PH1 PA PAA P111 PR2 M3 MC MCCC 50 1/1/2013 Monthly

PH2 PB PBB P222 PR3 M1 MA MAAA 60 1/30/2013 Semi-annually

FIGURE 6.16 PrescriptionFilling relation and some records of the relation.

99Logical Database Design

Example 6.13

Identify the functional dependencies that exist in the PrescriptionFilling relation in Figure 6.16.

Because PrescriptionID and MedicineID are the primary keys of the PrescriptionFilling relation, a given value of the primary key uniquely identifies one record of the relation, and each of the non-key attribute(s) has only one value in the record. That is, the pri- mary key attributes, PrescriptionID and MedicineID, uniquely determine all the non- key attributes. Hence, we have the following functional dependency:

PrescriptionID, MedicineID → PharmacyID, PharmacyName, Address, Phone, MedicineName, Description, Quantity, LastPickupDate, and RefillFrequency

For any relation, these always is a functional dependency that has the primary key attribute(s) as the determinant(s) and the non-key attribute(s) as the dependents.

In addition to the functional dependency with the primary key as the determinant(s), we identify the functional dependencies that have a smaller number of determinants than that in the functional dependency with the primary key as the determinant(s). The following two functional dependencies are identified:

PharmacyID → PharmacyName, Address, Phone

MedicineID → MedicineName, Description

Figure 6.18 gives a dependency diagram that shows the above three functional depen- dencies in the PrescriptionFilling relation.

Each of the three functional dependencies in Figure 6.18 can be changed to another functional dependency by making a dependent to a determinant. For example, the following functional dependency is created from FD2 in Figure 6.18 by changing PharmacyName from a dependent to a determinant:

PharmacyID, PharmacyName → Address, Phone

For normalization, however, we are not interested in functional dependencies that have the same or larger number of determinants than the number of determinants in the primary key of the relation. For the purpose of normalization, we are interested

A B

FIGURE 6.17 Dependency diagram showing a functional dependency of B depending on A.

PrescriptionFilling

FD1:

FD2: FD3:

PharmacyID PharmacyName Address Phone PrescriptionID MedicineID MedicineName Description Quantity LastPickupDate RefillFrequency

FIGURE 6.18 Functional dependencies in the PrescriptionFilling relation.

100 Developing Windows-Based and Web-Enabled Information Systems

in the  functional dependency with the primary key as the determinant(s) and all the non-key attributes as the dependents and other functional dependencies that have a smaller number of determinants. Hence, for the PrescriptionFilling relation, only three functional dependencies in Figure 6.18 are identified.

Among the three functional dependencies that exist in the PrescriptionFilling rela- tion as shown in Figure 6.18, FD3 is a partial functional dependency. In a partial func- tional dependency, a non-key attribute depends on some, but not all, of the primary key attributes. In FD3, two non-key attributes are determined by MedicineID, which is one of the two primary key attributes. FD2 is a transitive functional dependency. A transitive functional dependency involves no primary key attribute(s). In FD2, none of the PharmacyID, PharmacyName, Address, and Phone attributes are a primary key attribute.

The only functional dependency that should be present in a relation is the functional dependency with the primary key as the determinant and all the non-key attributes as the dependents. Additional functional dependencies in the relation, such as partial func- tional dependency, transitive functional dependency, and any other type of functional dependency, cause data redundancy and data anomalies and should be removed from the relation through normalization. For example, a functional dependency with a non-key attribute as the determinant and a primary key attribute as the dependent is neither a par- tial functional dependency nor a transitive functional dependency but should be removed because it causes data redundancy and data anomalies.

6.3.3 Normalization and Normal Forms

In general, if no functional dependencies other than the functional dependency with the primary key as the determinant(s) and all the non-key attributes as the dependents can be identified in a relation, the relation is in a normal form without data redundancy and data anomalies. If multiple functional dependencies are identified in a relation, normalization needs to be performed on the relation to remove the functional dependencies other than the one with the primary key as the determinant(s) and all the non-key attributes as the dependents and thus bring the relation to a normal form without data redundancy and data anomalies. That is, if there is only one functional dependency that has the primary key as the determinant(s) and all the non-key attributes as the dependents in a relation, there are no data redundancy and data anomalies in the relation, and no normalization needs to be performed. If there are multiple functional dependencies in a relation, normal- ization needs to be performed to remove those functional dependencies that do not have the primary key as the determinant(s) because those functional dependencies cause data redundancy and data anomalies. Normalization defines the first normal form (1NF), the second normal form (2NF), the third normal form (3NF), the Boyce–Codd normal form (BCNF), and so on.

A relation is in 1NF if each cell of the relation contains only one value. Example 6.14 illustrates 1NF.

Example 6.14

Figure 6.19 shows the PrescriptionFilling2 relation, which is the same as the PrescriptionFilling relation in Figure 6.16, except that the PrescriptionFilling2 relation

101Logical Database Design

has only one primary key attribute, PrescriptionID. Is the PrescriptionFilling2 relation in 1NF? If not, perform the normalization to bring the relation to 1NF.

Because the PrescriptionFilling2 relation uses only PrescriptionID as the primary key and the primary key should uniquely identify each record of the relation, there is only one record for PR1. A prescription can have more than one medicine. For example, the prescription, PR1, has two medicines, M1 and M2. Two medicines, M1 and M2, in PR1 have to be put in the same cell for the MedicineID attribute in the record for PR1. Hence, this cell contains two values. Similarly, the cells for the MedicineName, Description, Quantity, LastPickupDate, and RefillFrequency attri- butes in the record for PR1 also contain two values. Hence, the PrescriptionFilling2 relation is not in 1NF.

To bring the PrescriptionFilling2 relation to 1NF, we need to add an attribute that contains multiple values (e.g., MedicineID) to the primary key to separate multiple val- ues into different cells. By adding MedicineID to the primary key and, thus, using both PrescriptionID and MedicineID as the primary keys, as shown in Figure 6.16, each cell of the relation in Figure 6.16 contains only one value.

A relation is in 2NF if the relation is in 1NF and has no partial functional dependencies. Example 6.15 illustrates 2NF.

Example 6.15

Is the PrescriptionFilling relation in Figure 6.18 in 2NF? If not, perform the normaliza- tion to bring the relation to 2NF.

The PrescriptionFilling relation shown in Figure 6.18 has a partial functional depen- dency, FD3. The normalization of removing FD3 and bringing the relation to 2NF is performed by decomposing the PrescriptionFilling relation, specifically by taking out the attributes in FD3 and putting them into a new relation, Medicine, while keeping MedicineID in the PrescriptionFilling relation as the foreign key. Figure 6.20a shows the PrescriptionFilling and the Medicine relations after performing the normalization to take out the attributes in FD3 and put these attributes in the Medicine relation. Figure 6.20b shows two functional dependencies in the PrescriptionFilling relation and one functional dependency in the Medicine relation. The Medicine relation has only one functional dependency with the primary key (MedicineID) as the determinant and the non-key attributes (MedicineName and Description) as the dependents. There is no par- tial functional dependency in the Medicine relation. Hence, the Medicine relation is in 2NF. The PrescriptionFilling relation has two functional dependencies but no partial functional dependency. Hence, the PrescriptionFilling relation is also in 2NF.

A relation is in 3NF if the relation is in 2NF and has no transitive functional dependency. Example 6.16 illustrates 3NF.

PrescriptionFilling2 PharmacyID PharmacyName Address Phone PrescriptionID MedicineID MedicineName Description Quantity LastPickupDate RefillFrequency

PH1 PA PAA P111 PR1 M1, M2 MA, MB MAAA, MBBB 30, 20 1/1/2013, 1/3/2013 Monthly, Quarterly

PH1 PA PAA P111 PR2 M3 MC MCCC 50 1/1/2013 Monthly

PH2 PB PBB P222 PR3 M1 MA MAAA 60 1/30/2013 Semi-annually

FIGURE 6.19 PrescriptionFilling2 relation that is not in the 1NF.

102 Developing Windows-Based and Web-Enabled Information Systems

Example 6.16

Are the PrescriptionFilling and the Medicine relations in Figure 6.20b in 3NF? If not, perform the normalization to bring the relations to 3NF.

The Medicine relation has one functional dependency with the primary key as the determinant and all the non-key attributes as the dependents. As shown in Example 6.15, the Medicine relation is in 2NF. The Medicine relation does not have any transitive functional dependency. Hence, the Medicine relation is in 3NF.

The PrescriptionFilling relation has two functional dependencies. As shown in Example 6.15, the PrescriptionFilling relation is in 2NF. Because the PrescriptionFilling relation has a transitive functional dependency, FD2, the PrescriptionFilling relation is not in 3NF. We perform the normalization to take out all the attributes in FD2 and put them in a new relation, Pharmacy, while keeping PharmacyID in the PrescriptionFilling relation as a foreign key. Figure 6.21a shows the PrescriptionFilling and the Pharmacy relations along with the Medicine relation after the normalization. Figure 6.21b shows the dependency diagrams for these relations. Each relation is in 3NF and has only one functional dependency with the primary key as the determinant and all the non-key attributes as the dependents.

A relation is in BCNF if the relation is in 3NF and does not have any functional depen- dency whose determinant(s) is not the primary key. Example 6.17 illustrates BCNF.

Example 6.17

Figure 6.22a shows the Kid relation containing data about the programs that each kid participates in and the mentors assigned to each program. As shown by records in the Kid relation (see Figure 6.22a), a child can participate in multiple programs, multiple mentors can be assigned to a program, and a mentor can be assigned to only one program. Figure 6.22b shows the functional dependencies in the Kid rela- tion, FD1 and FD2. FD2 has the determinant that is not the primary key. FD2 is nei- ther a partial functional dependency nor a transitive functional dependency. Hence,

PrescriptionFilling

FD1:

FD2:

Medicine

(b)

PrescriptionFilling

Medicine

(a)

PharmacyID

MedicineID MedicineName Description

MedicineID MedicineName Description

PharmacyName Address Phone PrescriptionID MedicineID Quantity LastPickupDate RefillFrequency

PharmacyID PharmacyName Address Phone PrescriptionID MedicineID Quantity LastPickupDate RefillFrequency

FIGURE 6.20 Normalization to bring the PrescriptionFilling relation to the 2NF. (a) The PrescriptionFilling relation and the Medicine relation. (b) The dependency diagrams for the PrescriptionFilling relation and the Medicine relation.

103Logical Database Design

PrescriptionFilling

Pharmacy PharmacyID PharmacyName Address Phone

PharmacyID PharmacyName Address Phone

Medicine

PrescriptionFilling

(a)

Pharmacy

Medicine

(b)

PharmacyID PrescriptionID MedicineID

MedicineID MedicineName Description

MedicineID MedicineName Description

Quantity LastPickupDate RefillFrequency

PharmacyID PrescriptionID MedicineID Quantity LastPickupDate RefillFrequency

FIGURE 6.21 Normalization to bring the PrescriptionFilling relation to the 3NF. (a) The PrescriptionFilling relation, the Pharmacy relation, and the Medicine relation after normalization to 3NF. (b) The dependency diagrams for the PrescriptionFilling relation, the Pharmacy relation, and the Medicine relation.

Kid KidID Program Mentor

KidID ProgramMentor

(b) FD1:

FD2:

Kid

1 A X

2 A Y

3 A Y

3 B Z

4 B Z

(a)

Kid

(c)

KidID Program Mentor

FIGURE 6.22 Normalization to bring the Kid relation to the BCNF. (a) The Kid relation with some records. (b) The functional dependencies in the Kid relation. (c) The Kid relation after normalization to BCNF.

104 Developing Windows-Based and Web-Enabled Information Systems

the Kid relation is in 1NF, 2NF, and 3NF, but not in BCNF. The normalization of removing FD2 from the Kid relation is performed by making the Mentor attribute as a primary key attribute and the Program attribute as a non-key attribute, as shown in Figure 6.22c.

After normalizing a relation to 1NF, 2NF, 3NF, and BCNF, we bring the relation in a normal form, with only one functional dependency that has the primary key as the determinant(s) and all the non-key attributes as the dependents.

6.4 Summary

In this chapter, we first define a relation in a relational model. We then introduce methods of transforming an E-R model to a relational model, including the following:

• Representation of each entity by a relation • Representation of a composite attribute by including only the simple attributes

that make up the composite attributes • Representation of a multi-valued attribute by a new relation • Rule of “mandatory migrates to optional” to place a foreign key for a one-to-one,

optional-to-mandatory/mandatory-to-optional relationship • Rule of “one migrates to many” to place a foreign key for a one-to-many/many-to-

one relationship • Representation of a many-to-many binary relationship, a relationship of degree-3

or higher or an associative entity by a new relation • Representation of a weak entity by a relation with the key attributes of both the

weak the strong entities as the composite primary key

In this chapter, we also introduce several types of functional dependencies to detect data redundancy and data anomalies in an existing relation that is constructed incor- rectly. Steps of normalization are described to bring a relation to the 1NF, the 2NF, the 3NF, and the BCNF and, thus, correct problems of data redundancy and data anomalies in a relation.

EXERCISES

1. Transform the Show entity to a relation.

ShowTime NumberOfTicketsAvailable

Show

ShowDayShowID

105Logical Database Design

2. Transform the Movie entity to a relation.

Movie

MovieID

Title

Year

Awards

Description

Actors

Location

3. Transform the following Person entity to a relation.

ID Name Address

Phone Email

Contact

Person

DateOfHire EmHistory Salary

4. Transform the following E-R model to a relational model.

Movie

MovieID

Title

Year

Awards

Description

Actors

Location

Showroom RoomName

Location

Capacity

ShowTime NumberOfTicketsAvailable

Show

ShowDayShowID

PlayAssign

106 Developing Windows-Based and Web-Enabled Information Systems

5. The Buy relationship in the following E-R model is transformed into a relational model as follows. Is there any problem with this transformation? If so, discuss the problem(s) and make corrections.

Employee

ID Name Address

Phone Email

DateOfHire EmHistory Salary

Contact

Customer

Ticket

Buy

TicketNo

Price

BuyDate

Member

LoginName Password

Person

O

Type

Customer

ID Type TicketNo BuyDate

Ticket

TicketNo Price

6. Transform the E-R model in Exercise 5 to a relational model. 7. The transformation of an E-R model to a relational model is shown in the follow-

ing. Is this transformation correct? If not, explain what is incorrect and why it is incorrect, and make a correction.

DependentID StudentID

StudentName

Address

Student Has Dependent

DateOfBirth

107Logical Database Design

Student

StudentID StudentName Address

Dependent

DependentID DateOfBirth StudentID

8. Transform the E-R model of the movie theater database from Exercise 8 in Chapter 5 to the relational model of the movie theater database.

9. Transform the following E-R model to a relational model.

Customer

Phone Name Address

Room

RoomNo Type Size View Smoking Floor

ReservationMake

Assign

| | | <

ReservationNo Check-InTime Check-OutTime PaymentMethod Charges

ChargeID Amount

FinalBillAmount

Price

10. The PatientVisit relation keeps information about the patients’ visits to doctors. a. Identify the functional dependencies in the relation. b. Is this relation in 2NF, 3NF, and BCNF, and why? If not, show your steps of

performing the normalization to transform the relation into 2NF, 3NF, and BCNF. Show all referential integrity constraints in the relational model that result from the normalization.

c. Identify the functional dependency in each relation in the final relational model produced in b.

PatientVisit PatientID Name Address VisitDate DoctorID DoctorName P1209 J. Johns 134 West 3/3/02 D1256 A. Frank P1209 J. Johns 134 West 9/10/02 D1256 A. Frank P1215 J. Johns 200 West 3/3/02 D4523 D. Gomez P1221 F. Brown 223 South 6/7/02 D6712 G. Kelly P8912 F. Dong 45 East 9/12/02 D8917 M. Julius

108 Developing Windows-Based and Web-Enabled Information Systems

11. The MovieTicket relation keeps data about the movie tickets bought by customers. MovieTicket

MovieID MovieTitle BuyDate 12/31/2012 12/31/2012 1/1/2013 1/1/2013 1/1/2013 1/2/2013 1/2/2013 1/3/2013

1 1 1 1 1 1 2 2

AAA AAA AAA AAA AAA AAA BBB BBB

1 1 1 1 2 2 1 1

1/1/2013 1/1/2013 1/1/2013 1/1/2013 1/2/2013 1/2/2013 1/3/2013 1/3/2013

12 PM 12 PM 12 PM 12 PM 2 PM 2 PM 6 PM 6 PM

ShowDay ShowTime CustomerID CustomerName TicketNo ShowID 1 1

2

1 2 1 2

1 2 2 3 3 4 4

X X Y Y Z Z W W

3 4

a. Identify the functional dependencies in the relation. b. Is this relation in 1NF, 2NF, 3NF, and BCNF, and why? If not, show your steps

of performing the normalization to transform the relation into 1NF, 2NF, 3NF, and BCNF. Show all referential integrity constraints in the relational model that results from the normalization.

109

7 Database Implementation in Microsoft Access

Microsoft Access 2013 (called “Access” in the following text) is a database management system (DBMS) that is included in Microsoft Office to support the implementation of a database on computers based on a relational model of database. An Access database con- tains a collection of data objects: tables to store records of relations, queries to retrieve data selectively in various ways, and forms and reports to present retrieved data in user- preferred formats. This chapter describes the use of Access to implement data storage and data queries in a database. Access forms and reports are not included in this chapter because the graphical user interface for the data presentation in user-preferred formats is created using Microsoft Visual Studio and is described in Chapters 11, 12, 13, and 16. Visual Studio is used to develop a Windows-based or Web-enabled database application with an Access database embedded.

7.1 Tables for Data Storage

Each relation in a relational model is implemented by a table in Access. To create a table in Access, we first open Access and then click on “Blank Desktop Database,” which brings up a pop-up window, as shown in Figure 7.1.

For the prompt of the file name in the pop-up window, we can select a directory and a file name for a database and then click on the Create button, which brings up a Datasheet View of a default table, which is named “Table1.” As shown in Figure 7.2, the screen is divided into three areas: the menus and icons in the top area, the list of all Access objects in the left area, and the work space in the right area now showing the Datasheet View of Table1. We need to change to the Design View of the table for creating and defining the table. To change to the Design View of the table, we go to the View icon at the top-left corner of the screen (see Figure 7.2). Clicking on the down arrow in the View icon, we see two choices: Datasheet View and Design View. Selecting the Design View brings up a pop-up window with a prompt for the table name (see Figure 7.2). We enter “Patient” to create the table for the Patient relation in the relational model of the healthcare database in Figure 6.15.

Each attribute of the Patient relation becomes a data field of the Patient table. The Design View of the Patient table shows a default data field named “ID,” with AutoNumber as the data type of the ID field. We select the ID field and change “ID” to “PatientID,” as shown in Figure 7.3. We add other data fields of the Patient table.

7.1.1 Setting the Primary Key for a Table

There is a key symbol next to the PatientID field to indicate that the PatientID field is the primary key of the Patient table. In the Design menu, there is the Primary Key icon next to the View icon. To set the primary key of a table, we select a data field and then click on

110 Developing Windows-Based and Web-Enabled Information Systems

the Primary Key icon. We may deselect the data field as the primary key by clicking on the Primary Key icon again. If a table has two or more data fields as the primary key, we can select all of these data fields at the same time and click on the Primary Key icon to make these data fields together as the primary key.

7.1.2 Data Types of a Data Field

In Access, we can select one of the following data types for a data field:

• Short Text • Long Text • Number • Date/Time • Currency • AutoNumber • Yes/No • Object linking and embedding (OLE) object (Note that the data type of OLE object

can be used for object such as Microsoft Word document, Excel spreadsheet, etc.) • Hyperlink

FIGURE 7.1 Pop-up window in Access to select a directory and name a database file.

FIGURE 7.2 Screen in Access to save and name a table.

111Database Implementation in Microsoft Access

• Attachment • Calculated • Lookup

The data type of each data field in the Patient table is shown in Figure 7.3. Because the PolicyHolder field is a foreign key linked to the primary key, PatientID, the data type of the  PolicyHolder field is set to Lookup. Selecting Lookup as the data type of a data field brings up a pop-up window for the Lookup Wizard. We then use the Lookup Wizard to select the primary key that the data field looks up. Figure 7.4 shows the steps of using the Lookup Wizard for the PolicyHolder field of the Patient table to look up to the PatientID field of the Patient table. Using the Lookup data type for a data field that is a foreign key allows to select one of the values for the primary key as a value of the foreign key. This makes it easy to comply with the referential integrity constraint between a foreign key and its primary key.

7.1.3 Field Size Property of a Data Field

When we select the PatientID field, the Field Properties area at the bottom of the work- space shows the other properties of the PatientID field (see Figure 7.3). For the Field Size property, the default value is 255 characters for Short Text. We may change 255 characters for the PatientID field to a shorter length, e.g., 9 characters. For the data type of Number, if we click on the down arrow button in the value of the Field Size property, we see several

FIGURE 7.3 Design View of the Patient table in the healthcare database.

112 Developing Windows-Based and Web-Enabled Information Systems

FIGURE 7.4 Illustration of using the Lookup Wizard in Access to define the Lookup data type.

113Database Implementation in Microsoft Access

choices: Byte, Integer, Long Integer, Single, Double, Replication ID, and Decimal. For a data field with a data type of AutoNumber, there are two choices of values for the Field Size property: Long Integer and Replication ID.

7.1.4 Format Property of a Data Field

For certain data types, the Format property of a data field is available for us to choose among several different formats to store data. If the data type is Time/Date, clicking on the down arrow button in the value of the Format property shows us several format choices, as shown in Figure 7.5: General Date, Long Date, Medium Date, Short Date, Long Time, Medium Time, and Short Time. For the DateOfBirth field of the Patient table, we set the Format property to Short Date since we do not need the time but only the date for this data field. If the data type is Number, clicking on the down arrow button in the value of the Format property shows us several format choices: General Number, Currency, Euro, Fixed, Standard, Percent, and Scientific.

7.1.5 Input Mask Property of a Data Field

The Input Mask property of a data field masks a data field in a form to assist us in entering a value for the data field but does not change the way in which the value is stored in the data field. If the data type is Short Text, clicking on the down arrow button in the value of the Input Mask property brings up several pre-defined input masks for Social Security

FIGURE 7.5 Format property of the Date/Time data type.

114 Developing Windows-Based and Web-Enabled Information Systems

Number (SSN), Phone, Zip Code, Password, Data, and Time. For example, for the SSN field of the Patient table in Figure 7.3, we set the Input Mask for the SSN field of the Patient table to Social Security Number. With this setting, a form of SSN with “-” (see Figure 7.6) is displayed when we enter a value of SSN. Although an SSN is displayed like 111-11-1111, the stored value of the SSN is 111111111 without “-.” When performing a query (see discus- sions in Section 7.2) to look for a specific value of SSN, the specific value of SSN should be expressed in the query without “-”; otherwise, no match will be found because all values of SSN are stored without “-.”

If the data type is Date/Time, clicking on the down arrow button in the value of the Input Mask property shows us several pre-defined input masks: Long Time, Short Date, Short Time, Medium Time, and Medium Date.

7.1.6 Default Value Property of a Data Field

The Default Value property of a data field automatically sets the value of a data field to the default value if we do not enter another value to the data field. When we click on the button with a mark of “…” in the value of the Default Value property, the Expression Builder window is popped up to allow us to enter an expression for the default value (see Figure 7.7).

FIGURE 7.6 Illustration of the Input Mask property for the Short Text data type.

FIGURE 7.7 Expression Builder.

115Database Implementation in Microsoft Access

7.1.7 Validation Rule and Validation Text Properties of a Data Field

The Validation Rule property of a data field allows us to specify the range or type of values that the data field can take. For example, we may specify the following validation rule:

Like "PA*"

for the PatientID field of the Patient table to require that a PatientID value starts with “PA” followed by 0 or more characters. Note that “*” in “PA*” matches any string of 0 or more characters. This validation rule does not allow us to enter any value that does not comply with the pattern of “PA*”. The Validation Text property allows us to enter the text describ- ing the validation rule.

7.1.8 Required Property of a Data Field

The Required Property of a data field can have one of two possible values: Yes and No. If the Yes value is selected, we must enter a value for the data field. If the No value is selected, we do not need to enter a value for the data field. The data field as the primary key is auto- matically set to have the Yes value. A non–key data field has the No value as the default for the Required property.

7.1.9 Indexed Property of a Data Field

The Indexed property of a data field allows us to set an index on the data field. If an index is set on a data field of a table, a sorted order of all the records in the table by values of the indexed field is kept in the database. Hence, indexing a data field takes more memory space but allows finding a data record with a specific value of the indexed field efficiently without checking every record of the table. Hence, there is a trade-off between less search time and more memory space with regard to indexing a data field. Indexing a data field is justified if the data field is queried frequently. Clicking on the down arrow button in the value of the Indexed property shows three possible values of the Indexed property: No, Yes (Duplicates OK), and Yes (No Duplicates).

The data field(s) as the primary key of every table is automatically indexed with the value of Yes (No Duplicates) because the primary key should uniquely identify each record of the table, and thus, no duplicate values are allowed in the primary key. The value of Yes (Duplicates OK) can be used to set the Indexed property of a non–key data field if the data field is queried frequently. A non–key attribute has the default value of No for the Indexed property.

Example 7.1

Complete the design of the Patient table in the healthcare database. Figure 7.3 shows the design of the Patient table. Table 7.1 gives some properties of

some data fields in the Patient table.

7.1.10 Adding and Deleting Records of a Table

To add a record to a table, we need to use the Datasheet View of the table by clicking on the View icon in the Home or the Design menu. When we are in a Datasheet View of a table, we enter a record in a row of the table by entering a value in each cell.

116 Developing Windows-Based and Web-Enabled Information Systems

Example 7.2

Add two data records to the Patient table in the healthcare database. Figure 7.8 shows two data records that we add to the Patient table of the healthcare

database.

When adding data records to tables, we should pay attention to the values of a foreign key and its primary key. As discussed in Section 7.1.2, using the Lookup data type for a data field that is a foreign key makes it easy to maintain the referential integrity constraint.

To delete a record of a table, we select the row for the record and right-click in the row to bring up the pop-up menu that contains a menu item, Delete Record, as shown in Figure 7.9. Clicking on this menu item deletes the record.

The name of a table can be changed by right-clicking the table in the list of All Access Objects at the left of the screen and selecting the Rename item in the pop-up menu. The name of an Access object in other types (e.g., query) can be changed similarly.

TABLE 7.1

Data Fields and Field Properties of the Patient Table in the Healthcare Database

Data Field Property Value for the Property

SSN Input Mask SSN DateOfBirth Format Short Date DateOfBirth Input Mask Short Date E-Phone Input Mask Phone Number PlanID Data Type Lookup (to PlanID in the InsurancePlan table) EnrollmentDate Format Short Date EnrollmentDate Input Mask Short Date

FIGURE 7.8 Datasheet View of the Patient table with two records.

FIGURE 7.9 Deleting a record in a table.

117Database Implementation in Microsoft Access

7.2 Relationships of Tables

As relations are linked through primary key and foreign key pairings and referential integrity constraints in a relational model, tables in an Access database are linked through relationships of tables, which are based on primary key and foreign key pairings. To estab- lish a relationship between two tables for the pairing of a foreign key and a primary key, we select the Database Tools menu and click on the Relationships icon in this menu, which brings up the workspace for the layout of relationships and a pop-up window, as shown in Figure 7.10. The pop-up window allows us to add selected tables to the relationships layout. For example, we select the Patient and the InsurancePlan tables, click on the Add button to add these tables to the relationships layout, and then close the pop-up window (see Figure 7.10). We can bring back the pop-up window and add more tables to the rela- tionships layout by clicking on the Show Table icon in the Design menu, or right clicking in the area of the relationship layout.

The relationships layout now has the Patient and the InsurancePlan tables, as shown in Figure 7.11. The PlanID field in the Patient table is a foreign key, with its primary key being the PlanID field in the InsurancePlan table. We call the table with the primary key as the “parent table” and the table with the foreign key as the “child table.” Hence, there is a relationship between the Patient and the InsurancePlan tables through the pairing of PlanID in the Patient table and PlanID in the InsurancePlan table. To establish this rela- tionship, we drag PlanID in the Patient table over PlanID in the InsurancePlan table, or drag PlanID in the InsurancePlan table over to PlanID in the Patient table. This brings up the Edit Relationships window, as shown in Figure 7.11. This window shows two related data fields that are linked together and a list of checks for Enforce Referential Integrity, Cascade Update Related Fields, and Cascade Delete Related Records. We want to check Enforce Referential Integrity for enforcing the referential integrity constraint between the two related data fields. If we do not check Cascade Update Related Records and Cascade Delete Related Records, we are not allowed to delete a record or update a value of PlanID

FIGURE 7.10 Show Table window.

118 Developing Windows-Based and Web-Enabled Information Systems

in the parent table (the InsurancePlan table) if the value of PlanID exists in the PlanID field of the child table (the Patient table). If we check Cascade Update Related Fields and Cascade Delete Related Records, as shown in Figure 7.11, Access automatically deletes the related data record or updates the related value of PlanID in the child table when we delete a record in the parent table or update the value of PlanID in the parent table.

At the bottom of the Edit Relationships window in Figure 7.11, it is shown that the rela- tionship type is one-to-many. This one-to-many relationship is identified automatically by Access. Access uses the following rules to identify the type of a relationship.

• If the data field in the parent table is the primary key or has a unique index (i.e., hav- ing the Yes [No Duplicates] value for the Indexed property), and the related data field in the child table is a foreign key (not being the primary key or having no unique index), Access establishes a one-to-many relationship between these two tables, with the parent table on the “one” side and the child table on the “many” side. This rule is consistent with the “one migrates to many” rule for transforming a one-to-many rela- tionship to a relational model, as described in Chapter 6. This rule puts a foreign key in the table on the “many” side of the one-to-many relationship. The relationship of the Patient and the InsurancePlan tables in Figure 7.11 is a one-to-many relationship.

• If both related fields are the primary key field or have a unique index in their respective table, Access establishes a one-to-one relationship between these two tables. For example, if the parent table is for the relation representing a superclass entity and the child table is for the relation representing a subclass entity, the primary key of the parent table is the foreign key in the child table, which is also used as the primary key for the child table. The relationship of the Person and the Patient tables in Figure 7.12 is a one-to-one relationship.

• If the field in the parent table is the only primary key field and the child table has the composite primary key containing the related field, Access establishes a one-to- many relationship between the parent and the child tables, with the parent table on the “one” side and the child table on the “many” side. For example, if an entity in an entity-relationship (E-R) model has a multi-valued attribute, there are the parent table for the entity and the child table having the primary key of the parent table and the multi-valued attribute as the composite primary key. For one record in the parent

FIGURE 7.11 Edit Relationships window.

119Database Implementation in Microsoft Access

table, there may exist multiple records with the same value of the related field in the child table due to the multi-valued attribute. The relationship of the CareProvider and the ProviderSchedule tables is a one-to-many relationship that falls in this rule.

• If none of the related fields is a primary key or has a unique index, Access estab- lishes an indeterminate relationship between these two tables, and we cannot enforce referential integrity constraints for such a relationship.

In the Edit Relationships shown in Figure 7.11, there is the Join Type button. The join type of a relationship is described in Section 7.3.

To bring up the Edit Relationships window after we close it, we can right-click on the relationship line linking the related fields. This brings up a menu with two items: Edit Relationship and Delete. Selecting the Edit Relationship item brings up the Edit Relationships window. Selecting the Delete item deletes the relationship. We can also delete a relationship in the Relationships layout by selecting the relationship and pressing the Delete button on the keyboard. To remove a table in the Relationships layout, we select the table and press the Delete button on the keyboard. To add more tables to the Relationships layout for establish- ing more relationships, we go to the Database Tools menu, click on the Relationships icon, and then click on the Show Table icon to add more tables to the Relationship layout.

To save the Relationships layout, we click on the Save icon at the top-left corner of the Access screen or select the Save item in the File menu. We can also click on the Clear Layout icon in the Design menu to clear the entire relationships layout.

If we add two tables with the related data fields to the Relationships layout, and one of the related fields uses the Lookup data type to look up the values of another related field, a line linking the related fields is shown in the Relationships layout without us dragging one field over another field. However, the relationship type is not shown on the line until we bring the Edit Relationships window and check Enforce Referential Integrity.

Example 7.3

Establish the relationships of all the tables in the healthcare database. Figure 7.13 shows the relationships layout that includes all the tables and their rela-

tionship to implement the relational model of the healthcare database in Figure 6.15.

FIGURE 7.12 A one-to-one relationship in the Relationship layout.

120 Developing Windows-Based and Web-Enabled Information Systems

7.3 Queries for Data Retrieval

A query allows to retrieve selected data from a database. This section introduces the graph- ical user interface (GUI) of Access to construct a query. There are several types of queries in Access: Select query, Crosstab query, and Action query (e.g., Delete, Update, Append, or Make Table queries). We describe how to construct these queries in the following sections. Select queries include regular Selection queries, Select queries with parameters, and Select queries with Group-By.

7.3.1 Select Queries Using One Table

We can retrieve data from one table or multiple tables in a database. We show first how to construct a Select query to retrieve data from one table in Example 7.4.

Example 7.4

Construct a Select query to present the address and phone number of a pharmacy named “Super Pharmacy” in the healthcare database.

The Pharmacy table in the healthcare database has the name, address, and phone number of all pharmacies. Figure 7.14a shows records that we store in the Pharmacy table. Two of the three pharmacies have the name “Super Pharmacy.” We need to create a Select query to present the address and phone number of these two pharmacies.

Figure 7.15 shows the steps of creating this Select query. We first select the Create menu and then select the Query Design icon, which brings up the Design View of a query and

FIGURE 7.13 Relationships of all the tables in the healthcare database.

121Database Implementation in Microsoft Access

(a)

(b)

FIGURE 7.14 Select query to present selected data from one table. (a) Records in the Pharmacy table. (b) The result of execut- ing QueryPharmacy to present the selected data from the Pharmacy table.

Table pane

Design grid

FIGURE 7.15 Design View of a Select query using one table.

122 Developing Windows-Based and Web-Enabled Information Systems

the Show Table window. Using the Show Table window, we select the Pharmacy table and then close the Show Table window. We may change the name of this query from Query1 to another name, QueryPharmacy, by clicking on the Save icon at the top-left corner of the screen and then using the pop-up window to change the name of the query. We can also change the name of the query by right-clicking the query in the list of All Access Objects at the left side of the screen and selecting the Rename item in the pop-up menu.

Now, the Pharmacy table appears in the table pane in the top portion of the Design View for QueryPharmacy. The bottom portion of the Design View for QueryPharmacy is the design grid that has the rows for Field, Table, Sort, Show, and Criteria. Clicking on the down arrow in a column of the Field row, we see a list of the data fields in the Pharmacy table along with another field named “Pharmacy.*.” The asterisk (*) is a wildcard character that matches any string of characters. In Pharmacy.*, the asterisk matches every data field in the Pharmacy table and, thus, represents all the data fields in the Pharmacy table. In the first column of the Field row, we select the Address field. We can also drag the Address field in the Pharmacy table in the top portion of the Design View to the first column of the Field row to select the Address field in this query. In the second column of the Field row, we select the Phone field. Because we need to select the records of the Pharmacy table with the value of the PharmacyName field being “Super Pharmacy,” we select the PharmacyName field in the third column of the Field row and put in the selection criterion “Super Pharmacy” in the Criteria row of the PharmacyName column but do not check the Show row of the PharmacyName column. We can have the query result sorted in ascending order of values in the Address field by selecting Ascending in the Sort row of the Address column. We save the query design by clicking on the Save icon.

To execute this query, we click on the Run icon next to the View icon in the Design menu. The query result is shown in the Datasheet View of the query, as shown in Figure 7.14b.

The order of columns for data fields in the design grid of a query can be changed by dragging the column of a data field to a desired place. A data field in the design grid can be removed by selecting the column of the data field and pressing the Delete button on the keyboard. If we need to modify a query by replacing the existing table used by the query with another table, we can delete the existing table by selecting it in the table pane and pressing the Delete button. Right-clicking in the top portion of the Design View and select- ing the Show Table item in the pop-up menu allows us to add a new table to the Design View of the query.

Example 7.4 illustrates the selection of data fields and records from a table to display by a given order of values in one data field. More than one data field can be used to sort the query result.

A criterion can be placed on more than one data field to select records that have specific values of the data fields. Right-clicking the Criteria row of a column and selecting the Build item in the pop-up menu brings up the Expression Builder window (see Figure 7.7), which can be used to enter an expression as a selection criterion. The following are the commonly used operators in an expression:

• Comparison operators: >, >=, <, <=, = (equal), != (not equal) • Logical operators: AND, OR, NOT • BETWEEN … AND … • LIKE

123Database Implementation in Microsoft Access

Some examples of expressions are given as follows:

• Year = 2010 • GPA >= 3.0 • Class = “Junior” OR Class = “Senior” • Salary > 40,000 AND Type = “Full Time” • BETWEEN #1/1/2010# AND #12/31/2010# • BETWEEN #09:00:00 AM# AND #12:00:00 PM# • BETWEEN 1 AND 10 • LIKE “*Database*”

Note that a date or time is enclosed by the pound sign (#). “*Database*” matches any string of characters that contains “Database.” There are also build-in functions in the Expression Builder. Some commonly used date/time functions are as follows:

• Now(): returns the current date and time • Date(): returns the current date • Year(): returns the year in a date/time • Month(): returns the month in a date/time • Day(): returns the day in a date/time

For example, we put the expression = Year(Now()) in the Criteria row of a data field with a data type of Date/Time to select the records with the value of that data field falling in the current year.

7.3.2 Select Queries with Joins of Multiple Tables

This section shows how to construct a Select query to retrieve data from multiple tables.

Example 7.5

Design Query4 for patients in the healthcare database (see Chapter 26) to present the name, address, and phone number of all the healthcare providers who specialize in heart diseases and whose address is in Phoenix, Arizona.

This query needs the CareProvider table to select the healthcare providers who spe- cialize in heart diseases and needs the Person table to retrieve the name, address, and phone number of these healthcare providers. Figure 7.16 shows the query design. Figure 7.17 shows the query result in comparison with the data records in the Person and the CareProvider tables.

This query also shows how to change a column heading in the query result (see the Field row of the PersonName field in Figure 7.16). The column heading for the data field of PersonName is changed to Name, as seen in the query result in Figure 7.17.

When we put two tables in the table pane of a query design, we join the data fields of the two tables. There are three types of join: inner join, left outer join, and right outer join. In the table pane of the Design View of the query shown in Figure 7.16, right-clicking the line linking the ID field of the Person table and the ProviderID field of the CareProvider table brings up a

124 Developing Windows-Based and Web-Enabled Information Systems

pop-up menu with two items: Join Properties and Delete. Selecting the Join Properties brings up the Join Properties window, as shown in Figure 7.16. The window has three checks:

1. Only include rows where the joined fields from both tables are equal 2. Include ALL records from “Person” and only those records from “CareProvider”

where the joined fields are equal 3. Include ALL records from “CareProvider” and only those records from “Person”

where the joined fields are equal

(a)

(b)

(c)

FIGURE 7.17 Query result of a Select query, PAQ4, using multiple tables. (a) The records in the Person table. (b) The records in the CareProvider table. (c) The query result for PAQ4.

FIGURE 7.16 Design of a Select query, PAQ4, using multiple tables and the join properties.

125Database Implementation in Microsoft Access

The first check is for the inner join, which is the default join type. In Figure 7.16, the inner join of the Person and the CareProvider tables is checked. Because the inner join includes in the query result rows or records where the joined fields from both tables are equal, the inner join includes only the first two records of the Person table whose ID field has a value equal to a value of the ProviderID field in a record in the CareProvider table. The third record of the Person table is not included in the query result because there is no record in the CareProvider table that has a value of ProviderID equal to the value of ID in the third record of the Person table. For a record of the Person and the CareProvider tables with the same value of the joined fields, the inner join creates a record including all the data fields of the Person and the CareProvider tables in the query result. Figure 7.18a shows the result- ing two records of the inner join in the query result. After the join operation, the other query operations (e.g., selection of columns and rows, and sorting) specified in the design grid are executed on the join results to produce the query result. That is, when we put two or more tables in the table pane of a query design, the join operation is first performed, which is followed by other operations specified by the design grid.

The second and third checks in the Join Properties window in Figure 7.16 are for the outer joins. In the Join Properties window, the Person table is taken as the left table, and the CareProvider table is taken as the right table. The second check is for the left outer join, and the third check is for the right outer join. Because the left outer join includes all the records from the left table (the “Person” table in Figure 7.16) and only those records from the right table (the “CareProvider” table) where the joined fields are equal, the left outer join produces the join result, as shown Figure 7.18b. In Figure 7.18b, there are three records instead of two in Figure 7.18a because all the three records of the Person table are included in the join result. The record with the ID value of PA1, which does not have the equal value of ProviderID in the CareProvider table, is included in the join result of the left outer join but without the values of the data fields in the CareProvider table.

(a)

(b)

(c)

(d)

FIGURE 7.18 Effects of inner join, left outer join, and right outer join. (a) The join result of the inner join between the Person table and the CareProvider table. (b) The join result of the left outer join between the Person table and the CareProvider table. (c) The join result of the right outer join between the Person table and the CareProvider table. (d) The join result involving two tables that are not linked.

126 Developing Windows-Based and Web-Enabled Information Systems

Figure 7.18c shows the join result of the right outer join. This join result is the same as the join result of the inner join in Figure 7.18a because the ProviderID value in every record of the CareProvider table has the equal value in a record of the Person table. In the following text, the inner join is used as the default in a query design.

If two tables in the table pane of a query design do not have joined fields or are not linked, the join of the two tables produces all the combinations of records in the two tables. For example, if we remove the relationship between the Person and the CareProvider tables and put these two tables in the table pane of a query design, the join result includes 3 × 2 records, as shown in Figure 7.18d, because the Person table has 3 records and the CareProvider table has 2 records. If three or more tables are put in the table pane of a query design, the join result includes the data fields from all the tables.

7.3.3 Select Queries with Parameters and Calculated Fields

Many queries require the user of a database to enter a parameter. For example, a user enters the title of a book to search for the book at http://www.amazon.com. We give sev- eral examples of designing Select queries with parameters.

Example 7.6

Design Query1 for an employer in the healthcare database (see Chapter 26), which is to list the name of insurance companies and their insurance plans whose employer pre- mium is smaller than or equal to a given premium level.

For this query, the employer who wants to retrieve the data needs to enter a given premium level as a parameter. That is, this query takes a parameter input from the user of the database. Figure 7.19 shows the design of this query. It also shows the prompt for the input parameter when the query is executed.

Example 7.7

Design Query4 for healthcare providers in the healthcare database (see Chapter 26) to pres- ent appointment times and patient names for a given healthcare provider on a given date.

Figure 7.20 shows the query design. Note that the StartTime field appears in two columns—one for displaying StartTime and another for selecting the records with

FIGURE 7.19 Design View of a Select query, EQ1, with a parameter.

127Database Implementation in Microsoft Access

Day[StartTime] equal to the date embed by the user. The selection criterion for the PersonName field selects the records with a specific value of the CareProvider name entered by the user. The selection criterion for the PatientName field, Is Not Null, ensures that a specific patient has been scheduled at an appointment time.

Example 7.8 shows a query that uses a calculated field.

Example 7.8

Design Query2 for pharmacies in the healthcare database (see Chapter 26) to give the name and ID of all medicines whose inventory level falls below the minimum level at a given pharmacy.

Figure 7.21 shows the query design. In the design grid, the column with the heading of Difference is a calculated field that computes the difference between the values of two data fields, CurrentInventory and MinimumInventory, for a medicine. If the differ- ence is less than 0, the query presents the name and ID of the medicine.

FIGURE 7.20 Design View of a Select query, PRQ4, with two parameters.

FIGURE 7.21 Design View of a Select query, PHQ2, with a calculated field.

128 Developing Windows-Based and Web-Enabled Information Systems

7.3.4 Select Queries with Groupings of Records

This section shows how to construct a query that present aggregate data from the database.

Example 7.9

Design Query3 for healthcare providers in the healthcare database (see Chapter 26) to list all medicines that have been used to treat a given disease, sorted in descending order of use frequency.

Figure 7.22 shows the query design. The join of the four tables—the Medicine table, the Contain table, the Prescription table, and the Treatment table (shown in the table pane in Figure 7.22)—produces the records containing all the treatments of all the dis- eases and all the medicines needed to treat these diseases. The selection criterion for the Disease field selects those records for a given disease. The use frequency of a medicine for treating the disease is equal to the number of records that contain the medicine. To count the number of records that contain the same medicine, we need to group the records by the same medicine.

To enable the grouping of records by the medicine, we click on the Totals icon in the Design menu, as shown in Figure 7.22. This adds the Total row in the design grid of the query (see Figure 7.22). Clicking on the down arrow in the Total row of a column for a data field in the design grid shows the following choices for the Total row:

• Group By: use it if we want to group records by values of this data field • Sum: use it if we want to get the sum of values from this data field in each

group of records • Avg: use it if we want to get the average of values from this data field in each

group of records • Min: use it if we want to get the minimum of values from this data field in each

group of records • Max: use it if we want to get the maximum of values from this data field in

each group of records • Count: use it if we want to get the count of values from this data field in each

group of records

FIGURE 7.22 Design View of a Select query, PRQ3, with grouping and aggregate data.

129Database Implementation in Microsoft Access

• StDev: use it if we want to get the standard deviation of values from this data field in each group of records

• Var: use it if we want to get the variance of values from this data field in each group of records

• First: use it if we want to get the first value from this data field in each group of records

• Last: use it if we want to get the last value from this data field in each group of records

• Expression: use it if an expression is put in the data field as a calculated field • Where: use it if the data field is used for a selection criterion

We use Group By in the Total row of the MedicineName field in the design grid to group the records in the join result by MedicineName. We use Count in the Total row of the TreatmentID field to count how many treatment records each MedicineName group contains, give the heading of UseFrequency for this count, and sort the query result in descending order of UseFrequency. We use Where in the Total row of the Disease field, which is used to specify the selection criterion. Figure 7.23 shows the records in the

(a)

(b)

(c)

(d) (e)

FIGURE 7.23 Result of a Select query, PRQ3, with grouping and aggregate data. (a) Records in the Treatment table. (b) Records in the Prescription table. (c) Records in the Contain table. (d) Records in the Medicine table. (e) The query result for Example 7.9.

130 Developing Windows-Based and Web-Enabled Information Systems

Treatment table, the Prescription, the Contain, and the Medicine tables, and the query result when we enter Disease1, as the input parameter.

A selection criterion can be applied to aggregate data. For example, we can put the selec- tion criterion, >2, in the Criterion row of the UseFrequency column to display only the medicines whose use frequency is greater than 2.

Example 7.10

Design Query6 for insurance companies in the healthcare database (see Chapter 26) to give the average recovery time of a given disease.

Figure 7.24 shows the query design.

Example 7.11

Design Query2 for insurance companies in the healthcare database (see Chapter 26) to give the total number of patients who are 65 years or older and are treated by each healthcare provider on a given insurance company in the last year.

Figure 7.25 shows the query design. We use Group By in the Total row of the ProviderID field to group the records in the join result by ProviderID. To display the provider name in the query result, we include the Name field of the Person table and use Group By in the Total row of this field. Because in each group by ProviderID, there is only one provider name, this grouping by provider name does not change the grouping by ProviderID. For the CompanyName, DateOfVisit, and the DateOfBirth fields, which are used to specify the selection criteria, we use Where in the Total row. For selecting the join records with DateOfVisit falling in the last year, the selection criterion is

Year([DateOfVisit]) = Year(Now()) − 1.

FIGURE 7.24 Design View of a Select query, IQ6, with grouping and aggregate data.

131Database Implementation in Microsoft Access

The existing field name is enclosed in square brackets. For selecting the join records for patients who are 65 years or older, the selection criterion is

DateOfBirth <= Date() − 365*65, or equivalently Date() − DateOfBirth >= 365*65

which means that the difference between the current date and the date of birth is equal to or more than 65 years.

Example 7.12

Design Query1 for pharmacies in the healthcare database (see Chapter 26) to give the total number of patients who are treated for a given disease in a given month and the total amount of a given medicine used to treat the disease in that month.

Figure 7.26 shows the query design. Note that this query does not use Group By in the Total row of any field. This means that all the records in the join result are treated as one group and that aggregate data (i.e., the count of PatientID and the Sum of Quantity for medicine) are obtained from this group.

Example 7.13

Design Query5 for healthcare providers in the healthcare database (see Chapter 26) to give the total amount of payments that a given healthcare provider received from patient and insurance payments in the current year.

Figure 7.27 shows the query design. The first field in the design grid is a calculated field to compute the total amount of payments that the selected provider received from patient and insurance payments. For this field, we use Expression in the Total row.

FIGURE 7.25 Design View of a Select query, IQ2, with grouping and aggregate data.

132 Developing Windows-Based and Web-Enabled Information Systems

7.3.5 Crosstab Queries

A Crosstab query is similar to a Select query with Group By in two data fields but presents the query result in a table format. Example 7.14 illustrates a Crosstab query.

Example 7.14

Design Query3 for employers in the healthcare database (see Chapter 26) to present a table showing the total number of healthcare providers in the insurance plan of each insurance company.

Figure 7.28a shows the design of a Select query with Group By in the Total row for the CompanyName and the PlanName fields. Figure 7.28b shows the query result.

FIGURE 7.27 Design View of a Select query, PRQ5, with grouping and use of an aggregate data function in a calculated field.

FIGURE 7.26 Design View of a Select query, PHQ1, with aggregate data.

133Database Implementation in Microsoft Access

To create a Crosstab query, we first bring up the Design View of a Select query and then click on the Crosstab icon in the Design menu. This adds the Total and the Crosstab rows in the design grid, as shown in Figure 7.29a. Clicking on the down arrow button in the Crosstab row of a field brings up the following three choices:

• Row Heading • Column Heading • Value

Figure 7.29a shows the design of a Crosstab query, with CompanyName as the row heading, PlanName as the column heading, and the count of ProviderID as the value. In the Total row, we use Group By for the CompanyName field, Group By for the PlanName field, and Count for the ProviderID field. Figure 7.29b shows the query result. By com- paring the query results in Figures 7.28b and 7.29b, we can see that the Crosstab query presents the same data as the Select query with two groupings in Figure 7.28 but uses a different format to display the query result. The Crosstab query puts each value of the data field as the row heading in a row, each value of the data field as the column head- ing in a column, and the value of the data field as the value in the cell of a row and a column. In the Crosstab query result in Figure 7.29b, the cells for Company1 and Plan3, Company1 and Plan4, Company2 and Plan1, and Company2 and Plan2 do not have a value because Company1 does not offer Plan3 or Plan4 and Company2 does not offer Plan1 or Plan2.

(a)

(b)

FIGURE 7.28 Design View and Datasheet View of a Select query, EQ3, with two groupings and aggregate data. (a) The design of a Select query for EQ3. (b) The query result of the Select query in Part a.

134 Developing Windows-Based and Web-Enabled Information Systems

7.3.6 Embedded Select Queries

The Select queries in the previous sections retrieve data from tables in a database. A Select query can be designed to retrieve data from the result of another Select query, creating embedded queries. Examples 7.15 and 7.16 show how to design embedded queries.

Example 7.15

Design Query1 for insurance companies in the healthcare database (see Chapter 26) to give the total number of all patients, female patients, and male patients treated by each healthcare provider, sorted in alphabetical order of provider name.

The group of all patients, the group of female patients, and the group of male patients are three different groups. We cannot create three different groups in one query design. Instead, we create sub-queries (see Figure 7.30a–e) to create three dif- ferent groups and count for each group. Then, we create a query (see Figure 7.30f), which adds the sub-queries in the table pane of the query design and puts together the results of the sub-queries. Because the Gender field is in the Person table, which is linked to both the Patient and the CareProvider tables, using a selection criterion of letting the Gender field of the Person table equal to “F” will produce records in the join result containing female patients treated by female healthcare providers instead of female patients treated by all healthcare providers. To select female patients treated by all healthcare providers, we first create a sub-query to retrieve PatientID of female patients using the Patient and the Person tables (see Figure 7.30b). We then create a

(a)

(b)

FIGURE 7.29 Design View and Datasheet View of a Crosstab query. (a) The design of a Crosstab query for Example 7.14. (b) The query result of the Crosstab query in Part a.

135Database Implementation in Microsoft Access

(a)

(b)

FIGURE 7.30 Design View of embedded queries for IQ1. (a) The design of a subquery to count the number of all patients treated by each healthcare provider. (b) The design of a subquery to retrieve the PatientID of female patients.

136 Developing Windows-Based and Web-Enabled Information Systems

(c)

(d)

FIGURE 7.30 (Continued) Design View of embedded queries for IQ1. (c) The design of a subquery to produce the group and count of female patients treated by each healthcare provider. (d) The design of a subquery to retrieve the PatientID of male patients.

137Database Implementation in Microsoft Access

(e)

(f)

FIGURE 7.30 (Continued) Design View of embedded queries for IQ1. (e) The design of a subquery to produce the group and count of male patients treated by each healthcare provider. (f) The design of the query to put together the counts of all patients, female patients, and male patients treated by each healthcare provider using the three subqueries in Figure 7.30a, c, and e.

138 Developing Windows-Based and Web-Enabled Information Systems

(a)

(b)

FIGURE 7.31 Design View of embedded queries for IQ7. (a) The design of a subquery to compute the total amount of premi- ums for IQ7. (b) The design of a subquery to compute the total insurance payments for IQ7.

139Database Implementation in Microsoft Access

sub-query to produce the group and count of female patients by including the sub- query in the table pane along with the Person, Patient, CareProvider, and Visit tables (see Figure 7.30c). To add the sub-query in Figure 7.30b to the table pane, we right-click anywhere in the table pane area, select the Show Table item, click on the Queries tab in the Show Table window, and select the sub-query. Including PatientID of only female patients from the sub-query in the table pane of the sub-query in Figure 7.30c pro- duces the join result containing the records for female patients only due to the inner join of the tables and the sub-query in Figure 7.30b. Figure 7.30c and d shows two sub- queries working together to provide the group and count of male patients treated by each healthcare provider. Finally, the query in Figure 7.30f puts together the counts of all patients, female patients, and male patients treated by each healthcare provider using the three sub-queries in Figure 7.30a, c, and e. We establish the relationships of the three sub-queries in the table pane of the query in Figure 7.30f through the com- mon fields of PatientID in the three sub-queries to have the inner join of the query results from the three sub-queries.

Example 7.16

Design Query7 for insurance companies in the healthcare database (see Chapter 26) to give the total amount of premiums received, the total amount of insurance payments, and the total amount of profit for a given insurance company in a given year.

Data from different tables are needed to compute the total amount of premiums and the total amount of insurance payments. Hence, we create two sub-queries to compute the total amount of premiums and the total amount of insurance payments. Then, we design a query that puts together the results of the two sub-queries, link the two sub- queries to have the inner join through the CompanyName field, and compute the total amount of profit. Note that, in the sub-query in Figure 7.31a, to compute the total amount of premiums, we use a calculated field including the sum function and Expression in the Total row for this calculated field. Because the premiums are annual premiums, we do not need to specify a particular year when computing the total amount of premiums for the year.

(c)

FIGURE 7.31 (Continued) Design View of embedded queries for IQ7. (c) The design of the query to put together the results of the subque- ries in Parts a and b.

140 Developing Windows-Based and Web-Enabled Information Systems

7.4 Action Queries

An action query changes data stored in a database. We give examples of an Update query and a Delete query.

Example 7.17

Design Query8 for insurance companies in the healthcare database (see Chapter 26) to increase the member premium and the employer premium by 5%.

This Update query requires updating data in the InsurancePlan table. We go to the Create menu, click on the Query Design icon, add the InsurancePlan table through the  Show Table window, and then click on the Update icon in the Design menu to change the Select query to an Update query. The Update query has the Update To row. Figure 7.32 shows the query design. In the query design, we create two data fields, MemberPremium and EmployerPremium, and put the expression of updating each data field in the Update To row. When we execute the Update query, we get a prompt to remind us how many rows of the InsurancePlan table we are about to update and ask us to confirm if we want to proceed with the update. If we click the Yes button, the InsurancePlan table will be updated. Unlike executing a Select query, executing an Update query will not show any query result. Data are simply updated in the database without any query result to show.

Example 7.18

Design Query9 for insurance companies in the healthcare database (see Chapter 26) to delete a given insurance plan.

This Delete query requires deleting data in the InsurancePlan table. We go to the Create menu, click on the Query Design icon, add the InsurancePlan table through the Show Table window, and then click on the Delete icon in the Design menu to change the Select query to a Delete query. The Delete query has the Delete row. Figure 7.33 shows the query design. In the query design, we put all the data fields of the InsurancePlan table in one column of the design field by selecting InsurancePlan.* and use From in the Delete row of this column. When we execute the Delete query, we get a prompt to remind us how many rows of the specified table we are about to delete and ask us to

FIGURE 7.32 Design View of an Update query, IQ8.

141Database Implementation in Microsoft Access

confirm if we want to proceed with the deletion. If we click the Yes button, the affected rows of the InsurancePlan table will be deleted. Unlike executing a Select query, execut- ing a Delete query will not show any query result. Data are simply deleted in the data- base without any query result to show.

7.5 Summary

In this chapter, we introduce the use of tables in Access for data storage. Each relation in a relational model is implemented by a table in Access to store data. The properties of each data field in a table are described to define the primary key, the data type, the format, the input mask, the validation rule, and the required and indexed properties.

Each referential integrity constraint through primary key and foreign key pairings in a relational model is implemented by a relationship between two tables. The relationship type and the rules that Access uses to determine the relationship type are described. The use of the enforcement of referential integrity for a relationship is introduced.

This chapter also describes how to create various types of Select queries to retrieve data from a database, including Select queries using only one table, Select queries joining mul- tiple tables, Select queries with parameters and calculated fields, Select queries to produce groupings of records and compute aggregate data from a group of records, and Crosstab queries. Action queries, specifically Update and Delete queries, are described to enable us to update and delete data in a database.

EXERCISES

1. Create tables and relationships of tables in Access to implement the relational model of the movie theater database from Exercise 8 in Chapter 6.

2. Create queries for the movie theater database described in Exercise 5 in Chapter 5. 3. Suppose that tblMovie, tblShow, and tblShowroom in the movie theater database

from Exercise 1 have the following records.

FIGURE 7.33 Design view of a Delete query, IQ9.

142 Developing Windows-Based and Web-Enabled Information Systems

tblMovie:

tblShow:

tblShowroom:

a. How many records will be produced by the following query, given that the inner join is used for each relationship? Show these data records.

143Database Implementation in Microsoft Access

b. How many records will be produced by the following query, given that the inner join is used for each relationship? Show these data records.

4. For the movie theater database from Exercise 1, create a query that prompts for a date and return the movie show schedule for the selected date and the total num- ber of tickets per show as follows.

5. Suppose that tblCustomer and tblMember in the movie theater database from Exercise 1 have the following records.

tblCustomer:

tblMember:

144 Developing Windows-Based and Web-Enabled Information Systems

a. How many records will be produced by the following query, with the join properties shown in the following? Show these data records.

b. How many records will be produced by the following query, with the join properties shown in the following? Show these data records.

c. How many records will be produced by the following query, with the join properties shown in the following? Show these data records.

145Database Implementation in Microsoft Access

d. Records in tblTicket are shown in the following. How many records will be produced by the following query?

tblTicket:

6. The design and the data records of tblCustomer in the movie theater database from Exercise 1 are shown in the following. Use the design template to design a query that gives the name and the e-mail address of the customers whose address has the zip code of 85281.

146 Developing Windows-Based and Web-Enabled Information Systems

7. The records of tblShow in the movie theater database from Exercise 1 are shown as follows.

147Database Implementation in Microsoft Access

a. How many records are produced by the following query? List all the records in the query result.

b. How many records are produced by the following query? List all the records in the query result.

148 Developing Windows-Based and Web-Enabled Information Systems

c. How many records are produced by the following query? List all the records in the query result.

d. Can you design a query that lists the total count of shows by ShowDay and displays the specific ShowTime of each show through the same query? If yes, show your query design in the following; if no, explain why.

8. For the movie theater database from Exercise 1, design a query that prompts for a movie title and return the movie schedule for that movie and the total number of tickets available per show. The query result for the movie “The Thing” is shown as follows.

9. For the movie theater database from Exercise 1, create a query that prompts for an actor in a movie and a show period with a beginning date and an ending date and returns the schedule of the movie with the selected actor for the selected period.

149Database Implementation in Microsoft Access

The schedule is sorted in ascending order of the show day and the show time. Note that the user input for the prompt of an actor may contain only the first name or the last name of the actor, whereas an actor stored in the database has the actor’s full name. The format of the query result is shown as follows.

10. For the movie theater database from Exercise 1, create a Totals query that prompts for an actor in a movie and a show period with a beginning date and an ending date and returns the title of each movie including the selected actor and the total number of shows for each movie in the selected period. Note that the user input for the prompt of an actor may contain only the first name or the last name of the actor, whereas an actor stored in the database has the actor’s full name. The format of the query result is shown as follows.

151

8 Structured Query Language

Structured query language (SQL) is the most influential query language of relational data- bases. This chapter first reviews the background of SQL followed by a brief discussion on Backus–Naur form (BNF) and extended BNF (EBNF), a main notation technique used in illustrating SQL syntax. The basic SQL statements, including CREATE, UPDATE, DELETE, and QUERY data, are then explained with examples.

8.1 Introduction to SQL

SQL was initially developed by IBM as a part of the System R project in the early 1970s and was named structured English query language (SEQUEL). Later, it evolved to today’s acronym: SQL. In 1986, the American National Standards Institute first pub- lished the SQL standard, called “SQL-86,” followed by the International Organization for Standardization (ISO) in 1987. Since then, SQL has been enhanced several times. The first major revision was released in 1992, known as “SQL2” or “SQL-92,” which greatly enhanced the standard specification. The SQL syntax discussed in this chapter follows the SQL2 standard. SQL3 was released with support for object-oriented data manage- ment in 1999. In 2003, extensible markup language (XML)-related features were first introduced into SQL, resulted into version SQL-2003. As of today, there are seven revi- sions, and the most recent one is SQL-2011 (ISO 9075:2011), which was formally adopted in December 2011.

According to the SQL standard, SQL is composed of five main languages:

• Data definition language (DDL). DDL is used to create and modify the relation sche- mas within a database. Example commands are the CREATE, ALTER, and DROP statements.

• Data manipulation language (DML). DML is used to retrieve, store, modify, delete, insert, and update data in the database. Examples of this are the SELECT, UPDATE, and INSERT statements.

• Data control language (DCL). DCL is used to create roles, permissions, and referen- tial integrity constraints for a database. Updates that violate integrity constraints are disallowed. It is also used to control secured access to the database. Examples of DCL are the GRANT and REVOKE statements.

152 Developing Windows-Based and Web-Enabled Information Systems

• Transaction control language (TCL). TCL is used to manage different transactions occur- ring within a database. It can specify the beginning and the ending of a transaction, and undo transactions. The COMMIT and ROLLBACK statements are examples.

• Embedded and dynamic SQL. Embedded and dynamic SQL defines how SQL state- ments can be embedded within general-purpose programming languages, such as C, C++, Java, C#, VB.NET, etc.

This chapter will mainly discuss the first two languages: DDL and DML. We will visit DCL and TCL when we discuss MySQL in Chapter 9.

8.2 Backus–Naur Form

Before introducing the SQL syntax, it is necessary to understand BNF and EBNF, one main notation technique for context-free grammars, often used to describe the syntax of lan- guages in computing. In Table 8.1, the general symbols in BNF and EBNF used to describe SQL syntax are summarized.

For example, the EBNF statement: {mandatory_1 | mandatory_2} (optional,…) indicates that it is required to have either mandatory_1 or mandatory_2; in addition, there may be multiple optionals, separated by commas. Please remember the connotation of these sym- bols to interpret the SQL commands in the following sections.

8.3 SQL Syntax

We will adopt part of the healthcare system schema discussed in Chapter 5, the entity- relationship (E-R) and enhanced entity-relationship (EER) design, to illustrate the SQL syntax throughout the chapter. Specifically, the three relations that will be used here are as follows:

Drugs

Drug_ID Name Price Description T_001 Tylenol $17.99 100 325-mg tablets bottle T_003 Tylenol $8.99 T_002 Tri_Vitamin $14.99 1500-400-35 solution 50-mL bottle C_001 Claritin $35.99 10 tablets

TABLE 8.1

BNF and EBNF Symbols

Notation Description Example

{ } Required element {a} [ ] Term enclosed is optional [a]: a may or may not be used ( | ) Enclose groups of alternative terms (a | b | c): either a or b or c … Repetition: term enclosed is used zero or

more times

153Structured Query Language

Stores

Store_ID Name Phone_Number S_001 Walgreens 1-800-746-7287 S_002 CVS Pharmacy 1-800-925-4733 S_003 Fry’s Pharmacy 1-866-221-4141

Sales Transaction_ID Drug_ID Quantity Store_ID ST_001 T_001 25 S_001 ST_002 T_001 300 S_002 ST_003 C_002 150 S_001

Note: SQL uses the terms “table,” “row,” and “column” for the formal relational model termed “relation,” “tuple,” and “attribute,” respectively. The corresponding terms will be used interchangeably. In addition, uppercase is used for the SQL statements. However, most databases now accept both uppercase and lowercase.

8.3.1 SQL Data Definition Language and Data Types

The main SQL command for data definition is the CREATE statement, which can be used to create schemas, tables (relations), and domains.

8.3.1.1 CREATE SCHEMA

An SQL schema includes an authorization identifier to indicate the user or account who owns the schema, as well as descriptors for each element in the schema. It is identified by a schema name. These elements include tables, views, and domains that describe the schema. The earliest version of SQL did not include the concept of a relational database schema. All tables are considered to be part of a single schema. In SQL2, a schema was first introduced intending to group tables that belong to the same database application together.

To create a schema, the syntax is

CREATE SCHEMA schemaname [AUTHORIZATION username] [schema _ element […]]

The schemaname is the name for the schema. The AUTHORIZATION command is optional and is used to specify the user, username, who owns the schema. The schema_ element statement declares an optional list of elements (e.g., table, view).

Example 8.1

Create a schema named “myschema”:

CREATE SCHEMA myschema;

Example 8.2

Create a schema for user Joe:

CREATE SCHEMA myschema AUTHORIZATION joe

154 Developing Windows-Based and Web-Enabled Information Systems

Note: These two examples illustrate creating simple schemas. We will discuss the syn- tax of adding element descriptors for the schema when we learn how to create views, domains, and tables.

8.3.1.2 CREATE VIEW

The CREATE VIEW statement creates a new database view, which is an SQL query stored in the database.

The syntax for creating a view is

CREATE VIEW viewname [(column-list)] AS query [WITH [CASCADED|LOCAL|] CHECK OPTION]

The viewname statement is the name for the new view. Additionally, the column-list dec- laration is an optional list of names for the columns of the view, separated by commas. The query statement is any SELECT statement without an ORDER BY clause. Additionally, the column-list code must have the same number of columns as the select list in the query. The optional WITH CHECK OPTION clause is a constraint on updatable views. It affects SQL INSERT and UPDATE statements. The CASCADED and LOCAL specifiers apply when the underlying table for query is another view. The CASCADED statement requests that the WITH CHECK OPTION clause applies to all underlying views (to any level). On the other hand, the LOCAL statement requests that the WITH CHECK OPTION applies only to the current view. The LOCAL statement is the default.

Example 8.3

Create a view element that lists which drugs cost more than $20:

CREATE VIEW myviews AS SELECT Name FROM drugs WHERE price > 20

Example 8.4

Create a schema authorized to Joe, and create the above view element within the schema:

CREATE SCHEMA myschema AUTHORIZATION joe CREATE VIEW myviews AS SELECT Name FROM drugs WHERE price > 20

8.3.1.3 CREATE DOMAIN

A domain essentially defines the data type of an attribute. These optional constraints apply the attribute to be restricted to only a specific set of values. In SQL, a variety of built- in domain data types are supported (see Table 8.2).

The syntax for creating a domain is

CREATE DOMAIN [AS] domainname datatype [[NOT] NULL] [DEFAULT default-value] [CHECK (condition)]

The domainname command is the identifier of the domain. The datatype is the built-in SQL data type chosen from Table 8.2. The NULL clause allows one to specify the nullability

155Structured Query Language

of the domain. If it is not explicitly specified, the allow_nulls_by_default option is used. Furthermore, the CHECK clause specifies integrity constraints, that is, which attributes of the domain must satisfy a user-defined condition. The following are two examples of creat- ing a domain and making it an element of the schema.

Example 8.5

Create a domain named “store,” which holds a 10-character string and may be NULL:

CREATE DOMAIN store CHAR(10) NULL;

Example 8.6

Create a schema authorized to Joe, and create the above domain element within the schema:

CREATE SCHEMA myschema AUTHORIZATION joe CREATE DOMAIN store CHAR(10) NULL

8.3.1.4 CREATE TABLE

Tables are the basic units of organization and the storage of data in a database. Each table represents an entity such as a list of stores, drug sales, or purchase orders. The CREATE TABLE statement is the SQL command that adds a new table to a relational database.

The basic syntax is

CREATE TABLE tableName {(columnname datatype) [NOT NULL][UNIQUE] [DEFAULT defaultoption] [CHECK (searchcondition)] [,…]} [PRIMARY KEY (listofcolumns),] [FOREIGN KEY (listofforeignkeycolumns) REFERENCES parenttablename [(listofcandidatekeycolumns)]

TABLE 8.2

Domain Data Types

Data Types Description

char (n) A fixed-length character string with user-specified length n varchar (n) A variable-length character string with user-specified maximum length n. Int An integer smallint A small integer numeric (p, d) A fixed-point number with user-specified precision. The number consists of p digits, and d

of the p digits are to the right of the decimal point. For example, numeric (3, 1) allows 23.9 to be stored, but neither 239.0 nor 0.239

Real Floating-point number double precision Double-precision floating-point numbers float (n) A floating-point number, with precision of at least n digits Date A calendar date containing a year, month, and day in the form of YYYY/MM/DD Time The time of a day, in hours, minutes, and seconds. It is in the form of HH:MM:SS

156 Developing Windows-Based and Web-Enabled Information Systems

Here, the tablename statement is the name for the new table. The columnname clause is required to list the names of the columns within a table. Commas are used to separate multiple columns. The datatype clause is similar to the declaration of the DOMAIN state- ment and is obtained from standard SQL data types (as shown in Table 8.2). The optional constraints NULL and UNIQUE may also apply. The optional DEFAULT constraint is used to set a default value of the column. The optional CHECK constraint is again used to limit the range of permissible values that can be added into the column. The optional PRIMARY KEY constraint specifies the list of columns in the table that forms the primary key, separated by commas. The optional FOREIGN KEY constraint is used to list the col- umns within the table as the foreign key, followed by the REFERENCES constraint, which specifies the other table where the foreign key columns are linked.

Example 8.7

Use CREATE TABLE statement to create three relations: Drugs, Stores, and Sales:

CREATE TABLE Drugs ( Drug_ID char (20) NOT NULL, Name varchar (255), Price numeric (10, 2), Description varchar (255), CHECK (Price > 0), PRIMARY KEY (Drug_ID))

CREATE TABLE Stores ( Store_ID char (20) NOT NULL, Name varchar (255), Phone_Number varchar (10), PRIMARY KEY (Store_ID))

CREATE TABLE Sales ( Transaction_ID char (20) NOT NULL, Drug_ID char (20), Quantity int DEFAULT 0, Store_ID char (20), PRIMARY KEY (ST_ID) FOREIGN KEY (Store_ID) REFERENCE Stores (Store_ID) FOREIGN KEY (Drug_ID) REFERENCE Drugs (Drug_ID))

Example 8.8

Add the Drugs table element to the myschema created in Example 8.2:

CREATE SCHEMA myschema AUTHORIZATION joe CREATE TABLE Drugs ( Drug_ID char (20) NOT NULL, Name varchar (255), Price numeric (10, 2), Description varchar (255), CHECK (Price > 0), PRIMARY KEY (Drug_ID))

Now, we have learned to create schema, table. We will learn how to interact (retrieve, modify) the data in the next section.

157Structured Query Language

8.3.2 SQL Data Manipulation Language: Data Queries

8.3.2.1 Basic Structure of the SELECT Command

A relational database consists of a collection of relations (tables), each of which is assigned with a unique name. Each relation has a structure similar to that presented in Chapter 5. To create a query or a report that spans tables and to take in a subset of attributes, the SQL SELECT-FROM clause is used.

The basic syntax when creating an SQL SELECT-FROM query is

SELECT [DISTINCT | ALL] {* | [columnexpression [AS newname]] [,…]} FROM tablename [alias] [, …] [WHERE condition] [GROUP BY columnlist] [HAVING condition] [ORDER BY columnlist]

The SELECT clause specifies the list of columns desired in the result of a query. The FROM clause specifies the table(s) to be used in the query. The optional WHERE clause filters the rows based on the defined conditions. Additionally, the optional GROUP BY clause forms groups of rows with the same column value, while the optional HAVING clause filters the group subject to some conditions. The optional ORDER BY clause speci- fies the order of the output, with the default being descending order.

Example 8.9

Query all information about drugs in the Drugs relation:

SELECT * FROM Drugs;

The result is a relation consisting of all four columns with four rows as follows:

Drug_ID Name Price Description T_001 Tylenol $17.99 100 325-mg tablets bottle T_003 Tylenol $8.99 T_002 Tri_Vitamin $14.99 1500-400-35 solution 50-mL bottle C_001 Claritin $35.99 10 tablets

Example 8.10

Query all the names of all drugs stored in the Drugs relation:

SELECT Name FROM Drugs;

The result of the query is a relation consisting of a single column with the heading Name as follows:

Name

Tylenol Tylenol

Tri_Vitamin Claritin

158 Developing Windows-Based and Web-Enabled Information Systems

As observed, of the four queried rows, the drug “Tylenol” shows up twice. This result is the same as running the query:

SELECT ALL Name FROM Drugs;

To eliminate duplicates, we insert the DISTINCT statement after the SELECT clause.

Example 8.11

Query all the names of all drugs in the Drugs relation with no duplicate records:

SELECT DISTINCT Name FROM Drugs;

The result is a relation consisting of a single column with the heading Name. Since DISTINCT eliminates duplicate results, three records will be returned as follows:

Name

Tylenol Tri_Vitamin

Claritin

Example 8.12

Query which drugs would cost less than $20:

SELECT Name FROM Drugs WHERE Price < 20;

A relation consisting of one column with heading Name, two records will be returned when executing the query:

Name

Tylenol Tri_Vitamin

Example 8.13

Query information about a specific set of drugs that costs between $10 and $20:

SELECT * FROM Drugs WHERE Price <= 20 and Price >= 10;

or

SELECT * FROM Drugs WHERE Price BETWEEN 20 AND 10;

The result of the query is a relation consisting of four columns, and only three records will be returned. Note that the BETWEEN and AND clauses are inclusive conditions.

Drug_ID Name Price Description

T_001 Tylenol $17.99 100 325-mg tablets bottle T_003 Tylenol $8.99 T_002 Tri_Vitamin $14.99 1500-400-35 solution 50-mL bottle

159Structured Query Language

Example 8.14

Find the price as well as the name of the drug with the characters “Tyle” in their name:

SELECT Price FROM Drugs WHERE Name LIKE "%Tyle%";

After executing the query, the result is a relation consisting of two columns with the headings Name and Price, and two records:

Name Price

Tylenol $17.99 Tylenol $8.99

SQL has two special pattern matching symbols: specifically, the “%” character is used for a sequence of zero or more characters. As an example, the statement Name LIKE “%Tyle%” means to return any rows in which the Name column contains the string “Tyle” of any length. On the other hand, in the case that it is known, the Name string starts with “Tyle” or ends with “Tyle,” the commands Name LIKE “Tyle%” or Name LIKE “%Tyle” can be used, respectively.

Note: The pattern matching symbol may vary from different databases. For example, Microsoft Access database uses “*” as the wildcard character, while MySQL and Oracle databases both use the “%” character as the pattern matching symbol.

Example 8.15

Query information of all drugs where a description has not been filled:

SELECT * FROM Drugs WHERE Description IS NULL;

The result is a relation consisting of four columns, and one record will be returned as follows:

Drug_ID Name Price Description

T_003 Tylenol $8.99

Example 8.16

Obtain a list of the details of all drugs, arranged in ascending order of price:

SELECT * FROM Drugs ORDER BY Price ASC;

The query result is a relation consisting of four columns and four records, which are listed in the order of price from low to high:

Drug_ID Name Price Description

T_003 Tylenol $8.99 T_002 Tri_Vitamin $14.99 1500-400-35 solution 50-mL bottle T_001 Tylenol $17.99 100 325-mg tablets bottle C_001 Claritin $35.99 10 tablets

160 Developing Windows-Based and Web-Enabled Information Systems

Example 8.17

Assume that the drug “Tylenol” is having a 20% discount. Query the information of “Tylenol” after the promotion:

SELECT Drug_ID, Name, Price*0.8, Description FROM Drugs WHERE Name = "Tylenol"

A relation that has the same four columns as the Drugs relation, except for the column Price multiplied by 0.8, is queried from the Drugs relation.

Since the new price is the promotion price, we can use the AS clause to rename the column as follows:

SELECT Drug_ID, Name, Price*0.8 AS NEW_Price, Description FROM Drugs WHERE "Name" = "Tylenol"

Drug_ID Name New_Price Description T_001 Tylenol $14.39 100 325-mg tablets bottle T_003 Tylenol $7.19

We have illustrated the use of most clauses in the example except for GROUP BY and HAVING. These two clauses are closely tied to aggregate functions, which will be discussed in the next section.

8.3.2.2 Aggregate Functions

An aggregate function is a function that takes on a collection of values from a column and returns a single value that is a function of the input values. Some commonly used aggre- gate functions are as follows:

• AVG( ) returns the average value • COUNT( ) returns the total number of rows • MAX( ) returns the maximum value • MIN( ) returns the smallest value • SUM( ) returns the total sum of the value • FIRST( ) returns the first value • LAST( ) returns the last value

The input to SUM and AVG must be a collection of numbers, but other operators can operate on a collection of non-numeric data types, such as string. To illustrate these func- tions, we have the following examples.

Example 8.18

Count the number of drugs that cost more than $5.99:

SELECT COUNT (*) FROM Drugs WHERE Price > 5.99;

The corresponding result is

Count

4

161Structured Query Language

Example 8.19

Count the number of unique drugs in the Drugs relation, not counting duplicates:

SELECT COUNT (DISTINCT Name) FROM Drugs;

The result is

Count

3

Example 8.20

Find the number of sales and the total sales in the Sales relation in the store with Store_ID = S_001:

SELECT COUNT (*) AS Count, SUM (Quantity) AS Sum FROM Drugs WHERE Stored_ID = 'S_001';

The result of the query is a relation that consists of two columns, in which the AS clause is used to rename the returned columns:

Count Sum

2 175

Example 8.21

Get the minimum, maximum, and average sales from the Sales relation:

SELECT MIN (Quantity) AS Min, MAX (Quantity) AS Max, AVG (Quantity) AS Avg FROM Sales;

The result is a relation that consists of three columns. Again, the AS clause is used to rename the returned columns as follows:

Min Max Avg

25 300 158.3

Example 8.22

Query the number of transactions in each store and their total sales quantities:

SELECT Store_ID, COUNT (Drug_ID) AS Count SUM (Quantity) AS Sum FROM Sales GROUP BY Store_ID ORDER BY Store_ID;

This query first groups the records in the Sales relation by store. Next, within each store, the number of sales and the total amount of sales are summarized using the aggregate functions COUNT and SUM. The result of this query is a relation with three columns, specifically, Store_ID, Count, and Sum, with the records sorted in ascending order by Stored_ID.

162 Developing Windows-Based and Web-Enabled Information Systems

Store_ID Count Sum

S_001 2 175 S_002 1 300

Example 8.23

For stores with more than one transaction, find the number of store transactions and the corresponding sum of the total sales quantities:

SELECT Store_ID, COUNT (Drug_ID) AS Count SUM (Quantity) AS Sum FROM Sales GROUP BY Store_ID HAVING COUNT (Drug_ID) >1;

The result is similar to the preceding example query, except that the HAVING clause filters the store with less than two transactions:

Store_ID Count Sum

S_001 2 175

By now, we have learned how to use SELECT-FROM in the main query. SQL provides a mechanism for nested queries as follows.

8.3.2.3 Nested Subqueries

A subquery is a SELECT-FROM-WHERE expression that is nested within another query, such as the SELECT, INSERT, UPDATE, and DELETE statements. The following example illustrates the use of subquery in a SELECT query.

Example 8.24

Find the phone number of the stores that have sold “Tylenol”:

SELECT Store_ID, Phone_Number FROM Stores WHERE Store_ID IN (SELECT Store_ID FROM Sales WHERE Drug_ID IN (SELECT Drug_ID FROM Drugs WHERE Name LIKE "%Tylenol%"));

This nested subquery has three layers. First, the most inner query selects the Drug_ID of drugs that contain “Tylenol” in their name. The query result is the set (T_001, T_003). The next query selects the Store_ID of the stores that sold drugs with the Drug_ID just retrieved. The set operator, IN, is used. Finally, the outer query selects both Store_ID and Phone_ Number from the Stores relation for the specific Store_ID. The final query result is as follows:

Store_ID Phone_Number

S_001 1-800-746-7287 S_002 1-800-925-4733

As seen from this example, a subquery is used to query across multiple tables. In some sense, this is similar to a JOIN command in SQL.

163Structured Query Language

8.3.2.4 JOIN Query

The SQL JOIN command is used to query data from two or more tables, which is based on the relationship between certain columns in these tables. There are four types of JOIN operations:

• JOIN: This will return rows when there is at least one matching instance in both tables. In some database systems, it is called “INNER JOIN.”

• LEFT JOIN: This will return all rows from the left table, even if there are no matches in the right table. Alternatively, the LEFT OUTER JOIN clause is used in some databases.

• RIGHT JOIN: This will return all rows from the right table, even if there are no matches in the left table. The alternative clause, RIGHT OUTER JOIN, is used in some databases.

• FULL JOIN: This will return rows when there is a match in one of the tables.

Let us use the following examples to illustrate the use of JOIN command.

Example 8.25

JOIN (INNER JOIN): Find the names of the drugs being sold:

SELECT Name FROM Drugs JOIN Sales ON Drugs.Drug_ID = Sales.Drug_ID;

The result is a relation with one column, the heading is Name:

Name

Tylenol Tylenol Claritin

Note: Both the Drugs and Sales relations have the column Drug_ID. To avoid ambi- guity, we need to explicitly specify where the Drug_ID column is queried from. Often, we can use the alias for each table. For example, the above query can be rewritten as follows:

SELECT d.Name FROM Drugs d JOIN Sales s ON d.Drug_ID = s.Drug_ID;

This is also equivalent to

SELECT d.Name FROM Drugs d, Sales s, WHERE d.Drug_ID = s.Drug_ID;

The use of the alias d indicates that the Name is queried from the Drugs relation. Otherwise, the relational database will throw the “ambiguous” error.

164 Developing Windows-Based and Web-Enabled Information Systems

Example 8.26

LEFT JOIN (LEFT OUTER JOIN): Find all the drugs and their sales information, if any:

SELECT d.*, s.Quantity, s.Store_ID, s.Transaction_ID FROM Drugs d LEFT JOIN Sales s ON d.Drug_ID = s.Drug_ID ORDER BY d.Drug_ID;

Given this query, we define Drugs as the left relation. Therefore, all the records from the Drugs relation will be retrieved. The result of the query is a relation that is com- posed of seven columns. For the records that have no corresponding record in the Sales relation, the NULL value applies. For the drugs with Drug_ID being T_002, T_003, and C_001, the Quantity, Store_ID, and Transaction_ID have NULL values.

Drug_ID Name Price Description Quantity Store_ID Transaction_ID

T_003 Tylenol $8.99 NULL NULL NULL

T_002 Tri_Vitamin $14.99 1500-400-35 solution 50-mL bottle

NULL NULL NULL

T_001 Tylenol $17.99 100 325-mg tablets bottle 25 S_001 ST_001 T_001 Tylenol $17.99 100 325-mg tablets bottle 300 S_002 ST_002 C_001 Claritin $35.99 10 tablets NULL NULL NULL

Example 8.27

RIGHT JOIN (RIGHT OUTER JOIN): Find all the sales and their drug information, if any:

SELECT s.Transaction_ID, s.Store_ID, s.Quantity, d.* FROM Sales s RIGHT JOIN Drugs d ON s.Drug_ID = d.Drug_ID

Similar to the LEFT JOIN clause, the returned query for the drug with the ID being C_002 has NULL values.

Transaction_ID Quantity Store_ID Drug_ID Name Price Description

ST_001 25 S_001 T_001 Tylenol $17.99 100 325-mg tablets bottle ST_002 300 S_002 T_001 Tylenol $17.99 100 325-mg tablets bottle ST_003 150 S_001 C_002 NULL NULL NULL

Example 8.28

FULL JOIN: Query all drug details as well as their sales information per store:

SELECT d.*, s.Quantity, s.Store_ID, s.Transaction_ID FROM Drugs d FULL JOIN Sales s ON d.Drug_ID = s.Drug_ID;

The FULL JOIN statement returns all the records from the left relation (Drugs) as well as all the rows from the right relation (Sales). If there are records in Drugs that do

165Structured Query Language

not have matches in the Sales relation, or if there are records in Sales that do not have matches in the Drugs relation, these records will be listed with NULL for the respective columns. The result of the query is shown as follows:

Drug_ID Name Price Description Quantity Store_ID Transaction_ID

T_003 Tylenol $8.99 NULL NULL NULL T_002 Tri_Vitamin $14.99 1500-400-35 solution

50-mL bottle NULL NULL NULL

T_001 Tylenol $17.99 100 325-mg tablets bottle

25 S_001 ST_001

T_001 Tylenol $17.99 100 325-mg tablets bottle

300 S_002 ST_002

C_001 Claritin $35.99 10 tablets NULL NULL NULL C_002 NULL NULL NULL 150 S_001 ST_003

8.3.3 SQL Data Manipulation Language: Data Modification

Up until now, we have focused on the extraction of existing information from a data- base. In this section, we will learn how to add, remove, or change information using SQL commands.

8.3.3.1 INSERT

To insert data into a relation, we either specify a record to be inserted or write a query whose result is a set of records to be inserted.

The basic INSERT syntax is

INSERT INTO TableName [(columnList)] VALUES (dataValueList)

The optional columnlist statement specifies the list of columns within the table where the new records will be inserted. If it is not specified, SQL assumes that the list is in their original order specified in the CREATE TABLE command. Any missing data for specific columns must be declared as NULL, unless a DEFAULT value is specified for the column. The dataValueList statement is required to match the columnlist statement: the number of items in each list must be the same, the position of each item in each list must be the same, and data types in both lists must be matched perfectly.

Example 8.29

Add a new row into the Sales table when supplying data for all columns:

INSERT INTO Sales VALUES ('ST_006,' 'C_002,' 'S_003,' 140);

This query inserts a new record into the Sales relation:

Transaction_ID Drug_ID Quantity Store_ID

ST_006 C_002 140 S_003

166 Developing Windows-Based and Web-Enabled Information Systems

The updated Sales relation now has the following information (the new record is highlighted in bold):

Transaction_ID Drug_ID Quantity Store_ID

ST_001 T_001 25 S_001 ST_002 T_001 300 S_002 ST_003 C_002 150 S_001 ST_006 C_002 140 S_003

Example 8.30

Augment a new row into the Sales table, only supplying data for all mandatory columns:

INSERT INTO Sales (ST_ID) VALUES ('ST_006');

or

INSERT INTO Sales VALUES ('ST_006,' NULL, NULL, 0);

The query inserts a new record to the Sales relation, which has NULL for the Drug_ID and the Store_ID columns and 0 (default value) for the Quantity column as follows:

Transaction_ID Drug_ID Quantity Store_ID

ST_006 NULL 0 NULL

The updated Sales relation now has the following information (the new record is highlighted in bold):

Transaction_ID Drug_ID Quantity Store_ID

ST_001 T_001 25 S_001 ST_002 T_001 300 S_002 ST_003 C_002 150 S_001 ST_006 NULL 0 NULL

Example 8.31

Add a new row into the Drugs table for a new transaction, with the ST_ID being ST_003:

INSERT INTO Drugs SELECT Drug_ID FROM Sales WHERE ST_ID = "ST_003"

This query inserts a new record to the Drugs relation as follows:

Drug_ID Name Price Description

C_002 NULL NULL NULL

167Structured Query Language

The updated Drugs relation now has the following information (the new record is highlighted in bold):

Drug_ID Name Price Description

T_001 Tylenol $17.99 100 325-mg tablets bottle T_003 Tylenol $8.99 T_002 Tri_Vitamin $14.99 1500-400-35 solution 50-mL bottle C_001 Claritin $35.99 10 tablets C_002 NULL NULL NULL

8.3.3.2 UPDATE

In some situations, we may want to change a value in a record without changing all values in the record. For this purpose, the UPDATE statement can be used.

The syntax of the UPDATE statement is

UPDATE TableName SET columnName1 = dataValue1 [, columnName2 = dataValue2...] [WHERE searchcondition]

The TableName can be the name of the base table or an updatable view. The SET clause specifies the names of one or more columns that are to be updated with a new dataValue. The dataValue must be compatible with the data type for the corresponding column. The optional WHERE clause is used to filter the records to be updated based on the searchCon- dition. If the WHERE clause is omitted, all the records in the table will be updated.

Example 8.32

Give all drugs a 15% discount:

UPDATE Drugs SET Price = Price*0.85;

Example 8.33

Apply a 15% discount to “Tylenol” drugs only:

UPDATE Drugs SET Price = Price*0.85 WHERE Name = "Tylenol"

Similar to INSERT, a SELECT-FROM subquery may apply to specify the records to be updated.

8.3.3.3 DELETE

The DELETE statement is used to delete the whole record instead of some values of any particular column.

The DELETE syntax is applied as

DELETE FROM TableName [WHERE searchCondition]

168 Developing Windows-Based and Web-Enabled Information Systems

The TableName is the name of the base table or an updatable view in which the records are to be deleted. Similar to the UPDATE clause, the optional WHERE clause is used to filter the records to be deleted based on the conditions. If it is omitted, then all the records in the relation will be removed. However, the relation still exists in the database. To remove the relation, the DROP statement is used, which is discussed in the next section.

Example 8.34

Delete all the records in the Sales relation:

DELETE FROM Sales

Example 8.35

Delete all sales with the Drug_ID being “C_002”:

DELETE FROM Sales WHERE Drug_ID = "C_002"

8.3.4 SQL Data Manipulation Language: Relation Modification

In a database, the created relation can be changed or removed using the ALTER and DROP statements.

8.3.4.1 ALTER TABLE

The SQL ALTER TABLE statement is used to add, edit, or modify a column from an exist- ing table.

The basic syntax is

ALTER TABLE TableName [ADD columnname datatype [NOT NULL][UNIQUE]] [DROP columnname] [ADD [constraint]] [DROP [constraint]] [ALTER [column] SET DEFAULT DefaultOption] [ALTER [column] DROP DEFAULT]

The TableName is the table to be modified. This statement includes multiple optional statements, specifically, the ADD column with a specified datatype is to add a new column, the DROP statement is to drop the existing column, the ADD/DROP constraint (e.g., pri- mary key, foreign key) are add or remove key relations, and the ALTER statement is to change the default value for the column.

Here we only show one simple example as follows:

Example 8.36

Add a column named “Transaction_Date” in the Sales relation:

ALTER TABLE Sales ADD Transaction_Date date

169Structured Query Language

8.3.4.2 DROP TABLE

Relations can easily be removed with the DROP statement. The simple syntax is

DROP TABLE TableName;

The DROP statement can also be used to remove the entire database as

DROP DATABSE DatabaseName;

In this section, we only briefly illustrate the relation modification. For advanced developer, please refer to specific database manual for comprehensive discussions on this topic.

8.4 Summary

This chapter reviews the history of SQL. An important notation technique, BNF, and its extension EBNF are introduced. Following the notation, basic SQL DDL statements are explained, including CREATE SCHEME, CREATE VIEW, CREATE DOMAIN, and CREATE TABLE. In addition, DML syntax, including the SELECT-FROM-WHERE clause, INSERT, UPDATE, DELETE, INSERT, DROP, ALTER, CREATE, and JOIN, are explained with examples. The use of aggregate function within the SELECT statement and the use of nest subqueries are also discussed with illustration examples.

EXERCISES

Review Exercises

1. SQL was initially developed by the World Wide Web Consortium (W3C). True or false?

2. DDL is used to modify tables within a database schema. True or false? 3. The SELECT statement is an example of a DML. True or false? 4. The referential integrity constraints for a database are manipulated by the TCL. 5. The {} character denotes a required element in the EBNF. 6. The following database elements that are modified by CREATE statements are: a. Domains b. Schemas c. Tables d. All of the above e. None of the above 7. Which of the following is a valid declaration to authorize username “Chandler” to

access the Sales schema? a. CREATE AUTHORIZATION chandler; b. CREATE SCHEMA sales; c. CREATE SCHEMA sales AUTHORIZE chandler;

170 Developing Windows-Based and Web-Enabled Information Systems

d. INSERT AUTHORIZATION chandler; e. None of the above 8. Which of the following is not an optional statement in the CREATE VIEW clause? a. Query b. columnlist c. CHECK d. CASCADED e. LOCAL 9. Which of the following CREATE VIEW SQL code is not valid? a. CREATE VIEW AS SELECT * FROM Book_ Authors WITH CHECK OPTION; b. CREATE VIEW Author_ M AS SELECT * FROM Book_ Authors; c. CREATE VIEW Author_M FROM Book_Authors WITH LOCAL CHECK OPTION; d. CREATE VIEW AS SELECT * FROM Book_ Authors WITH CASCADED CHECK

OPTION; e. All of the above are valid. 10. A domain essentially declares the data type of the entire table. True or false? 11. Declare a CREATE DOMAIN command to store the prices of computer spare

parts. The price must be positive and can have, at most, two digits to the right and, at most, four digits to the left of the decimal point.

12. Declare a CREATE DOMAIN command to store the type of blood code named “Blood_Type_C.” The type of blood code is a three-character field with default type “O+” and should not be NULL.

13. The CREATE TABLE statement is used to create new tables to hold entities within a database. True or false?

14. Suppose that we want to create a “Blood Donors” table that contains the “Blood_ Type_C” domain from Review Exercise 12. To create a table with an attribute with a domain as a reference, we declare:

a. CREATE TABLE Blood_Donors (Donor_ M varchar(255), Blood_Type DOMAIN); b. CREATE TABLE Blood_Donors (Donor_M varchar(255), Blood_Type Blood_ Type_C); c. CREATE TABLE Blood_Donors (Donor_ M varchar(255), Blood_Type Varchar(3)

NOT NULL FOREIGN KEY(Blood_Type) REFERENCE Blood_Type_C(Blood_Type)); d. CREATE TABLE Blood_Donors (Donor_M varchar(255), Blood_Type DOMAIN

Varchar(3)); e. CREATE TABLE Blood_Donors (Donor_ M varchar(255), Blood_Type AS Blood_

Type_C); 15. The SELECT ALL FROM Table_name; is equivalent to a. SELECT ALL COLUMNS FROM Table_name; b. SELECT * FROM Table_name; c. SELECT FROM Table_name; d. FROM Table_name SELECT; e. None of the above. SELECT ALL FROM Table_name; is an erroneous statement.

171Structured Query Language

16. The SELECT DISTINCT statement almost always returns less number of records as compared to the SELECT ALL statement. True or false?

17. The BETWEEN and statements are equivalent to the WHERE clause using the less than “<” and the greater than “>” commands. True or false?

18. Given the LIKE “Fac%” statement, the following column values will be returned except:

a. Faculty b. Fac c. Fa d. All of the following will be returned. e. None of the following will be returned. 19. Given the LIKE “%201_” statement, the following column values will be returned

except: a. Code2013 b. 201334 c. 2013 d. All of the following will be returned. e. None of the following will be returned. 20. There is no need to declare the ORDER BY clause as ASC (ascending). True or

false? 21. Given a table named Faculty with Faculty_ M as the faculty name and Faculty_R as

the faculty rank, write a query that will return an alphabetical listing of faculty members sorted by descending alphabetical faculty rank then by ascending fac- ulty name.

22. Given the following JOINs on two tables with much more left table instances as compared to right table instances, which JOIN maximizes the COUNT statement?

a. LEFT JOIN b. FULL JOIN c. JOIN d. RIGHT JOIN e. All join statements would return the same count. 23. Given the following JOINs on two tables with many more left table instances as

compared to right table instances, which JOIN minimizes the AVG statement? a. LEFT JOIN b. FULL JOIN c. JOIN d. RIGHT JOIN e. All join statements would return the same count. 24. The following are valid aggregate functions except: a. AVG(Sales_Price) b. LAST(Name_Code)

172 Developing Windows-Based and Web-Enabled Information Systems

c. COUNT(Sales_Price) d. SUM(Supplier_Address) e. All of the above are valid aggregate functions. 25. Suppose that a table named “Goods_Receipt” with “Receipt_N” as the goods

receipt number and “Receipt_D” as the receipt data created in that order, where receipt number is numeric and is not null, is given. Which of the following is a valid INSERT statement?

a. INSERT INTO Goods_Receipt VALUES(‘10-30-2013’, ‘75774’) b. INSERT INTO Goods_Receipt(Receipt_D, Receipt_N) VALUES(‘10-30-2013’, ‘75774’) c. INSERT INTO VALUES(‘10-30-2013’, ‘75774’) WHERE VALUE = Goods_Receipt d. INSERT INTO VALUES(‘10-30-2013’, ‘75774’) WHERE Goods_Receipt(Receipt_D,

Receipt_N) e. None of the above 26. You can update a non-existent instance. True or false? 27. Given a table where column_1 is required and column_2 is optional, you can use

the DELETE command to delete the value of column_2. True or false?

Practice Exercises

For practice questions 1 to 6, consider the following dataset:

Faculty Fac_N Fac_First_M Fac_Last_M Fac_Rank_C 13443 Iris Martinez ASCP 66234 Aura Matias PROF 45463 Lorelie Grepo ASTP 44556 Mickey Mancenido ASTP

Courses Course_C Course_M Course_Units IDOE Design of Experiments 3 IBDM Business Intelligence and Data Mining 3 MSCM Supply Chain Management 3 SRES Research Methods 1 SSIM Simulation 4

Faculty_Courses Course_C Fac_N Stud_Count Sem_C MSCM 13443 34 2012-1 MSCM 66234 32 2012-2 IDOE 44556 56 2012-1 SRES 13443 12 2012-1 IBDM 66234 40 2012-2 MSCM 44556 32 2013-1

173Structured Query Language

1. Create a query to count the number of students that took Supply Chain Management. Name the column “Student_Count.”

2. Create a query to handle the following desired query result:

Fac_Last_M Num_Students

Mancenido 88 Martinez 46 Matias 72

3. Create a query for the following report:

Course Description Rank Last Name Semester Units

Business Intelligence and Data Mining Prof Matias 2012-2 3 Design of Experiments ASTP Mancenido 2012-1 3 Research Methods ASCP Martinez 2012-1 1 Supply Chain Management PROF Matias 2012-2 3 Supply Chain Management ASCP Martinez 2012-1 3 Supply Chain Management ASCP Mancenido 2013-1 3

4. The report in question 3 seems to lack the actual description of the faculty rank. Create a table named “Rank_Description” with the attributes: • Fac_Rank_C: A four-character attribute with default value INST and cannot be

NULL. Additionally, this is the primary key of the table. • Fac_Rank_ M: A variable character attribute that describes the rank. • Min_Sal_Grade: A numeric 1-digit number that can take values from 1 to 9,

indicating the minimum salary grade of the rank.

5. Write an SQL command to populate the table in question 4 with the following data:

Fac_Rank_C Fac_Rank_M Min_Sal_Grade

INST Instructor 1 ASTP Assistant Professor 3 ASCP Associate Professor 5 PROF Professor 7 UPRF University Professor 9

6. Write a nested query without join statements for the following report:

First Name Last Name Rank

Iris Martinez Associate Professor Aura Matias Professor Lorelie Grepo Assistant Professor Mickey Mancenido Assistant Professor

174 Developing Windows-Based and Web-Enabled Information Systems

For questions 7 to 12, consider the following tables:

Purchase_Orders PO_N PO_D Supplier_N PO_Pay_Terms 610557 2/27/2013 1335 NET30 610558 2/27/2013 2652 2/10NET20 610559 2/27/2013 1335 COD 610560 2/28/2013 1226 2/10NET20 610561 3/01/2013 2652 2/10NET20

Purchase_Order_Items PO_N Item_N Item_Q Item_Price 610557 36796 15 664.25 610557 36224 21 224.54 610559 36624 100 0.65 610560 36547 1 10887.10 610561 36869 224 336.65

7. The PO_Pay_Terms will be used in different tables; hence, a domain could be use- ful to handle this in the event of future changes. This attribute must contain a maximum of 10 characters; the default value should be “NET30,” which corre- sponds to net 30 days; and should not be NULL. Write an SQL code to create a domain named “Pay_Terms” subject to the aforementioned specifications.

8. Create a “Suppliers” table with the following attributes: • Supplier_N: A four integer–digit primary key attribute used to uniquely iden-

tify a supplier and should not be NULL. • Supplier_ M: A string that denotes the name of a supplier with a maximum of

255 characters. • Supplier_Tier: A single-digit number between 1 and 5 indicating the supplier’s tier. • Default_Pay_Terms: An attribute that follows the domain “Pay_Terms” defined

in question 6.

9. Populate the table generated in question 8 with the following information:

Supplier_N Supplier_M Supplier_Tier Default_Pay_Terms

1335 MacLaren’s Irish Pub 4 NET30 2652 Central Perk 3 NET30 1226 Beltway Coffee Unknown NET30

175Structured Query Language

10. Create a query to for this report:

Supplier_M PO_N Item_N Item_Q MacLaren’s Irish Pub 610557 36796 15 MacLaren’s Irish Pub 610557 36224 21 Central Perk 610558 NULL NULL MacLaren’s Irish Pub 610559 36624 100 Beltway Coffee 610560 36547 1 Central Perk 610561 36869 224

11. A round of supplier reviews has lapsed, and the Beltway Coffee supplier’s master data need to be updated (see bold font).

Supplier_N Supplier_M Supplier_Tier Default_Pay_Terms 1335 MacLaren’s Irish Pub 4 NET30 2652 Central Perk 3 NET30 1226 Beltway Coffee 3 2/10NET20

Create a single query to update the record. 12. A new attribute is needed for the Purchase_Orders relation to denote whether the

purchase order instance is paid. Write the corresponding SQL code to add the fol- lowing attributes: • Pay_Status: A three-character string indicating the status of the purchase order

payment, with the default value “NYP” indicating not yet paid.

177

9 MySQL

Most relational database systems (RDBSs) adopt the basic syntax of Structured Query Language (SQL) data definition language (DDL) and data manipulation language (DML) discussed in Chapter 8, but may have some minor variations. In this chapter, a widely used open-source database system, MySQL, is introduced. First, the use of DDL and DML in MySQL is illustrated, followed by the discussions on the use of transaction control lan- guage (TCL) and data control language (DCL). Some unique features of MySQL, specifi- cally, the utilities, the tools, and some SQL variations of MySQL, are also introduced. Since MySQL products are enhanced regularly, this chapter only covers the key aspects of this database system. When using a particular system version, it is recommended to consult the user manuals available online (http://www.mysql.com) for specific details.

9.1 Introduction

Currently, the most popular RDB management systems are Oracle (Oracle Corporation), SQL Server (Microsoft), DB2 (IBM), and MySQL (Oracle Corporation). As an open-source database server, MySQL is on the top of the list chosen by the users who are looking for a free or inexpen- sive database management system. Some notable advantages of MySQL include the following:

• Ease of use. MySQL is a high-performance but relatively simple database system and is much less complicated to set up and administer than larger systems such as Oracle and DB2.

• MySQL supports the SQL standard, which is the language of choice of all modern RDBS systems.

• Capability. In MySQL, many clients can connect to the server at the same time. Furthermore, developers can access MySQL interactively using several interfaces to enter queries and view the results. Examples of these interfaces are command- line clients, Web browsers, or X Window system clients. In addition, a variety of interfaces are available for programming languages, such as Perl and Java, just to name a few. MySQL also provides a number of connectors to support open database connectivity (ODBC), Java database connectivity (JDBC), and application platforms (e.g., .NET).

• Connectivity and security. MySQL is fully networked, and the database can be accessed from anywhere using the telecommunication network. MySQL has access control to ensure security issues. In addition, it supports encrypted connec- tions using the AQ protocol.

• Multi-platform. MySQL runs on Unix, Windows, Linux, and Mac.

178 Developing Windows-Based and Web-Enabled Information Systems

• Small size. MySQL has a modest distribution size, especially compared to the large disk space footprint of some commercial database systems.

• Availability and cost. MySQL was initially developed by Sun Microsystems as an open-source RDBS and was later acquired by Oracle Corporation. Currently, Oracle Corporation still supports MySQL as an open-source software, thus mak- ing it freely available under the terms of the general public license (GPL).

In the remaining of the chapter, we will explain how to work with MySQL.

9.2 Get Ready to Work with MySQL

Although MySQL runs across multiple operating systems (OSs), not all systems are equally good for supporting MySQL. To learn more about the supported systems, please check out the MySQL website (http://www.mysql.com).

Currently, some general MySQL tools are available as open distributions from the website:

• MySQL Community Server is a freely downloadable version of the open-source database. The supported platforms include Windows (×86, 32-bit, and 64-bit), Mac OS X, Linux, and Sun Solaris. There are different formats for download. It is rec- ommended to download the binary distribution for easy installation. Binary dis- tribution is also tuned for optimal and stable MySQL performance.

• MySQL Cluster is a real-time transactional database designed to serve intensive read and write workloads under high throughput conditions. It is available in both open-source and commercial edition versions.

• MySQL Workbench is a visual database design application that can be used to design, manage, and document database schema. It is also available in both open- source and commercial variants.

• MySQL Connectors is a list of available-for-download database drivers compatible with the ODBC and JDBC industry standards. • ODBC Connector is a standardized database driver for Windows, Linux, Mac

OS, and Unix platforms. • .NET Connector is a standardized database driver for .NET platforms and

development. • J Connector is a standardized database driver for Java platforms and development. • C++ Connector is a standardized database driver for C++ development. • Python Connector is a standardized database driver for Python platforms and

development. • C Connector is a client library for C development (libmysql).

• MySQL Installer is a package that includes most MySQL tools such as MySQL Server, Workbench, Connectors, and MySQL for Excel and Visual Studio. However, unlike the tools listed above, the MySQL installer currently runs on Windows- based systems only. Note: Although the installer is 32-bit, it will run on both 32- and 64-bit binary OS versions.

179MySQL

Other than open distributions, MySQL has some commercial tools such as the MySQL Enterprise Edition and the MySQL Cluster Carrier Grade Edition (CGE).

In this chapter, the installation and configuration of MySQL using the MySQL installer for Windows is mainly illustrated. For Mac OS X, only the installation procedure of the MySQL Community Server is explained. Since the Workbench installation for Mac OS X is similar to Windows, we will not discuss it in detail. For learning purposes, the use of the command-line clients to practice SQL commands and the use of the Workbench tool to handle the database management are recommended.

9.2.1 MySQL Installation on Windows

To run MySQL on Windows systems, the Windows OS needs to be 2000, XP, Vista, Server 2008, Windows 7, Windows 8, or newer. Generally, one should install MySQL in Windows using an account that has administrator rights to have full access to the system environ- ment. In addition, the system should support Transmission Control Protocol/Internet Protocol (TCP/IP) to run the MySQL Server.

There are different ways to install MySQL. New users are recommended to use the MySQL Installer, which provides wizards to install and configure the MySQL Server on Windows. The detailed procedure is outlined as follows:

Step 1: Download and launch the MySQL Installer. The process for starting the wiz- ard depends on the content of the installation package being downloaded. If there is a setup.exe or .msi file present, just double-click on it to start the installation process.

Step 2: Once the installation is initiated, the welcome screen is shown, as seen in Figure 9.1. There are three options: Install MySQL Products, About MySQL, and

FIGURE 9.1 MySQL Installer Welcome screen.

180 Developing Windows-Based and Web-Enabled Information Systems

Resources. If the Installer detects some MySQL components (e.g., connectors) available in the system, the installation will be launched as an update procedure.

To Install MySQL products for the first time, it is required to accept the license agreement before launching the installation process.

Step 3: Next, the setup type needs to be determined based on the installation prefer- ence. For the purposes of this chapter, we may choose the “Developer Default” setting. This includes MySQL Server, MySQL Workbench, MySQL Visual Studio plugin, MySQL Connectors, examples, and tutorial. In addition, the installation and data paths are defined here. The MySQL Installer will check the system for necessary external requirements (e.g., .NET framework for Visual Studio sup- portability) and then download and install missing components onto the system (Figure 9.2).

Step 4: Next, a comprehensive list of components to be installed will be popped up, as seen in Figure 9.3. Click Execute. This will start the installation.

Step 5: After the installation, the configuration will be initiated (Figure 9.4). This is to enable the TCP/IP networking component and specify the port number for the communication. If enabled, the MySQL client can access the server via the Internet. For example, let the IP address of the server be 149.169.43.22; the network-enabled communication is then set up over 149.169.43.22:3306 (assuming that the default port number 3306 is used). Otherwise, only localhost connections are allowed. The “Advanced Configuration” option provides additional logging options to configure. This includes defining file paths for the error log, general log, etc. (Figure 9.5).

FIGURE 9.2 MySQL Installer: Choosing a Setup Type.

181MySQL

FIGURE 9.3 MySQL Installer: Installation Progress.

FIGURE 9.4 MySQL Installer: MySQL Server Configuration 1/3.

182 Developing Windows-Based and Web-Enabled Information Systems

Step 6: Next, the password is required, with the username being “root” by default. At this point, it is optional to create additional users. There are several differ- ent pre-defined user roles that each have different permissions. For example, the “Backup Admin” user role has minimal rights needed to back up any database, the “DB Admin” role grants the rights to perform all tasks, and the “DB Designer” only has the right to create and reverse-engineer any database schema, although anonymous accounts can be created at this step, but it is not recommended as it may lead to some security issues.

Step 7: The last step is to configure the Windows Service name, how the MySQL Server should be loaded at startup, and how the Windows Service for the MySQL Server will be run (see Figure 9.6).

After the installation, it is recommended to test the MySQL Server to ensure that the server is up and running, the user is able to connect to the server, and information can be retrieved from the server. One simple way for testing is to use the MySQL command line client. As seen in Figure 9.7, the prompt window asks for the password (the username being “root”).

Upon successful login, type the command “show databases;” to see what databases exist in the MySQL Server.

The list of installed databases may vary, but it will always include the mysql and the information_schema databases. The mysql database includes useful information such as  help topics and general logs related to the MySQL Server. The information_schema database stores information about all other databases that the MySQL Server maintains. These databases contain read-only tables; thus, INSERT, UPDATE, or DELETE operations do not apply.

FIGURE 9.5 MySQL Installer: MySQL Server Configuration 2/3.

183MySQL

9.2.2 MySQL Installation on Mac OS X

Unlike MySQL for Windows, the MySQL Community Server and the MySQL Workbench need to be downloaded and installed separately for a Mac OS. The open distributions for the MySQL Server are available in two different forms: the native package installer format and the tar package format. In this section, we will learn about the use of the native pack- age installer, which uses the native Mac OS X installer to walk through the installation. Note that administrator privileges are required to perform the installation.

Step 1: Download and open the MySQL package installer, which is provided on a disk image (.dmg). This includes the main MySQL installation package, the MySQLStartupItem.pkg installation package, and the MySQL.prefPane. Double- click the disk image to open it.

Step 2: Double-click the MySQL Installer package. It will be named according to the version of MySQL being downloaded.

FIGURE 9.6 MySQL Installer: MySQL Server Configuration 3/3.

FIGURE 9.7 MySQL Command Line Client login.

184 Developing Windows-Based and Web-Enabled Information Systems

Step 3: A pop-up window will show up, indicating the start of the installation. Step 4: Next, a copy of the installation instructions and other important information

relevant to the installation are displayed. Step 5: Since the MySQL Community Server is downloaded and being installed, a

copy of the relevant GNU General Public License (GPL) is then displayed. Step 6: Next, select the drive to use to install the MySQL startup item. The drive must

have a valid, bootable, Mac OS X installed. Step 7: In the last step, some details of the installation, including the space required

for the installation, need to be confirmed before the installation launches.

During the installation process, a symbolic link from usr/local/mysql to the version/ platform specific directory will be created automatically. The installation of MySQL Workbench for Mac OS X follows the same procedure as that of MySQL Community Server.

9.3 Working with MySQL Command Line Client

To better practice SQL command, it is recommended to start with MySQL Command Line Client. Note that, in MySQL, each command needs to end with a semicolon. The system recognizes this character and triggers the execution of the code entered before it. In the following sections, the database example discussed in Chapter 8 will be used to illustrate the MySQL statements (Figure 9.8).

Before we demonstrate the use of MySQL, we will first understand the data types used in MySQL which is discussed in the following.

9.3.1 Data Types

MySQL supports standard data types, including numbers, string values, temporal values such as dates and times, and NULL values (some are defined in Table 8.2) with some variations.

9.3.1.1 Numeric Data Types

MySQL specifies numbers as integers (with no fractional part) or floating point values (with a fractional part). Integers can be specified in decimal or hexadecimal format. The general numeric types supported by MySQL are summarized in Table 9.1.

FIGURE 9.8 MySQL Command Line Client: show databases.

185MySQL

9.3.1.2 String (Character) Data Type

A string is a sequence of character values, such as ‘Phoenix,’ or ‘patient seeks treatment.’ Either single or double quotes can be used to surround a string value. However, the American National Standards Institute (ANSI) SQL standard specifies single quotes, so statements written using single quotes would be more portable to other database engines. The string character types are summarized in Table 9.2. There are several escape sequences within strings, and these can be used to indicate special characters (see Table 9.3). Each sequence begins with a backslash character (\) to signify a temporary escape from the usual rules for character interpretation. Note that a NUL byte is not the same as the NULL value; NUL is a zero-valued byte, whereas NULL is the absence of a value. We will discuss more on NULL in MySQL in the later sections of this chapter.

9.3.1.3 Date and Time (Temporal) Data Types

Dates and times are values such as '2013-07-15' or '17:34:43'. MySQL also supports com- bined date/time values, such as '2013-07-15 17:34:43'. Take special note of the fact that

TABLE 9.1

Numeric Types and Ranges in MySQL

Type Meaning Range

tinyint A very small integer Signed values: −128 to 127 (−27 to 27–1) smallint A small integer Unsigned values: 0 to 255 (0 to 28–1) mediumint A medium-sized integer Signed values: −32,768 to 32,767 (−215 to 215–1) int A standard integer Unsigned values: 0 to 65,535 (0 to 216–1) bigint A large integer Signed values: −8,388,608 to 8,388,607 (−223 to 223–1) float (m, d)

A single-precision floating-point number with M digits and D decimals

Unsigned values: 0 to 16,777,215 (0 to 224–1)

double (m, d)

A double-precision floating-point number with M digits and D decimals

Signed values: −2,147,683,648 to 2,147,483,647 (−231 to 231–1)

decimal (m, d)

A floating-point number with M digits and D decimals

Unsigned values: 0 to 4,294,967,295 (0 to 232–1)

TABLE 9.2

String (Character) Type and Maximum Size

Type Meaning Maximum Size

char (m) A fixed-length character string M bytes varchar (m) A variable-length character string M bytes tinyblob A very small binary large object (BLOB) 28–1 bytes blob A small BLOB 216–1 bytes mediumblob A medium-sized BLOB 224–1 bytes longblob A large BLOB 232–1 bytes tinytext A very small text string 28–1 bytes text A small text string 216–1 bytes mediumtext A medium-sized text string 224–1 bytes longtext A large text string 232–1 bytes enum ('value1'...)

An enumeration; column values may be assigned one enumeration member

65,525 members

set ('value1'...) A set; column values may be assigned multiple set members 64 members

186 Developing Windows-Based and Web-Enabled Information Systems

MySQL follows the ANSI SQL standard (known as the ISO 8601 format) and represents Sdates in year-month-day order, which is YYYY-MM-DD. Although we can display date val- ues any way we want by using the DATE _ FORMAT( ) function, the default display format lists the year first, and input values must be specified with the year first (Table 9.4).

9.3.1.4 NULL Value

The NULL data type is a “typeless” value and is usually used to mean “no value,” “unknown value,” “missing value,” “out of range,” “not applicable,” “none of the above,” and so on. The user can insert NULL values into tables, retrieve them from tables, and test whether a value is NULL. However, the user cannot perform arithmetic operations on NULL values; if one tries, the result is still NULL.

9.3.2 Practice Data Definition Language in MySQL

9.3.2.1 Statements for Database Operations

MySQL provides several database-level statements: USE for selecting a default database, CREATE DATABASE for creating databases, DROP DATABASE for removing them, and ALTER DATABASE for modifying global database characteristics.

9.3.2.1.1 Creating A Database

The CREATE DATABASE statement is used to create a database, with syntax as follows:

TABLE 9.3

String Escape Sequences

Sequence Meaning

\0 NUL (ASCII 0) \' Single quote \" Double quote \b Backspace \n Newline (linefeed) \r Carriage return \t Tab \\ Backslash \Z Ctrl-Z (Windows EOF character)

TABLE 9.4

Date and Time

Type Meaning Range

date A date value, in “YYYY-MM-DD” format “1000-01-01” to “9999-12-31” time A time value, in “hh:mm:ss” format “−838:59:59” to “838:59:59” datetime A date and time value, in “YYYY-MM-DD

hh:mm:ss” format “1000-01-01 00:00:00” to “9999-12-31 23:59:59”

timestamp A timestamp value, in YYYYMMDDhhmmss format

19700101000000 to some time in the year 2037

year A year value, in YYYY format 1901 to 2155 for YEAR(4)

187MySQL

CREATE DATABASE db_name;

When creating a database, the following constraints must be considered: the name must conform to standards, the database must not already exist, and the creator must have suf- ficient privileges to create it.

Example 9.1

Create a database named healthcaresystem:

9.3.2.1.2 Selecting A Database

The USE statement selects a database to make it the default (current) database for a given connection to the server. The user must have access privileges for the database in question; otherwise, the user cannot select it. The syntax for selecting a database is

USE db_name;

Example 9.2

Choose the database named healthcaresystem:

Note: When a connection to the server terminates, any notion of what the default database is also disappears. That is, if we connect to the server again, it does not remem- ber what database we had selected previously. We will have to select the database again.

9.3.2.1.3 Altering Databases

The ALTER DATABASE statement makes changes to a database’s global characteristics or attributes. For example,

ALTER DATABASE db_name DEFAULT CHARACTER SET charset;

The charset statement should be the name of a character set supported by the server, such as latin1 (cp1252 West European) or binary. Please refer to the MySQL online manual for a comprehensive list of character sets supported by the server.

Example 9.3

Change the default character set from latin1 to latin2:

9.3.2.1.4 Dropping Databases

Dropping a database is just as easy as creating one, assuming that we have sufficient privi- leges. The syntax for deleting databases is

DROP DATABASE db_name;

188 Developing Windows-Based and Web-Enabled Information Systems

However, the DROP DATABASE statement is something that the user should use with caution. It removes the database and all the tables within it. After the database is dropped, it is gone forever. If only some or all of the tables within the database need to be removed while keeping the database, the DROP TABLE statement should be used instead.

Example 9.4

Remove the database named healthcaresystem:

9.3.2.2 Statements for Relation Operations

In general, MySQL follows the ANSI SQL standard to operate on tables. These include CREATE TABLE for creating tables, DROP TABLE for removing the table from the database, and ALTER TABLE for changing table characteristics. Note that, in the previous illustra- tive examples, the query results show “0 rows affected” since the table is empty with no records yet.

9.3.2.2.1 Creating Tables

To create a table, the CREATE TABLE statement is used.

Example 9.5

Create the Drugs table from Chapter 8:

Note: In this example, the decimal data type (Table 9.1) instead of the numeric data type is used for the Price column.

9.3.2.2.2 Altering Table Structure

ALTER TABLE is a versatile statement in MySQL. The basic use of this statement is to rename tables, add or drop columns, change column types, and add or drop constraints.

Example 9.6

Rename the Drugs relation with a new table name New_Drugs:

Example 9.7

Change the data type of the column Price from decimal to float:

189MySQL

Example 9.8

Remove the Primary key constraint from the Drugs relation:

9.3.2.2.3 Dropping Tables

Dropping a table is much easier than creating it because there is no need to specify any- thing about its contents. For example,

Example 9.9

Delete the Drugs relation:

In MySQL, if the user is not sure whether a table exists, but the user wants to drop it if it does, IF EXISTS can be used in the statement. For example,

Example 9.10

Drop the relation named “Sales” by using IF EXISTS.

Note: If the Sales relation does not exist, without using IF EXISTS, MySQL will throw an error message:

9.3.3 Practice Data Manipulation Language in MySQL

Similar to DDL, MySQL adopts SQL DML for data manipulations. These include INSERT INTO for adding new records, UPDATE for changing the records, SELECT for querying the records, and DELETE for removing the records. The concept of nested query also applies.

9.3.3.1 Adding Records to a Table

The INSERT INTO statement is used to add new records into an existing table.

Example 9.11

Add the four records into the Drugs relation.

Drug_ID Name Price Description

T_001 Tylenol $17.99 100 325mg Tablets Bottle T_003 Tylenol $8.99 T_002 Tri_Vitamin $14.99 1500-400-35 Solution 50ml Bottle C_001 Claritin $35.99 10 Tablets

190 Developing Windows-Based and Web-Enabled Information Systems

Note: An easy way to populate multiple rows into a table is to create a text file (e.g., Notepad, and Microsoft Word is NOT recommended) containing all the INSERT state- ments. The user can copy the statement, and in the command-line window, go to Edit | Paste; a series of executions are triggered to have the records added into the table one by one. Please take note of the difference between symbols ' ' (MySQL) and ‘ ’ (text editor). Inappropriate use of the symbols will throw an error message “Error 1064 (42000): You have an error in your SQL syntax.”

9.3.3.2 Querying Tables

To query table(s), the SELECT statement is used. We have previously learned the USE state- ment for selecting the database. If the user does have access to a database, the tables can be queried directly without selecting the database explicitly. This is done by referring to table names with the database name. For example,

Example 9.12

Retrieve the contents of the Drugs table in the healthcaresystem database without selecting the database first:

It is, however, much more convenient to refer to tables without having to specify a data- base qualifier. Thus, the USE statement is recommended before executing queries. Another benefit of the USE statement is that the user can issue the statements to switch among data- bases as long as the access privileges are granted.

Note: By default, MySQL returns the results in descending order over the first column Drug_ID.

9.3.3.3 Updating Tables

When the values of existing records in a table need to be changed, the UPDATE statement is used.

Example 9.13

Update the price of all drugs in the Drugs relation with 15% discount:

Note: The MySQL UPDATE statement has a notable difference from standard SQL. That is, if the user accesses a column from the table to be updated, MySQL UPDATE uses

191MySQL

the current value (after the update) of the column instead of the original value of the column. For example, in the statement:

UPDATE table_1 SET col1 = col1 + 1, col2 = col1;

col2 will be set with the new col1 value after the update. As a result, both two columns have the same values.

9.3.3.4 Deleting Records

The DELETE statement is used to remove records from an existing table.

Example 9.14

Remove all records from the Drugs relation:

Note: The DELETE statement will only delete records. Even if the user has all the records removed from the relation, the empty table still exists in the database.

9.3.4 MySQL Transaction Control Language

A transaction consists of a sequence of query and/or update statements. By default, a con- nection to the MySQL Server begins with the AUTOCOMMIT mode enabled, which auto- matically commits every SQL statement being executed. When the user wants to issue a sequence of DML statements and commit them or roll them back all together, the user needs to switch AUTOCOMMIT off and end each transaction with the COMMIT or the ROLLBACK command when appropriate. The COMMIT statement will commit the current transaction and make the update performed by the transaction permanently in the data- base. After the transaction is committed, a new transaction is automatically started. The ROLLBACK statement causes the current transaction to be rolled back. That is, it undoes all the updates performed by the SQL statements in the transaction. Thus, the database state is restored to what it was before the first statement of the transaction was executed. The following examples show two transactions, in which the first one is committed, while the second one is rolled back.

Example 9.15

Use of commit. In this example, we first create a relation named “Drugs.” Next, we insert a new record into the relation. The COMMIT statement indicates that it is the end of the transaction.

192 Developing Windows-Based and Web-Enabled Information Systems

Note: In MySQL, the “--” (double-dash) character is used as a comment syntax. Take note that the second dash should be followed by at least one whitespace or control char- acter (such as a space, tab, newline, and so on).

Example 9.16

Use of rollback. In this example, we first turn off AUTOCOMMIT. Next, we insert three new records into the relation then delete the new record that we added in Example 9.15. The ROLLBACK statement undoes the three new insert commands and one delete; thus, the relation is restored to the original state after Example 9.15.

9.3.5 MySQL Data Control Language

The account information of MySQL is stored in the tables of the MySQL database. There are a number of administrative statements available in MySQL. For example, CREATE USER is used to add new users to the system, GRANT is used to offer the privileges to exist- ing users, REVOKE is used to revoke privileges from existing users, and DROP USER is used to remove a user from the system. Only the system administrator has the authority to use these statements. Here, we only use one simple example to illustrate the DCL in MySQL as follows:

Example 9.17

First, create a new user named ‘Joe1,’ and full privilege to the database named “health- caresystem” is also granted. Next, revoke the privileges of “Joe1” and delete ‘Joe1’ from the system.

193MySQL

9.3.6 MySQL Utilities

To better study MySQL, there are some important utilities we need to know, for example, SHOW, DESCRIBE, and HELP. We will discuss briefly in this section.

9.3.6.1 SHOW Statement

MySQL provides a SHOW statement that displays information about databases and the cor- responding tables within them. The SHOW command is helpful for keeping track of the contents of the databases and for reminding the users about the structure of the tables.

Example 9.18

List the current databases managed by the server:

Example 9.19

Obtain a listing of tables in the healthcaresystem database:

Note: Alternatively, if the USE statement is used prior to this command, the simple type “show tables” will get the same results.

Example 9.20

Display information about the columns in the Drugs table within the healthcaresystem database:

Example 9.21

List the tables in the MySQL database containing “help” in the name:

194 Developing Windows-Based and Web-Enabled Information Systems

9.3.6.2 DESCRIBE Statement

The DESCRIBE statement provides information about the columns in a table. It is an alter- native to the SHOW COLUMN FROM clause.

Example 9.22

Display the information about the columns in the Drugs table in the healthcaresystem database using the DESCRIBE statement:

9.3.6.3 HELP Statement

The HELP statement searches the help tables from the MySQL database for the given string and displays the result of the search. Note that the search string provided by the user is not case sensitive.

Example 9.23

List the information related to “show tables:”

So far, we have learned the use of command-line clients to practice SQL commands. Next, we will learn another MySQL tool, Workbench, and its basic functionalities.

9.4 Working with MySQL Workbench

The MySQL Workbench is a unified visual tool for database architects, developers, and administrators. MySQL Workbench provides the platform for data modeling and SQL development, as well as comprehensive administration tools for server configuration and user administration. Figure 9.9 illustrates the three main functionalities of the MySQL Workbench. Each will be briefly introduced in the following sections.

195MySQL

9.4.1 Data Modeling

The data modeling module enables the users to create models of the database schema graphically, reverse- and forward-engineer a schema and a live database, and edit all aspects of the database using the comprehensive Table Editor. The Table Editor provides easy-to-use facilities for editing tables, columns, indexes, triggers, partitioning, options, inserts and privileges, routines, and views. We will use the following example to show how to work with data modeling in MySQL Workbench.

Example 9.24

Use of data modeling to create the healthcaresystem database from scratch.

Step 1: Click on the Add Diagram icon in the Model Overview window. A new Enhanced Entity-Relational (EER) diagram tab will then appear. We can create any number of EER diagrams just as we may create any number of schemas.

Step 2: Click the EER diagram tab to navigate to the canvas used to graphically manipulate database objects. The vertical toolbar is on the left side of this page. When using the toolbar, we are able to create relations and columns for each rela- tion, and define the relationship constraints between the relations (see Figure 9.10).

Step 3: Click on the MySQL model tab. We will then see the physical schemas panel. The MySQL database schema has automatically populated the three relations created in the EER diagram.

Step 4: If we want to grant privileges to a given user, we will work with the Schema Privileges panel, where we can add new and assign roles to users.

Step 5: Next, click on the Database | Forward engineer. We will be asked to con- nect to the server and populate the database schema to the server. A database called “healthcaresystem” with three relations is created.

Next, we can use the SQL development tools to populate the data to the database.

FIGURE 9.9 Launching MySQL Workbench.

196 Developing Windows-Based and Web-Enabled Information Systems

9.4.2 SQL Development

The MySQL Workbench provides the capability to create and manage connections to data- base servers, configure connection parameters, and execute SQL queries on the database connections using the built-in SQL editor. Since we have extensively discussed the use of SQL commands in MySQL Command Line Client, we only show one screenshot on the use of Workbench for populating data to the Drugs relation (Figure 9.11).

The easiest way is to navigate to the database and right-click on the relation. The Table Editor is used to edit data and commit changes using a simple grid format. For example, “edit the table” is chosen to create new records, “alter the table” is chosen to make changes to an existing table, and “drop the table” is chosen to remove the table.

FIGURE 9.10 Creating an EER diagram for the healthcaresystem database.

FIGURE 9.11 MySQL Workbench: SQL development.

197MySQL

The visual SQL editor allows the users to build, edit, and run queries; create and edit data; and view and export results. Multiple queries can be executed simultaneously, and results can be viewed on individual tabs in the Results window. The export facility allows the users to export results data to common formats, while the History panel pro- vides a complete session history of queries and statements showing what queries were run and when. The users can easily retrieve, review, rerun, append, or modify previ- ously executed SQL statements. In addition, the SQL Snippet panel allows the users to save and easily reuse common Selects, DML, and DDL code. The most common queries can be placed in snippets so that they can quickly be retrieved and reused later by sim- ply double-clicking in the snippet list.

9.4.3 Server Administration

A server instance is created to provide a way of connecting to a server to be managed. The server administration tool enables the users to create and administer these server instances (Figure 9.12).

The administrator functionality is grouped into several tabs:

• Management: • Server Status: This includes CPU Utilization, Memory Usage, Connection

Usage, Traffic, Query Cache Hit Rate, and Key Efficiency. • Startup/Shutdown: This allows the users to start and stop the MySQL Server

and view the startup message log. • Status and System Variables: This displays system and status variables. • Server Logs: This displays server log file entries.

• Configuration: This allows the users to view and edit the MySQL configuration file (my.ini or my.cnf) using the GUI.

• Security: This allows the users to create user accounts and assign roles and privileges.

• Connection: This displays connections to the MySQL Server. • Data Export/Restore: This is an important utility to import different data files (e.g.,

MS Excel, .csv) into MySQL and export MySQL schema objects/tables to an SQL file.

FIGURE 9.12 MySQL Workbench: server administration.

198 Developing Windows-Based and Web-Enabled Information Systems

9.5 Summary

MySQL is one of the top choice open-source RDBS. This chapter briefly discusses how to install and configure the MySQL Server. Two important MySQL tools are introduced: the MySQL Command Line Client and the MySQL Workbench. To serve the purpose of learn- ing SQL, the MySQL Command Line Client tool is the first focus to demonstrate (1) how to use DDL and DML commands in MySQL, (2) how to manage and control MySQL transac- tions using TCL and DCL, and (3) how to use some important MySQL utilities including the SHOW, DESCRIBE, and HELP statements. To serve the purpose of learning database management, data modeling, and MySQL Workbench, a GUI tool is then briefly reviewed.

We only cover the basics of MySQL. Advanced readers are referred to the online manu- als for more information.

EXERCISES

1. MySQL is an open-source database management system. True or false? 2. Oracle Corporation originally established MySQL. True or false? 3. The following are advantages of MySQL except: a. MySQL runs on Linux and Unix. b. MySQL supports SSL protocol. c. MySQL can connect to .NET applications. d. MySQL has many features at a trade-off of large installation size. e. All of the above are advantages of MySQL. 4. The MySQL __________ is a MySQL version that handles high-volume transac-

tions per time period. 5. Which of following are functions of the MySQL Workbench? a. Design a database schema. b. Normalize data stores. c. Document the database schema. d. All of the above are valid functions. e. Only a and c are valid. 6. The following are valid connectors of MySQL except: a. C++ b. Python c. J d. C e. All of the above are valid connectors of MySQL. 7. The __________ is used to enable a MySQL client to access the server over the

Internet. 8. MySQL supports floats for numbers with no fractional part and integers for num-

bers with fractions. True or false? 9. Escape sequences are started using the backslash (\) character. True or false?

199MySQL

10. MySQL follows the “MM-DD-YYYY” format, which is the ISO 8601 standard for- mat for dates. True or false?

11. The NULL value is interpreted as a value of zero when performing arithmetic operations in MySQL. True or false?

12. When creating databases in MySQL, which of the following need to be considered? a. The user must have sufficient privileges to create a database. b. The database name must conform to standards. c. The database name must be unique. d. All of the above. e. None of the above. 13. MySQL can set a default database to access using the USE command even after

restarting MySQL. True or false? 14. The IF EXISTS command can be used to avoid error messages in MySQL for the

DROP statement. True or false? 15. By default, the AUTOCOMMIT is enabled when a new connection is started. True

or false? 16. Given a database for the purchasing, purchasing_records, write a code in MySQL

to list all tables within the database. 17. The __________ and __________ are two alternative commands to list the columns

of a table in MySQL. 18. The Table Editor can accomplish the following except: a. Edit databases b. Modify tables c. Edit columns d. Edit user privileges e. All of the above

201

10 Object-Based Database Systems

Relational database systems (RDBS) originated from the IBM R project in the 1970s and gained great popularity in business applications. As of today, RDBS products are still the dominant choice of database systems for information management problems. However, with the advancement of computer software and hardware, several limitations of RDBS products began to attract a lot of attention in the late 1980s. First, structured query language (SQL)-92 only supports numbers and strings, but several applications began to include complex data objects such as geographic points, images, and digital signal data. Second, relational tables are flat and do not support nested structures efficiently. Third, RDBS do not take advantage of object-oriented (OO) approaches, which have been widely accepted in industry software engineering practices. This chapter introduces some improvements and advancements made to RDBS, specifically OODBMS and object RDBMS (ORDBMS).

10.1 Object-Oriented Database Management System

10.1.1 Object-Oriented Database Concepts

An OO database is an extension of the OO programming language (OOPL) technique for persistent data management. It is a system that integrates database capabilities with OOPL capabilities.

As in OOPL, an object in an OO database typically has two components: its state (value) and its behavior (operations), except that it usually has a complex data structure as well as several specific operations defined by the developer. Objects in an OOPL exist only during program execution and are, hence, called “transient” objects. On the other hand, an OO database can extend the existence of objects so that they can be stored permanently. As a result, the objects persist beyond program termination and can be retrieved later and shared to other programs. These types of objects are called “persis- tent” objects.

One major advantage of an OO database over a relational database is that it preserves the direct relationship between real-world and the database objects. Additionally, it keeps the integrity and identity of objects consistently, and these objects can be easily identified and operated upon. In addition, complex data structures are often used to organize the information for better describing the objects. In contrast, relational data- bases represent the complex object structures over many scattered relations and tuples (records), which result only in an implicit mapping between the real-world entities and its database representations.

To learn OO databases, some important concepts are reviewed first as follows.

202 Developing Windows-Based and Web-Enabled Information Systems

10.1.1.1 Objects and Identities

Each real-world entity is modeled as a database object. The object needs a unique identity to be stored in the system. This unique identity is implemented via a unique, system- generated object identifier, termed as “OID.” An OID is used internally by the system instead of the developers to identify each object uniquely and create and manage inter- object references. Using OIDs over keys for object identification presents at least one advantage, that is, the developers do not need to be concerned about selecting the appro- priate keys for the objects since the OID is generated by the system. On the other hand, an OID is criticized as having little semantic meaning to the users. Interestingly, Cattell (1991) argues that, although the keys are important for the user, they present some difficulties. For example, short keys (e.g., social security number [SSN]) may not contain significant semantic information, whereas longer keys (first and last names, book titles) tend to be extremely inefficient if used as foreign keys.

10.1.1.2 Complex Objects

A complex object is an object that includes the set of attributes associated to it. The value of an attribute can be an object or a set of objects. This characteristic enables an arbitrary complex object to be defined from other objects. Complex objects are built by applying constructors to simpler objects, which are data types such as integers, real numbers, character strings, Booleans, and other basic data types supported by the system. These constructors are composed of several other types: sets, tuples, lists, bags, and arrays. Typically, the set constructor is used to define a set of objects identified by OIDs, for example, {OID1, OID2, … OIDn}. The tuple constructor is used to represent the property of the objects, for example, {a1: OID1, a2: OID2, … an: OIDn}, where a1 is an attribute name for the object. The list constructor is similar to a set constructor except that the OIDs in a list are ordered. The bag constructor is similar to a set constructor except that all ele- ments in a set must be distinct, whereas a bag can have duplicate elements. The array constructor is a one-dimensional array of elements. The main difference between a list and an array is that a list can have an arbitrary number of elements, whereas an array has a maximum size.

10.1.1.3 Encapsulation

Encapsulation is derived from the concepts of abstract data types and the information hid- den in OOPL. Each object consists of two elements that can be accessed and manipulated by other objects: an interface and an implementation. By definition, the interface element is the specification of the set of operations that can be invoked on the object, and it is the only visible part to the users. On the other hand, the implementation element contains the data of the object and the methods that provide the implementation of each operation.

Let us revisit the Patient relation in the healthcaresystem database to illustrate the encapsulation concept. In a relational database, a set of patients is represented by a table in which each row represents each patient. This relation can be queried using SQL data manipulation language statements (e.g., SELECT-FROM). There exists a clear separation between the query statements (operations) and the data in a relational database. Moreover, the structures of the database objects are visible to the users and the external programs. Hence, the SQL statements are generic and may be applied to any relation in the database. In an OO database, however, the entity Patient is defined as an object consisting of a data

203Object-Based Database Systems

component (similar to the table structure defined in a relational database) and an oper- ational component, for example, GetPatientHistory. Both the data and the operational components are stored in a database. The external users of the Patient object are only pre- sented with information on the interface of the object (e.g., the name and the parameters of each operation), whereas the internal data structure and the associated methods are hidden from the users. As a result, the encapsulation implements the idea of revealing only pertinent information to users.

10.1.1.4 Classes and Inheritance

In an OO database, all objects sharing the same set of attributes and methods are grouped together into classes. A class can be defined as another instance of one or more existing classes and will inherit the attributes and methods of such classes. This class is often referred to as a “subclass,” whereas the classes from which it has been defined are referred to as a “superclass.” For example, we define a Person class as

class Person { attribute string FirstName; attribute string LastName; attribute date DateOfBirth; attribute string Address; attribute string SSN; short Age( ) }

In this class, the FirstName, LastName, SSN, Address, and DateOfBirth are stored as attributes, and the Age function is a method to calculate the Age from the value of the DateOfBirth attribute and the current date. To illustrate the concept of a subclass, let us assume a new class, Patient, which has similar attributes to that of the Person class to be created. The Patient class can be defined as

class Patient { attribute string FirstName; attribute string LastName; attribute date DateOfBirth; attribute string Address; attribute string SSN; short Age( ) string GetPatientHistory( ) string ReactionToTreatment( ) }

It is observed that the Patient class has the same attributes and an Age function as compared to the Person class, with two added functions: GetPatientHistory and ReactionToTreatment. Given a Person class defined previously, the Patient class can be redefined as

class Patient extends Person { string GetPatientHistory( ) string ReactionToTreatment( ) }

204 Developing Windows-Based and Web-Enabled Information Systems

10.1.1.5 Overloading, Overriding, and Late Binding

Overloading, overriding, and late binding are also known as “polymorphism of operations.” This concept allows different methods to be associated with a single operation name, leaving the system to determine which method should be used to execute a given operation.

10.1.2 Object-Oriented Database Design and Modeling

The idea of implementing OO concepts into a database design was inspiring. However, the practical implementation of this concept is not an easy task, mainly due to the lack of existing standards. In 1991, a consortium of OODBMS vendors, called “Object Data Management Group” (ODMG), was formed to work toward standardizing OO database management. In 1993, the first standard, ODMG 1.0, was released, and the final version, ODMG 3.0 Standard, was released in 2001. The ODMG standards outlined three com- ponents: object modeling, object definition language (ODL), and object query language (OQL), which are explained in the following sections.

10.1.2.1 Object Modeling

The object model is designed to be the foundation for object database systems upon which the ODL and OQL are defined. It supports the concepts discussed previously, including the representation of complex objects, encapsulation, and inheritance. Furthermore, object mod- eling has some common concepts with relational modeling (e.g., entity-relational model) but with some fundamental differences. The key comparisons are provided as follows, and the key terms in an OO data model and its counterpart in the relational model are summarized in Table 10.1.

10.1.2.1.1 Object versus Entity and Tuple

An OODM object resembles the entity and the tuple in the relational models, with addi- tional characteristics such as behavior, inheritance, and encapsulation. These characteris- tic make OO modeling much more natural than relational modeling.

TABLE 10.1

Summary of OO Data Models

OO Data Model Description

Corresponding Relational Model Element

Class An entity that has a well-defined role in real-world application, which is described by its state, behavior, and identity.

Entity set

Object An instance of a class that encapsulates data and behavior. Entity State Consists of an object’s properties and the values of the

properties. Attributes

Operation A function that is provided by all the instances of a class. None or not available Method Implements the operations on the objects such as query and

update. None or not available

Association A named relationship between or among object classes, for example, unary and binary.

Degree of relationship

Multiplicity A specification that indicates how many objects participate in a given relationship.

Cardinality

Class diagram Illustrates the static structure of an OO data model: the classes, their structure, and interrelationship.

E-R diagram

205Object-Based Database Systems

10.1.2.1.2 Class versus Entity Set and Table

The concept of a class can be associated with the relational models’ concepts of entity sets and tables. However, the concept of a class is more powerful in that it allows the descrip- tion not only of the data structure, but also of the behavior (represented by method) of the class objects.

10.1.2.1.3 Encapsulation and Inheritance

Classes are organized in class hierarchies. An object belonging to a class inherits all the properties of its superclasses. Additionally, encapsulation means that the data representa- tion and the methods’ implementation are hidden from other objects and the end user, whereas in a relational model, data are directly accessible from the external environment.

10.1.2.1.4 Object ID

Object ID is not supported in the relational model.

10.1.2.1.5 Relationship

The main property of any data model is found in its representation of relationships among the data components. The relational model uses a value-based relationship approach. This means that a relationship among entities is established through a common value in one or several of the entity attributes. In contrast, the object model uses the object ID, which is identifier based, to establish relationships among objects, and such relationships are independent of the state of the object. The relationships in an object model can either be inter-class references or class hierarchy inheritance.

Example 10.1

Present the healthcaresystem database learned in Chapter 5 using an object model. The solution is shown in Figure 10.1. As seen, the Patient class has five attributes—

FirstName, LastName, DateOfBirth, Address, and SSN—and three methods— Age( ), GetPatientHistory( ), and ReactionToTreatment( ). The TreatmentPlan class has two attributes—Drug_Description and TreatmentDate—and one method— PatientOutcome( ). The Drug class has three attributes—Drug_ID, Name, and Description—and one method—Sales( ). Each Patient may have multiple TreatmentPlans, and each TreatmentPlan may be applied to multiple Patients. In each TreatmentPlan, multiple drugs may be prescribed, and each Drug may be prescribed in multiple TreatmentPlans. In the next section, we will learn the use of ODL to define the classes and their relationships.

Patient +FirstName +LastName +DateofBirth +Address +SSN +Age() +GetPatientHistory() +ReactionToTreatment()

+Drug_Prescription +Treatment_Date

* *Take * *Prescribe

Taken by Prescribed by

TreatmentPlan Drug

+PatientOutcome()

+Drug_ID +Name +Description +Sales()

FIGURE 10.1 Object data model for part of the healthcaresystem database.

206 Developing Windows-Based and Web-Enabled Information Systems

10.1.2.2 Object Definition Language

Similar to SQL-92 data definition language, which specifies a logical schema for a relational database, object definition language allows the users to specify a logical schema for an OO database. ODL is designed to support the semantic constructs of the ODMG object model and is independent of any particular programming language. Table 10.2 summarizes the keywords used to define a class, extent, attribute, struct, and relationship in ODL. An opera- tion is specified without the use of a keyword; a parenthesis is simply added after its name.

In the example shown in Figure 10.1, the ODL for the three classes are as follows. Since each Patient can take multiple TreatmentPlans, the keyword set is used after the relationship keyword. Another keyword is extent, which is simply used to refer to the class set.

class Patient { ( extent Patients) attribute string FirstName; attribute string LastName; attribute date DateOfBirth; attribute string Address; attribute string SSN; relationship set <TreatmentPlan> takes inverse TreatmentPlan::taken_by short Age( ) string GetPatientHistory( ) string ReactionToTreatment( ) }

class TreatmentPlan { ( extent TreatmentPlans) attribute string Drug_Prescription; attribute date Treatment_Date; relationship set <Patient> taken_by inverse Patient::takes relationship set <Drug> prescribes inverse Drug::prescribed_by string PatientOutcome( ) }

TABLE 10.2

Some Important ODL Keywords

Keywords Description

Class A class is specified using the “class” keyword. Extent The extent of a class is the set of all instances of the class within the database. Attribute An attribute is specified using the “attribute” keyword and can either be OID or literal. The

literal types include atomic literals (string, char, Boolean, float, short, long), collection literals (set, bag, list, array, and dictionary), and structural literals (Date, Interval, Time, Timestamp, and user-defined structures).

Struct A user-defined structure is specified using the “struct” keyword. Relationship The “relationship” keyword is used to specify the relationship between the objects. ODL

supports only unary and binary relationships. Inverse The “inverse” keyword is used to specify the relationship in the reverse direction.

207Object-Based Database Systems

class Drug { ( extent Drugs) attribute string Drug_ID; attribute string Name; attribute string Description; relationship set <TreatmentPlan> prescribed_by inverse TreatmentPlan::prescribes string SalesTransaction( ) }

Let a user-defined struct Size for the patient class be as follows:

struct Size { float weight; float height; short BMI; }

The Patient class can then be modified with the added reference to the user-defined struct Size:

class Patient { ( extent Patients) attribute string FirstName; attribute string LastName; attribute date DateOfBirth; attribute string Address; attribute string SSN; attribute SIZE patient_size; relationship set <TreatmentPlan> takes inverse TreatmentPlan::taken_by short Age( ) string GetPatientHistory( ) string ReactionToTreatment( ) }

Compared to the example from Chapter 5, fundamental differences between an OO database schema and a relational database schema are clearly observed. In the following section, the query language for an OO database, OQL, is introduced.

10.1.2.3 Object Query Language

OQL is the query language designed to work closely with the programming language for which an ODMG binding is defined. Hence, an OQL query embedded into any OOPL can return objects that match the type system of that language. The OQL syntax for queries is similar to the syntax of SQL-92, with additional features for ODMG concepts, such as object identity, complex objects, operations, inheritance, polymorphism, and rela- tionships. In an OQL query, the developer needs to query from the extent of the class. For example, if one wants to query for patient information, the query is executed on the Patients class.

208 Developing Windows-Based and Web-Enabled Information Systems

Example 10.2

Query from one class: Retrieve the Address of the patient whose name is “Joe Smith”:

SELECT p.Address FROM Patients p WHERE p.FirstName = "Joe" and p.LastName = "Smith";

Example 10.3

Including an operation in a query: Retrieve the name of all patients who are 40 years or older:

SELECT p.FirstName, p.LastName FROM Patients p WHERE p.age >= 40;

Example 10.4

Use of aggregate functions: Find the number of patients being treated with the corre- sponding code:

SELECT count (*) FROM Patients p

Example 10.5

Query from multiple classes: Retrieve the name of the drug used in treating the patient named “Joe Smith” without duplicates:

SELECT DISTINCT d.Name FROM Drugs d, Patients p, Treatments t d.prescribed_by t p.takes t WHERE p.FirstName = "Joe" and p.LastName = "Smith"

Example 10.6

Use of subquery: Retrieve the name of the patient who has a “Positive” response to the treatment and is over 40 years:

SELECT x.FirstName, x.LastName FROM ( SELECT p FROM Patients where p.ReactionToTreatment = "Positive") as x WHERE x.age >= 40

Note that only a subset of OQL queries are illustrated above. Please refer to ODMG 3.0 Standard (2000) for more standard OQL features.

10.1.3 Summary of OODBMS

In summary, OODBMS yield several benefits over relational database systems. Most of these benefits are based on the previously explored OO concepts. Database systems such as

209Object-Based Database Systems

hierarchical, network, and relational systems have been very successful in traditional busi- ness applications. However, new business operations require more complex structures for objects, new data types for storing images or large textual items, and the need to define non- standard application-specific operations. OO databases were proposed to meet these needs. These databases offer the flexibility to handle some of the requirements without being lim- ited by the data types and query languages available in traditional database systems.

However, OODBMS is not without limitations. First, OODBMS vendors have discovered the difficulty of tying the database design closely to the application design. Maintaining and evolving an OODBMS-based information system has also been an arduous under- taking. In addition, the OO data model lacks the mathematical foundations present in relational models, which make an OODBMS not well understood as compared to RDBS.

Inspired by the OO concept and noting some complementary features of OO and rela- tional models, there has been an effort to combine the two methods in developing large information systems, which produced the ORDBMS.

10.2 Object RDBMS

10.2.1 Object-Relational Model

The object-relational (OR) data model enhances a relational model with objects. The main task of the model is to provide a flexible framework for organizing and manipulating data- base objects that correspond to real-world objects.

To achieve this, the OR model extends the relational model by incorporating attributes that are atomic in nature as well as those that have a structured data type. Similar to relational databases, an OR database is composed of a group of tables made of rows, and all rows may consist of values from structurally identical specific data types instead of only atomic types. Like an OO database, the structured types are the types built from atomic types via the use of constructors. This extension enables an OR database to have nested relations. Let us use the following example to illustrate the concept.

Example 10.7

Design a nested relation schema for the Patient relation that incorporates an attribute called “TreatmentPlan,” representing all the treatments that the patient has had.

For simplicity, let the Patient relation have three original attributes: Name, Address, and SSN. Furthermore, let the TreatmentPlan have two attributes: Drug_Description and TreatmentDate. An example of the design is shown in Figure 10.2. As seen, Name, Address, and SSN have atomic values, whereas TreatmentPlan has a relation as a structured type. It includes two attributes, Drug_Description and Treatment_Date, which are atomic in nature.

The detailed schema for this example, thus, is written as

Patient (Name (FirstName, LastName), Address, SSN, TreatmentPlan (Drug_ Description, Treatment_Date))

Since the drug “Claritin” was prescribed to two patients, it appears twice in the rela- tion. Based on the relational database design principle, this violates Boyce–Codd nor- mal form (BCNF) which we have discussed in the earlier chapter. To address this issue, the OR database introduces reference idea, which allows the tuples to point to other tuples to avoid duplicates (shown in Figure 10.3).

210 Developing Windows-Based and Web-Enabled Information Systems

The schemas of the two relations, thus are:

TreatmentPlans (Drug_Description, Treatment_Date) Patient (Name (FirstName, LastName), Address, SSN, TreatmentPlan ({*TreatmentPlans}))

We use the Patient (… {*TreatmentPlans}) statement to describe that in Patient rela- tion, in which a reference to a single tuple with a relation schema named TreatmentPlans exists. Note that this is different from decomposing relations as is done in relational data- base design. The TreatmentPlan attribute is still an attribute in Patient relation since the structured type is defined in a separate relation and is referenced in the Patient relation.

Note: This example does not discuss the use of methods as part of the OR schema. In practice, the SQL-99 standard and OR database allow the same ability as the ODL in defin- ing methods associated with any relation.

10.2.2 Object-Relational Query Language

The OR database adopts the query-centric approach as done in a relational database. All data handling uses declarative SQL statements. Since an OR database supports structured data types, the query statements are capable to deal with complex objects (e.g., Name) instead of atomic data types only (e.g., integer, varchar). Indeed, SQL-99 adds an extensive type system to the earlier SQL standards and allows structured types and type inherence. All these extensions can serve for OR database queries.

Example 10.8

Create Patient relation, that is,

Patient (Name (FirstName, LastName), Address, SSN, TreatmentPlan ({*TreatmentPlans}))

Name Address SSN TreatmentPlan Drug_Description Treatment_Date FirstName LastName

FirstName LastName

Joe Smith Flamingo Dr. Chandler, AZ

322- 18- 9533

Tyler O’Neil Riggs Rd. Gilbert, AZ

905- 35- 5217

FIGURE 10.3 Illustration of the reference concept used in the OR database.

Name Address SSN TreatmentPlan FirstName LastName

FirstName LastName

Joe Smith Flamingo Dr. Chandler, AZ

322-18- 9533

Drug_Description Treatment_Date Claritin 02/25/2013 Claritin 04/10/2013

Tyler O’Neil Riggs Rd. Gilbert, AZ

905-35- 5217

Drug_Description Treatment_Date Claritin 02/25/2013 Claritin 05/01/2013

FIGURE 10.2 Nested relation for Patients and associated TreatmentPlan.

211Object-Based Database Systems

TreatmentPlans (Drug_Description, Treatment_Date)

First, let the following two structured types be defined as

CREATE TYPE PersonName AS ( FirstName varchar (20), LastName varchar (20))

CREATE TYPE TreatmentPlan AS ( Drug_Description varchar (255), Treatment_Date date)

Next, the Patient relation is defined as

CREATE TABLE Patients ( Name PersonName NOT NULL, Address varchar (255), SSN char (9), TreatmentPlan TreatmentPlan NOT NULL, Primary Key (SSN))

As seen, an OR table is structurally very similar to its relational counterpart, and the same data integrity and physical organization rules can be enforced over it. The major difference is specifying column types. In the OR table, atomic data types such as varchar and char are the same as in relational tables, but the other two columns (PersonName, TreatmentPlan) are completely new. From an OO perspective, these data types correspond to class names and can be user defined.

In terms of query, it is very similar to SQL. For example, if the user wants to get the infor- mation from the patient named “Joe Smith,” the query can be written as

SELECT FROM Patients WHERE Name = PersonName ("Joe Smith")

In this section, the basics of OR queries are briefly illustrated. Advanced readers can refer to SQL-99 for a comprehensive discussion on the OR query language features.

10.2.3 Summary of ORDBMS

In summary, ORDBMS synthesizes the features of RDBS with the best ideas of OODBMS. Although ORDBMS reuse the relational model as SQL interprets it, the OR data model is opened up in novel ways. New data types and functions can be implemented using general- purpose languages such as C and Java. In other words, ORDBMS allow the developers to embed new classes of data objects into the relational data model abstraction. Moreover, ORDBMS schemas have additional features that are not present in an RDBS schema. Several OO structural features such as inheritance and polymorphism are included in the OR data model. Second, ORDBMS adopt the RDBS query-centric approach to data management. All data access in an ORDBMS is handled with declarative SQL statements. There is no procedural, or object-at-a-time, navigational interface. ORDBMS persist with the idea of a data language that is fundamentally declarative and, therefore, mismatched with procedural OO host languages. This significantly affects the internal design of the ORDBMS engine, and it has profound implications for the interfaces that the developers use to access the database. Third, from a systems architecture point of view, ORDBMS

212 Developing Windows-Based and Web-Enabled Information Systems

are implemented as a central server program rather than as a distributed data architec- ture typical in OODBMS products. However, ORDBMS extends the functionality of RDBS products significantly, and an information system using an ORDBMS can be deployed over multiple machines.

10.3 Summary

This chapter gives a basic overview of two advanced database systems: the OO database and the OR database. Inspired by the success of OO programming, the OO database was designed to incorporate complex data objects (e.g., images, pictures). However, it has not been widely accepted due to the lack of mathematical rigor and the labor intensiveness of maintaining the system. As a result, most relational database vendors choose to develop an OR database. Some fundamentals such as data modeling and queries of both the two systems are briefly discussed in this chapter.

EXERCISES

1. The two components of an object are its current values and its methods. True or false? 2. Objects are transient objects, that is, they only exist when the program is closed.

True or false? 3. The following are advantages of an OO database over a relational database except: a. It makes object identification easier. b. It provides an implicit mapping of real-world entities. c. It preserves a direct relationship between a real object and a database object. d. All of the above are valid advantages of an OO database. e. None of the above is a valid advantage. 4. The following are properties of OIDs except: a. An internally generated ID to act as a primary key for an entity b. No need for primary keys to uniquely identify each element c. Has less semantic meaning d. Used to create and manage inter-entity relationships e. All of the above are valid properties 5. A complex object can be composed of other complex objects. True or false? 6. In encapsulation, the ____________ element is a set of operations that is visible to

the users, while the ____________ element is hidden. 7. A superclass can inherit the following from a subclass: a. Attributes b. Data points c. Methods d. All of the above e. Only a and c

213Object-Based Database Systems

8. The ____________ of operations concept allows different methods to be associated with a single operation.

9. An OODM object incorporates encapsulation as compared to an RDBS entity. True or false?

10. An operation in OODM is equivalent to a query in RDBS. True or false? 11. Using an ODL, an operation is specified using a keyword “define.” True or false? 12. An OQL query can be integrated into any OOPL to query objects. True or false? 13. An OR database extends the relational model by treating atomic attributes as

objects. True or false? 14. What is the main advantage of using an OQL over relational databases?

For questions 14 and 15, consider the following schema:

Person (Person_M(First_M, Last_M), Person_Address(Number, Street, City, State, Zip)) Background (Academic_Inst, Year_Completed, Degree) Courses (Course_M, Course_Units)

15. Write an ODL code to create a table called “Faculty_Member” that inherits all the attributes from a person schema with academic backgrounds. Additionally, the new table should include a Faculty_N integer attribute to act as a primary key and a new attribute called “Specialization” to denote a faculty’s academic specialization.

16. Write an ODL code to create a table named “Faculty_Courses” from these sche- mas. This table should list down all courses as well as the faculty members teach- ing the courses, the number of students for the courses, and the semester and year that the courses are taught.

Section III

Windows Application Development

Microsoft Visual Studio is a collection of tools that developers can use to design, develop, debug, and deploy Windows applications and Web applications. It is based on the .NET framework and includes several different programming languages such as Visual Basic, Visual C#, Visual C++, Visual F#, JScript, and ASP.NET. Figure III.1 shows the basic com- ponents of Visual Studio, where common language runtime manages the execution of a programming project writing in any language; .NET framework class library is a library of pre-defined class objects including Windows forms, controls, and more; and ADO.NET is an object model to enable the access to data from diverse sources.

Section III of this book covers Visual Basic programming language (VB.NET) and Visual Basic for Application (VBA). VB.NET is an object-oriented programming language work- ing with objects. Example objects are form objects, button objects, text box objects, and

Windows forms

Web forms

XML Web services

ASP.NET

.NET framework

Data access and XML classes (ADO.NET)

.NET framework class library

Common language runtime (CLR)

Operating system

FIGURE III.1 NET framework.

216 Windows Application Development

label objects, just to name a few. VB is also known as an event-drive programming lan- guage in that the developers may write the program code to respond to the events con- trolled by end users. Examples include “click a button.” The properties and events of the objects are covered in Chapter 11, followed by Visual Basic programming in Chapter 12. Next, the application of ADO.NET in VB.NET for Windows application with database con- nectivity is covered in Chapter 13. Chapters 14 and 15 focus on VBA, explaining the objects in VBA and covering the programming basics of the use of ADO.NET in Excel application, respectively.

217

11 Windows Forms and Controls in Microsoft Visual Studio

Visual Studio 2013 is an integrated development environment that integrates and manages software development tools in a single environment, including various programming lan- guages (e.g., Visual Basic, Visual C++, and Visual C#), database development, Windows appli- cation development and Web application development. We use Visual Basic in this text. Using Visual Studio 2013, we can develop a Windows-based application that runs on a computer and is used by one user at a time, or a Web-enabled application that can be used by multiple users via the Internet. Chapters 11 to 13 focus on developing a Windows-based application. Chapter 16 describes the development of a Web-enabled application. A Windows-based application consists of a graphical user interface (GUI), which is described in this chapter; functions implemented through Visual Basic programming, which is described in Chapter 12; and connection with database, which is described in Chapter 13. The following sections describe Windows forms and controls in the GUI of a Windows-based application.

11.1 Visual Basic Development Environment

At the first time of using Visual Studio 2013, we get the window shown in Figure 11.1 when we start Visual Studio 2013. We select Visual Basic for Development Settings, and then click on the Start Visual Studio button. This brings up the Start Page shown in Figure 11.2, which allows us to select either starting a new project or opening an existing project. We select Start a New Project, which brings up the New Project window shown in Figure 11.3. We use the New Project window to name our new project “Healthcare Information System.” The New Project window has Windows Forms Application set as default, and we take this default application type for our project. After we click the OK button in the New Project window, the Visual Basic development environment, as shown in Figure 11.4, is brought up. This development environment has the menus and shortcut icons at the top and several windows that assist the development. On the left side, there are two tabs for Toolbox and Data Sources. The Toolbox tab contains controls and other tools that we can use to create a GUI. The Data Sources tab initially has no data sources and will con- tain databases and other types of data sources when they are added to the project. If the Toolbox tab is closed, it can be brought up again by selecting Toolbox in the View menu. If the Data Source tab is closed, it can be brought up again by selecting Other Windows in the View menu and then Data Sources in the pop-up menu. In the Design window, there is the Design View of Form1, which is a default form. The Solution Explorer window gives the list of components in the project. At the beginning, the list includes only three compo- nents in the project: My Project, App.config, and Form1.vb. The Properties window shows properties of the selected object in the Design window, which is Form1 in Figure 11.4.

218 Developing Windows-Based and Web-Enabled Information Systems

FIGURE 11.1 Starting window of Visual Studio 2013.

FIGURE 11.2 Start Page window.

219Windows Forms and Controls in Microsoft Visual Studio

FIGURE 11.3 New Project window.

FIGURE 11.4 Visual Basic development environment.

220 Developing Windows-Based and Web-Enabled Information Systems

If we go to the File menu now and select Save All in the File menu or click the Save All icon at the top, we see the Save Project window (see Figure 11.5). We can use the Browse button to select a folder on the computer to keep the project folder. The project folder con- tains the Visual Studio Solution file with .sln and the folder with various components of the project. If we close Visual Studio and then double-click the Visual Studio Solution file, we can bring up the development environment for the project again.

11.2 Windows Forms

Among many properties of a Windows form including those in the Properties window, the name and the text of a Windows form often need to be changed. For example, Form1 in Figure 11.4 has the name of Form1.vb and the text of Form1. The name appears in the Solution Explorer window, in the tab of the Design window, and in the Properties window, as shown in Figure 11.4. The text appears as the heading at the top of the Windows form, as shown in Figure 11.4. We can change the name of a Windows form. For example, to rename “Form1.vb” to “FormMain.vb,” we right-click Form1.vb in the Solution Explorer window to bring up a pop-up menu, select Rename, and change the name of “Form1.vb” to “FormMain.vb.” The text of a Windows form can be changed by finding the Text prop- erty in the Properties window and changing the value of the Text property from “Form1” to “Healthcare Information System.” Resizing a Windows form can be performed in the Design window as we resize windows in the Windows system.

The first form of a Windows Form application is set as the default startup form of the application. We can check the setup of the startup form by selecting the Project menu. The last item of the Project menu is for the properties of the application. For example, the last item of the Project menu is shown as “Healthcare Information System Properties” for the Healthcare Information System application. Selecting this item brings up a tab for Healthcare Information System in the Design window, as shown in Figure 11.6. In Figure 11.6, the value of the startup form property shows FormMain. The project prop- erties can also be brought up by double-clicking My Project in the Solution Explorer window.

When we run an application, the startup form will appear first. To run an application, we go to the Debug menu. When we select Start Debugging in the Debug menu, an appli- cation is executed. For the Healthcare Information System application, FormMain with the heading “Healthcare Information System” appears when we run this application. To stop debugging, we go to the Debug menu and select Stop Debugging. We can run an application without opening Visual Studio by going into the bin folder inside the project folder, opening the Debug folder in the bin folder, and double-clicking the application’s

FIGURE 11.5 Save Project window.

221Windows Forms and Controls in Microsoft Visual Studio

executable file in the Debug folder. We can move the application’s executable file in the Debug folder to the desktop to create an icon on the desktop and then double-click the icon to execute the application.

We can add more Windows forms to our application. To add a Windows form, we select the Project menu and select Add Windows Form in this menu. This brings up the Add New Item window with Windows Form selected, as shown in Figure 11.7. We name the new Windows form, for example, FormEmployers, and click the Add button. Now, we see FormEmployers added in the Solution Explorer window. We change the Text property of FormEmployers to Employers so that the heading shows Employers.

With two Windows forms in the Healthcare Information System application and FormMain as the startup form, we want to close FormMain and open FormEmployers when we click on FormMain, and then we want to close FormEmployers and open FormMain when we click on FormEmployers. This requires adding code to both Windows forms. First of all, we change from the Design View to the Code View of FormMain by either double-clicking on FormMain in the Design View or selecting Code in the View menu. Figure  11.8 shows the Code View of FormMain, which presents the code of FormMain. The top portion of the Code View has two boxes with drop-down arrows. The left box has a list of objects. The right box has a list of events for the selected object in the left box. The code of FormMain contains a default subroutine that begins with the keywords, Private Sub, and ends with the keywords, End Sub. The keywords, Private Sub, are fol- lowed by the name of the Windows form and the Load event that triggers the execution of the subroutine. The Load event means the loading of the Windows form. This subroutine is executed when the FormMain_Load event occurs. Hence, Visual Basic programming in Visual Studio 2013 is an event-driven programming to have the code written and executed in response to events.

FIGURE 11.6 Properties window of a Windows Form application.

222 Developing Windows-Based and Web-Enabled Information Systems

FIGURE 11.7 Add New Item window.

FIGURE 11.8 Code View of a Windows form.

223Windows Forms and Controls in Microsoft Visual Studio

Many events may occur on a Windows form. When we select the Click event from the list of events, another subroutine is added to the code of FormMain for FormMain_Click. This new subroutine is executed when we run the application and click on FormMain. We add the following code in the subroutine for FormMain_Click:

Me.Hide() FormEmployers.Show()

The keyword, Me, denotes this form, that is, FormMain. Hide() is a function to hide this form. The code FormEmployers.Show() shows FormEmployers. We add a similar code to FormEmployers by double-clicking FormEmployers.vb in the Solution Explorer window, opening the Code View of FormEmployers, and adding a subroutine for FormEmployers_ Click and the following code in this subroutine:

Me.Hide() FormMain.Show()

With the above codes of FormWelcome and FormEmployers, we can click on FormMain to switch to FormEmployers and click on FormEmployers to switch to FormMain when we run the application.

Example 11.1

Add FormPatients, FormProviders, FormInsuranceCompanies, FormPharmacies, and FormUpdates, with Patients, Providers, Insurance Companies, Pharmacies, and Updates as the value of the Text property for these forms, respectively, to the Healthcare Information System application.

The steps in adding these forms are given as follows:

1. Select Add Windows Form in the Project menu, which brings up the Add New Item window.

2. Change the name of the Windows form to FormPatients in the Add New Item window, and click the Add button.

3. Change the value of the Text property in the Properties window to Patients. 4. Click the Save All icon to save the change. 5. Repeat Steps 1 to 4 to add FormProviders, with Providers as the value of the

Text property. 6. Repeat Steps 1 to 4 to add FormInsuranceCompanies, with Insurance

Companies as the value of the Text property. 7. Repeat Steps 1 to 4 to add FormPharmacies, with Pharmacies as the value of

the Text property. 8. Repeat Steps 1 to 4 to add FormUpdates, with Updates as the value of the Text

property.

11.3 GUI Controls

Example 11.2 shows how to add GUI controls to a Windows form.

224 Developing Windows-Based and Web-Enabled Information Systems

Example 11.2

Add the GUI controls, as shown in Figure 11.9, including one label, six radio buttons, and one button, to FormMain. Add code so that checking one of the six radio buttons and then clicking on the Next button switches the Windows form from FormMain to the Windows form corresponding to the selected radio button.

By double-clicking FormMain.vb in the Solution Explorer window, we bring the Design View of FormMain in the Design window. After clicking on the Toolbox tab, we see several categories of tools, including All Windows Forms, Common Controls, and so on. Expanding Common Controls gives us the list of common controls, as shown in Figure 11.10.

We select Label from the list of common controls, drag and drop it to FormMain, and use the Properties window to set the value of the Text property to “Select one of the following.” We can change the font and the size of this label by setting the Font and the Size properties in the Properties window. Other appearance features of this label can also be changed using the Properties window.

We then drag and drop RadioButton from the list of common controls to FormMain and use the Properties window to set the value of the Text property to Employers and the value of the Name property to RadioButtonEmployers. We add the other five radio but- tons for Patients, Healthcare Providers, Insurance Companies, Pharmacies, and Updates in the same way, and set the name of these five radio buttons to RadioButtonPatients, RadioButtonProviders, RadioButtonCompanies, RadioButtonPharmacies, and Radio- ButtonUpdates, respectively. No matter how many radio buttons we add to this Windows form, the radio button function lets the user check only one of these radio buttons.

We now drag and drop Button to FormMain and use the Properties window to set the value of the Text property to Next and the value of the Name property to ButtonNext. There are many other types of common controls that give other types of GUI functions for a developer to choose from.

Double-clicking ButtonNext brings up the Code View of FormMain and adds a sub- routine for the event ButtonNext_Clicked. The code that we enter into this subroutine is shown in Figure 11.11. The code hides FormMain, examines which of the radio buttons is checked, and shows the Windows form corresponding to the checked radio button.

Example 11.3 shows how to take the user’s input using TextBox and displays an output message using MessageBox.

FIGURE 11.9 GUI of FormMain in the Healthcare Information System.

225Windows Forms and Controls in Microsoft Visual Studio

FIGURE 11.10 List of common controls in the Toolbox tab.

FIGURE 11.11 Code for the Click event of ButtonNext in FormMain of the Healthcare Information System.

226 Developing Windows-Based and Web-Enabled Information Systems

Example 11.3

Create a Windows form that asks the user to enter the user’s first and last names and then displays a message to welcome the user, as shown in Figure 11.12a.

The design of the Windows form in Figure 11.12a contains two labels to prompt the user to enter the first and last names, two textboxes (with the names of TextBoxFirstName and TextBoxLastName) for the user to enter the first and last names, and a button to click to display the welcome message box. Figure 11.12b shows the code for the Click event of the Welcome button. TextBoxFirstName.Text returns what the user enters in the text box for the first name, and TextBoxLastName.Text returns what the user enters in the text box for the last name. The following code puts together the values in the Text property of TextBoxFirstName and TextBoxLastName and the string “Welcome,” using the string operator and displays the entire message in the message box.

MessageBox.Show("Welcome," & TextBoxFirstName.Text & " " & TextBoxLastName.Text)

(a)

(b)

FIGURE 11.12 Design View and Code View of FormWelcome for Example 11.3. (a) Design View of FormWelcome and the message box to welcome a user. (b) Code View of FormWelcome.

227Windows Forms and Controls in Microsoft Visual Studio

11.4 GUI of the Healthcare Information System Application

Figure 7.13 shows all the tables in the healthcare database implemented in Access. Chapter 7 gives some example queries for the healthcare database. Chapter 26 gives the complete list of queries for employers, patients, healthcare providers, insurance companies, and phar- macies. The Healthcare Information System application lets a user select and view data from the queries and update data in the tables through GUI. Figure 11.13 shows the GUI design of the Healthcare Information System application, which includes a hierarchy of Windows forms. The startup Windows form at the top level of the hierarchy on the left side of Figure 11.13, FormMain, organizes data from the queries into several categories for employers, patients, providers, companies, and pharmacies using five radio buttons. Checking one of the five radio buttons takes the user to one of the five Windows forms at the middle level of the hierarchy—FormEmployers, FormPatients, FormProviders, FormCompanies, and FormPharmacies. Each of the five Windows forms gives a list of data from the list of queries in Chapter 26 for the selected category (employers, patients, providers, companies, or pharmacies). Selecting a data item in the list takes the user to the Windows form at the bottom level of the hierarchy, which presents the data. FormMain also contains a radio button for updating data in the tables. Checking the radio button for Updates takes the user to a Windows form at the middle level of the hierarchy, which includes a list of the tables. Selecting a table takes the user to the Windows form at the bottom level of the hierarchy, which presents the data in the selected table. The user can change the table data directly in the Healthcare Information System application. Note that Query 8 and Query 9 for insurance companies are action queries of updating and deleting data. Updating and deleting data in a table will be done on the table data by taking the route through FormUpdates. Hence, there are no radio buttons in FormCompanies corre- sponding to Query 8 and Query 9 for insurance companies.

Example 11.4

Complete the GUI of the Healthcare Information System application that has been started in Example 11.2 to create all the Windows forms, the GUI elements of each Windows form, and the code for making transitions among these Windows forms, as shown in Figure 11.13, except the Windows forms at the bottom level of the hierarchy, which present data from the queries and the tables.

Figures 11.9 and 11.11 give the Design View and the Code View of FormMain at the top level of the hierarchy. Figures 11.14 to 11.19 give the Design View and the Code View of the Windows forms at the middle level of the hierarchy for the Healthcare Information System application.

11.5 Execution of a Windows-Based Application Outside Visual Studio

We can execute a Windows-based application without launching Visual Studio by double- clicking the executable application file in the project folder of the application. For exam- ple, we first go to the bin folder in the Healthcare Information System application folder and then open the Debug folder in the bin folder. When we double-click the executable application file, Healthcare Information System.exe, in the Debug folder, the Healthcare

228 Developing Windows-Based and Web-Enabled Information Systems

FormMain

FormEmployers

FormPatients

FormProviders

FormCompanies

FormPharmacies

FormUpdates

Form presenting Query 1 data for Employers

Form presenting Query 2 data for Employers

Form presenting Query 3 data for Employers

Form presenting Query 1 data for Patients

Form presenting Query 2 data for Patients

Form presenting Query 3 data for Patients

Form presenting Query 4 data for Patients

Form presenting Query 1 data for Healthcare Providers

Form presenting Query 1 data for Insurance Companies

Form presenting Query 2 data for Insurance Companies

Form presenting Query 3 data for Insurance Companies

Form presenting Query 4 data for Insurance Companies

Form presenting Query 5 data for Insurance Companies

Form presenting Query 6 data for Insurance Companies

Form presenting Query 7 data for Insurance Companies

Form presenting Query 1 data for Pharmacies

Form presenting Query 2 data for Pharmacies

Form presenting Query 3 data for Pharmacies

Form presenting data of the Person table

Form presenting data of the Patient table

Form presenting data of the InsurancePlan table

Form presenting data of the InsuranceCompany table

Form presenting data of the Claim table

Form presenting data of the ClaimService table

Form presenting data of the Provider table

Form presenting data of the ProviderSchedule table

Form presenting data of the ProviderLanguage table

Form presenting data of the Include table

Form presenting data of the Visit table

Form presenting data of the Treatment table

Form presenting data of the Prescription table

Form presenting data of the Contain table

Form presenting data of the Medicine table

Form presenting data of the Pharmacy table

Form presenting data of the Keep table

Form presenting Query 2 data for Healthcare Providers

Form presenting Query 3 data for Healthcare Providers

Form presenting Query 4 data for Healthcare Providers

Form presenting Query 5 data for Healthcare Providers

FIGURE 11.13 Hierarchy of Windows forms in the Healthcare Information System application.

229Windows Forms and Controls in Microsoft Visual Studio

Information System application is executed. All files in the Debug folder can be moved to another location on the computer, for example desktop, so that the user can double-click on the icon of the executable file on the desktop to execute the application.

11.6 Summary

This chapter describes the following:

• Setting up the Visual Basic development environment in Visual Studio 2013 • Changing the properties of a Windows form, adding a Windows form, setting up the

startup Windows form, and adding code to make transitions among Windows forms • Adding common controls, such as label, radio button, button, and text box, to a

Windows form to create a GUI of a Windows form • Adding event-driven programming code

The next chapter gives more details of Visual Basic programming in Visual Studio 2013.

FIGURE 11.14 Design View and Code View of FormEmployers in the Healthcare Information System application.

230 Developing Windows-Based and Web-Enabled Information Systems

EXERCISES

1. Create a GUI of the movie theater application using Visual Studio 2013 to include all the Windows forms, which are similar to those at the top and the middle levels in the hierarchy of Windows forms for the Healthcare Information System applica- tion shown in Figure 11.13. The list of the Windows forms is given as follows: • FormWelcome at the top level: Contains five radio buttons for movies, producers,

customers, statistics, and updates, and the Next button • FormMovies at the middle level to appear after a user selects the radio button for

movies in FormWelcome: Contains five radio buttons for Query 2, Query 3, Query 4, Query 8, and Query 9 in the movie theater database from Exercise 2 in Chapter 7, and the Back and the Next buttons

• FormProducers at the middle level to appear after a user selects the radio button for pro- ducers in FormWelcome: Contains one radio button for Query 7 in the movie the- ater database from Exercise 2 in Chapter 7, and the Back and the Next buttons

FIGURE 11.15 Design View and Code View of FormPatients in the Healthcare Information System application.

231Windows Forms and Controls in Microsoft Visual Studio

• FormCustomers at the middle level to appear after a user selects the radio button for customers in FormWelcome: Contains two radio buttons for Query 6 and Query 10 in the movie theater database from Exercise 2 in Chapter 7, and the Back and the Next buttons

• FormStatistics at the middle level to appear after a user selects the radio button for statistics in FormWelcome: Contains two radio buttons for Query 1 and Query 5 in the movie theater database from Exercise 2 in Chapter 7, and the Back and the Next buttons

• FormUpdates at the middle level to appear after a user selects the radio button for updates in FormWelcome: Contains the radio buttons for all the tables in the movie theater database from Exercise 1 in Chapter 7, and the Back and the Next buttons

Add code to the Click event of the Back and the Next buttons in the above Windows forms to make transitions among these Windows forms.

FIGURE 11.16 Design View and Code View of FormProviders in the Healthcare Information System application.

232 Developing Windows-Based and Web-Enabled Information Systems

FIGURE 11.17 Design View and Code View of FormCompanies in the Healthcare Information System application.

233Windows Forms and Controls in Microsoft Visual Studio

FIGURE 11.18 Design View and Code View of FormPharmacies in the Healthcare Information System application.

234 Developing Windows-Based and Web-Enabled Information Systems

FIGURE 11.19 Design View and Code View of FormUpdates in the Healthcare Information System application.

235

12 Visual Basic Programming in Microsoft Visual Studio

In Chapter 11, we add some Visual Basic programming code to subroutines for events of objects. This chapter describes various elements used to put together statements in Visual Basic code, including variables, constants, control structures, and operators used in expres- sions. We give examples of Visual Basic code that use these elements.

12.1 Variables and Data Types

A variable must be first declared before being used in a Visual Basic statement. The syntax for the variable declaration is

Dim <variable1, …, variableN> As <data type> [= <initial value>]

where Dim and As are the keywords. The keywords of Visual Basic appear in blue color in Visual Studio. The initial value in the square brackets is optional. We use square brackets to indicate an optional element. Some commonly used data types are as follows:

• Integer • Long • Double • String • Char • Boolean • Date • DateAndTime

where Long is used for a large integer, and Double is used for a value with decimals.

Example 12.1

Declare two variables, i and j, as Integer, with an initial value of 1, and declare the vari- able, note, as String.

The declaration of the variables is given as

Dim i, j as Integer = 1 Dim note as String

236 Developing Windows-Based and Web-Enabled Information Systems

The syntax of declaring an array is

Dim <array1, …, arrayN> (<size of the 1st dimension>, …, <size of the mth dimension>) As <data type>

Example 12.2

Declare the variable, students, as a one-dimensional array that keeps names of up to 100 students, and declare the variable, schoolstudents, as a two-dimensional array that keeps names of up to 100 students in each of 9 to 12 grades.

The declaration of the two arrays are given as

Dim students (99) as String Dim schoolstudents (3, 99) as String

Note that the size of the array for students is 99 because the index of an element in the array goes from 0 to 99. Hence, schoolstudents (3, 99) has four rows for four grades (9 to 12 grades) and 100 columns.

The ith element of the students array is referred to as students(i). The element in the ith row and the jth column of the schoolstudents matrix is referred to as schoolstudents(i, j).

Table 12.1 lists a number of functions to check and convert data types. Example 12.3 illustrates the use of some conversion functions.

Example 12.3

Create Visual Basic statements that use an InputBox function to take the user’s input of two numbers, perform the division of the two numbers, and use a MessageBox to display the integer result of the division.

The Visual Basic statements are given as

Dim Number1, Number2, Result as Integer Number1 = InputBox("Enter Number1:", "Division") Number2 = InputBox("Enter Number2:", "Division") Result = CInt(Number1/Number2) MessageBox.Show("The division result is:" & Result, "Division")

Figure 12.1 shows the two InputBoxes and the MessageBox.

TABLE 12.1

List of Functions to Check and Convert Data Types

Function Description

IsNumeric(<object name>) Return True if the object is numeric; False if otherwise. IsDate(<object name>) Return True if the object is a date; False if otherwise. IsArray(<object name>) Return True if the object is an array; False if otherwise. CInt(<object name>) Convert the object to an integer. CLng(<object name>) Convert the object to a long integer. CDbl(<object name>) Convert the object to a real value. CStr(<object name>) Convert the object to a string. CBool(<object name>) Convert the object to a Boolean value. CDate(<object name>) Convert the object to a date.

237Visual Basic Programming in Microsoft Visual Studio

The syntax of the InputBox function is

<variable> = InputBox(<prompt>, [title], [default response], [x coordinate of the InputBox on the screen], [y coordinate of the InputBox on the screen])

The syntax of the MessageBox is

MessageBox.Show(<message>, [title], [type], [icon])

The types of MessageBox are MessageBoxButtons.OKCancel, MessageBoxButtons. YesNoCancel, and MessageBoxButtons.AbortRetryIgnore, to include different sets of responses for the user to select. The icons of MessageBox are MessageBoxIcon.Warning, MessageBoxIcon.Information, MessageBoxIcon.Error, and MessageBoxIcon.Question to show four different icons in the MessageBox.

FIGURE 12.1 Examples of InputBoxes and MessageBox.

238 Developing Windows-Based and Web-Enabled Information Systems

12.2 Constants

A constant can be declared for a commonly used value, for example, sale tax rate. The syntax of declaring a constant is

Const <constant name> As <data type> [= <value>]

Example 12.4

Declare a constant for the sale tax rate, which is equal to 9%. The declaration of this constant is given as

Const tax As Double = 0.09

12.3 Operators and Mathematical Functions in Expressions

Table 12.2 gives some of the commonly used arithmetic operators, comparison operators, logical operators, string operators, and mathematical functions in Visual Basic, along with examples.

12.4 Control Structures

There are two control structures, branch and loop, which allow the program control to change from the sequential execution of Visual Basic statements. The branch structures include the If–Then–Else structure and the Select–Case structure. The syntax of the If– Then–Else structure is given as

If <condition> Then <body code> [Elseif <condition> Then <body code>] [Else <body code>] End If

Examples 12.5, 12.6, and 12.7 illustrate the use of the If–Then–Else structure.

Example 12.5

The code for the Click event of the Next button in Figure 11.19 uses the If–Then–Else structure to switch from the current form to another form based on which radio button is selected. Create a code that handles only one radio button. If the user selects the radio button, the program switches from the current form to another form. If the user does not select the radio button, the program does nothing.

239Visual Basic Programming in Microsoft Visual Studio

The code is given as

If RadioButtonU1.Checked Then Me.Hide() FormU1.Show() End If

Example 12.6

Create a code that is similar to the code in Example 12.5 to handle one radio button. If the user selects the radio button, the program switches from the current form to another form. If the user does not select the radio button, the program displays a message to prompt the user to select the radio button.

The code is given as

If RadioButtonU1.Checked Then Me.Hide() FormU1.Show()

TABLE 12.2

Operators and Some Mathematical Functions

Operator or Function Description Example

+ Addition a + b – Subtraction a − b * Multiplication a * b / Division a / b ^ Exponential a ^ b (a to the power of b) Mod Modulo a Mod b = Equal a = b <> Not equal a <> b > Greater than a > b >= Greater than or equal to a >= b < Less than a < b <= Less than or equal to a <= b AND Logical and a AND b OR Logical or a OR b NOT Logical not NOT a XOR Logical exclusive or a XOR b & Concatenate two strings a & b String.Length Return the length of a string String.Length(a) Math.Abs Return the absolute value Math.Abs(a) Math.Sqrt Return the square root Math.Sqrt(a) Math.Pow Return the power Math.Pow(a, b) (return a to the

power of b) Math.Log Return the natural logarithm Math.Log(a) Math.Log10 Return the base 10 logarithm Math.Log10(a) TimeOfDay Return the current time TimeOfDay DateString Return the current date DateString Year Return the year of a date as an integer Year(DateString) Month Return the month of a date as an integer Month(DateString)

240 Developing Windows-Based and Web-Enabled Information Systems

Else MessageBox("Please select the radio button.") End If

Example 12.7

Create a code that is similar to the code in Example 12.5 but handles two radio but- tons. If the user selects the radio button, the program switches from the current form to another form. If the user does not select the radio button, the program does nothing.

The code is given as

If RadioButtonU1.Checked Then Me.Hide() FormU1.Show() Elseif RadioButtonU2.Checked Then Me.Hide() FormU2.Show() End If

The syntax of the Select–Case structure is given as

Select Case <test expression> Case Is <expression> <body> … [Case Is <expression> <body code>] Case Else <body code> End Select

Example 12.8 illustrates the use of the Select–Case structure.

Example 12.8

Write a Visual Basic code for Example 12.6 using the Select–Case structure. The code is given as

Select Case RadioButtonU1.Checked Case True Me.Hide() FormU1.Show() Case Else MessageBox("Please select the radio button.") End Select

The loop structures include the Do–Loop–While, the Do–While–Loop, the Do–Loop– Until, the Do–Until–Loop, and the For–Next structures. The syntax and example of each loop structure are given as follows:

The syntax of the Do–Loop–While structure is

Do <body code> Loop While (<condition>)

241Visual Basic Programming in Microsoft Visual Studio

This loop structure goes into the body of the loop and then checks the condition. If the condition is true, the program goes back into the body of the loop; otherwise, the program gets out of the loop.

The syntax of the Do–While–Loop structure is

Do While (<condition>) <body code> Loop

This loop structure first checks the condition. If the condition is true, the program goes into the body of the loop; otherwise, the program gets out of the loop.

The syntax of the Do–Loop–Until structure is

Do <body code> Loop Until (<condition>)

This loop structure goes into the body of the loop and then checks the condition. If the condition is true, the program gets out of the loop; otherwise, the program goes back into the body of the loop.

The syntax of the Do–Until–Loop structure is

Do Until (<condition>) <body code> Loop

This loop structure first checks the condition. If the condition is true, the program gets out of the loop; otherwise, the program goes back into the body of the loop.

The syntax of the For–Next structure is

For <counter> = <first value> To <last value> [Step <increment value>] <body code> Next

This loop structure goes into the body of the loop for a number of times from the first value to the last value of the counter.

The above five loop structures are illustrated in Examples 12.9, 12.10, 12.11, 12.12, and 12.13, respectively. The code in each of the five examples implements the same function that takes two input integers, a and b, from the user, computes base a to the power of b, and displays the result.

Example 12.9

Create a Visual Basic code for the power function using the Do–Loop–While structure. The code is given as

Dim a, b As Integer Dim result As Integer = 1 a = InputBox("Enter the base number.") b = InputBox("Enter the power number.") Do result = result * a b = b − 1

242 Developing Windows-Based and Web-Enabled Information Systems

Loop While (b > 0) MessageBox.Show("The result is:" & result)

Example 12.10

Create a Visual Basic code for the power function using the Do–While–Loop structure. The code is given as

Dim a, b As Integer Dim result As Integer = 1 a = InputBox("Enter the base number.") b = InputBox("Enter the power number.") Do While (b > 0) result = result * a b = b − 1 Loop MessageBox.Show("The result is:" & result)

Example 12.11

Create a Visual Basic code for the power function using the Do–Loop–Until structure. The code is given as

Dim a, b As Integer Dim result As Integer = 1 a = InputBox("Enter the base number.") b = InputBox("Enter the power number.") Do result = result * a b = b − 1 Loop Until (b = 0) MessageBox.Show("The result is:" & result)

Example 12.12

Create a Visual Basic code for the power function using the Do–Loop–Until structure. The code is given as

Dim a, b As Integer Dim result As Integer = 1 a = InputBox("Enter the base number.") b = InputBox("Enter the power number.") Do Until (b = 0) result = result * a b = b − 1 Loop MessageBox.Show("The result is: " & result)

Example 12.13

Create a Visual Basic code for the power function using the For–Next structure. The code is given as

'i is the counter for the For–Next loop Dim a, b, i As Integer Dim result As Integer = 1 a = InputBox("Enter the base number.") b = InputBox("Enter the power number.")

243Visual Basic Programming in Microsoft Visual Studio

For i = 1 to b result = result * a Next MessageBox.Show("The result is:" & result)

The first line of the above node is a comment line that starts with '.

12.5 Summary

This chapter gives the syntax and examples of the various elements used to put together a Visual Basic code, including variable declaration; constant declaration; and some of the commonly used arithmetic, comparison, logical, and string operators, mathematical func- tions, and branch and loop control structures. The next chapter describes how to embed an Access database in a Windows-based application.

EXERCISES

1. Create a Windows form containing a currency converter that takes an amount in one currency and converts the amount to another currency.

2. Create a Windows form containing a simple calculator that can perform addition, subtraction, multiplication, and division.

3. Create a Window form that lets a user enter an integer number, computes and displays the factorial of this number.

4. Create a Visual Basic code that sorts and rearranges values in an array of 100 inte- gers to a nondecreasing order of these values.

245

13 Database Connection in Microsoft Visual Studio

This chapter describes how to connect to an Access database in a Windows-based appli- cation so that we can present data from tables and queries in the Access database in the Windows-based application. We first introduce how to add an Access database as a data source to an application. We then bring data from tables and simple Select queries to an application. We also describe other capacities such as handling Crosstab and Select queries with a parameter, displaying related data from parent and child tables, and writing Visual Basic code to retrieve, add, and delete data of an Access database.

13.1 Adding an Access Database as a Data Source

We can select Add New Data Source in the Project menu or open the Data Sources win- dows and click the Add New Data Source icon to start the procedure of adding an Access database as a data source in a Windows-based application. Figure 13.1 shows the proce- dure, which includes the following steps in the Data Source Configuration Wizard:

1. Select Database in the Choose a Data Source Type window. 2. Select Dataset in the Choose a Database Model window. 3. Click the New Connection button in the Choose Your Data Connection window. 4. Select Microsoft Access Data File and click the Continue button in the Choose

Data Source window. 5. Use the Browse button to select an Access database file on the computer (e.g.,

Healthcare Access Database.accdb), click the Test Connection button to see a mes- sage showing “Test connection succeeded,” and click the OK button, all in the Add Connection window, which shows Microsoft Access Database File labeled as “OLB DB.”

6. Click the Next button in the Choose Your Data Connection window, which now shows the selected Access database file.

7. Click the Yes button in the message box to copy the database file to the project folder.

8. Click the Next button in the Save the Connection String to the Application Configuration File window showing that the connection to Healthcare Access Database.accdb is saved as “Healthcare_Access_DatabaseConnectionString.”

9. Select both Tables and Views, expand Tables to see the list of tables, expand Views to see the list of queries, and click the Finish button, all in the Choose Your Database Objects window, which shows Healthcare_Access_DatabaseDataSet as the Dataset name.

246 Developing Windows-Based and Web-Enabled Information Systems

(a)

(b)

FIGURE 13.1 Procedure of adding an Access database as a data source in a Windows-based application. (a) Choose a data source type. (b) Choose a database model.

247Database Connection in Microsoft Visual Studio

(c)

FIGURE 13.1 (Continued) Procedure of adding an Access database as a data source in a Windows-based application. (c) Choose a data connection.

248 Developing Windows-Based and Web-Enabled Information Systems

(d)

(e)

FIGURE 13.1 (Continued) Procedure of adding an Access database as a data source in a Windows-based application. (d) Add connection. (e) Choose a data connection.

249Database Connection in Microsoft Visual Studio

(f )

(g)

FIGURE 13.1 (Continued) Procedure of adding an Access database as a data source in a Windows-based application. (f) Save the connec- tion string. (g) Choose database objects.

250 Developing Windows-Based and Web-Enabled Information Systems

In Step 9, we see the list of all the tables in the Healthcare Access database and a list of the following, but not all, queries in the Healthcare Access database:

• IQ1 • IQ1a • IQ1b • IQ1bb • IQ1c • IQ1cc • IQ3 • PAQ4 • PRQ1

The queries that are in the Healthcare Access database but are not included in the list of queries for Views are a Crosstab query (EQ3), a number of Select queries with parameter(s), and action queries. Section 13.5 describes how to bring a Crosstab query and a query with parameter(s) into a Windows-based application. Section 13.6 shows how to update data of an Access database in a Windows-based application.

FIGURE 13.2 Tables and views in the data set for the Healthcare Access database.

251Database Connection in Microsoft Visual Studio

When we open the Data Sources window, we see the list of the Healthcare_Access_ DatabaseDataSet, which includes all the tables and the queries in the Healthcare Access database, excluding Crosstab queries, Select queries with parameter(s), and action queries, as shown in Figure 13.2.

13.2 Presenting Data from Tables and Simple Select Queries

Data from a table can be brought to a Windows form by dragging the table from Healthcare_ Access_DatabaseDataSet in the Data Sources window and dropping it in the Windows form, which creates a DataGridView to present data from the table.

Example 13.1

When the user selects the Person radio button in FormUpdates of the Healthcare Information System application, the program brings up FormU1 to present data from the Person table. Add the Person table to FormU1.

Figure 13.3a shows the Design View of FormU1 after we drag the Person table from Healthcare_Access_DatabaseDataSet in the Data Sources window and drop it to FormU1, which adds the Toolstrip and the DataGridView. The Toolstrip shows a number indicating the selected record, the total number of records in the DataGridView, navigation buttons to go through records back and forth, the Add New button to add a new record, the Delete but- ton to delete a selected record, and the Save Data button. The DataGridView is a graphical user interface (GUI) control (named “PersonDataGridView” in the Properties window) on FormU1 for presenting data from the Person table in Healthcare_Access_DatabaseDataSet. Figure 13.3b shows data of the Person table presented in the DataGridView when we exe- cute the Healthcare Information System application.

A query available in Healthcare_Access_DatabaseDataSet in the Data Sources window can be dragged and dropped to a Windows form to present the query result in the Windows form.

Example 13.2

Add the query, PRQ1, to FormPR1 in the Healthcare Information System application to present the query result showing patients whose annual physical exam is overdue.

Figure 13.4 shows the Design View of FormPR1 after we drag the query, PRQ1, from Healthcare_Access_DatabaseDataSet in the Data Sources window and drop it to FormPR1.

13.3 Architecture of Database Connection

As seen in Figure 13.3, five objects related to the database connection are added to a win- dow below the Design View window of FormU1 when we drag and drop the Person table in Healthcare_Access_DatabaseDataSet to FormU1. These five objects are as follows:

• Healthcare_Access_DatabaseDataSet • PersonBindingSource

252 Developing Windows-Based and Web-Enabled Information Systems

• PersonTableAdapter • TableAdapterManager • PersonBindingNavigator

Similarly, five objects are added to a window below the Design View window of FormPR1 when we drag and drop the query, PRQ1, in Healthcare_Access_DatabaseDataSet to FormPR1, as shown in Figure 13.4. These five objects are as follows:

(a)

(b)

FIGURE 13.3 FormU1 to present the data of the Person table. (a) Design View of the FormU1 in the Healthcare Information System application to present data of the Person table. (b) Data of the Person table presented when the applica- tion is executed.

253Database Connection in Microsoft Visual Studio

• Healthcare_Access_DatabaseDataSet • PRQ1BindingSource • PRQ1TableAdapter • TableAdapterManager • PRQ1BindingNavigator

Figure 13.5 shows how these objects work together to enable the connection between a database and a data presentation in a DataGridView on a Windows form. When we make the connection to the Healthcare Access database, as described in Section 13.1, Healthcare_Access_DatabaseDataSet is created to include data tables that keep data in objects (e.g., tables and queries) of the Healthcare Access database in the memory for the Healthcare Information system application. That is, the data tables in Healthcare_Access_ DatabaseDataSet are the memory storage of data from the Healthcare Access database for the Healthcare Information System application.

FIGURE 13.4 Design View of Form PR1, showing the query, PRQ1, to present patients whose annual physical exam is overdue.

Database

TableAdapter

Dataset DataGridView Bind

Fill

Update

BindingSource

FIGURE 13.5 Architecture of database connection.

254 Developing Windows-Based and Web-Enabled Information Systems

The path and other information of the connection are stored in the ConnectingString, which can be found by double-clicking App.config in the Solution Explore window to open App.config, which contains the following:

ConnectionString="Provider=Microsoft.ACE.OLEDB.12.0;DataSource=&quot;|Data Directory|\HealthcareAccessDatabase.accdb&quot;"

There is a TableAdapter for each object (e.g., table or query) in the Healthcare Access database for making the connection between the database object and the correspond- ing object in Healthcare_Access_DatabaseDataSet. A TableAdapter supports filling a data table in Healthcare_Access_DatabaseDataSet with the query result by executing a Structured Query Language (SQL) statement of a query on the database. A TableAdapter also supports propagating updates to a data table in Healthcare_Access_DatabaseDataSet to the corresponding table in the database by executing an SQL statement for Update. A TableAdapter holds an SQL statement of a query and two default methods: Fill and GetData. The Fill method executes the SQL statement of a query and fills a data table in Healthcare_Access_DatabaseDataSet with the query result. The GetData method executes an SQL statement for Update, which propagates updates to a data table in Healthcare_ Access_DatabaseDataSet back to the corresponding table in the database.

A TableAdapter and its Fill and GetData methods can be viewed by double-clicking Healthcare_Access_DatabaseDataSet in the Solution Explorer window to bring up the Design View window of Healthcare_Access_DatabaseDataSet. The Design View win- dow of Healthcare_Access_DatabaseDataSet can also be brought up by selecting a TableAdapter below the Design View of a Windows form containing a DataGridView, right-clicking on the right arrow button at the top-right corner of the TableAdapter, and selecting Edit Queries in DataSet Designer. Figure 13.6 shows a part of the Design View of Healthcare_Access_DatabaseDataSet, which includes the TableAdapters for all the tables in the Healthcare Access database, relationships of the tables, and certain queries. Right- clicking a TableAdapter in the Design View of Healthcare_Access_DatabaseDataSet brings up a pop-up menu. Selecting Configure in the pop-up menu brings up the TableAdapter

FIGURE 13.6 Part of the Design View of Healthcare_Access_DatabaseDataSet.

255Database Connection in Microsoft Visual Studio

Configuration Wizard (see Figure 13.7), which shows the SQL statement of a query held by the TableAdapter.

Each DataGridView has a BindingSource as the value of the Data Source property of the DataGridView to bind the DataGridView to the corresponding data table in Healthcare_ Access_DatabaseDataSet for bringing data in the data table in Healthcare_Access_ DatabaseDataSet to the DataGridView. When a Windows form containing a DataGridView is loaded, the subroutine for the Load event of the Windows form object contains the code to execute the Fill method of the TableAdapter, which is to execute the SQL statement of the query held by the TableAdapter and fill the DataGridView with the query result. Figure 13.8 shows the code for the Load event of FormU1 to execute the Fill method of the

FIGURE 13.7 SQL statement of a Select query held by PersonTableAdapter for the Person table in the Healthcare Access database.

FIGURE 13.8 Code of FormU1 in the Healthcare Information System application.

256 Developing Windows-Based and Web-Enabled Information Systems

PersonTableAdapter. The code is automatically added to the Load event of FormU1 when we drag and drop the Person table from Health_Access_DatabaseDataSet. The code is executed to present data of the Person table in Healthcare_Access_DatabaseDataSet when FormU1 is loaded.

A BindingNavigator enables the user to navigate and manipulate data in the DataGridView through the Toolstrip. Figure 13.8 shows the code for the Click event of the Save Data icon in the Toolstrip, PersonBindingNavigatorSaveItem_Click. The code is automatically added when we drag and drop the Person table from Health_Access_DatabaseDataSet. The code uses the UpdateAll method of the TableAdapterManager to update the Person table in Healthcare_Access_DatabaseDataSet.

There are two components in the list of project components in the Solution Explorer win- dow: Healthcare Access Database.accdb and Healthcare_Access_DatabaseDataSet.xsd. Double-clicking Healthcare Access Database.accdb brings up the Server Explorer window. We can also bring up the Server Explorer window by selecting Server Explorer in the View menu. As shown in Figure 13.9, the Data Connections section of the Server Explorer win- dow contains Healthcare Access Database.accdb, which includes the following:

• All the tables in the Tables folder • Queries that are also available in the Healthcare_Access_DatabaseDataSet in the

Views folder • A Crosstab query (EQ3) and two action queries (IQ8 and IQ9) in the Stored

Procedures folder • Queries with parameter(s) in the Functions folder

Hence, the Server Explorer window shows all the objects of the Healthcare Access data- base in Healthcare Access Database.accdb, whereas the Data Sources window shows the data tables and views in HealthcareAccessDatabaseDataSet. Because there are no data tables in Healthcare_Access_DatabaseDataSet for the queries in the Stored Procedures and the Functions folders shown in the Server Explorer, we cannot directly add the queries in these two folders to a Windows form.

There are two copies of the Healthcare Access database in the project folder of the Healthcare Information System application, with one copy in the Debug folder of the bin folder (the database file in the execution environment) and the other copy along with forms and other objects of the application in the folder with the name of the project (the database

FIGURE 13.9 Server Explorer window.

257Database Connection in Microsoft Visual Studio

file in the design environment). When we execute the Healthcare Information System application, Healthcare_Access_DatabaseDataSet is bound to the copy of the Healthcare Access database in the execution environment, and any database changes made by the application are made to this copy of the Healthcare Access database. However, when we open the Healthcare Information System application file in Visual Studio to design the application, the copy of the Healthcare Access database in the design environment is used.

13.4 Displaying Related Data in Parent and Child Tables

In the Healthcare Access database, data in the Person table (the parent table) are related to data in the CareProvider table (the child table) through the common data fields (ID in the Person table and ProviderID in the CareProvider table). Data in the parent and child tables can be displayed in a way to relate them together. Example 13.3 shows the procedure of displaying related data in parent and child tables and the display of related data.

Example 13.3

With data of the Person table added to FormU1 in Example 13.1, add data of the CareProvider table to FormU1 so that data of the Person table and data of the CareProvider table are displayed in a synchronized way.

The following steps are taken:

1. Double-click FormU1.vb in the Solution Explorer window to bring up FormU1 in the Design View.

2. Double-click the Person table in the Data Sources window to expand the Person table and bring up a list of the data fields in the Person, the CareProvider, and the Patient tables, which are related to the Person table (see Figure 13.10a).

3. Drag and drop the CareProvider table in the expanded list for the Person table to FormU1 (see Figure 13.10a). Note that dragging and dropping the CareProvider table to FormU1 adds CareProviderBindingSource and CareProviderTableAdapter to FormU1 (see Figure 13.10a).

Figure 13.10b shows the display of related data from the Person and the CareProvider tables. Note that only one record of the CareProvider table corresponding to the selected record in the Person table is displayed. If the selected record in the Person table (e.g., the record with the ID value of PA1) does not have a corresponding record in the CareProvider table, nothing is displayed in the CareProvider table. Hence, when we use the Toolstrip to navigate one record to another record in the Person table, the record displayed in the CareProvider table changes accordingly.

Because the ID field of the Person table and the ProviderID field of the CareProvider table are the common data fields and have the same value, we do not need to display the ProviderID field. We can delete the ProviderID field by selecting the CareProvider table in FormU1 and clicking the right arrow at the top-right corner of the CareProvider table to bring up a pop-up menu (see Figure 13.10a). Selecting Edit Columns in the menu brings up the Edit Columns window, as shown in Figure 13.11. By selecting ProviderID in the Selected Columns area and clicking the Remove button, we remove ProviderID from the CareProvider table on FormU1.

In Figure 13.10a, the CareProvider table is in the list of data tables in the Data Sources window and is also embedded in the expanded list of data fields and child table(s) for

258 Developing Windows-Based and Web-Enabled Information Systems

(a)

(b)

FIGURE 13.10 Use of the Data Sources window to display related data in parent and child tables. (a) Design View of FormU1 with related data from the Person table and the CareProvider table. (b) Display of related data from the Person table and the CareProvider table when the application is executed.

259Database Connection in Microsoft Visual Studio

the Person table. To display related data of parent and child tables, we have to drag and drop the child table embedded in the parent table. For example, if we drag and drop the CareProvider table in the list of data tables in the Data Sources window (not the embedded child table) to FormU1, data of the Person and the CareProvider tables in FormU1 will not be related. That is, the selected record of the Person table and the displayed record of the CareProvider table are not necessarily for the same person and do not necessarily have the same value of the common data fields.

13.5 Displaying Data Fields Using Controls Other Than DataGridView

In Examples 13.1 and 13.3, all records and data fields of the Person table are displayed in the DataGridView control on FormU1 after we drag and drop the Person table to FormU1. Using the Edit Columns window shown in Figure 13.11, we can display only the selected data fields of a table in the DataGridView control. We can also display the selected data fields and only one record of a table using a control for an individual data field rather than the DataGridView control. Example 13.4 shows the procedure for this.

Example 13.4

Add the PatientID, SSN, DateOfBirth, Employer, and Position fields of the Patient table to FormU2 in the Healthcare Information System application using the TextBox control to display each data field.

FIGURE 13.11 Edit Columns window.

260 Developing Windows-Based and Web-Enabled Information Systems

The following steps are taken:

1. Double-click the Patient table in the Data Sources window to expand it and bring up a list of data fields and child table(s) for the Patient table (see Figure 13.12a).

2. Select the PatientID field and click the down arrow next to the field to view a list of controls that can be used to display the data field. Although the default control for PatientID is set to TextBox as shown at the left side of the field (see Figure 13.12a), another control can be selected if needed.

3. Drag and drop PatientID to FormU2. 4. Repeat Step 3 for the SSN, DateOfBirth, Employer, and Position fields.

(a)

(b)

FIGURE 13.12 Display of individual data fields and one record at a time from the Patient table in the Healthcare Information System application. (a) Design View of FormU2 and the Data Sources window. (b) Display of individual data fields in one record from the Patient table.

261Database Connection in Microsoft Visual Studio

Figure 13.12b shows the display of one record containing the values for these data fields from the Patient table. We can navigate from one record to another record of the Patient table using the Toolstrip.

13.6 Handling Crosstab and Select Queries with Parameters

EQ3 in the Healthcare Access database is a Crosstab query and is taken as a stored pro- cedure after we add the Healthcare Access database as a data source to the Healthcare Information System application. As a stored procedure, EQ3 is not included in Healthcare_ Access_DatabaseDataSet and cannot be directly dragged and dropped to a Windows form to display the query result. To display the query result of a Crosstab query in an Access database such as EQ3, we can create a regular Select query that embeds the Crosstab query. The regular Select query is included in Healthcare_Access_DatabaseDataSet of the data- base in a Windows-based application. Example 13.5 illustrates how to include a Crosstab query in an Access database in a Windows-based application.

Example 13.5

Create a Select query, EQ3-Select, which embeds EQ3 in the Healthcare Access database; update Healthcare_Access_DatabaseDataSet in the Healthcare Information System application; and add EQ3-Select to FormE3.

Figure 13.13a shows the Design view of EQ3, the Crosstab query, in the healthcare Access database. Figure 13.13c shows the query result of EQ3. We open the healthcare Access database in the project folder of the Healthcare Information System applica- tion, and create EQ3-Select by adding EQ3 to the table pane and putting EQ3.* in the data field. Figure 13.13b shows the Design View of EQ3-Select. The query result of EQ3-Select is the same as that shown in Figure 13.13c. To update Healthcare_ Access_DatabaseDataSet in the Healthcare Information System application, we click the Configure Data Source with Wizard icon in the Data Sources window, select Continue with Wizard in the Choose DataSet Editor window, check Views in the Choose Your Database Objects window, and click the Finish button. Figure 13.14 shows the updated Healthcare_Access_DatabaseDataSet, which includes EQ3-Select. We drag and drop EQ3-Select from Healthcare_Access_DatabaseDataset to FormE3 to display the query result.

All the other queries in the Healthcare Access database but not in the Healthcare_Access_ DatabaseDataSet in the Healthcare Information System application are queries that request parameter input(s) from the user. Such queries with parameter(s) are shown as functions in the Server Explorer window under Healthcare Access Database.accdb. Because a query with parameters is not included in the DataSet of an Access database, we need to create the query using the Query Builder in Visual Studio. Example 13.6 shows how to create a query with a parameter input from the user using the Query Builder in Visual Studio.

Example 13.6

Create EQ1 using the Query Builder in Visual Studio and display the query result of EQ1 in FormE1 of the Healthcare Information System application.

262 Developing Windows-Based and Web-Enabled Information Systems

(a)

(b)

(c)

FIGURE 13.13 Creation of a Select query that embeds a Crosstab query. (a) Design View of EQ3, a Crosstab query. (b) Design View of EQ3-Select. (c) Query result of EQ3 and EQ3-Select.

263Database Connection in Microsoft Visual Studio

Figure 13.15 shows the Design View of EQ1 in the healthcare Access database. The procedure of creating EQ1 in Visual Studio includes the following steps:

1. We double-click Healthcare_Access_DatabaseDataSet.xsd in the Solution Explorer window to bring up the Design View of Healthcare_Access_DatabaseDataSet.xsd. The Design View of Healthcare_Access_DatabaseDataSet.xsd contains all the

FIGURE 13.14 Healthcare_Access_DatabaseDataSet in the Data Sources window including EQ3-Select and the Design View of FormE3.

FIGURE 13.15 The Design View of EQ1 in the healthcare Access database.

264 Developing Windows-Based and Web-Enabled Information Systems

tables, relationships of the tables, and individual queries that are taken as Views and included in Healthcare_Access_DatabaseDataSet.

2. We right-click the Design View area of Healthcare_Access_DatabaseDataSet. xsd to bring up a pop-up menu and select Add and then TableAdapter. This brings up the TableAdapter Configuration Wizard, as shown in Figure 13.16. In the Choose Your Data Connection window of the TableAdapter Configuration Wizard, we keep Healthcare_Access_DatabaseConnectionString and click the Next button.

3. In the next Choose a Command Type window of the TableAdapter Configura- tion Wizard, we select Use SQL Statements and click the Next button.

4. In the next Enter an SQL Statement window of the TableAdapter Configura- tion Wizard, we click the Query Builder button, which brings the Add Table and the Query Builder windows. We use the Add Table window to add the InsuranceCompany and the InsurancePlan tables to the Query Builder, and then close the Add Table window.

The Query Builder window contains four areas: the table pane, the design grid, the SQL statement, and the query result. The table pane and the design grid are similar to those in the Design View of a query in Access except a dif- ferent orientation of the Design grid. Hence, we design a query as we do it in Access. The Design grid has the following: • Column to include a data field in the query • Alias to give a display name for the column • Output to indicate whether we want to display the column in the query

result • Sort Type and Sort Order to determine how to sort the query result • Filter to specify the selection criterion.

Right-clicking the table pane brings up a pop-up menu. Selecting Add Group By in the pop-up menu adds Group By in the Design grid.

5. We select CompanyName for the first row in the Design grid and check Output to display CompanyName in the query result.

6. We select PlanName for the second row in the Design grid and check Output to display PlanName in the query result.

7. We select EmployerPremium for the third row in the Design grid and put the selection criterion:

<= ?

in Filter to prompt for the user input of a given level of the employer premium. This completes the query design. As we build the query in the Design grid in Steps 5 to 7, the SQL statement for the query is automatically changed to reflect the query design in the Design grid. The SQL statement is shown as follows after Step 7:

SELECT InsuranceCompany.CompanyName, InsurancePlan.PlanName FROM (InsuranceCompany INNER JOIN InsurancePlan ON InsuranceCompany.CompanyID = InsurancePlan.CompanyID) WHERE (InsurancePlan.EmployerPremium <= ?)

8. After we click the Execute Query button, the query result is shown. 9. We click the OK button in the Query Builder window. The Enter an SQL state-

ment window now shows the completed SQL statement. 10. We click the Next button.

265Database Connection in Microsoft Visual Studio

(a)

(b)

FIGURE 13.16 Procedure of using the Query Builder in Visual Studio to create EQ1, a query with a parameter input. (a) Choose a data connection. (b) Choose a command type.

266 Developing Windows-Based and Web-Enabled Information Systems

(c)

(d)

FIGURE 13.16 (Continued) Procedure of using the Query Builder in Visual Studio to create EQ1, a query with a parameter input. (c) Enter an SQL statement. (d) Add table.

267Database Connection in Microsoft Visual Studio

(e)

(f )

FIGURE 13.16 (Continued) Procedure of using the Query Builder in Visual Studio to create EQ1, a query with a parameter input. (e) Query builder. (f) Enter an SQL statement.

268 Developing Windows-Based and Web-Enabled Information Systems

11. In the Choose Methods to Generate window, we select the Fill and GetData methods and click the Next button.

12. In the Wizard Results window, we click the Finish button. 13. Now, the Design View of Healthcare_Access_DatabaseDataset includes the

new data table and the new table adaptor. We double-click the names of the new data table and the new table adaptor to rename them to EQ1 and EQ1TableAdapter, respectively.

14. Now, the Data Sources window includes EQ1. 15. We close the Design View of Healthcare_Access_DatabaseDataSet.xsd,

bring up the Design View of FormE1, and drag and drop EQ1 from the Data Sources window to FormE1. Figure 13.17 shows the Design View of FormE1. In addition to the DataGridView control and the Toolstrip, the label, EmployerPremium, the text box for the user to enter a value for a level of EmployerPremium, and the Fill button are added to FormE1. When FormE1 is loaded during the execution of the application, we enter a value in the text box for EmployerPremium. After we click the Fill button, the query result is displayed in the DataGridView.

When using the query builder to design a query in Visual Studio, some features of the query design in Access do not work in Visual Studio. For example, in Access, the asterisk (*) is used to match any string, whereas in Visual Studio, the percentage (%) should be used to match any string. In Access, a pair of double quotes is used to enclose a string, whereas in Visual Studio, a pair of single quotes is used to enclose a string. Hence, the selection criterion such as “PA*” in Access should be ‘PA%’ in Visual Studio.

(g)

FIGURE 13.16 (Continued) Procedure of using the Query Builder in Visual Studio to create EQ1, a query with a parameter input. (g) Choose methods to generate.

269Database Connection in Microsoft Visual Studio

In Access, brackets with a prompt, e.g., [Enter a provider name], are used to create a parameter input in a query design. In Visual Studio, the question mark (?) is used in a place where a parameter input is needed (see the query design in Example 13.6). Some time-related functions such as Year() and Month() in Access also do not work in Visual Studio. In Visual Studio, the function datepart() can be used to extract a specific part of a date/time. Examples of extracting the year, quarter, month, day, day of year, week, hour, minute, and second of the current date.time using datepart() in Visual Studio are given below:

Year: datepart(‘yyyy’, now()) Quarter: datepart(‘q’, now()) Month: datepart(‘m’, now())

(h) (i)

FIGURE 13.16 (Continued) Procedure of using the Query Builder in Visual Studio to create EQ1, a query with a parameter input. (h) Wizard results. (i) Data sources.

270 Developing Windows-Based and Web-Enabled Information Systems

(a)

(b)

FIGURE 13.17 Design View of FormE1 in the Healthcare Information System application and the display of the query result. (a) Design View of FormE1 showing the query result of EQ1, a query with a paramater input. (b) The display of the query result when FormE1 is executed.

271Database Connection in Microsoft Visual Studio

Day: datepart(‘d’, now()) Day of year: datepart(‘y’, now()) Week: datepart(‘ww’, now()) Hour: datepart(‘h’, now()) Minute: datepart(‘n’, now()) Second: datepart(‘s’, now())

13.7 Updating Data to an Access Database

The Toolstrip associated with a DataViewGrid control contains the following:

• Add new button for the user to add a new row of data in the DataGridView control

• Delete button for the user to delete a row of data in the DataGridView control • Save Data button to save changes in the DataGridView control to the DataSet and

update the database using the DataSet

Figure 13.18 shows the code for the Save Data button in FormU3, which is automati- cally generated when we drag and drop the InsurancePlan table from Healthcare_Access_ DatabaseDataSet to FormU3. In the subroutine for the Click event of the Save Data button, the first line of code:

Me.Validate()

validates and saves changes made in the DataGridView control to the Healthcare_Access_ DatabaseDataSet. The second line of code:

Me.InsurancePlanBindingSource.EndEdit()

stops editing to Healthcare_Access_DatabaseDataSet. The third line of code:

Me.TableAdapterManager.UpdateAll(Me.Healthcare_Access_DatabaseDataSet)

executes the Update SQL statement held by TableAdapterManager to propagate the con- tent of the Healthcare_Access_DatabaseDataSet to the healthcare Access database in the Debug folder of the bin folder. Hence, the user can use the Toolstrip to update data to a database.

FIGURE 13.18 Code for the Save Data button in the Toolstrip.

272 Developing Windows-Based and Web-Enabled Information Systems

13.8 Visual Basic Code to Retrieve, Add, and Delete Data of an Access Database

Instead of using the DataGridView control and the Toolstrip, we can use other Windows controls and code to add a row of data, delete a row of data, update a value, and then update a database with changes as shown in Figure 13.19. Figure 13.19a shows a Windows form with the InsurancePlan table dragged and dropped from Healthcare_Access_ DatabaseDataset, along with the Toolstrip. On the left side below the InsurancePlan table on the Windows form, GUI controls including labels, textboxes for PlanID, PlanName, Description, CompanyID, MemberPremium, and EmployerPremium, and the button for Enter the New Plan, are added. Figure 13.19b gives the Visual Basic code to be exe- cuted for adding a new data record to the InsurancePlan table in Healthcare_Access_ DatabaseDataSet and in the healthcare Access database, when the user enters the values for PlanID, PlanName, Description, CompnanyID, MemberPremium, and EmployerPremium in the textboxes and clicks the button for Enter the New Plan. The comments in the code explain the purpose of each code block. In the middle part below the InsurancePlan table on the Windows form, GUI controls, including the labels, three textboxes for PlanID, MemberPremium, and EmployerPremium and the button for Update Premiums, are added. Figure 13.19c gives the Visual Basic code to retrieve a data record with the value of PlanID entered by the user and update the value of MemberPremium and the value of EmployerPremium, when the user enters the parameter inputs and click on the button for Update Premiums. On the right side below the InsurancePlan table on the Windows form, GUI controls, including the labels, a textbox and a button for Delete the Plan, are added. Figure 13.19d gives the Visual Basic code to delete the data record with the value of PlanID entered by the user, when the user enters the parameter input and clicks on the button for Delete the Plan.

Example 13.7 shows how to add the login function to the Healthcare Information System application.

Example 13.7

Add the login function to the Healthcare Information System application as follows. When the user selects the radio button for Updates and clicks on the Next Button on the Windows form, FormMain, the user is taken to the Windows form, FormLogin, which allows the user to enter a username and a password. The application checks to see if the username and the password matches any pair of username and password kept by the application. If there is a match, the user is taken to FormUpdates, which allows the user to access the list of the tables in the healthcare Access database for updating data of the tables. If there is no match, a message is displayed to tell the user that the access is denied.

We first use Access to add a new table, Login, to the healthcare Access database. Figure 13.20a shows a data record with a username and a password stored in the Login table. We then open the Healthcare Information System application in Visual Studio. Since we add the Login table in the healthcare Access database, we need to update the healthcare Access database embedded in the Healthcare Information System applica- tion. To update the healthcare Access database in the Healthcare Information System application, we use the Solution Explorer window to delete the existing healthcare Access database and the existing Hearlthcare_Access_DatabaseDataSet, and then use the Data Sources window to add a new data source using the updated healthcare

273Database Connection in Microsoft Visual Studio

(a)

(b)

FIGURE 13.19 Design View of FormU3 in the Healthcare Information System application and the code for the three buttons for updating the InsurancePlan table in the Healthcare Access database. (a) Design View of FormU3. (b) Code for the Enter the New Plan button.

274 Developing Windows-Based and Web-Enabled Information Systems

Access database. Next, we add the Windows form, FormLogin, to the Healthcare Information System application. We add GUI controls, including the labels, two text- boxes for username and password, the button for Login, and the button for Back as shown in Figure 13.20b. We also drag and drop the Login table from Healthcare_ Access__DatabaseDataSet to FormLogin. This adds five objects to FormLogin: HealthcareAccessDatabaseDataSet, LoginBindingSource, LoginTableAdapter, TableAdapterManager, and LoginBindingNavigator (see Figure 13.20b). Although we need LoginTableAdapter in the Visual Basic code for the Login button, we do not really need the Login table and the Toolstrip on FormLogin. Hence, we select the Login table on the Windows form and press the Delete button on the keyboard to delete the Login form. We right-click on LoginBindingNavigator and select Delete in the pop-up menu to delete LoginBindingNavigator and the Toolstrip. We now double- click the Login button in the Design view of FormLogin to add the Visual Basic code (see Figure 13.20c) to check the match of the entered username and password with any pair of username and password stored in the Login table of the healthcare Access database. We double-click the Back button in the Design view of FormLogin to add the code for this button (see Figure 13.20c). Note that in the code for the Login button, the underline, __, is a line breaker to break one line of code into code written in two lines. At last, we update the Visual Basic code (see Figure 13.20d) for FormMain so that when the user selects the radio button for Updates and clicks on the Next button, the user is taken to FormLogin instead of FormUpdates.

(d)

(c)

FIGURE 13.19 (Continued) Design View of FormU3 in the Healthcare Information System application and the code for the three buttons for updating the InsurancePlan table in the Healthcare Access database. (c) Code for the Update Premiums button. (d) Code for the Delete the Plan button.

275Database Connection in Microsoft Visual Studio

(b)

(a)

(c)

FIGURE 13.20 (a) The Login table added to the healthcare Access database and a data record stored in the Login table. (b) The Windows form, FormLogin, added to the Healthcare Information System application. (c) The Code view of FormLogin.

276 Developing Windows-Based and Web-Enabled Information Systems

13.9 Summary

This chapter describes how to add an Access database as a data source in a Windows- based application using Visual Studio 2013. The architecture of the database connection is explained. With the established database connection, we show how to display data of a table, the query result of a simple Select query, a Crosstab query, and a query with parameter(s) in a Windows form. We also show how to update a database through a Windows-based application and how to write Visual Basic code to retrieve, add, and delete data of an Access database.

EXERCISES

1. Complete the Windows-based application for the Healthcare Information System by adding all the Windows forms at the bottom level in the hierarchy of Windows forms for the Healthcare Information System application shown in Figure 11.13. These Windows forms display data from all the tables and queries in the health- care Access database. Add code to the Click event of the Back button in these Windows forms.

2. Complete the Windows-based application for the Movie Theater application that is started in Exercise 1 in Chapter 11 by adding all the Windows forms to display data of the tables and the query results of the queries. Add code to the Click event of the Back button in these Windows forms.

(d)

FIGURE 13.20 (Continued) (d) The Code view of FormMain.

277

14 Working with VBA in Excel

VBA, which stands for Visual Basic for Applications, is a programming language devel- oped by Microsoft Corporation. VBA is embedded in several Microsoft Office products, including Word, Excel, PowerPoint, and Visio to augment their capabilities. Other prod- ucts, such as the Arena system simulation software, also provide an interface to handle VBA to further enhance their functionalities. While VBA is implemented in many applica- tions, this chapter will primarily focus on Microsoft Excel 2013 to illustrate VBA funda- mentals. After familiarizing with VBA’s object models in Excel, this chapter will cover how to work with VBA, which includes Visual Basic Editor (VBE), an editor environment for VBA development; Windows Forms and Controls in VBA; various VBA procedures; and some VBA programming essentials. Lastly, the use of VBA functions and the integration of Excel’s built-in worksheet functions into VBA are discussed.

14.1 Introduction to VBA

Previous chapters of the book (Chapters 11 and 12) introduced both the VB.NET and the VB programming languages. Although VBA and VB share a lot in common for syntax, VB is used to develop stand-alone applications, whereas VBA is more suited to be embedded in an existing application (e.g., Excel) and interfaces with the application. Taking Excel as an example, one notable advantage of VBA is that it can automate some highly repetitive tasks in Excel. It can also enhance the application by executing custom-built functions that are not handled in Excel. This benefit, however, does not come with ease. This chapter will cover how to write programs in VBA to fully leverage its benefits. One notable disad- vantage of VBA though is that it is a moving target. As Microsoft continuously upgrades Office products, any VBA program may need modifications to be compatible with a newer version of the application.

Before we delve into VBA details, let us first create a simple macro in Excel.

Example 14.1

For the Excel spreadsheet shown in the following, we want to create a macro to copy the drug information for “Tylenol” into a separate spreadsheet within the same workbook. We can accomplish this task without a macro; however, it is tedious and the procedure is composed of repetitive tasks. In this example, we will demonstrate the use of a Macro to automate the task.

Recording a Macro Step 1: Open the workbook entitled Chapter14_Macro.xlsx, as shown in

Figure 14.1.

278 Developing Windows-Based and Web-Enabled Information Systems

Step 2: Click View → Macro → Record Macro. The following screen is then shown. Please specify the name for the macro (e.g., GetTylenolInfo), pro- vide the required shortcut key (e.g., Ctrl + Shift + T), add a description if needed, and click OK (Figure 14.2).

Step 3: After clicking OK, Excel now records any steps made by the user. Now, it is ready to record the manual steps for the macro. Within the current worksheet (“Sheet1”), first select the range of cells named “Name.” Then, click Data → Filter. Next, click on the Filter symbol at the cell “Name,” uncheck “(Select All),” check “Tylenol,” then click OK. The screenshot of the filter is shown in Figure 14.3.

FIGURE 14.1 Example Excel Spreadsheet.

FIGURE 14.2 Start recording a macro.

279Working with VBA in Excel

Step 4: After clicking OK, the spreadsheet now contains only the information related to “Tylenol.” Copy the rows containing the “Tylenol” information using the Ctrl + C command.

Step 5: Add a new spreadsheet using the Insert New Spreadsheet option, and paste the information as seen in Figure 14.4 using the Ctrl + V command starting from cell A1.

Step 6: Click View → Macro → Stop Recording. Since this is a macro-enabled Excel, please save the Excel workbook with extension .xlsm (Excel macro- enabled workbook).

FIGURE 14.3 Use Filter in Excel.

FIGURE 14.4 Paste filtered information on a new spreadsheet.

280 Developing Windows-Based and Web-Enabled Information Systems

Testing the Macro To test the macro we just created, we execute the procedure as follows:

Step 1: Click the spreadsheet containing the original drug information. Data → Unclick Filter. This will get us back to the original spreadsheet listing all the drug information.

Step 2: Press Ctrl + Shift + T. This shortcut key is used to trigger the macro to automate the Filter–Select–Copy–Paste procedure. There are other ways to trigger the macro, which will be discussed in the following sections.

Examining the Macro To take a look at the macro code just created, use the following steps:

Step 1: Click View → Macro → View Macro. A macro window will then pop up, listing all the macros available in Excel in a drop-down list.

Step 2: Choose the macro to be examined and click Edit Macro. The following code is shown in a new window (this window is called “Code window,” and we will discuss this in Section 14.3.1).

Sub GetTylenolInfo() 'GetTylenolInfo Macro 'This Macro is used to obtain the drug information for Tylenol only. ' 'Keyboard Shortcut: Ctrl + Shift + T ' Columns("B:B").Select Selection.AutoFilter ActiveSheet.Range("$B$1:$B$16").AutoFilter Field:=1,

Criteria1:="Tylenol" Range("A1:D7").Select Selection.Copy Sheets.Add After:=Sheets(Sheets.Count) ActiveSheet.Paste Columns("D:D").EntireColumn.AutoFit End Sub

The GetTylenolInfo( ) macro (also known as a “sub-procedure”) consists of several statements. Any statement that is preceded by an apostrophe (') is considered as a comment and is subsequently ignored by the VBA compiler. Excel executes the state- ments line by line, top to bottom. Recall in Chapter 12 that we have learned the VB programming essentials. Hence, we should be familiar with the objects, methods, and dot operators in the aforementioned code. However, the macro shown above incorpo- rates new objects from Excel, such as Columns, Selections, Ranges, Sheets, etc. These will be discussed in Section 14.2. Here, we will briefly review what each statement does. First, the columns (“B:B”) representing the range for “Name” is selected. The AutoFilter method for the Selection object, in this case, the selected columns, is trig- gered. The AutoFilter method sets the selection criteria as “Tylenol,” that is, the work- sheet will be filtered showing only “Tylenol” information. After filtering, the range (“A:D”) is selected and copied. A new worksheet is added, and the selected data are pasted on the new sheet.

This example shows that a Macro recorder is a powerful tool to generate a code for a VBA project. In digesting the program, it is observed that the VBA code includes many objects. Generally, a VBA project consists of Microsoft Excel Object, Forms, Modules, and sometimes, a Class Module. This chapter will cover Excel Object Model, Forms, and Modules. Advanced readers may refer to VB programming resources for creating and using Class Modules.

281Working with VBA in Excel

14.2 Excel Object Models

All Excel-based VBA projects start with an Excel Application as an object and are com- posed of other objects. These objects, in turn, contain more sub-objects. Simply put, the Excel Object Model involves an object hierarchy that can be presented as a tree structure, with the root of the tree being the Excel Application object. A basic object hierarchy in an Excel Application is shown in Figure 14.5.

Figure 14.5 comes across an important concept—collections. By definition, a collection is a group of objects of the same type and is, itself, an object. In an Excel Application object, some commonly used collections are Workbooks, which is a collection of all cur- rently opened Workbook objects; Worksheets, which is a collection of all Worksheet objects contained in a particular Workbook object and usually contains items such as range and formulas; Sheets, which is a collection of all sheets contained in a particu- lar Workbook object, consists not only of worksheets, but also of other types of sheets, such as Chart sheet; and Charts, which is a collection of all Chart objects contained in a particular Workbook object, which takes up an entire worksheet but not charts that are inserted as part of a worksheet.

Following the tree structure, the Excel Object Model hierarchy can be explained as follows: the Excel Application object contains multiple Workbooks collection objects, and within each Workbook, there may be more than one Workbook object. Within each Workbook object there are Worksheets collection objects, Sheets collection objects, and Charts col- lection objects. Worksheets may contain multiple Worksheet objects, whereas Charts may contain multiple Chart objects. As for Sheets collection, it may contain Worksheet and/or Chart objects. Each Worksheet object contains multiple Range objects.

The developers can work with an entire collection of objects. However, more often, the developer works with a specific object within the collection. In this case, referring to an object is important. The following examples illustrate how to reference an object in an Excel Application.

Application

Workbooks

Workbook

Worksheets

Worksheet

Range

Sheets

Charts

Chart

FIGURE 14.5 Basic Excel Object Model hierarchy.

282 Developing Windows-Based and Web-Enabled Information Systems

Example 14.2

Get to the Workbook object named “Chapter14.xlsm” and use the Application object:

Application.Workbooks("Chapter14.xlsm")

Example 14.3

Get the Worksheet object named “Sheet1” under the workbook named “Chapter14. xlsm”:

Application.Workbooks("Chapter14.xlsm").Worksheets("Sheet1")

or

Application.Workbooks("Chapter14.xlsm").Worksheets(1)

if Sheet1 is the first worksheet in the collection.

Example 14.4

Get to the value of cell “A1” on the Worksheet object named “Sheet1” from the work- book named “Chapter14.xlsm”:

Application.Workbooks("Chapter14.xlsm").Worksheets("Sheet1"). Range("A1").value

Clearly, dot operators are extensively used here. Take a look at Example 14.4, the Application code is referring to an application object. The Workbooks code, on the other hand, is a Workbooks collection object belonging to the Application object. “Chapter14. xlsm” is used to identify the specific Workbook object within the Workbooks collection. For the workbook named “Chapter14.xlsm,” the Worksheets object is a collection object belonging to it, and “Sheet1” is used to identify the specific Worksheet object within the Worksheets collection. This Worksheet object contains a Range object. The value property of the Range object (“A1”) is obtained.

Indeed, the above examples are pieces of VBA codes, and we have thus been exposed to VBA programming for Excel. In the remaining of the chapter, we will learn more about VBA basics.

14.3 VBA for Excel

14.3.1 Visual Basic Editor

Visual Basic Editor (VBE) is a separate application where the developer can write and edit a VBA code. The quickest way to activate VBE is to press Alt + F11 when Excel is active. To return to Excel, press Alt + F11 again. Figure 14.6 shows the VBE window, with some key windows identified.

Alternatively, the Developer tab can be used to navigate to VBE. However, the Developer tab is hidden by default. To include it first in the ribbon, use the following steps: File → Options → Customize Ribbon → select the Developer tab.

283Working with VBA in Excel

14.3.1.1 Project Explorer Window

The Project Explorer window displays a tree diagram that consists of every workbook cur- rently open in Excel (including add-ins and hidden workbooks). For each workbook, the Microsoft Excel objects are displayed. These include Sheets, Modules, Class Modules, and UserForms.

14.3.1.2 Code Window

A Code window (sometimes known as a “Module window”) displays the VBA code of a particular project. Every object in a project has an associated Code window. The macro recorded in Example 14.1 is stored in a module, and the VBA code can be directly modified in the code window.

14.3.1.3 Property Window

The Property window has the same functions as that in VB.NET. It represents the associ- ated properties of the selected object. Each control, such as Form, Label, Command Button, etc., has its own unique set of properties. When we click over different objects, a different list appears in the property window in response to the selected control.

14.3.1.4 Immediate Window

The Immediate window is most useful for executing VBA statements directly and for debugging the code.

Project explorer Code window

Property window Immediate window

Object browser

FIGURE 14.6 VBE window.

284 Developing Windows-Based and Web-Enabled Information Systems

14.3.1.5 Object Browser Window

The VBE includes another tool, known as the Object Browser. As the name implies, this tool allows the developers to browse through the objects available. To access the Object Browser, press F2 when VBE is active. A window as in Figure 14.7 will show.

The drop-down list at the top left contains a list of all currently available object librar- ies. To browse through VBA objects, select the VBA Project option from the drop-down list.

For all the windows discussed above, if they are not shown when VBE is launched, please choose View and activate the windows as needed.

14.3.2 Graphical User Interface: Forms and Controls

Recall that the main objects used to help a user interact with the computer are called “Windows controls.” Chapter 11 covers various Windows forms and controls. The Windows forms and controls of VBA are similar to those in VB.NET, with some variations. For example, there are two ways to access controls with Excel.

14.3.2.1 Controls in Excel

In Excel, we can add or position controls in a document. In this case, the graphical user interface is built upon the controls on the Excel spreadsheet. This VBA project may not contain any Forms object. To use various controls with Excel, first click the Developer tab on the Ribbon, then click Insert in the Control section (see Figure 14.8).

This would display the list of controls available in Microsoft Excel. The controls appear in two sections: Form Controls and ActiveX Controls. If we are working on a spreadsheet in Excel, it is recommended to use the controls in the ActiveX Controls section. In other cases, Form controls may be used. Every control in the Form Controls or the ActiveX Controls section has a specific name. The ones shown in Figure 14.8 are the most widely used and are summarized in Table 14.1.

Other than the ActiveX controls listed above, one can always click the More Controls button to add more controls available in the computer system.

FIGURE 14.7 Object Browser.

285Working with VBA in Excel

14.3.2.2 Controls with Forms

An alternative way to work with a control is with Windows forms, in which case the VBA project will then have the Forms object. If the VBE is activated, we can add or position controls on the UserForm by right clicking Microsoft Excel Objects → Insert → UserForm.

FIGURE 14.8 Adding controls to Excel documents.

TABLE 14.1

Controls with Excel or VBA

Form controls Name ActiveX controls Name

Command button Command button

Combo box Combo box

Check box Check box

Spin button Spin button

List box List box

Option button Option button

Scroll bar Scroll bar

Frame Text box

Aa Label Label

Image

Toggle button

More Controls button

286 Developing Windows-Based and Web-Enabled Information Systems

A UserForm window, as shown in Figure 14.9, along with a toolbox equipped with vari- ous controls will appear. The controls in the toolbox are the same as those in the form control section discussed previously. Since this approach is very similar to what is covered in VB.NET programming, please use Chapter 11 as a guideline to work with Windows Form and Controls. These include: (1) set the properties of the controls using the Property window, (2) change the properties of the controls in running time, and (3) use the control’s event procedure. Please remember that the VBA code for the forms and controls will reside in the Forms object.

14.3.3 Modules: Sub and Function Procedure

In a VBA project, the Module object is a very important object. Modules are mainly used to store procedures that can be accessed either from within that module or from anywhere in the project. The VBA code usually resides in the procedure and there are two most com- mon types of procedures: Subs and Functions.

A Sub procedure is a group of VBA statements that performs an action within Excel. Most of the macros in VBA are Sub procedures. A Sub procedure can be considered as a command, that is, execute the Sub procedure and something happens. Every Sub proce- dure starts with the keyword Sub and ends with an End Sub statement.

Example 14.5

Here is an example of a Sub procedure:

Sub ShowMessage() MsgBox("Hi there") End Sub

FIGURE 14.9 Adding controls to Windows forms.

287Working with VBA in Excel

This example shows a procedure named ShowMessage. A set of parentheses follows the procedure’s name. In most cases, these parentheses are empty. However, we may pass arguments to Sub procedures from other procedures. In this case, the arguments will be listed between the parentheses.

A Function procedure is a group of VBA statements that takes in one or more argu- ments, performs a calculation, and returns a single value. Excel includes many worksheet functions, for example, SUM and VLOOKUP. These pre-defined functions can be used in formulas. The same goes for the Function procedures written by the developers using VBA. Every Function procedure starts with the keyword Function and ends with an End Function statement.

Example 14.6

Here is an example of a Function procedure:

Function Perimeter(radius) Perimeter = 2 * radius * WorksheetFunction.Pi() End Function

This function, named “Perimeter,” takes one argument named “radius,” which is enclosed in parentheses. Functions can have any number of arguments or none at all. When the function is executed, it returns a single value, Perimeter, which is the perimeter of the circle for a given radius. In this example, a worksheet function Pi is called, and it returns the value of Pi (3.14) to calculate the perimeter.

There are two ways to execute a Function procedure: execute it from another procedure (a Sub or another Function procedure) or use it in a worksheet formula. Unlike the Sub procedure, the Excel Macro recorder cannot be used to record a Function procedure. The function procedure has to be coded by the developers manually.

14.3.4 Modules: Variables

Other than a Sub procedure and a Function procedure, modules are also used to store vari- ables, constants, and declarations to be accessed from anywhere within the project. VBA’s main purpose is to manipulate data. Some data, such as worksheet ranges, reside in objects. Other data are stored in variables. A variable is simply a named storage location in the com- puter’s memory. We assign a value to a variable using the equals sign operator (“=”). A vari- able’s value may change over the course of the procedure. Sometimes, we may need to refer to a value or string that never changes, we use constant. The variables and constants in a VBA project are used in a similar manner. Hence, we will only focus on the discussion of variables in this chapter.

Naming variables in VBA is rather flexible, although there are a few rules: (1) the name can use letters, numbers, and some punctuation characters; however, the first character must be a letter; (2) no space or period is allowed in a variable’s name; (3) no special charac- ters such as #, $, %, &, and ! are allowed in the name; (4) VBA-reserved words, such as Sqr, Sub, With, For, etc., are not allowed to be used as variable names.

VBA has a variety of built-in data types (e.g., Integer, Date) that are the same as those in VB.NET. When a variable is declared without explicitly being assigned to a data type, the default data type “variant” is used. Data stored as a variant will change its type depending on the code, and VBA will automatically handle the conversion; however,

288 Developing Windows-Based and Web-Enabled Information Systems

remember that this comes with the cost of speed and memory. To force the developers to declare all the variables, the following statement can be added at the beginning of the VBA Module:

Option Explicit

Keep in mind that the Option Explicit statement only applies to the module in which it resides. If there is more than one VBA module in a project, such statement needs to be added for each module. To ensure that the Option Explicit statement is inserted automati- cally whenever a new VBA module is created, the Require Variable Definition option can be turned on by choosing Tools → Options.

Recall that a workbook can have any number of VBA modules, and a VBA module can have any number of Sub and Function procedures. A variable’s scope determines which modules and procedures can use the variables, and this information is summarized in Table 14.2.

14.3.4.1 Procedure-Only Variables

Variables declared with this scope can be used only in the procedure in which they are declared. When the procedure ends, the variable no longer exists, and Excel frees up its memory. If we execute the procedure again, the variable is re-initiated, but its previ- ous value is lost. The most common way to declare a procedure-only variable is with a Dim statement placed between a Sub statement and an End Sub statement (or between a Function and an End Function statement). The Dim keyword is short for “dimension,” which simply means that we are setting aside memory for a particular variable. We usu- ally place Dim statements immediately after the Sub or Function statement and before the procedure’s code. The following example shows some procedure-only variables declared by using Dim statements:

Sub MySub( ) Dim x As Integer, y As Integer Dim DOB As Date Dim Name As String Dim Price As Long Dim Description ' ... [The procedure's code goes here.] ... End Sub

Notice that the last Dim statement in the example does not declare a data type; it declares only the variable itself. The effect is that the variable Description is a “variant.” Also, please

TABLE 14.2

Variable’s Scope

Scope How the Variable Is Declared

Procedure only By using a Dim or a Static statement in the procedure that uses the variable Module only By using a Dim statement before the first Sub or Function statement in the module All procedures in all modules

By using a Public statement before the first Sub or Function statement in a module

289Working with VBA in Excel

note that VBA does not allow us to declare a group of variables to be a particular data type by separating the variables with commas. For example, although valid, the following state- ment does not declare all the variables as integer:

Dim x, y As Integer

Instead, only y is declared to be an integer; x is declared as variant. If a variable is declared with procedure-only scope, other procedures in the same mod-

ule can use the same variable name, but each instance of the variable is unique to its own procedure. In general, variables declared at the procedure level are the most efficient vari- ables because VBA frees up the memory that they use when the procedure ends.

14.3.4.2 Module-Only Variables

To make a variable available to all procedures in a module, the variable needs to be declared before the module’s first Sub or Function statement. This is done in the Declaration section, where the Option Explicit statement is located. With this declaration in place, the variable has a module-only scope and, thus, can be used in any procedures within the module, and the variable retains its value from one procedure to another.

14.3.4.3 Public Variables

To make a variable available not only within the module, but also to all procedures in the VBA project, the variable needs to be declared at the same location as the module-level variable, but with the Public keyword. Here is an example:

Public Price As Long

The Public keyword makes the Price variable available to any procedure in the work- book. Often, for a large VBA project, a better way to manage public scope variables is to pull all public declarations in a separate module.

14.3.4.4 Static Variables

Often, when a procedure ends, all of the variables are reset. Static variables are a special case because they retain their values even when the procedure ends (the project is still executing). A static variable is declared at the procedure level using the Static keyword. A static variable may be useful, for example, if we want to track the number of times that we execute a pro- cedure, a static variable can be used here, and it increases by one each time the procedure is executed. Since static variables are procedure level, their values are not available to any other procedures in the same module and in other modules in the workbook.

14.4 Programming with VBA and Excel

14.4.1 Excel: Range Object

In a typical VBA Excel project, much of the programming focuses on the Range object.

290 Developing Windows-Based and Web-Enabled Information Systems

14.4.1.1 Range

A Range object represents a range contained in a Worksheet object. It can be as small as a single cell or all cells in a worksheet (A1:IV 65536, or 16,777,216 cells). We can refer to a Range object with a code as follows:

Range("A:D")—Columns A to D

or if the range has a variable name (created by selecting Insert → NameBox, and entering the name of the range), we use

Range("Drug_List")

Unless Excel is informed otherwise, it assumes that the Range object is from the active worksheet. If anything other than a worksheet is active (such as a chart sheet), the range reference fails, and an error message will show. Certainly, a range outside the active sheet can be referenced by qualifying the Range object with a worksheet name and a workbook name. For example:

Worksheets("Sheet1").Range("A1:C5")

This statement refers to the range on “Sheet1” for the active workbook. Alternatively,

Workbooks("Healthcaresystem.xlsm").Worksheets("Sheet1").Range("A1:C5")

can be used to refer to a range from a different workbook.

14.4.1.2 Cells

Rather than using the VBA Range keyword, a range can be referenced via the Cells prop- erty. It may be confusing, but note that Cells is a property that VBA evaluates instead of an object. To use the Cells property appropriately, two arguments need to be provided: rows and columns. For example, to refer to cell C2 on “Sheet1,” the expression is as follows:

Worksheets("Sheet1").Cells(2, 3)

The Cells property can also be used to refer to a multi-cell range such as:

Range(Cells(1, 1), Cells(10, 10))

This expression refers to a 100-cell range that extends from cell A1 (row 1, column 1) to cell J10 (row 10, column 10). It is equivalent to

Range("A1:J10")

The advantage of using the Cells property to refer to ranges is that we can use actual numbers as the cell’s argument, which may make looping an easier task.

14.4.1.3 Offset

The Offset property provides another handy means for referring to ranges. When we record a macro to handle multiple cells, often, the Offset property is used. This property,

291Working with VBA in Excel

which operates on a Range object and returns another Range object, lets us refer to a cell, which is a particular number of rows and columns away from the current Range cell. Like the Cells property, the Offset property takes two arguments. The first argument rep- resents the number of rows to offset, while the second argument represents the number of columns to offset. The following expression refers to a cell one row below cell A1 and two columns to the right of cell A1. In other words, this refers to cell C2. This example is illustrated as follows:

Range("A1").Offset(1, 2)

The Offset property can also use negative arguments. A negative row offset refers to a row above the range. A negative column offset refers to a column to the left of the range.

14.4.1.4 Some Useful Properties and Methods for the Range Object

The Range object has dozens of operable properties. Some Range properties are read-only properties, which means that we cannot change their current values. Some Range proper- ties are both readable and writable. Table 14.3 summarizes some of the more commonly used Range properties. For complete details, consult the Help system in VBE.

In addition, some commonly used Range object methods are listed in Table 14.4.

14.4.2 VBA Functions and Excel Functions

To develop a powerful application, we need to integrate VBA with Excel seamlessly. By seamlessly, we mean to utilize the data stored in the Excel worksheet and the variables declared and used in the module. Seamless integration also means to take advantage of

TABLE 14.3

Some Useful Range Object Properties

Property Description Example

Value Represents the value contained in a cell. It is a read–write property.

Worksheets(“Sheet1”). Range(“A1”).Value

Text Returns a string representing the text as displayed in a cell. It is a read-only property.

Worksheets(“Sheet1”). Range(“A1”).Text

Count Returns the number of cells in a Range object. It is a read-only property.

Range(“A1:C3”).Count

Column Returns the column number of a single cell range. It is a ready-only property.

Sheets(“Sheet1”).Range(“A1”). Column

Row Returns the row number of a single cell range. It is a read-only property.

Sheets(“Sheet1”).Range(“A1”).Row

Address Displays the cell address for a Range object in an absolute notation (e.g., $A$1). It is a ready-only property.

Range(Cells(1, 1), Cells(5, 5)). Address

HasFormula Returns True if all cells in the Range object contains a formula, False if all cells in the Range object contains no formula, and Null if mixture.

Range(“A1:A2”).HasFormula

Formula Represents the formula in a cell. It is a read–write property.

Range(“A1”).Formula = “=SUM(A2:A12)”

NumberFormat Represents the number format of the Range object. It is a read–write property.

Columns(“A:A”). NumberFormat=”0.00%”

292 Developing Windows-Based and Web-Enabled Information Systems

existing Excel built-in functions, VBA built-in functions, and the functions that we write in VBA.

14.4.2.1 VBA Functions

VBA provides numerous built-in functions. Some of these functions take input arguments, while some do not. For example, the Date(') function returns the current date and, thus, requires no input argument, whereas the Len(myString) function returns the length of the string and requires a string argument (e.g., myString). The best source to learn the proper- ties of a particular VBA function is the Excel Visual Basic Help system. To query the details of a particular function, type the function name into a VBA Module, move the cursor any- where in the text, and press F1.

What deserves more discussion here is that, among the VBA built-in functions, a few go beyond simply performing the tasks and returning a single value. If used appropriately, these functions can greatly enhance the functionalities of the application. A few examples of these built-in functions are discussed as follows:

• MsgBox: The MsgBox command displays a dialog box containing a message and a button. For example,

MsgBox("Hi, there")

This will return the following window:

TABLE 14.4

Some Useful Range Object Methods

Method Description Example

Select Selects a range of cells. Note that we need to ensure that the worksheet is activated.

Sheets(“Sheet1”).Activate Range(“A1:C12”).Select

Copy This method is applicable to the Range object. Sub CopyRange() Range(“A1:A12”).Select Selection.Copy Range(“C1”).Select ActiveSheet.Paste End Sub

Paste This method is applicable to the Worksheet object.

Clear Deletes the contents of a range and all the cell formatting. Columns(“D:D”).Clear Delete Deletes a range; Excel will shift the remaining cells around

to fill up the range that we deleted. Rows(“1:1”).Delete

293Working with VBA in Excel

• InputBox: The InputBox command displays a simple dialog box that asks the user for some input. The function returns whatever the user enters into the dialog box. For example,

Dim UserName As String

UserName = InputBox("Please enter your UserName.")

When we execute the VBA, the following window will pop up. Let us type “Joe Smith” in the input box, and the variable UserName will be assigned with the value being “Joe Smith.”

• Shell: The Shell command executes another program. The function returns the task ID (a unique identifier) of the other program (or an error if the function can- not start the other program). The following is an example:

Dim ApplicationID As String ApplicationID = Shell("C:\Program Files (×86)\Internet Explorer\

iexplore.exe") MsgBox(ApplicationID)

When we execute the program, the Shell function will launch Internet Explorer with the path stated as “C:\Program Files (×86)\Internet Explorer\iexplore.exe,” and the ApplicationID will be returned.

14.4.2.2 Excel Functions

Although VBA offers a collection of built-in functions, we take note of the important concept that the various Excel’s worksheet functions can be incorporated into the particular VBA project. VBA makes Excel’s worksheet functions available through the WorksheetFunction object, which is contained in the Application object. Therefore, any statement that uses a worksheet function must use the Application and/or the WorksheetFunction qualifier.

For example, the following three expressions will work exactly as running Excel’s native func- tions. To calculate the sum of Range(“A2:A12”) and write it back to Cells(1, 1) (the range of A1):

Cells(1, 1) = Application.WorksheetFunction.Sum(Range("A2:A12")) Cells(1, 1) = WorksheetFunction.Sum(Range("A2:A12")) Cells(1, 1) = Application.Sum(Range("A2:A12"))

The first statement states the Application.WorksheetFunction, the second statement uses the WorksheetFunction qualifier, and the third statement omits the Worksheet but uses the Application qualifier.

294 Developing Windows-Based and Web-Enabled Information Systems

The WorksheetFunction object contains the worksheet functions available to the VBA procedures. To see the list of these functions, please use the following steps to display a complete list of worksheet functions available in VBA:

Step 1: In VBE, press F2. The Object Browser appears. Step 2: In the Project/Library drop-down list, select Excel. Step 3: In the list labeled “Classes,” select WorksheetFunction.

The members of the list show all the worksheet functions that can be used in a VBA project (Figure 14.10).

In a VBA project, we may get confused with VBA’s built-in functions and Excel’s func- tions. A good practical rule is to first check whether VBA has the functions to meet the need. If not, we then check out Excel’s functions. If all fail, we may have to write our own function by using VBA.

14.4.3 Some Important Events in Excel

Similar to VB programming, where the event-handling procedure is an important proce- dure for Windows forms and controls, there are also several important events in Excel that are very helpful for running a VBA project. These events reside in the Code window of an Excel Object Module (not the standard VBA Module).

A list of some useful workbook-related events is provided in Table 14.5, and a list of some worksheet-related events is provided in Table 14.6.

One example of using a workbook-related event is as follows: When the workbook is opened, some initializations (e.g., initialize variables, set the con-

stant) can be done under this event procedure:

FIGURE 14.10 Comprehensive list of WorksheetFunction.

295Working with VBA in Excel

Private Sub Workbook_Open() ' Our code in initialization goes here. End Sub

A good example of using a worksheet-related event is as follows: Let us assume that the user is asked to input data to the Excel spreadsheet in Cells(1, 1)

and that the data entered must be greater than 100. A piece of VBA codes can be written under the event of Worksheet_Change( ).

Private Sub Worksheet_SelectionChange() If Cells(1, 1) < 100 Then MsgBox("Sorry. The number is less than 100!") End If End Sub

By now, we have touched upon some basics of VBA. Next, we will work on a mini project using VBA.

14.5 Mini Project

14.5.1 Introduction

An industrial engineering student would want to query a subset of the top Institute of Science Index (ISI)-indexed industrial engineering (IE) journals. Consider the data from the “Chapter14MiniProjectRawData.xlsx” Excel data sheet. The data consist of the name of the journal; its publisher; its corresponding IE field specialization; and its 2012 ISI impact

TABLE 14.6

Partial List of Worksheet-Related Events

Worksheet Event Description

Activate The worksheet is activated. BeforeDoubleClick A cell in the worksheet is double-clicked. BeforeRightClick A cell in the worksheet is right-clicked. Calculate The worksheet is recalculated. Change A change is made to a cell in the worksheet. Deactivate The worksheet is deactivated.

TABLE 14.5

Partial List of Workbook-Related Events

Workbook Event Description

Activate The workbook is activated. Open The workbook is opened. WindowActivate The workbook window is activated. BeforeClose The workbook is closed. BeforeSave The workbook is saved. Deactivate The workbook is deactivated.

296 Developing Windows-Based and Web-Enabled Information Systems

factor, which is used to measure a journal’s relative importance. A simple Excel VBA- enabled spreadsheet could be used to handle this scenario.

14.5.2 Spreadsheet Specifications

The specifications of the VBA enabled spreadsheet are as follows:

• Upon opening the spreadsheet, an overview page as in Figure 14.11 is shown to the user.

• A “command button” control with the caption “Execute Search Query” would act as a starting point to query the journals. Upon clicking the button, the following UserForm, as presented in Figure 14.12, is shown to the user. • The publisher option acts like a filter to return only journals that matched

the selection of the user. Additionally, the publisher option is a “combo box”

FIGURE 14.12 Sample Search Parameters form.

FIGURE 14.11 Sample Welcome screen of the VBA spreadsheet.

297Working with VBA in Excel

control where the drop-down list is the list of unique publishers queried from the data.

• The filed option is similar to the publisher option, which acts like a filter to return only journals that matched the selection of the user. Furthermore, the field option is a “combo box” control where the drop-down list is the list of unique fields queried from the data.

• The minimum impact factor is a text box that accepts only numeric values. This serves as a filter that queries journals that meet the minimum require- ment impact factor provided by the user.

• The Sort Options panel is an “option button” control that determines how the data are to be sorted. By default, the data are sorted by descending impact factor. Additionally, the user may choose to sort by publisher or by field alphabetically.

14.6 Summary

This chapter gives a nutshell on developing a VBA project in Excel. First, a Macro recorder tool is demonstrated on how to generate a VBA code. By digesting the VBA code, the use of object and its related properties and methods is noted. This chapter then briefly overviews some objects in an Excel VBA project. Since we have extensively studied Windows forms objects and control objects in the earlier chapters, this chapter mainly focuses on Excel objects. Specifically, the Range object is discussed in detail. The module concepts, includ- ing the sub procedure and the function procedure, are then explained. Given the unique- ness of VBA in Excel, some important VBA functions, Excel functions, and Excel-related event procedures are reviewed, which, together, could help develop a strong application.

EXERCISES

1. The Record Macro feature of Microsoft Excel can be accessed using the Insert → Macro → Record Macro navigation option. True or false?

2. A shortcut key is a required step in recording a macro and must be unique across macros. True or false?

3. Which of the following Excel file types allows the execution permission of macros by default?

a. *.xls b. *.xlsx c. *.xlxm d. xlsm.* e. None of the above 4. Which of the following ways are valid procedures in inspecting a recorded macro? a. View → Macros → View Macro → Edit Macro b. Developer → Macros → View Macro → Edit Macro c. Insert → Macros → View Macro → Edit Macro

298 Developing Windows-Based and Web-Enabled Information Systems

d. All of the above are valid ways to inspect a macro. e. Only a and b are valid ways to inspect a macro. 5. A workbook is composed of the following objects except: a. Worksheets b. Workbooks c. Sheets d. Charts e. All of the above are valid objects within a workbook. 6. The following are examples of valid Excel collections except: a. Charts b. Sheets c. Worksheets d. Workbooks e. All of the above are valid collections. 7. Write a VBA code to refer to the value of cell “A10” of the worksheet named

“DataSheet” within the Excel file “Medicines.xlsm” using the appropriate VBA object hierarchy.

8. Write a VBA code to refer to the values of a 10 × 1 array starting from the “A1” cell. However, the name of the worksheet is unknown, but on the other hand, we know that it is the eighth sheet within the “Medicines.xlsm” Excel file.

9. The ___________ window displays a tree diagram for every workbook in Excel. a. Project Explorer b. Code c. Property d. Immediate e. None of the above 10. The Object Browser has the following properties: a. Shows all VBA classes, objects, and their corresponding members b. Can be accessed using the F3 key c. Can also be accessed using the View → Object Browser from the Excel menu d. All of the above are true. e. None of the above is true. 11. VBA controls are accessed through the Excel Developer tab. True or false? 12. User-defined forms can be added to an Excel spreadsheet using the following

except: a. VBE: Insert → UserForm b. Right-click on the VBA project within the Project Browser: Insert → UserForm c. Excel: Insert → UserForm d. All of the above are true. e. Only a and b are true.

299Working with VBA in Excel

13. The two common types of subroutines in VBA are Procedures and Functions. True or false?

14. A function can have multiple input arguments or could have none at all. True or false?

15. VBA variables have the following properties except: a. An equals sign operator is used to assign a value to a variable. b. The first character to name a variable should be a letter. c. Reserved words cannot be named as a variable. d. The default type of a variable is the variant type. e. A variable’s range determines which modules can see the variable. 16. A procedure variable exists only within the procedure even if the procedure ends.

True or false? 17. Using the code: “Dim string1,string2 As String,” “string2” is declared as a string

while “string1” is declared as a: a. String b. Variant c. Dim d. Integer e. Error, and will not compile 18. The primary advantage of the Cell property over the Range property is that it can

be used within a loop. True or false? 19. Using the Cell property, write a code to refer to the values of a 10 × 2 array starting

from the “A1” cell of the “Drugs” worksheet within the “Medicines.xlms” Excel file.

20. Using the Offset property, write a code to refer to the value of a cell two rows to the left and one cell below from cell “D2” of the “Drugs” worksheet within the “Medicines.xlms” Excel file.

21. From the selected values from Review Exercise 19, write the VBA code to select and copy the values.

22. From Review Exercise 21, paste the copied values into a new sheet called “DrugsCopy” starting from cell “A1.”

23. Built-in Excel functions are accessed using the Application.WorksheetFunction command. True or false?

24. The following are valid events in Excel except: a. Workbook_Activate() b. UserForm_Initialize() c. Button_Click() d. All of the above are valid events. e. None of the above is a valid event.

301

15 Database Connectivity with VBA

After learning the fundamentals of Visual Basic for Applications (VBA) in Chapter 14, this chapter will cover how to use ADO.NET in Excel VBA to connect to external data sources. These include databases such as a Microsoft Access database or a MySQL database. In addition, connecting to Web sources such as uniform resource locator (URL) and exten- sible markup language (XML) documents is also briefly discussed.

15.1 ActiveX Data Object

ActiveX Data Object (ADO) is an object model that enables the developers to access data stored in a variety of data sources. Commonly, the data source could be a database, but it could also be a text file, an Excel spreadsheet, or an XML file. Since different data sources use different protocols, there is no single set of classes that allows the developers to accom- plish this task universally. ADO.NET is a set of object-oriented libraries that support the communication of Excel VBA to the right data source using the right protocols. In ADO. NET, the libraries are called “data providers” and are usually named after the protocol or data source type that they allow the developers to interact with. Table 15.1 lists the data providers that are included in the ADO.NET framework.

15.2 ADO.NET Objects in Data Provider Libraries

ADO.NET includes many objects that can be used to work with data. The five core objects are Connection, Command, DataReader, Dataset, and DataAdapter. In VBA, the Connection, Command, and Recordset (instead of Dataset) objects are commonly used. We will briefly discuss the objects and illustrate their uses through some examples.

15.2.1 Connection Object

The Connection object is used to create an open connection to a database via which the developers can access and manipulate the records within the database. The connection object needs to specify the database server, database name, username, password, and other parameters that are required for initiating the connection.

302 Developing Windows-Based and Web-Enabled Information Systems

Example 15.1

Create a connection object to connect to a MySQL database named “healthcaresystem.” The syntax is as follows:

cntMyConnection.ConnectionString="DRIVER={MySQL ODBC 5.2 Driver}; Server=localhost; Database=healthcaresystem;Uid=root; pwd=12345678; OPTION=3;"

Since we want to connect to MySQL database, parameters such as the database driver, server name, database name, user ID, and password need to be specified and assigned to the ConnectionString property of the object named “cntMyConnection.”

Example 15.2

Create a connection object to connect to the Access database named “healthcaresys- tem.” The syntax is as follows:

cntMyConnection.ConnectionString="Provider=Microsoft.Jet.OLEDB.4.0; Data Source=c:/healthcaresystem.accdb"

When connecting to an Access database, only the database provider (Microsoft Jet Engine) and the data source (note that there is a space between the two) that speci- fies the location of the database are needed to define the ConnectionString property of cntMyConnection.

The connection object is then used by command objects to ensure that the commands are applied to the right database.

15.2.2 Command Object

The command object is used to execute a single query on a database. To execute a set of commands that work on a group of data, the DataAdapter object is used. The query can perform actions like creating, adding, retrieving, deleting, or updating records. The data queried will be returned as a Recordset object, that is, the retrieved data can be manipu- lated by the properties, collections, methods, and events of the Recordset object.

Example 15.3

Create a command object based on the cntMyConnection, that is, the connection object created in Example 15.1. The syntax is

mysqlcommand.CommandText="Select * from drugs" mysqlcommand.ActiveConnection=cntMyConnection mysqlcommand.Execute

TABLE 15.1

ADO.NET Data Providers

Provider Name Data Source Description

OLEDB data provider For data sources exposed to an OLEDB interface, that is, Access ODBC data provider For data sources exposed by using ODBC SQL data provider For interacting with the Microsoft SQL Server Oracle data provider For Oracle databases. The .NET Framework Data Provider for Oracle supports

Oracle client software version 8.1.7 and later.

303Database Connectivity with VBA

15.2.3 Recordset Object

The Recordset object is used to hold a group of records from a database. It consists of a set of records and columns. When Recordset is first opened, the current record pointer will point to the first record, and the Beginning of File (BOF) property is true and the End of File (EOF) property is false. If there are no records, the BOF and the EOF are true.

Example 15.4

An example of using the Recordset object to retrieve information from the Drugs table based on the connection created previously is as follows:

src="SELECT * FROM Drugs" rstFirstRecordset.Open Source:=src, ActiveConnection:=cntMyConnection

In this example, the rstFirstRecordset is the Recordset object. The query command is assigned to a variable src. The ActiveConnection parameter is set to cntMyConnection (Example 15.1).

With the basic understanding of ADO.NET objects used in VBA, some examples are used to demonstrate the applications in detail.

15.3 Illustrative Example 1: Connecting to MySQL

Let us use the healthcaresystem MySQL database created in Chapter 9 to demonstrate how to connect VBA to an external database using ADO.NET.

Before starting, we need to consider the following:

1. Be sure that the MySQL database is installed on the computer and runs appropriately.

2. Download the database driver for MySQL from http://www.mysql.com. As dis- cussed in Chapter 9, there is a wide variety of connectors available. In this exam- ple, we need to download the open database connectivity (ODBC) connector, which is the standardized database connection driver. Do make sure that the cor- rect driver is downloaded (e.g., Windows vs. Mac, 32 bit vs. 64 bit). After installing the driver, a better way to check the database driver is to go to Control Panel → Administrative Tools → Data Source (ODBC). As shown in Figure 15.1, the system has the MySQL ODBC 5.2a Driver installed for the MySQL database. This infor- mation will then be used in the VBA project.

3. Activate Excel Visual Basic Editor, choose Tools → References → Microsoft ADOs 2.8 Library (see Figure 15.2). There may be a number of ADO object libraries avail- able. We may choose any versions of the libraries that would provide the needed database connection capability.

304 Developing Windows-Based and Web-Enabled Information Systems

Now, we will add a module to the VBA project. In the code window, we type the following:

1. Sub querydatafromMySQL() 2. Dim cntMyConnection As ADODB.Connection 3. Set cntMyConnection=New ADODB.Connection 4. cntMyConnection.ConnectionString="DRIVER={MySQL ODBC 5.2a Driver};Server=localhost;Database=healthcaresystem;Uid= root; pwd=12345678; OPTION=3;"

FIGURE 15.1 ODBC Data Source Administrator.

FIGURE 15.2 Adding ADO Library Reference to a VBA project.

305Database Connectivity with VBA

This example retrieves data from the Drugs table in a MySQL database named health- caresystem. This Drugs table has four columns: Drugs_ID, Name, Price, and Description. In this example, we use both Connection and Recordset objects. Let us go through the codes in detail.

Lines 1, 29: This is a Sub Procedure with Sub–End Sub. Lines 2 to 3: Declare an object of Connection class named cntMyConnection. Next, a

new instance of the object is created. Line 4: Specifies the details of the database connection string, including the data-

base driver, in this example, MySQL ODBC 5.2a (this information is gathered from Data Source [ODBC]) is used; the name of the server, localhost, and connec- tion details including the username and the password are also included. We also include OPTION=3 in the MySQL connection string. This option setting allows us to direct the MySQL server to behave in a specific manner for the duration of each connection. The “3” setting corresponds to the field length, which implies that there is no need to optimize the column width, and for the row, simply return the matching rows. We can find more information about this setting at the MySQL Reference Manual.

5. cntMyConnection.Open 6. Dim rstFirstRecordset As ADODB.Recordset 7. Set rstFirstRecordset=New ADODB.Recordset 8. Dim src as String 9. src="SELECT * FROM Drugs" 10. rstFirstRecordset.Open Source:=src, ActiveConnection:= cntMyConnection 11. Worksheets("DatabaseConnection").Activate 12. Dim column as Integer, row as Integer 13. With ActiveSheet 14. For column=1 to rstFirstRecordset.Fields.Count 15. .cells(1, column)=rstFirstRecordset(column −1). name 16. Next column 17. row=2 18. While Not rstFirstRecordset.EOF 19. For column=1 to rstFirstRecordset.Fields.Count 20. .Cells(row, column)=rstFirstRecordset (column −1).Value 21. Next column 22. rstFirstRecordset.MoveNext 23. row=row + 1 24. Wend 25. End With 26. Set rstFirstRecordset=Nothing 27. cntMyConnection.Close 28. Set cntMyConnection=Nothing 29. End Sub

306 Developing Windows-Based and Web-Enabled Information Systems

If a remote database is to be connected, the syntax is as follows:

DRIVER={MySQL ODBC 5.2 Driver};Server=myserverAddress;Database=myDatabase; Uid=myUsername; pwd=myPassword; OPTION=3;"

Furthermore, if the transmission control protocol (TCP)/internet protocol (IP) port needs to be specified when connecting to a remote database, the syntax is as follows:

DRIVER={MySQL ODBC 5.2 Driver};Server=myserverAddress;Port=3306; Database=myDatabase; Uid=myUsername; pwd=myPassword; OPTION=3;"

Line 5: The Open method of the cntMyConnection object is called. This opens the connection to the database.

Lines 6 to 7: Similar to lines 2 to 3, a new object from the Recordset class named “rst- FirstRecordset” is first declared then instantiated.

Lines 8 to 9: Declare src as a string data type. An SQL statement is then assigned to src.

Line 10: The Open method of the rstFirstRecord object is triggered. There are two parameters: the SQL statement, which is src (line 9), and the active connection to the database, cntMyConnection, which is established in line 5.

Line 11: This is a statement for the Excel object. It makes the worksheet named “DatabaseConnection” active.

Line 12: Declare two integer variables: column representing the columns and row representing the rows. These two variables are used as increment in the following loop statements.

Lines 13, 25: For the active sheet, we use the With–End With Statement. The advan- tage of using this statement pair is that, when the Cells property is called, we do not need to spell out the name of the worksheet every time.

Lines 14 to 16: Use the For Loop here to get the field names from the “Drugs” table. The property rstFirstRecordset.Fields.Count is used to indicate the number of col- umns retrieved. The property rstFirstRecordset(column).name is used to indicate the name of the specific column. This information is written back to the Excel worksheet in the first row, from the first column to the rstFirstRecordset.Fields. Count column. In this example, we move from column 1 to column 4. Note that, in the Excel object, the index starts with 1, whereas in the ADO.NET object, the index starts with 0. As a result, to retrieve the name held by rstFirstRecordset, we will use (column −1) to list the name for columns 1 to 4.

Line 17: Assign the row to be 2, that is, from now on, the data retrieved from the Recordset will be populated to the spreadsheet from the second row onward.

Lines 18, 23, 24: Use the While–End Loop to retrieve the records one by one. The rstFirstRecordset.EOF property is used to check if all the records (rows) are retrieved from the Recordset. The row variable has an increment of one in each iteration.

Lines 19 to 21: Similar to lines 14 to 16, for the specific row, data are retrieved for each column and written onto the worksheet.

Line 22: The method MoveNext of the RecordSet object is called. This will move the cursor from the current row to the next row.

307Database Connectivity with VBA

Lines 26 to 28: These statements are used to release the system resources. As we are done with the database transaction, we will set the Recordset and the connection object to Nothing to release the computer resource.

In this example, we demonstrate the query from the database. To update the database, for example, performing INSERT, DELETE, and UPDATE operations, the statement in line 9 needs to be changed in response to the appropriate SQL commands. Be cautious when using the INSERT statement—the data types need to be matched.

For example, a new drug information is added to the worksheets (“new_drugs”) on Range (“A1:A4”). The command statement to add the Drug_ID information is defined as follows:

src="INSERT INTO Drugs (Drug_ID) Values (' & worksheets ("new_drugs"). Range("A1") & ')"

In this case, the Drug_ID is defined as a character data type. If Drug_ID is defined as a numeric data type, the corresponding statement is as follows:

src="INSERT INTO Drugs (Drug_ID) Values ( & worksheets("new_drugs"). Range("A1") & )"

The difference between the two statements is the use of the single quote pair ‘ ‘. In the first example, Drug_ID is a character data type, and it should be encoded with ‘ ‘ when being inserted into the table, whereas in the second example, the numeric data type does not need a single quote pair for the data variable.

15.4 Illustrative Example 2: Connecting to an Access Database

There are two ways to connect Excel and Access. The easiest approach is to go to Data → From Access. Following the procedure, we will be able to populate the data from the Access database. However, more often than that, a VBA project using ADO.NET can be developed to connect to an Access database programmatically and dynamically. In this case, the pro- cedure is similar to that in MySQL, with a couple of issues deserving special attention. First, Microsoft is making it confusing to us in terms of using 32-bit versus 64-bit Windows operating system, and 32-bit versus 64-bit Office product. If we are working on an older version of MS Office, we need to modify the codes in lines 4 and 5 as follows:

ConnectionString="Provider=Microsoft.Jet.OLEDB.4.0; Data Source=myAccessdatabase" cntMyConnection.open ConnectionString

If we are working on a newer version of Office (e.g., Access 2013), the syntax will be as follows:

ConnectionString="Driver={Microsoft Access Driver (*.mdb, *.accdb)}; DBQ=myAccessdatabase;" cntMyConnection.open ConnectionString

The main difference is that it is “Driver” instead of “Provider” following the ODBC con- nection syntax, and the database query (DBQ) parameter is used in replacing “Data Source.” Thus, the correct Access Driver needs to be downloaded and registered as that in MySQL.

308 Developing Windows-Based and Web-Enabled Information Systems

15.5 Illustrative Example 3: Connecting to XML

XML stands for extensible markup language, also known as a semi-structured database. It is becoming a standard platform for data exchange over the Internet; thus, many appli- cations are providing the interface to XML, including Microsoft Office products. More details of XML will be discussed in Chapters 17 and 18. Here, we will learn how Excel interacts with an XML document.

Importing XML to a worksheet is a relatively simple procedure: choose Data → From Other Sources → From XML Import. Following the wizard, the data in XML can be mapped to the cells in a worksheet. On the other hand, exporting data from Excel cells to an XML document may need some VBA programming. Simply exporting an Excel worksheet to an XML docu- ment may generate a messy XML file that is difficult to read. The following example shows how to use ActiveX data objects database (ADODB) objects to query from an Excel worksheet and save the data into an XML document. Here, we need to use the document object model (DOM) object, which will be explained in Chapter 18. As shown in Figure 15.3, references to the Microsoft ADO Library as well as the Microsoft XML Library need to be added.

1. Sub ExporttoXML() 2. Dim xmlMyConnection As ADODB.Connection 3. Set xmlMyConnection=New ADODB.Connection 4. Dim xmlMyRecordset As ADODB.Recordset 5. Set xmlMyRecordset=New ADODB.Recordset 6. Dim xmlMyXML As DOMDocument 7. Set xmlMyXML=New DOMDocument 8. Dim xmlMyWorkbook As String 9. xmlMyWorkbook=Application.ThisWorkbook.FullName 10. xmlMyConnection.Open "Provider=Microsoft.Jet.OLEDB.4.0;" & _

FIGURE 15.3 Adding references to a VBA project.

309Database Connectivity with VBA

Lines 1, 21: This is a Sub Procedure with Sub–End Sub. Lines 2 to 5: Initiate two ADODB objects: the xmlMyConnection object and the xmlMy-

Recordset object. Lines 6 and 7: Create an XML DOM Document object named xmlMyXML (refer to

Chapter 18 for DOM concepts). Lines 8 and 9: Get the name of the Excel workbook and assign it to a string variable

xmlMyWorkbook. Line 10: The open method for the connection object named xmlMyConnection is trig-

gered. Provider and Data Source information about the source data (the Excel workbook) need to be specified.

Line 11: Execute the open method for xmlMyRecordset. Other than the connection param- eter and the status of the connection, we specify the “query”–”Select * from [Sheet1$]”, which will search for all the data contained in Sheet1 and assign the set of records to the Recordset.

Lines 12 to 15: Use the If–Then–End If loop to read the Recordset. The EOF property is used to check if the Recordset has records. If yes, the cursor is moved to the first record, and all the records are saved to the DOM object.

Line 16: Save the XML file as “output.xml” in the same directory as the Excel workbook. Lines 17 to 20: Release all the system resources. Let us use the Excel in Figure 15.4 as the example. The exported output.xml is shown as

follows. Please refer to Chapter 18 for more details on the XML document.

"Data Source=" & xmlMyWorkbook & ";" & _ "Extended Properties=excel 8.0;" & _ "Persist Security Info=False" 11. xmlMyRecordset.Open "Select * from [Sheet1$]", xmlMyConnection, adOpenStatic 12. If Not xmlMyRecordset.EOF Then 13. xmlMyRecordset.MoveFirst 14. xmlMyRecordset.Save xmlMyXML, adPersistXML 15. End If 16. xmlMyXML.Save (ThisWorkbook.Path & "\Output.xml") 17. xmlMyRecordset.Close 18. Set xmlMyConnection=Nothing 19. Set xmlMyRecordset=Nothing 20. Set xmlMyXML=Nothing 21. End Sub

FIGURE 15.4 Exporting Excel to an XML example.

310 Developing Windows-Based and Web-Enabled Information Systems

<xml xmlns:s="uuid:BDC6E3F0-6DA3-11d1-A2A3-00AA00C14882" xmlns:dt="uuid:C2F41010-65B3-11d1-A29F-00AA00C14882" xmlns:rs="urn:schemas-microsoft-com:rowset" xmlns:z="#RowsetSchema"> <s:Schema id="RowsetSchema"> <s:ElementType name="row" content="eltOnly"> <s:AttributeType name="Drug_ID" rs:number="1" rs:nullable="true" rs:maydefer="true" rs:writeunknown="true"> <s:datatype dt:type="string" dt:maxLength="255"/> </s:AttributeType> <s:AttributeType name="name" rs:number="2" rs:nullable="true" rs:maydefer="true" rs:writeunknown="true"> <s:datatype dt:type="string" dt:maxLength="255"/> </s:AttributeType> <s:AttributeType name="price" rs:number="3" rs:nullable="true" rs:maydefer="true" rs:writeunknown="true"> <s:datatype dt:type="float" dt:maxLength="8" rs:precision="15" rs:fixedlength="true"/> </s:AttributeType> <s:AttributeType name="description" rs:number="4" rs:nullable="true" rs:maydefer="true" rs:writeunknown="true"> <s:datatype dt:type="string" dt:maxLength="255"/> </s:AttributeType> <s:extends type="rs:rowbase"/> </s:ElementType> </s:Schema> <rs:data> <z:row Drug_ID="T_001" name="Tylenol" price="17.989999999999998" description="100 325mg Tablets Bottle"/> <z:row Drug_ID="T_003" name="Tylenol" price="8.9900000000000002"/> <z:row Drug_ID="T_002" name="Tri_Vitamin" price="14.99" description="1500-400-35 Solution 50ml Bottle"/> <z:row Drug_ID="C_001" name="Claritin" price="35" description="10 Tablets"/> </rs:data> </xml>

15.6 Summary

This chapter covers an important data object model: ADO.NET. Within ADO.NET, the Connection object used to connect to the database (or file), the Command object used to hold the query command, and the Recordset object used to contain the retrieved data from the database are discussed. In addition, three illustrative examples are provided: Excel VBA connecting to MySQL, Excel VBA connecting to Access, and Excel VBA connecting to the XML document.

EXERCISES

1. An OLEDB object is an object model used by Microsoft Excel to connect to a Microsoft Access Database. True or false?

2. ADO.NET libraries are usually named by their data source type. True or false?

311Database Connectivity with VBA

3. When using VBA, the following objects are used in a VBA database connection except:

a. Dataset b. Connection c. Command d. Recordset e. All of the above are used in VBA 4. When using a connection object, the following are commonly included in the dec-

laration except: a. Password user account accessing the database b. Username of the user accessing the database c. Database server trough an IP address or a local drive directory d. All of the above are true e. None of the above is true 5. A DataAdapter is an example of a command object. True or false? 6. In VBA, a Recordset supports the database disconnected mode. True or false? 7. A Recordset object is a two-dimensional object composed of records and column

attributes. True or false? 8. Write an instruction to find the currently installed MySQL connection driver ver-

sion within a Windows PC. 9. Write an instruction to activate the Microsoft ADOs Library in Excel 2010. 10. A loop is typically used to scroll through Recordset objects. True or false? 11. Given a list called “DrugRecordset” queried from a table consisting of 14 drug

records, how many times will the subroutine PrintData be executed in the follow- ing code?

While DrugRecordset.BOF Call PrintData DrugRecordset.MoveNext Wend

12. The RecordsetObject.EOF is used to check if the pointer is positioned at the start of a Recordset to start the retrieval of data. True or false?

Section IV

Web Application Development

Section IV of this book covers the development of Web applications and Web services. A Web application is a network application that runs using the server–client protocol for Web. It is becoming increasingly popular due to the ubiquity of the Internet, Web brows- ers, and Web servers. In Chapter 16, the application of ASP.NET in Microsoft Visual Studio 2013 to develop Web applications is covered. To learn Web services, an important backbone standard, extensible markup language (XML) is described. Chapters 17 and 18 discuss the fundamentals of XML, followed by discussions on developing Web services in Microsoft Visual Studio.

315

16 Web Applications in Microsoft Visual Studio

A Web application is a network application involving a hypertext transfer protocol (HTTP) server and HTTP clients. Internet Explorer is a common HTTP client software package for the Windows environment. Internet information server (IIS) is a common HTTP server software package for the Windows environment. IIS is integrated in Visual Studio 2013. An HTTP server is a software system that hosts and stores Web pages. An HTTP client is a software system that reads and displays Web pages. Hypertext markup language (HTML) is used to construct the appearance of a Web page, while there is program code such as Visual Basic code to provide the back-end functions of a Web page. An HTTP server and an HTTP client work together through request–response cycles. In each request–response cycle, the HTTP client sends a request for an HTML file using universal resource locator (URL) to locate the HTTP server and the HTML file held by the HTTP server, and the HTTP server receives the request and responds to send the requested file to the HTTP client.

In this chapter, we first show how to start building a Web application project in Visual Studio 2013. We then introduce the use of HTML and Visual Basic code to construct a Web page. We also show how to incorporate an Access database into a Web application.

16.1 Starting a Web Application in Microsoft Visual Studio

Example 16.1 shows how to start building a Web application for the Healthcare Information System–Web using Visual Studio 2013.

Example 16.1

Create a Web application, the Healthcare Information System–Web, including a Web form that is similar to the Windows form in the Healthcare Information System shown in Figure 11.9.

After starting Visual Studio 2013, we see Start Page. We close Start Page and select New Website in the File menu. In the New Website window (see Figure 16.1), we select Visual Basic for Templates, select ASP.NET Empty Website, and click the Browse but- ton, which brings up the Choose Location window. In ASP.NET, ASP stands for active server pages. In the Choose Location window, we select File System on the left side of the window so that we can navigate folders in the file system and select a folder in which a project folder is created. For example, we can enter: C:\Users\nongye\ Documents\My\yenong\Healthcare Information System–Web in which yenong is an existing folder and Healthcare Information System–Web is the name that we give to our new project folder held in the yenong folder. We then click the Open button in the Choose Location window, which brings up a message telling us that the folder, C:\Users\nongye\Documents\My\yenong\Healthcare Information System–Web, does not exist and asking us if we want to create this folder. We click the Yes button, which brings us back to the New Website window. We click the OK button, which brings up

316 Developing Windows-Based and Web-Enabled Information Systems

the same development environment that we use to develop a Windows-based applica- tion in Chapter 11. The development environment has the design area along with the Toolbox tab, the Solution Explorer window, the Properties window, etc.

Next, we select Add New Item in the Website menu, which brings up the Add New Item window (see Figure 16.2). We select Web Form, change the name of the Web file to FormMain.aspx, and click the Add button, which brings up the Design View of the Web form kept in the Web file. We select the Web form and use the Properties window to change the ID of the Web form to FormMain. We want to add to this main Web form for the Healthcare Information System–Web application the same graphical user inter- face controls as those in FormMain (see Figure 11.9) in the Windows-based application “Healthcare Information System.” We drag and drop a Label from the Standard part of Toolbox to the Web form and use the Properties window to change the value for the Text property of the label. Next, we drag and drop a RadioButtonList from the Standard part of ToolBox to the Web form. Selecting the RadioButtonList bring up a right arrow button. Clicking the right arrow button and then selecting Edit Items in the pop-up menu brings up the ListItem Collection Editor window (see Figure 16.3). Using the Add button, we can add six items. We then select each ListItem, use ListItem Properties to change the text of ListItem, and click the OK button. We drag and drop a button from the Standard part of Toolbox to the Web form and use the Properties window to change the text of the button.

Figure 16.4 shows the Design View of the Web form. Switching from the Design tab to the Source tab at the bottom of the screen, we see the Source View of the Web form, which shows the Web page in HTML (see Figure 16.5). Going back to the Design View and double-clicking the Next Button brings up the Code View of the Web form (see Figure 16.6).

We right-click FormMain.aspx in the Solution Explorer and select Set as Start Page in the pop-up menu to set FormMain.aspx as the Start page of the Web application. To run

FIGURE 16.1 New Website window.

317Web Applications in Microsoft Visual Studio

FIGURE 16.2 Add New Item window.

FIGURE 16.3 ListItem Collection Editor window.

318 Developing Windows-Based and Web-Enabled Information Systems

the Web application, we select Start Debugging in the Debug menu. Figure 16.7 shows the display of the Web page when we run the application.

After we close Visual Studio and, thus, the website project, we can open the website project again by launching Visual Studio and selecting Open Website in the File menu.

The Design View of the Web form in Figure 16.4 and the Code View of the Web form in Figure 16.6 are similar to the Design View and the Code View of a Windows form described in Chapter 11. The Source View of the HTML file of the Web page is explained in the next section.

When a user clicks a button on a Web page at a client side, the Web page is automati- cally posted back for server-side processing. Many events may occur on a Web page at a client side, for example, mouse movement. For all controls on a Web page except the button

FIGURE 16.4 Design View of a Web form, FormMain, in the Healthcare Information System–Web.

FIGURE 16.5 Source View of a Web form, FormMain, in the Healthcare Information System–Web.

319Web Applications in Microsoft Visual Studio

control, the AutoPostBack properties of controls are set to False because the performance of the Web application will be degraded if the Web page is posted back to the server for every event. We can use the Properties window to change the AutoPostBack property of a control to True for enabling the automatic post-back.

16.2 Hypertext Markup Language

Figure 16.5 shows the HTML file of FormMain.aspx. HTML uses tags to mark Web data. For example, in Figure 16.5, we have

<head runat="server"></head> <body></body>

to mark two major parts of a Web page: head and body. The head in Figure 16.5 consists of a title marked by

<title></title>

The body in Figure 16.5 consists of the Web form marked by

<form id="FormMain" runat="server"></form>,

FIGURE 16.6 Code View of a Web form, FormMain, in the Healthcare Information System–Web.

FIGURE 16.7 The display of the Web page for FormMain.aspx when the website application, Healthcare Information System– Web, is executed.

320 Developing Windows-Based and Web-Enabled Information Systems

a label marked by

<asp:Label ID="Label1" runat="server" Text="Select one of the following:"></asp:Label>,

a RadioButtonList with several ListItems, and a button. As illustrated by the above examples of tags, we can use tag attributes to specify the

details of tags, such as Text="Select one of the following:" in the tags for the label and id="FormMain" in the tags for the Web form.

The HTML file in Figure 16.5 is made of tags that can be read by a Web client to know what to display on a Web page. The tags in the HTML file in Figure 16.5 are automati- cally generated when we create the Web form, and drag and drop several controls to the Web form. Hence, Visual Studio allows us to create a Web page using Toolbox with- out knowing the syntax of HTML tags and manually entering HTML tags. However, we can manually add tags in the HTML file of the Web page. For example, we can make the text for the label "Select one of the following:" in bold by adding the bold tags as follows:

<asp:Label ID="Label1" runat="server" Text="<b>Select one of the following:</b>"></asp:Label>

which has the bold tags embedded in the text tags. In Figure 16.7, the title of the Web page shows the default text “localhost” because there is no text in the title tags: <title> </title>. We can change the title of the Web page as follows:

<title>Healthcare Information System</title>

We can also add a hyperlink for another Web page as follows.

<a href="http://www.asu.edu">Arizona State University</a>

Figure 16.8 gives the new HTML file of Main.aspx after adding the above tags. Figure 16.9 shows the display of the Web page with "Select one of the following:" in bold, "Healthcare Information System" as the title of the Web page, and the hyperlink. When we click the hyperlink, we are directed to the website of Arizona State University.

16.3 Multiple Web Forms

After Example 16.1, we have one Web form, FormMain, in the Healthcare Information System–Web application. In Example 16.2, we add FormEmployers, FormPatients, FormProviders, FormCompanies, FormPhamarcies, and FormUpdates at the middle level of the hierarchy shown in Figure 11.13 and show how to move from one Web form to another Web form.

321Web Applications in Microsoft Visual Studio

Example 16.2

Add FormEmployers, FormPatients, FormProviders, FormCompanies, Form Phamarcies, and FormUpdates in the Healthcare Information System–Web applica- tion and link these forms to FormMain created in Example 16.1.

Following the steps in Example 16.1, we create FormEmployers, FormPatients, FormProviders, FormCompanies, FormPhamarcies, and FormUpdates, which are shown in Figures 16.10 to 16.15, respectively. Figure 16.16 gives the Code View of FormMain that redirects from FormMain to FormEmployers, FormPatients, FormProviders, FormCompanies, FormPhamarcies, or FormUpdates, depending on which item of the RadioButtonList is selected. As shown in Figure 16.16, the code for redirecting from one page to another page is

FIGURE 16.8 Updated HTML file for Main.aspx.

FIGURE 16.9 Display of the Web page in Figure 16.7 after manually adding new tags.

322 Developing Windows-Based and Web-Enabled Information Systems

FIGURE 16.10 Design View of FormEmployers.aspx in the Healthcare Information System–Web application.

FIGURE 16.11 Design View of FormPatients.aspx in the Healthcare Information System–Web application.

FIGURE 16.12 Design View of FormProviders.aspx in the Healthcare Information System–Web application.

FIGURE 16.13 Design View of FormCompanies.aspx in the Healthcare Information System–Web application.

323Web Applications in Microsoft Visual Studio

Response.Redirect("<a given Web form>")

For example, the code for redirecting from FormMain to FormEmployers is

Response.Redirect("FormEmployers.aspx")

The tags for a hyperlink can also be used to redirect from one Web page to another Web page. For example, we can add the following hyperlink tags in the HTML file of FormMain.aspx to redirect from the Web page for FormMain.aspx to that for FormEmployers.aspx.

<a href=FormEmployers.aspx>Employers</a>

FIGURE 16.14 Design View of FormPharmacies.aspx in the Healthcare Information System–Web application.

FIGURE 16.15 Design View of FormUpdates.aspx in the Healthcare Information System–Web application.

324 Developing Windows-Based and Web-Enabled Information Systems

16.4 Database Connection

Example 16.3 shows how to connect to an Access database in a Web application and dis- play data of a table in the database on a Web form.

Example 16.3

Add FormU1 to the Healthcare Information System–Web application, add code to redi- rect from FormUpdates to FormU1, and display data of the Person table in the healthcare Access database on FormU1.

Following the steps in Examples 16.1 and 16.2, we add FormU1 and code to redirect from FormUpdates to FormU1. To add the connection to the healthcare Access database, we bring up Server Explorer by selecting Server Explorer in the View menu. We right- click Data Connection and select Add Connection in the pop-up menu (see Figure 16.17). In the Add Connection window, we select the healthcare Access database file using the Browse button and click the OK button. Healthcare Access Database.accdb now appears in Server Explorer under Data Connection.

We double-click Healthcare Access Database.accdb in Server Explorer and double- click the Tables folder to expand and view the list of tables. We right-click the Person table and select Retrieve Data in the pop-up menu. We drag and drop the Person table in Server Explorer to the Design View of FormU1. This adds the GridView control and an AccessDataSource to FormU1. When we run the application and go to FormU1, we have the data of the Person table displayed. Figure 16.18a shows the Design View of FormU1, and Figure 16.8b gives the display of data in the Person table on FormU1.

We can select the GridView on FormU1, click the right arrow button at the top-right corner, select AutoFormat in the pop-up menu, and use the AutoFormat window to select a different display format for the GridView.

FIGURE 16.16 Code View of FormMain, including the code for redirecting to other forms.

325Web Applications in Microsoft Visual Studio

We click the right arrow for the GridView on FormU1 again; check Enable Paging, Enable Sorting, Enable Editing, Enable Deleting, and Enable Selection; and run the application. Figure 16.19 shows the display of FormU1 after checking Enable Paging, Enable Sorting, Enable Editing, Enable Deleting, and Enable Selection. We now have the data of the Person table displayed in two pages now with Enable Paging. When we click the heading of a given data field, data are sorted in ascending order of the values in the data field. Clicking the heading of the data field again changes the order of the values from ascending to descending. When we click Edit for a row, we can update the value of any data field in the row. When we click Delete for a row, we delete the row. Clicking Select for a row selects the row (Figure 16.19).

Data in each of the other tables in the Healthcare Access database can be displayed in the corresponding Web form in the Healthcare Information System–Web application by following the steps described in Example 16.3.

To display the query result of a query on an Access database, we need to design the query in an AccessDataSource. Example 16.4 shows how to design a query in an AccessDataSource.

Example 16.4

Add FormPA4 to the Healthcare Information System–Web application and display the query result of PAQ4 in the healthcare Access database on FormPA4. Figure 16.20 shows the query design of PAQ4 in Microsoft Access. This query presents a list of healthcare providers in Phoenix, Arizona, who are specializing in heart diseases.

Following the steps in Example 16.1, we add FormPA4 to the Healthcare Information System–Web application and add code that redirects from FormPatients to FormPA4 and from FormPA4 back to FormPatients. As shown in Figure 16.20, PAQ4 uses two

FIGURE 16.17 Server Explorer and Add Connection window for making connection to an Access database.

326 Developing Windows-Based and Web-Enabled Information Systems

(b)

(a)

FIGURE 16.18 (a) Design View of FormU1 and (b) the display of FormU1 when the Healthcare Information System–Web appli- cation is running.

327Web Applications in Microsoft Visual Studio

(b)

(a)

FIGURE 16.19 (a) Design View of FormU1 and (b) display of FormU1 when the Healthcare Information System–Web applica- tion is running after checking Enable Sorting, Enable Sorting, Enable Editing, Enable Deleting, and Enable Selection.

328 Developing Windows-Based and Web-Enabled Information Systems

tables: the CareProvider table and the Person table. We drag and drop one of the two tables, the CareProvider table, to FormPA4. This adds the GridView control and the AccessDataSource to FormPA4. We click the right arrow at the top-right corner of the GridView control and select Configure Data Source. Figure 16.21 shows the procedure of designing a query and its structured query language (SQL) using Configure Data Source. In the Choose a Database window, the healthcare Access database file is already selected, and we click the Next button. In the Configure the Select Statement window, we check Specify a Custom SQL Statement or Stored Procedure and click on the Next button. In the Define Custom SQL Statements or Stored Procedures window, we click on the Query Builder button. In the Query Builder window, we design the query as we did in Chapter 13. We right-click in the table pane area and select Add Table in the pop-up menu. In the Add Table window, we select the Person table, click the Add button, and click the Close button to go back to the Query Builder window, which now has the CareProvider and the Person tables in the table pane. Following the same steps described in Chapter 13, we enter the query design in Figure 16.20 in the Query Builder window, click the Execute Query button to see the query result, and click the OK button. The Define Custom SQL Statements or Stored Procedures window now has the completed SQL statement for the query, and we click on the Next button.

Note that, in Visual Studio 2013, we use

LIKE '%Phoenix%Arizona%'

in the selection criterion for the Address field, whereas in Access, we use

LIKE "*Phoenix*Arizona*"

for the same selection criterion. Hence, in Visual Studio 2013, % is the wild card to match any string, and a string is enclosed in ' ' rather than in " ". Visual Studio 2013 uses the operator “+” to put strings together into one string, whereas Access uses the operator “&” to do it. For example,

LIKE '%' + ? + '%'

matches any string that contains a string entered by a user, where “?” prompts the user to enter a string.

FIGURE 16.20 Query design of PAQ4 in the healthcare Access database.

329Web Applications in Microsoft Visual Studio

(a)

(b)

FIGURE 16.21 Procedure of designing a query and its SQL statement for a Web application. (a) Choose a Database window. (b) Configure the Select Statement window.

330 Developing Windows-Based and Web-Enabled Information Systems

(c)

(d)

FIGURE 16.21 (Continued) Procedure of designing a query and its SQL statement for a Web application. (c) Define Custom Statements or Stored Procedures window I. (d) Query Builder window.

331Web Applications in Microsoft Visual Studio

(e)

(f )

FIGURE 16.21 (Continued) Procedure of designing a query and its SQL statement for a Web application. (e) Define Custom Statements or Stored Procedures window II. (f) Test Query window.

332 Developing Windows-Based and Web-Enabled Information Systems

We click the Finish button in the Test Query window. We get a message asking us if we want to regenerate the GridView column fields, and we click the Yes button. We right-click the right arrow of the GridView and check Enable Paging and Enable Sorting. Figure 16.22 shows the Design View of FormPA4 and the display of FormPA4 when the application is running.

The query result for each of the other queries in the healthcare Access database can be displayed in the corresponding Web form in the Healthcare Information System–Web application by following the steps described in Example 16.4.

Example 16.5 shows how to display related data using a DetailsView control.

Example 16.5

Add data of the CareProvider table to FormU1 so that related data of the Person table and the CareProvider data are displayed.

We drag and drop the CareProvider table from Server Explorer to FormU1. We click the right arrow button and select Configure Data Source in the pop-up menu. In the Choose the Database window, we click the Next button. In the Configure the Select Statement window, we check Specify Columns from a Table or View, and click the Where button (see Figure 16.23). We configure the query in the Add WHERE clause window, as shown in Figure 16.23; click the Add button; and then click the OK button. We click the Next button in the Configure the Select Statement window and click the Finish button in the Test Query window. We run the application. On the Web page for FormU1, we click Select for the row with the ProviderID value of PR1, and we see the corresponding record in the CareProvider table displayed.

(b)

(a)

FIGURE 16.22 (a) Design View and (b) display of FormPA4 in the Healthcare Information System–Web application.

333Web Applications in Microsoft Visual Studio

(a)

(b)

FIGURE 16.23 Configuring the query to display related data. (a) Configure the Select Statement window. (b) Add WHERE Clause window I.

334 Developing Windows-Based and Web-Enabled Information Systems

16.5 Summary

This chapter describes how to develop a Web application using Visual Studio 2013. We introduce three parts of a Web page—design, code, and HTML source—and explain the components of an HTML source file. We show how to add multiple Web forms for multiple Web pages to a Web application and redirect the control from one Web form to another Web form. We then show how to include an Access database in a Web application and dis- play data from a table and related data of multiple tables in the Access database on a Web form. We describe how to design a query such that the query result can be displayed on a Web form (Figure 16.24).

EXERCISES

1. Develop a Web application as shown in the following that does the currency conversion.

2. Develop a Web application for a simple calculator that can perform addition, sub- traction, multiplication, and division.

(c)

FIGURE 16.23 (Continued) Configuring the query to display related data. (c) Add WHERE Clause window II.

335Web Applications in Microsoft Visual Studio

FIGURE 16.24 Design View and display of FormU1 showing related data.

336 Developing Windows-Based and Web-Enabled Information Systems

3. Complete the Web application for the Healthcare Information System–Web to include all the forms and the display of data from each table and each query that are contained in the Windows-based application completed in Exercise 13.1.

4. Complete the Web application for the Movie Theater Information System–Web to include all the forms and the display of data from each table and each query that are contained in the Windows-based application completed in Exercise 13.2.

337

17 Working with XML (I)

Extensible markup language (XML) is a markup language that defines the structures of documents. Like hypertext markup language (HTML), XML originated from standard generalized markup language (SGML). This chapter first briefly reviews the concepts of markup languages, specifically, SGML and HTML. Then, the basic XML elements are dis- cussed. Furthermore, two important concepts of XML, well-formed and validated XML, are explained in detail.

17.1 What Is Markup Language?

A markup language is designed to process, define, and present information in text format. In 1974, SGML was first created from an IBM document-sharing project, and it became an International Organization for Standardization (ISO) standard (ISO 8879) in 1986. As the Internet grew, especially the World Wide Web (WWW), two popular adaptations of SGML emerged: one is HTML, a standard to describe document layout and display pages, and the other one is XML, a standard to describe document semantics, thus enabling data exchange over the Internet.

17.1.1 Standard Generalized Markup Language

An SGML document is composed of several basic elements. An element is used to define the document structure and to present information contained within the document. Moreover, it should be noted that different SGML elements can be given different names. Since SGML was initially designed to manage and format large documents subject to fre- quent revisions, it is only concerned with annotating the syntax instead of the semantics of each element. Additionally, the most important rule in SGML is that all elements need to be explicitly marked. Specifically, the start-tag and end-tag pair is used to bracket off the element, as the way of using parentheses or quotation marks in conventional punctuation. An example SGML is shown in Table 17.1.

This example shows that the document type is a Book, and it consists of a number of book chapters. Each Chapter has a title and several sections, and each section has several paragraphs. This structure is represented by start-tag and end-tag pairs as follows:

• <Book> </Book>

• <title> </title>

• <section> </section>

• <paragraph> </paragraph>

338 Developing Windows-Based and Web-Enabled Information Systems

From this example, it is learned that SGML rules emphasize the use of start and end tags to enclose each element. It does not, however, specify strict rules on how the tags should be  placed. For example, whether a <title> can appear in places other than preceding the  <section>, or whether a <paragraph> can appear outside a <section>, is not strictly required.  In such cases, the beginning and the end of every element must be explicitly marked because there are no identifiable rules as to where an element can appear. As a result, SGML is not able to extract the semantic meaning of the elements.

17.1.2 Hypertext Markup Language

The first popular derivative of SGML is HTML, which was inspired by SGML’s tagging concept. It is the main markup language for displaying Web pages as well as its basic build- ing block. A typical Web browser does not display an HTML document’s tags. Instead, it uses the tags to interpret and display the corresponding HTML Web page. Some standard HTML tags are headings (e.g., h1), formatting (e.g., b for bold), font (e.g., font size), and image (e.g., img src) to name a few. An HTML document example is shown in Table 17.2, with its corresponding display using a Web browser shown in Figure 17.1.

As shown in the example, HTML follows the tagged structure recommended by SGML. The tags <html>, <head>, <body>, and <h1> are standard and can be interpreted by any Web browser. On the other hand, HTML is less strict on the syntax of its code. For example, HTML is not case sensitive in the sense that <h1> and <H1> will be treated the same way. HTML does not require the start-tag and end-tag pairs to be closed: an HTML document having <h1> without being closed by </h1> is still accepted by the Web browser.

The two aforementioned examples illustrate that SGML is used to define the structure for document sharing without concerning the semantics of the elements, while HTML is

TABLE 17.1

Example SGML Document (ch17_SGMLExample.sgml) <Book> <Chapter ID="17"> <title> Working with XML </title> <section> <paragraph>SGML stands for Standard Generalized Markup Language.</paragraph> <paragraph>XML stands for Extensible Markup Language.</paragraph> </section> </Chapter> </Chapter ID="18"> …. </Chapter> </Book>

TABLE 17.2

Example HTML Document (ch17_HTMLExample.html) <html> <head> <title>Hello HTML</title> </head> <h1>Hello World!</h1> </html>

339Working with XML (I)

used as a standard to define the layout for document presentation. However, there was no standardized way to describe and share data contained within the document. Motivated by the success of SGML and HTML, the WWW Consortium (W3C) initiated the process to develop a standard data sharing mechanism and came up with the first XML recom- mendation in 1998.

17.1.3 Extensible Markup Language

XML is developed to describe the content of a document in a standardized, text-based for- mat that can easily be transported over the Internet. One notable advantage of XML is its extensibility. That is, the developer has the freedom to extend data elements provided by the basic XML standard as long as it follows the XML syntax and is validated to be correct. As such, the structure of an XML document would be usually complex for any human being to digest. However, there are a number of tools available (some are discussed in Chapter 18) to make XML validation and interpretation easy even in its most complex form. Compared to HTML, XML has much more rigid rules in its syntax. In addition, an XML document’s structure and content need to be validated based on some validation standard, which includes data type definition (DTD) and XML schema. Although optional, data validation is becoming increasingly important, especially for new XML document developments.

17.2 XML Basics: Elements and Attributes

The W3C XML Recommendation is strict about the format requirements. This provides a major difference between an SGML and an XML document. These requirements are then used to define a well-formed XML document that is composed of elements, attributes, and text.

17.2.1 Elements

Elements are the basic building blocks of an XML document. An element in an XML starts with a “< >” tag and ends with a “</>” tag, and it looks like the following:

<name_of_the_element> text </name_of_the_element>

FIGURE 17.1 Display of the example HTML (ch17_HTMLExample.html).

340 Developing Windows-Based and Web-Enabled Information Systems

The names of the elements are developer defined and must adhere to the following basic rules:

• Element names must begin with a letter or an underscore (_) and cannot start with “xml” (reserved by XML).

• Element names cannot contain spaces. Underscores are usually used to replace space.

• Element names can contain any number of letters, digits, and underscores. • A colon (:) is not recommended as it is generally reserved for namespaces (dis-

cussed later in the chapter). • A dot (.) is also not recommended as it could be troublesome for object operation. • Hyphen (-) is not recommended as it may confuse the program with the subtrac-

tion (−) operator. • All names are case sensitive for start and end tags, which must match perfectly.

EXERCISE 17.1

Correct and incorrect examples of XML element names <1stChapter> </1stChapter> Incorrect since the element begins with invalid characters <first Chapter> </first Chapter> Incorrect since no space is allowed in the element name <first_Chapter> </first_Chapter> Correct <_firstChapter> </_firstChapter> Correct <xmlbook> </xmlbook> Will generate errors since “xml” is reserved by the namespace <title> </Title> Incorrect since XML is case sensitive <!title> </!title> Will generate errors since the declaration has an invalid name <Chapter.title> </Chapter.title> Will generate errors since the declaration has an invalid name

<Chapter:title> </Chapter:title> Incorrect; reference to undeclared namespace prefix

17.2.2 Attributes

XML elements can have attributes that provide additional information about an element. It is presented as part of an element’s starting tag as follows:

<name_of_the_element name_of_the_attribute="value"> </name_of_the_ element>

The naming rules of attributes follow those for element names, with some additional requirements, including the following:

• Attribute values must be quoted. Either single or double quotes can be used. If the value itself contains one type of quote (e.g., double quote), the other type of quote (e.g., single quote) shall be used for the attribute value. For example,

<gangster name="George 'Shotgun' Ziergler"> </gangster>

• Attributes cannot contain multiple values.

341Working with XML (I)

There are no explicit rules on when to use attributes or when to use elements. In the con- text of documents, attributes are part of the markup, while elements are part of the basic document structure. Sometimes, ID references are assigned to elements to identify each element. These metadata (data about data) are recommended to be stored as an attribute, and the data itself should be stored as element. For example,

<Chapter ID="C1"> text </Chapter>

is usually preferred by the developer than

<Chapter> <ID> C1 </ID> </Chapter>

The concepts of ID references will be discussed more in the DTD section later in this chapter.

17.2.3 Text

Text is located between the starting and closing tags of an element, and usually represents the actual data associated with the elements:

<element attribute="value"> text </element>

Note that the text is not constrained by the same syntax rules of elements and attributes, so any text can be used for the elements. The data type of the text (e.g., String, Time, Date) is usually specified in a validation standard via a DTD or an XML schema.

17.2.4 Empty Elements

Sometimes, elements with no attribute or text can also be represented as follows:

<element> </element> or <element/> or <element attribute="value"/>

This is usually used to accommodate a predefined data structure.

17.2.5 Well-Formed XML

Other than syntax rules on elements, attributes, and text, structural rules are defined to ensure a well-formed XML. Let us use the following example to explain the structure of a well-formed XML document (Table 17.3).

Example 17.1

Line 1: XML document declaration: Most XML documents start with a <?xml?> element, which refers to an XML document declaration. In the W3C XML 1.0 specification, this is not required for a well-formed XML. However, it is very useful to determine the version of XML and the encoding type of the source data. In this example, line 1 indicates that the version of the XML document is “1.0” and that the encoding type is “UTF-8.”

342 Developing Windows-Based and Web-Enabled Information Systems

The XML version is used to identify version-specific syntax rules to parse the XML document. As of today, there are two versions of XML: version 1.0 and version 1.1. Version 1.1 improves upon XML 1.0 by fixing a few errors and by making the support for Unicode stronger (Table 17.4).

The encoding type refers to the character set (e.g., ASCII, Unicode). Since the XML Recommendation was developed by W3C, an international organization, Unicode was chosen as the standard text format to accommodate the world’s language. Commonly used values for encoding attributes are “UTF-8” and “UTF-16,” where “UTF” stands for “universal character transformation format,” and the number “8” or “16” refers to the number of bits that the character is stored in. Using either UTF-8 or UTF-16, XML editors, generators, and parsers can identify and work with all major world languages and alphabets.

Line 2 and line 15: Root element: Every XML document starts with a root element, and in this example, we have: <Book>…</Book>. The start-tag <Book> is in line 2, and the end-tag </Book> is in line 15. Other elements and text values can be nested under the root element; however, the root element must be the first in the document and it must be unique.

Line 3 to line 8: Under the root element, the first nested sub-element is <Part>. This ele- ment has an attribute ID (“ID”), which has a value of “P1,” while the text for this element is “This is the first part.” Underneath the first sub-element Part, there are two next-level nested sub-elements of type: Chapter (Chapter 1 and Chapter 2). The first sub-element

TABLE 17.3

Example XML Document (ch17_SimpleXMLExample.xml) 1. <?xml version="1.0" encoding="UTF-8"?> 2. <Book> 3. <Part ID="P1">This is the first part 4. <Chapter ID="C1">This is the first chapter. 5. </Chapter> 6. <Chapter ID="C2">This is the second chapter. 7. </Chapter> 8. </Part> 9. <Part ID="P2">This is the second part 10. <Chapter ID="C3">This is the third chapter. 11. </Chapter> 12. <Chapter ID="C4">This is the fourth chapter. 13. </Chapter> 14. </Part> 15. </Book>

TABLE 17.4

XML Specifications and Timeline

Specification Recommendation

XML 1.0 Feb. 10, 1998 XML 1.0 (2nd Edition) Oct. 06, 2000 XML 1.0 (3rd Edition) Feb. 04, 2004 XML 1.0 (5th Edition) Nov. 26, 2008 XML 1.1 Feb. 04, 2004 XML 1.1 (2nd Edition) Aug. 16, 2006

Source: w3c, 2014. http://www.w3.org.

343Working with XML (I)

<Chapter> has an attribute ID (“ID”) having a value of “C1,” and the text for this element is “This is the first Chapter.” Moreover, the end tag has to be nested appropriately to close the element, specifically, we write </Chapter>. The second <Chapter> sub-element has an attribute ID (“ID”) having a value of “C2,” and the text is “This is the second Chapter.” followed by the closing tag </Chapter>, which needs to be prior to the </Part> (line 8) sub-element.

Line 9 to line 14: Similar to lines 3 to 8, this is another instance of a nested sub-element <Part> under the root element. It has an attribute ID having a value of “P2,” while its corresponding text is “This is the second part.” In addition, it has two sub-elements Chapter (Chapter 3 and Chapter 4).

As seen from the example, a well-formed XML needs to follow rigid syntax and struc- tural rules. Specifically,

• A well-formed XML document must have a corresponding end tag for all its start tags.

• Proper nesting of elements within each other in an XML document is a must. • In each element, each attribute needs to be distinctly specified. • An XML document can contain only one root element. • The root element shall appear only once in an XML document, and it cannot

appear as a sub-element within any other element.

EXERCISE 17.2

Well-formed XML versus not well-formed XML <Book> </Book> Not well formed since XML is case sensitive; start and closing

tags do not match <Part ID=1> </Section> Not well formed since the attribute value needs to be enclosed

within quotation marks <Part ID="1"> Not well formed since closing tag is missing <Part ID="P1" ID="P2"/> Not well formed since attribute name needs to be unique <Part id="P1" ID="P2"/> Well formed! XML is case sensitive (note that, this, however, is

not a good idea) <Part ID="P1"> <Chapter ID="C1"/> </Part>

Well formed! Attribute name needs to be unique within one start tag or empty-element tag

<Part> <Chapter> </Part> </Chapter>

Not well formed since the elements are not properly nested

<!---correct way of comments ---> Well formed! This follows the SGML comment tag format

17.3 XML Namespaces

The namespace is a great concept implemented in XML to support extensibility. Since XML allows developers to choose their own element names, duplicated names may appear. A namespace is implemented to identify and differentiate identical names from different vocabularies in an XML document. This is done by prefixing a name to an element and an

344 Developing Windows-Based and Web-Enabled Information Systems

attribute so that elements with the same name but different prefixes will not be in conflict with each other. The namespace declaration is defined as follows:

<prefix:localname xmlns:prefix='namespace URI'/> or <prefix:localname xmlns:prefix='namespace URI'></prefix:localname/> or <prefix:localname xmlns:prefix='namespace URI'>children</ prefix:localname/>

Note that, in defining namespace declarations, a reserved xmlns: prefix is used. The attribute name after the xmlns: prefix identifies the name for the defined namespace. An example namespace declaration is as follows:

<eb:textbook xmlns:eb= http://www.crcpress.com/engineeringbooks>

The name of the namespace is eb, which has a uniform resource identifier (URI) http:// www.crcpress.com/engineeringbooks. All elements from this namespace will have the prefix eb, for example, eb:industrialengineering, eb:mechanicalengineering.

Let us take a look at two XML examples shown in Table 17.5. Apparently, if these two XML documents are put together, the XML parser will not

be able to process this document since both have <Book> and <Title> elements. Using namespace, the problem is solved as shown in Table 17.6.

TABLE 17.5

Two XML Examples Without Using Namespace

Example I: CRC Press Publishes Engineering Textbooks <?xml version="1.0" encoding="UTF-8"?> <Book> <Title>Discrete Event Simulation </Title> </Book>

Example II: CRC Press Publishes Computing, Information Technology Textbooks <?xml version="1.0" encoding="UTF-8"?> <Book> <Title>Developing Information Systems for Windows and Web Applications </Title> </Book>

TABLE 17.6

Using Namespace <?xml version="1.0" encoding="UTF-8"?> <eb:Book xmlns:eb="http://www.crcpress.com/engineeringbooks"> <eb:Title>Discrete Event Simulation </eb:Title> </eb:Book> <cit:Book xmlns:cit="http://www.crcpress.com/computingitbooks"> <cit:Title>Developing Information Systems for Windows and Web Applications </cit:Title> </cit:Book>

345Working with XML (I)

Although namespaces are optional components of basic XML documents, they are rec- ommended if the XML documents have any current or future potential of being shared with other XML documents that may have the same element names. In addition, XML- based technologies such as XML schemas, simple object access protocol, and Web services description language make heavy use of namespaces to identify the encoding types and the important elements of the structure.

17.4 XML Validation

Given a well-formed XML, the document needs to be validated against a pre-defined struc- ture. XML validation encompasses element data type verification, proper sub- element nesting, and correct element sequencing, just to name a few. This validation process is based on either a DTD or an XML schema.

17.4.1 DTD and XML Validation

DTD is the traditional way to (1) define the legal building blocks of an XML document and (2) define the specific structure and format of the document using the building blocks. Thus, DTD has been commonly used to validate an XML document.

Example 17.2

For the XML example in Table 17.3, create the DTD document to define its structure (Table 17.7).

Line 1: The Book element has one sub-element—Part. The “+” sign means that there may be more than one Part in the Book.

Line 2: The Part element has one sub-element—Chapter. The “+” sign means that there may be more than one Chapter within each Part.

Line 3: The element Part has an attribute “ID.” Each instance of the Part element is required (#REQUIRED) to have this attribute and must have regular character data (CDATA).

Line 4: The element Chapter can contain any data as text. Line 5: The element Chapter has an attribute “ID.” Each instance of the Chapter element

is required (#REQUIRED) to have this attribute and must have regular character data (CDATA).

This example illustrates the basic DTD document that can be used to validate an XML document. Next, how the detailed DTD structures are defined is explained.

TABLE 17.7

DTD for Example XML from Table 17.3 (ch17_DTDonSimpleXMLExample.dtd) 1. <!ELEMENT Book (Part+)> 2. <!ELEMENT Part (Chapter+)> 3. <!ATTLIST Part ID CDATA #REQUIRED> 4. <!ELEMENT Chapter ANY> 5. <!ATTLIST Chapter ID CDATA #REQUIRED>

346 Developing Windows-Based and Web-Enabled Information Systems

17.4.1.1 DTD Structure

DTD element structure: There are a number of different syntaxes to define element struc- tures in DTD documents. However, all of these structures use the !ELEMENT statement and are mandatory.

• <!ELEMENT element-name>: A statement that declares an element. As it is unspecified, the sub-elements can be in any order.

Example 17.3

DTD example: <!Element Chapter> XML example: <Chapter> text </Chapter>

• <!ELEMENT element-name EMPTY>: Declares an empty element. It is an empty element having no sub-elements, but attributes may be defined for the element.

Example 17.4

DTD example: <!ELEMENT Appendix EMPTY> XML example: <Appendix/>

Example 17.5

DTD example: <!ELEMENT Appendix EMPTY>

<!ATTLIST Appendix Appendix_ID>

XML example: <Appendix Appendix_ID="Appendix for Chapter 1"/>

Note: <!ATTLIST> is DTD attribute which will be discussed later in the chapter.

• <!ELEMENT element-name (#PCDATA)>: A statement that implies that an element that can contain only text data, but no sub-elements. Note that the non-markup text for the element is commonly taken as parsed character data (PCDATA) in DTD. The text defined as PCDATA will be processed and analyzed for the content by the XML parser.

Example 17.6

DTD example: <!ELEMENT Section (#PCDATA)> XML example: <Section> text </Section>

• <!ELEMENT element-name (child)>: Means that the element has one sub-element.

Example 17.7

DTD example: <!ELEMENT Book (Title)> XML example: <Book> <Title> text </Title> </Book>

• <!ELEMENT element-name (child1, child2…)>: A statement that declares an element that has more than one sub-element that follow the order specified.

347Working with XML (I)

Example 17.8

DTD example: <!ELEMENT Book (Title, Part)> XML example: <Book> <Title> text </Title> <Part> text </Part> </Book>

DTD element cardinality: After the element structure is defined, the element cardinality is used to define how many times each element may appear within the current element. These include the following:

• x: x can appear once.

Example 17.9

DTD example: <!Element Book (Title)>. This DTD cardinality statement means that there is only one Title within a book element. XML example: <Book> <Title> text </Title> </Book>

• x+: x can appear one or more times.

Example 17.10

DTD example: <!Element Part (Chapter+)>. This implies that there can be one or more Chapters within a Part element. XML example: <Book> <Chapter> chapter text </Chapter> <Chapter> more chapter text </Chapter> </Part>

• x*: x can appear zero or more times.

Example 17.11

DTD example: <!Element Chapter (Section*)>. This means that there can be zero or more than one Sections within a Chapter element. XML example: <Chapter> <Section> section text </Section> <Section> more section text </Section> </Chapter>

or </Chapter> <Section> </Chapter>

• x?: x can appear once or not at all.

Example 17.12

DTD example: <!Element Chapter (Appendix?)>. This statement means that some Chapters may have one Appendix or some Chapters may have no Appendix.

348 Developing Windows-Based and Web-Enabled Information Systems

XML example: <Chapter> <Appendix> appendix text </Appendix> </Chapter> </Chapter> <Appendix> </Chapter>

• x, y: Means that x is followed by y.

Example 17.13

DTD example: <!Element Book (Title, Part)>. This means that the Book element has two sub-elements, with the first sub-element being the Title element followed by the Part element. XML example: <Book> <Title> title text </Title> <Part> title text </Part> </Book>

Note that this is the same as Example 17.8.

• x | y: x or y, but not both.

Example 17.14

DTD example: <!Element Books (EngineeringBook|ComputerITBook)>. This implies that the Book element is either an EngineeringBook or a ComputerITBook, but not both. XML example: <Book> <EngineeringBook> text </EngineeringBook> </Book>

or

<Book> <ComputerITBook> text </ComputerITBook> </Book>

DTD attribute declaration: In a DTD, attributes are declared with an ATTLIST declaration along with their corresponding data type. CDATA (character data) is the most common data type to declare for an attribute. This data type is not parsed by the XML processor. Other data types include (1) ID: the value is a unique ID; (2) IDREF: the value is the ID of another element; (3) IDREFS: the value is a list of other IDs; (4) NMTOKEN: the value is a valid XML name; (5) NMTOKENS: the value is a list of valid XML names; and (6) ENTITY: the value is an entity. While there are a number of ways to declare an attribute, some commonly used ones are provided as follows:

• <!ATTLIST element-name attribute-name CDATA>: A declaration that implies that the attribute is non-parsed character data.

Example 17.15

DTD example: <!ATTLIST Book Author> XML example: <Book Author="William Johnson">

• <!ATTLIST element-name attribute-name CDATA "default value">: Means that the attribute is non-parsed character data with the default-value given.

349Working with XML (I)

Example 17.16

DTD example: <!ATTLIST square size "10"> XML example: <square size="100"> (Note: If no size is specified, it has a default value of 10.)

• <!ATTLIST element-name attribute-name CDATA #REQUIRED>: Used in cases where the default value is not available and there is still a need to have values set for the attribute. The #REQUIRED keyword is usually used to handle this requirement.

Example 17.17

DTD example: <!ATTLIST Book Total_Pages #REQUIRED> XML example: <Book Total_Pages="500">

• <!ATTLIST element-name attribute-name CDATA #IMPLIED>: Means that the attribute is optional.

Example 17.18

DTD example: <!ATTLIST Book Co-Authors #IMPLIED> XML example: <Book/>

• <!ATTLIST element-name attribute-name CDATA #FIXED "value">: A declaration that the attribute is optional in the XML document but is specified in the DTD. The keyword #FIXED can be used to prevent XML developers to change its value.

Example 17.19

DTD example: <!ATTLIST Book Publisher #FIXED "CRC Press"> XML example: <Book Publisher="CRC Press"> or <Book/>

• <!ATTLIST element-name attribute-name ID #REQUIRED attribute- name IDREFS>: Implies that the ID attribute value of each element in an XML docu- ment must be distinct, so the ID attribute value is an object identifier. Furthermore, an attribute of type IDREF must contain the ID value of an element in the same document.

Example 17.20

DTD example: <!ATTLIST individual individual_id ID #REQUIRED parent_ id IDREFS> XML example: <individuals> <individual individual_id="e10001" parent_id="e10002 e10003"> <first_name>Bart</first_name> <last_name>Simpson</last_name> </individual> <individual individual_id="e10002"> <first_name>Homer</first_name> <last_name>Simpson</last_name> </individual> <individual individual_id="e10003"> <first_name>Marge</first_name> <last_name>Simpson</last_name> </individual> </individuals>

350 Developing Windows-Based and Web-Enabled Information Systems

In this example, the individual with ID (individual_id) “e10001” is the child of parents with ID e10002” and “e10003.” This relationship is described using two attributes: ID and IDREF.

• <!ENTITY entity-name "entity-value">: Enables the reference to data that are outside of the current document. An ENTITY has three parts: an ampersand (&), an entity name, and a semicolon (;).

Example 17.21

DTD example: <!ENTITY copyright http://www.crcpress.com/entities.dtd> XML example: <author> &copyright; </author>

17.4.1.2 Internal versus External DTD

A DTD can be declared inline inside an XML document or as an external reference. To include a DTD inside an XML document, the <!DOCTYPE> element is declared. Taking the examples from Table 17.3 (XML) and Table 17.7 (DTD), an XML document with an internal DTD is shown in Table 17.8.

On the other hand, we can also create a DTD document externally and add a reference in the XML document. Public and private settings are two ways to refer to a DTD document externally.

Let us create a DTD document “Ch17_DTDonSimpleXMLExample.dtd” from Table 17.7. Now, we can specify that we are using an external private DTD by using the SYSTEM key- word in the <!DOCTYPE> element, such as this:

<!DOCTYPE Book SYSTEM "Ch17_DTDonSimpleXMLExample.dtd">

TABLE 17.8

Example XML Document with Internal DTD <?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE Book [ <!ELEMENT Book (Part+)> <!ELEMENT Part (Chapter+)> <!ATTLIST Part ID CDATA #REQUIRED> <!ELEMENT Chapter ANY> <!ATTLIST Chapter ID CDATA #REQUIRED ]> <Book> <Part ID="S1">This is the first part. <Chapter ID="C1">This is the first chapter. </Chapter> <Chapter ID="C2">This is the second chapter. </Chapter> </Part> <Part ID="S2">This is the second part. <Chapter ID="C3">This is the third chapter. </Chapter> <Chapter ID="C4">This is the fourth chapter. </Chapter> </Part> </Book>

351Working with XML (I)

The corresponding XML document is written as shown in Table 17.9. In this example, it is assumed that the DTD is located in the same folder as the XML

document. In case that the DTD is located in a different place, a uniform resource locator (URL) can be used to point to the DTD as:

<! DOCTYPE document SYSTEM "http://www.crcpress.com/dtds/Ch17_ DTDonSimpleXMLExample.dtd">

To create and use a public external DTD, the keyword PUBLIC is used instead of SYSTEM in the <!DOCTYPE> DTD. In addition, a formal public identifier (FPI) needs to be created, which is a quoted string of text that is made up of four fields separated by “//.” An example of an FPI is “-//W3C//DTD HTML 4.01//EN.” The first field indicates whether the DTD is written for a formal standard. For DTDs that are created on our own, this field should be written as “-.” If a non-official standard body has created the DTD, the “+” character is used. For formal standard bodies, this field is a reference to the standard itself (such as ISO/IEC 19775:2003). The second field holds the name of the group or person respon- sible for the DTD. Note that the name used should be unique (e.g., W3C just uses W3C). Furthermore, the third field specifies the type of the document that the DTD is used for and should be followed by a unique version number of some kind (e.g., XML 1.1). The fourth field specifies the language in which the DTD is written for (EN for English).

When a public external DTD is used, the <!DOCTYPE> element is defined as:

<!DOCTYPE rootname PUBLIC FPI URI>

Let us assume that we create a public DTD using a made-up FPI as follows: -//CRCPRESS// My DTD Version 1.0//EN. This document uses Ch17_DTDonSimpleXMLExample.dtd as the external DTD, as in the previous example, but as we can see, it treats the DTD as a public external DTD, complete with its own FPI (Table 17.10).

While DTD is still in use for XML validation, it has several major limitations. First, it is not written in XML syntax; as a result, the developers have to learn a new syntax to develop

TABLE 17.9

Referring to an External DTD in Private <?xml version="1.0" encoding="UTF-8"> <!DOCTYPE Book SYSTEM "Ch17_DTDonSimpleXMLExample.dtd"> <Book> <Part> … </Part> </Book>

TABLE 17.10

Referring to an External DTD in Public <?xml version="1.0" encoding="UTF-8"> <!DOCTYPE Book PUBLIC "-//CRCPRESS//My DTD Version 1.0//EN" "http://www.crcpress.com/dtds/Ch17_DTDonSimpleXMLExample.dtd"> <Book> <Part> … </Part> </Book>

352 Developing Windows-Based and Web-Enabled Information Systems

the DTD. Second, the DTD does not support namespaces, which limits its extensibility. In addition, there are no constraints imposed on the type of character data (e.g., Date, Time). To address these issues, an XML schema is recommended by W3C as an updated docu- ment format for XML validation.

17.4.2 XML Schema and XML Validation

XML schema is an XML-based alternative to DTD that is used to describe the structure of an XML document and, thus, is utilized to validate it. The XML schema language is also referred to as “XML schema definition” (XSD).

17.4.2.1 XML Schema Data Types

One drawback of DTDs is that it has limited support for pre-defined data types. An XML schema can support all DTD data types, including CDATA (replaced by primitive String data type), ID, IDREF, IDREFS, ENTITY, ENTITIES, NMTOKEN, NMTOKENS, and NOTATION. In addition, other data types are supported in a different format. Some of the primary types of data used in a schema are summarized in Table 17.11.

17.4.2.2 XML Schema Namespaces

One major advantage of XML schema over DTD is that XML schema supports namespaces as part of its document. For example, the <schema> element is the root element of every XML schema and is usually defined as shown in Table 17.12.

In this example,

Line 1: Indicate that the elements and data types used in the schema element (the root element) come from http://www.w3.org/2001/XMLSchema namespace, with the prefix being xs:, for example, xs:schema.

Line 2: Under the root element, the developer defines the elements coming from “http://www.crcpress.com” namespace.

TABLE 17.11

XML Schema Data Types

Name Description

String Any well-formed XML string Date Date value in the format YYYY-MM-DD Time Time value in the format HH:MM:SS dateTime Combined date and time value in the format YYYY-MM-DD HH:MM:SS Number Any numeric value up to 18 decimal places Decimal Any decimal value number Float Any 32-bit floating point–type real number Double Any 64-bit floating point–type real number Integer Any integer Byte Any signed 8-bit integer Int Any signed 32-bit integer Long Any signed 64-bit integer Boolean Standard binary logic, in the format of 1, 0, true, or false

353Working with XML (I)

Line 3: Indicate that the default name space is “http://www.w3schools.com.” Line 4: Indicate that any element declared in this XML schema must be namespace

qualified. Line 5: The closing tag of the root element schema to ensure that it is a well-

formed XML itself.

17.4.2.3 XML Schema Elements and Attributes

Within the schema root element, the details of every element contained in the XML docu- ment are defined. There are two types of elements: simple and complex. Each element type is defined as follows.

17.4.2.3.1 Simple Elements

A simple element is an XML element that contains only text and does not contain sub- elements or attributes. The text can be of any different types, including some listed in Table 17.11, or customized as defined by the developer. The syntax for defining a simple element is

<xs:element name="xxx" type="yyy"/>

An example XML schema is

<xs:element name="person" type="xs:string"/> <xs:element name="age" type="xs:integer"/>

A simple element may have a default value of a specified fixed value. For example,

<xs:element name="size" type="xs:integer" default="10"/> <xs:element name="publisher" type="xs:string" fixed="CRCPress"/>

17.4.2.3.2 Complex Elements

On the other hand, a complex element is an XML element that contains nested elements and/or attributes.

• The syntax for defining a complex element that only contains nested elements is

<xs:element name="xxx"> <xs:complexType> <xs:sequence> <xs:element name="xxx-xxx" type="xs:yyy"/> <xs:element name="xxx-xxx-xxx" type="xs:yyy"/> </xs:sequence> </xs:complexType> </xs:element>

TABLE 17.12

Schema Element in an XML Schema Document 1. <xs:schema xmlns:xs=http://www.w3.org/2001/XMLSchema 2. targetNamespace=http://www.crcpress.com 3. xmlns=http://www.w3schools.com 4. elementFormDefault="qualified"> … 5. </xs:schema>

354 Developing Windows-Based and Web-Enabled Information Systems

• The syntax for defining a complex element that only contains attributes is

<xs:element name="xxx"> <xs:complexType> <xs:attribute name="xxx-xxx" type="xs:yyy"/> </xs:complexType> </xs:element>

• For a complex element containing simple contents (text and attribute) only, the simpleContent element can be used. simpleContent will use an extension or a restriction, and the syntax is defined as follows:

<xs:element name="xxx"> <xs:complexType> <xs:simpleContent> <xs:extension base="basetype"> .... .... </xs:extension> </xs:simpleContent> </xs:complexType> </xs:element>

or

<xs:element name="xxx"> <xs:complexType> <xs:simpleContent> <xs:restriction base="basetype"> .... .... </xs:restriction> </xs:simpleContent> </xs:complexType> </xs:element>

• The syntax for defining a complex element containing both sub-elements and attributes will be a combination of the two discussed above. In addition, the mixed attribute must be set to “true.” As an example, the syntax is

<xs:element name="xxx"> <xs:complexType> <xs:sequence> <xs:element name="xxx-xxx" type="xs:yyy"/> <xs:element name="xxx-xxx-xxx" type="xs:yyy"/> </xs:sequence> <xs:attribute name="xxx-xxx-xxx" type="xs:yyy"/> </xs:complexType> </xs:element>

17.4.2.3.3 XML Schema Extensibility

One additional function provided by XML schema is the use of the <any> and <anyattribute> elements. The <any> element enables developers to extend the XML

355Working with XML (I)

document with elements not specified by the schema. To serve the same purpose, the <anyAttribute> element enables the developers to extend the XML document with attri- butes not specified by the schema as well.

17.4.2.3.4 XML Schema Indicator

Similar to DTD element cardinality, XML schema introduces Indicators to control how ele- ments may appear in XML documents. The syntax is as follows:

<xs:element name="xxx"> <xs:complexType> <xs:orderindicator> <xs:element name="xxx-xxx" type="xs:yyy"/> <xs:element name="xxx-xxx-xxx" type="xs:yyy"/> </xs:orderindicator> </xs:complexType> </xs:element>

The orderindicator can be one of the three: all, choice, and sequence. The all indicator means that the sub-elements can appear in any order and that each sub-element must occur only once. On the other hand, the choice indicator means that only one of the sub-elements can appear, while the sequence indicator implies that the sub-elements must appear in the order specified. Often, other types of indicators can be added into the schema to define the number of times that an element can appear. These occurindicators include minOccurs and maxOccurs. The default value for both indicators is “1.” The syn- tax that includes this type of indicator is as follows:

<xs:element name="xxx"> <xs:complexType> <xs:indicatortype> < xs:element name="xxx-xxx" type="xs:yyy" occurindicators

="zzz"/> <xs:element name="xxx-xxx-xxx" type="xs:yyy"/> </xs: indicatortype> </xs:complexType> </xs:element>

17.4.2.3.5 Other XML Schema Element and Data Type Restriction

Other than the elements listed above, there are several other types of elements that define constraints on other elements in the schema, as summarized in Table 17.13.

17.4.3 Referencing XML Schema for Validation

Let us use an example to explain how XML schema is referenced for validation.

Example 17.22

For the XML example from Table 17.3, create the XML schema. The result is shown in Table 17.14. Line 1: Different from DTD, an XML schema is based on XML syntax. So, it starts with

an XML declaration. Line 2: Declare the root element of the XML, in this example, it is a Book.

356 Developing Windows-Based and Web-Enabled Information Systems

Line 3 to line 11: The complex element Part is defined. The Part element is defined as a mixed element since it includes several Chapter sub-elements and one attribute named ID. The number of Chapters is unbounded. In addition, the attribute ID is required.

Line 12 to line 21: The complex element Chapter is defined. The Chapter element has an attribute named ID, which is required. Moreover, the text of the Chapter element uses

TABLE 17.13

XML Schema Element Restrictions

Restriction Description

fractionDigits Maximum number of decimal places for a value. Integers are 0 length Number of characters or, for lists, number of list choices maxExclusive Maximum up to, but not including, the number specified maxInclusive Maximum including the number specified maxLength Maximum number of characters or, for lists, number of list choices minExclusive Minimum down to, but not including, the number specified minInclusive Minimum including the number specified minLength Minimum number of characters or, for lists, number of list choices pattern Defines a pattern and sequence of acceptable characters totalDigits Number of non-decimal, positive, non-zero digits whiteSpace How line feeds, tabs, spaces, and carriage returns are treated when the document is parsed

TABLE 17.14

XML Schema for Example XML from Table 17.3 1. <?xml version="1.0" encoding="UTF-8"?> 2. <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"> 3. <xs:element name="Part"> 4. <xs:complexType mixed="true"> 5. <xs:sequence> 6. <xs:element name="Chapter" maxOccurs="unbounded"/> 7. </xs:sequence> 8. <xs:attribute name="ID" use="required"> 9. </xs:attribute> 10. </xs:complexType> 11. </xs:element> 12. <xs:element name="Chapter"> 13. <xs:complexType> 14. <xs:simpleContent> 15. <xs:extension base="xs:string"> 16. <xs:attribute name="ID" use="required"> 17. </xs:attribute> 18. </xs:extension> 19. </xs:simpleContent> 20. </xs:complexType> 21. </xs:element> 22. <xs:element name="Book"> 23. <xs:complexType> 24. <xs:sequence> 25. <xs:element name="Part" maxOccurs="unbounded"/> 26. </xs:sequence> 27. </xs:complexType> 28. </xs:element> 29. </xs:schema>

357Working with XML (I)

a string data type. Note that appropriate closing tags are required to ensure that XML schema itself is a well-formed XML.

Line 22 to line 28: The complex element Book is defined. The Book element includes a number of Part sub-elements.

Line 29: This line is the closing tag for the root element schema. To reference this XML schema in an XML document, we need to add the following at

the top of the XML document right after the XML document declaration (note that, here, we use a make-up URI http://www.CRCPress.com):

<B ook xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:noName spaceSchemaLocation="http://CRCPress.com/BookExampleXSD.xsd">

In this case, the namespace declaration referencing to http://www.w3.org/2001/ XMLSchema-instance resolves to an actual document at that location, which is a brief description of the way that the W3C schema should be referenced, and to a link to the actual schema that describes schema data types, elements, and other schema descrip- tions based on the current W3C Recommendation. The noNamespaceSchemaLocation value tells us that there is no pre-defined namespace for the schema, but the location of the schema is http://www.CRCPress.com/BookExampleXSD.xsd.

17.5 XML Editor Tools

There are a number of XML editors available online for creating and validating XML. In this chapter, all the examples are created using XMLSpy (http://www.altova.com/). There are also a number of applications from database vendors that are providing interfaces to XML. Some of these will be discussed in the next chapter.

17.6 Summary

This chapter covers the basics of XML. Starting with SGML, the markup language prin- ciples are discussed. The major differences between SGML, HTML, and XML are then explained. Noting that XML is developed to exchange data over the Internet, which, thus, requires a strict rule to define the syntax, the basic building blocks of XML, that is, ele- ment, attribute, and text are explained. Next, the concept of a well-formed XML is intro- duced. Lastly, DTD and XML schema are introduced and illustrated for XML validation.

REVIEW EXERCISES

1. SGML elements are used to define the structure of an SGML document. True or false?

2. An HTML will not be displayed correctly when <body> is closed with </Body>. True or false?

3. HTML rules emphasize the use of start and end tags while SGML does not. True or false?

4. List three characters that are not included in naming XML elements and give cor- responding examples.

358 Developing Windows-Based and Web-Enabled Information Systems

5. An XML attribute can be placed in an XML element’s end tag. True or false? 6. The following are ways to declare an empty element except: a. <name> </name> b. </name> c. </name> d. All of the above are valid ways to declare an empty element. e. None of the above are valid ways to declare an empty element. 7. Namespaces are optional when XML documents are shared with other XML doc-

uments of different element names. True or false? 8. XML DTDs are used for the following except: a. Defining blocks that provide a good structure for an XML document b. Formatting the document c. Defining which elements can have sub-elements d. Defining which element attributes are required e. All of the above are valid DTD purposes 9. One major advantage of XML schemas over DTDs is that XML schema supports

namespaces. True or false? 10. Differentiate a simple element as compared to a complex element and provide

examples. 11. Given the attribute declaration: <!ATTLIST Prod _ N CDATA #REQUIRED>, the

following XML document codes are valid except: a. <Product Prod _ N="121123"> b. <Product Prod _ N="12-1223"> c. <Product> Prod _ N="12-1223" </Product> d. <Product Prod _ N=" _ 121123 _ "> e. All of the above are valid XML codes for the attribute declaration. 12. Given the element declaration: <!ELEMENT Food _ Item(Expiration?)>, the

following XML document codes are valid except: a. <Food _ Item Expiration=""> </Food _ Item> b. <Food _ Item> </Food _ Item Expiration=""> c. <Food _ Item><Expiration> </Food _ Item> d. <Food _ Expiration> </Food _ Expiration> e. None of the above is a valid XML code for the element declaration. 13. Which element is used to declare a DTD inside an XML document? a. <!DOCTYPE> b. <!DTD> c. <!DTDTYPE> d. <!SYSTEM> e. <!XMLTYPE> 14. Enumerate two limitations of DTDs that gave rise to XML schemas.

359Working with XML (I)

15. Give an example of an FPI with the following properties: (1) A Non-Standards Institution: the Institute of Operations Research and Management Science (INFORMS) generated the XML, (2) INFORMS is responsible for DTD and (3) is used for the INFORMS Journal of Computing new article format with current ver- sion 1.1 with (4) English as its language.

16. Given the simple XML schema definition: <xs:element name="Sales _ Order" type="xs:string" default="Standard">, which of the following statements is true?

a. We are declaring an attribute named "Sales_Order" with default value "Standard".

b. We are declaring an element named "Sales_Order" with attribute Type composed of integers and default value "Standard".

c. We are declaring a string element named Sales_Order with attribute Type and default value "Standard".

d. We are declaring an attribute named Sales_Order with sub-attribute Type and default value "Standard".

e. The XML schema will not compile since this is a complex XML schema definition. 17. Valid XML schema order indicators are: a. Sequence b. Series c. All d. All of the above e. Both a and c 18. Given the XML element declaration: <xs:element name="interest _ rate"

type="xs:float" default="2.21"/>, which of the following is a valid XML implementation of the schema?

a. <interest _ rate>2.21<interest _ rate> b. <interest _ rate>3.67%</interest _ rate> c. <interest _ rate rate="3.67"/> d. <interest _ rate>3.67</interest _ rate> e. Both b and d are valid implementations. 19. The <any> element extends the XML document with attributes not specified by

the schema. True or false? 20. Given the following complex element declaration:

<xs:element name="sales_order"> <xs:complexType> <xs:attribute name="sales_order_n" type="integer"/> </xs:complexType> </xs:element>

Which of the following are valid declarations of XML schema? a. <sales _ order sales _ order _ n=1221232></sales _ order> b. <sales _ order sales _ order _ n=454><sales _ item>Text</sales _

item></ sales _ order>

360 Developing Windows-Based and Web-Enabled Information Systems

c. <sales _ order sales _ order _ n=112.94></sales _ order> d. All of the above e. Both a and b

PRACTICAL EXERCISES

1. The following is an XML document that describes a single Purchase Order with multiple line items of fine jewelry.

1. <?xml version="1.0" encoding="UTF-16"?> 2. <Purchase_Order> 3. <Billing_Info> 4. <Bill_Name>Irish Maravillas</Bill_Name> 5. <Bill_Street>1975 East University</Bill_Street> 6. <Bill_City>Tempe</Bill_City> 7. <Bill_State_Code>AZ</Bill_State_Code 8. <Bill_Zip_Code>85281</Bill_Zip_Code> 9. </Billing_Info> 10. <Ship_Info> 11. <Ship_Name>Jennifer Jareau</Ship_Name> 12. <Ship_Street>3600 Wilshire Blvd.</ship_Street> 13. <Ship_City>Los Angeles</Ship_City> 14. <Ship_State_Code>CA</Ship_State_Code> 15. <Ship_Zip_Code>90010</Ship_Zip_Code> 16. </Billing_Info> 17. <Purchase_Order> 18. </Purchase_Order> 19. <Items> 20. <Item itemNum=N28812> 21. <Product.Name>Diamond Necklace</Product.Name> 22. <Quantity=3</Quantity> 23. <_UnitPrice>99.85</_UnitPrice> 24. </Item> 25. <Item itemNum=R15632> 26. <Product.Name>Star Trek Ring</Product.Name> 27. <Quantity>1</Quantity> 28. <_UnitPrice>1985.90</_UnitPrice> 29. </Item> 30. </Items> 31. </Purchase_Order>

Identify which lines of code do not adhere to the guidelines set forth by WC3 in writing well-formed XML documents.

2. Consider the following e-mail message with ID="823324": 1. From: gilbert@hatanalytics.com 2. To: catherine@hatanalytics.com, nick@hatanalytics.com 3. cc: warrick@hatanalytics.com 4. Subject: Installation of ERP Server 5. Content: We will be performing maintenance in our servers and, as such, our

access to the cloud-based information systems will be intermittent.

Create an XML document for this e-mail message.

361Working with XML (I)

3. Consider the form. This form is a report of a comprehensive examination that is filled out after a student takes the comprehensive exam.

Arizona State University Report of Comprehensive Exam Results

Name: Aaron Mendez Student Number: 2002-36001 Program: MS Computer Science �esis Title: Addressing Issues on Large-Scale Pairwise Comparison Matrices Committee Member Results:

Department Chair: Michelle Lazaro College Dean: Haley Hotchner

ID Committee Position Name Result 1 Co-Chair Erin Strauss High Pass 2 Co-Chair Conrad Grissom Pass 3 Member William Li Pass

Create a well-formed XML document for this form. 4. A student’s transcript of records in XML is shown as follows:

1. <TORForm> 2. <Title Text_Align="Center"> 3. Transcript of Records 4. </Title> 5. <Stud_M Stud_N="1202703393"> 6. <First_M>Theodore</First_M> 7. <Middle_M>Evelyn</Middle_M> 8. <Last_M>Mosby</Last_M> 9. </Stud_M> 10. <Deg_Program>BS Systems Engineering</Deg_Program> 11. <Semester Sem_ID="2314" Year="2012" Term="Fall"> 12. <Course Course_N="74336"> 13. <Course_M>Information Systems</Course_M> 14. <Units>3</Units> 15. <Grade>A</Grade> 16. </Course> 17. <Course Course_N="73979"> 18. <Course_M>Stochastic Optimization</Course_M> 19. <Units>3</Units> 20. <Grade>B+</Grade> 21. </Course> 22. </Semester> 23. <Semester Sem_ID="2314" Year="2013" Term="Spring"> 24. <Course Course_N="78557"> 25. <Course_M>Research Methods</Course_M> 26. <Units>1</Units> 27. <Grade>C</Grade> 28. </Course> 29. </Semester> 30. </TORForm>

Create a document-like representation for the content of the transcript of records.

362 Developing Windows-Based and Web-Enabled Information Systems

5. The following is an example of a DTD for a newly hired employee form.

1. <!DOCTYPE EmpForm[ 2. <!ELEMENT Emp_M (First_M, Middle_M, Last_M)> 3. <!ATTLIST Emp_M Emp_N CDATA #REQUIRED> 4. <!ELEMENT First_M CDATA> 5. <!ELEMENT Last_M CDATA> 6. <!ELEMENT Address (Street_M, City_M, State_M)> 7. <!ELEMENT Acad_Background (Degree+)> 8. <!ATTLIST Degree Degree_N CDATA #REQUIRED> 9. <!ELEMENT Degree (University, Course, GradYear)> 10. <!ELEMENT University CDATA> 11. <!ELEMENT Course CDATA> 12. ]>

Generate a well-formed XML document that is validated by this DTD. The example should have at least two Academic Degrees.

6. Consider the following XML schema:

1. <?xml version="1.0" encoding="UTF-8"?> 2. <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"> 3. <xs:element name="Service" type="Service_Contract"/> 4. <xs:complexType name="Service_Contract"> 5. <xs:sequence> 6. < xs:element name="Service_Desc"

type="xs:string"/> 7. < xs:element name="Service_Cost"

type="CostType"/> 8. < xs:element name="Service_Pay_Terms"

type="xs:string" default="NET30"/> 9. < xs:element name="Service_Item"

type="Service_Items"/> 10. </xs:sequence> 11. < xs:attribute name="Service_N" type="xs:integer"

use="required"/> 12. <xs:attribute name="Service_Start_D" type="xs:date"/> 13. <xs:attribute name="Service_End_D" type="xs:date"/> 14. </xs:complexType> 15. <xs:complexType name="Service_Items"> 16. <xs:sequence> 17. < xs:element name="Service_M"

type="xs:string" maxOccurs="unbounded"/> 18. < xs:element name="Service_Cost"

type="xs:float"/> 19. </xs:sequence> 20. < xs:attribute name="Service_Item_N"

type="xs:integer" use="required"/> 21. </xs:complexType> 22. <xs:simpleType name="CostType"> 23. <xs:restriction base="xs:float"> 24. <xs:minInclusive value="100.00"/> 25. <xs:maxInclusive value="unbounded"/> 26. </xs:restriction>

363Working with XML (I)

27. </xs:simpleType> 28. </xs:schema>

Generate a well-formed XML document that is validated by this XML schema. The example should have at least three Service Items.

7. The following is an example of a dental patient invoice in DTD format:

1. <!DOCTYPE Patient_Invoices[ 2. <!ELEMENT Patient_Invoices(Patient_Invoice+)> 3. < !ELEMENT Patient_Invoice(Patient_Info, Insurance_Info?, Dental_

Procedure+)> 4. <!ATTLIST Patient_Invoice Invoice_N CDATA #REQUIRED> 5. <!ELEMENT Patient_Info (Patient_M, Phone, Address)> 6. <!ATTLIST Patient_Info Patient_SSN CDATA #REQUIRED > 7. <!ELEMENT Patient_M(First_M, Middle_M, Last_M)> 8. <!ELEMENT Phone EMPTY> 9. <!ATTLIST Phone Phone_N CDATA Phone_Type CDATA> 10. <!ELEMENT Insurance_Info (Company_M, Plan_N, Phone, Address)> 11. <!ELEMENT Dental_Procedure(Procedure_M, Cost)> 12. <!ATTLIST Dental Procedure Procedure_N CDATA #REQUIRED> 13. ]>

Construct an XML schema from this DTD document. 8. Generate an XML schema for the following business rules: a. A bank is composed of many accounts. b. There are two types of accounts. Savings and Checking accounts. Each account

must have a unique 10-digit account number and is linked to a single customer account. It is composed of the following elements: available balance, current balance, and account tier type. An account can have a maximum daily with- drawal limit and a minimum maintaining balance.

c. Each account has many transactions. Each transaction has the following ele- ments: posting date, transaction details, and amount (positive for deposits, negative for withdrawals). A transaction has a required attribute called “trans- action number.”

365

18 Working with XML (II)

Data independence, that is, the separation of content from its presentation, makes extensi- ble markup language (XML) an ideal framework for data exchange. Since XML is designed to describe the data presented in the document, an application that sends and receives XML would need interfaces to interpret the data before it is integrated into the applica- tions. This chapter first introduces two application programming interfaces (APIs) for parsing XML documents: document object model (DOM) and simple API for XML (SAX). The XML query language is then briefly reviewed. The interface between some commonly used database applications and XML is illustrated.

18.1 XML Parsers

An XML parser is a tool that manages the data content of an XML document. It identifies and converts XML elements into either nested nodes in a tree structure (e.g., DOM) or document events (e.g., SAX).

The W3C DOM is a standardized API used to define the logical structure of an XML doc- ument and the way the document is accessed and manipulated. Because an XML document is hierarchically structured, the DOM defines a tree structure to represent the entire XML document with the DOM root node matching the root element in the XML document. In addition, every XML element is matched to a DOM element node, an XML text is matched to a DOM text node, an XML attribute is matched to a DOM attribute node, and an XML comment is matched to a DOM comment node. These DOM nodes are properly located in the tree that correspond to their structural relationship defined in the XML document.

Let us revisit the Book example from Chapter 17 (shown in Table 18.1). The DOM tree is illustrated in Figure 18.1. As seen in the figure, the root element of the XML is <Book>, which has a sub-element named <Part>. Furthermore, the <Part> sub-element has a sub- element named <Chapter>. The <Part> and <Chapter> elements are connected to an attri- bute node and a text node, respectively. Working with the DOM tree, developers have the freedom to modify, delete, and add nodes to the tree whenever it is necessary. In this simple example, only the parent–child relationship is illustrated. There is another type of relationship called the “sibling,” which defines a relationship between two element nodes having the same parent. For example, assume that the <Book> element has another sub- element called <Title> at the same level as <Part>. A corresponding <Title> element node will be added at the same level as <Part> in the DOM tree.

Any DOM-based parser will first convert an XML into DOM objects before the data are retrieved and manipulated. Many applications (such as Web browsers) have a built-in parser to make the conversion. To understand how a DOM is used as an XML parser, we first summarize some important DOM interfaces (Table 18.2), the methods for Documents

366 Developing Windows-Based and Web-Enabled Information Systems

Text: �is is…

Element: <Chapter>

Attribute: “ID”

Text: �is is…

Element: <Part>

Attribute: “ID”

Root element: <Book>

Parent Child

Parent Child

FIGURE 18.1 DOM tree representation of the XML document.

TABLE 18.1

Example XML Document 1. <?xml version="1.0" encoding="UTF-8"?> 2. <Book> 3. <Part ID="P1">This is the first part. 4. <Chapter ID="C1">This is the first chapter. 5. </Chapter> 6. <Chapter ID="C2">This is the second chapter. 7. </Chapter> 8. </Part> 9. <Part ID="P2">This is the second part. 10. <Chapter ID="C3">This is the third chapter. 11. </Chapter> 12. <Chapter I D="C4">This is the fourth chapter. 13. </Chapter> 14. </Part> 15. </Book>

TABLE 18.2

Some DOM Interfaces

DOM Interfaces Description

Document Represents the XML documents’ top-level node, which provides access to all the document nodes

Node Represents an XML document node NodeList Represents a read-only list of Node objects Element Represents an element node; derives from Node Attr Represents an attribute node; derives from Node Text Represents a text node; derives from Node Comment Represents a comment node; derives from CharacterData ProcessingInstruction Represents a processing instruction node; derives from Node DATASection Represents a CDATA section; derives from Text

367Working with XML (II)

(Table 18.3), the methods for Nodes (Table 18.4), the methods for Elements (Table 18.5), and the several types of Nodes (Table 18.6).

The DOM-based parsers have been written in a variety of programming languages, including Java, Javascript, C#, and VB.NET. In this chapter, two illustrative examples are provided: the first one is written in Javascript, and the second one is written in VB.NET. These examples are used to illustrate the use of some of the interfaces, the methods of which are listed above.

TABLE 18.3

Some Document Methods

Document Methods Description

createElement Creates an element node createAttribute Creates an attribute node createTextNode Creates a comment node createProcessingInstruction Creates a processing instruction node createCDATASection Creates a CDATA section node getDocumentElement Returns the document’s root element appendChild Appends a child node getChildNodes Returns the child nodes createXmlDocument Parses an XML document write Outputs the XML document

TABLE 18.4

Some Node Methods

Node Methods Description

appendChild Appends a child node cloneNode Duplicates the node getAttributes Returns the node’s attributes getChildNodes Returns the node’s child nodes getNodeName Returns the node’s name getNodeType Returns the node’s type (e.g., element, attribute, text, etc.) getNodeValue Returns the node’s value getParentNode Returns the node’s parent hasChildNodes Returns true if the node has child nodes removeChild Removes a child node from the node replaceChild Replaces a child node with another node setNodeValue Sets the node’s value insertBefore Appends a child node in front of a child node

TABLE 18.5

Some Element Methods

Element Methods Description

getAttribute Returns an attribute’s value getTagName removeAttribute

Returns an element’s name Removes an element’s attribute

setAttribute Sets an attribute’s value

368 Developing Windows-Based and Web-Enabled Information Systems

Example 18.1

A DOM-based parser written in Javascript (Table 18.7). In this JavaScript example,

Line 1: Import (i.e., specifies the location of) the classes needed by the program. In this example, some classes from the package org.apache.xerces.parsers. DOMParser are imported.

Line 5: Declare and create an instance of a DOM parser object.

TABLE 18.7

Example DOM (JavaScript) 1. import org.apache.xerces.parsers.DOMParser; 2. public class DOMWriter { 3. public static void main( String[] args) { 4. try { 5. DOMParser parser=new DOMParser(); 6. parser.parse("Chapter 17.xml"); 7. Document document=parser.getDocument(); 8. processElement(document.

getDocumentElement()); 9. } catch (SAXException e) { 10. e.printStacktrace(); 11. } catch (IOException e) { 12. e.printStacktrace(); 13. } 14. } 15. private static void processElement( Element

element) { 16. NodeList nodes=element.getChildNodes(); 17. System.out.println("<"+element.

getNodeName()+">"); 18. for (int i=0; i < nodes.getLength(); i++) { 19. Node node=nodes.item(i); 20. if (node.getNodeType()==Node.ELEMENT_NODE)

{ 21. processElement((Element)node); 22. } else { 23. System.out.println( node.getNodeValue()); 24. } 25. } 26. System.out.println("</"+element.

getNodeName()+">"); 27. } 28. }

TABLE 18.6

Some Node Types

Node Types Description

Node.ELEMENT_NODE Represents an element node Node.ATTRIBUTE_NODE Represents an attribute node Node.TEXT_NODE Represents a text node Node.COMMENT_NODE Represents a comment node Node.PROCESSING_INSTRUCTION_NODE Represents a processing instruction node

369Working with XML (II)

Line 6: The DOM parser is assigned to read the “Chapter 17.xml” document. Line 7: Declare and create an instance of a Document object, called document.

This object will start to parse the XML document via calling the getDocu- ment method.

Line 8: First, the document.getDocumentElement method will get the root ele- ment of the XML document, in this example, the <Book> element. Next, a func- tion processElement is called.

Line 16: For the root element, <Book>, its child elements are obtained by calling the method element.getChildNodes and are then assigned to an instance of the NodeList.

Line 17: For the root element, its name (in this example, Book) is obtained by call- ing the method element.getNodeName. The output is the root element name with “<” added at the left and “>” added at the right.

Line 18: The number of child elements for the root element is obtained by calling nodes.getLength.

Line 19: This is the start of a FOR loop for each child element (the index is used in nodes.item(i)).

Line 20: If the node is an element node, a recursive call of processElement is triggered. This will parse through the tree from the root element to the leaf elements.

Line 21: If the node is not an element node, but instead, it is an attribute node or a text node, the value of the node is obtained by calling the node.getNodeValue function.

Line 26: Finally, as the whole XML document is parsed through, it is closed with a closing tag, which is presented by the value of the root element adding “</” at the left and “>” at the right.

The output of this DOM is an XML document shown in Table 18.1 but without line 1, which is an XML declaration. This example only illustrates parsing through the XML document that mainly uses the getNodeValue function; other methods can be called to make changes (e.g., delete the Node) as the original XML document is parsed through.

Example 18.2

A DOM-based parser written in VB.NET. The Microsoft .NET platform provides its own classes to support a DOM parser and,

thus, defines Microsoft XML core services. Note that the functionalities of the basic interfaces and methods listed in Tables 18.2 to 18.6 are the same, but the names may vary. In this example, we illustrate a Windows application. Please refer to Chapters 11 and 12 for details on Windows forms, controls, and VB.NET coding. Here, we will briefly explain the interfaces and methods related to the parser (Table 18.8).

In this VB.NET example,

Lines 1 to 2: Import (i.e., specify the location of) the classes needed by the pro- gram. For example, some classes from package System.xml and System.IO are imported.

Lines 3 to 4: This is a VB.NET Windows application with a Button control on the Windows form.

Line 5: Declare and create an instance of a DOM document. Line 6: Declare xmlnode as XmlNodeList object, a list of node objects (functions as

NodeList from Table 18.2). Line 10: Load the XML document. Line 11: The root node of the XML is obtained and is assigned to the xmlnode

instance.

370 Developing Windows-Based and Web-Enabled Information Systems

Lines 12 to 16: The start of the FOR loop. Here, the element name is specified, that is, <Part>. As a result, for each <Part> element within the XML, its value and the information from its sub-elements are retrieved, and a message box is used to pop up the outputs.

In summary, both DOM parser examples indicate that the DOM is centered on a tree structure. DOM-based parsers will always scan through the entire XML document, no matter how much it is actually needed by the developer. By contrast, SAX is an event- based API that is known to be faster than DOM as it retrieves only related data from the document.

However, DOM will store the entire document’s data in memory, and data can be quickly accessed since the SAX parser will pass the data to the application as it is found. There have always been discussions on DOM versus SAX. One recommendation is that SAX performs well for read-only access, while DOM works well if the developers need to alter, delete, and add data into the XML tree. Since SAX is not a W3C standard, we will not discuss this topic in detail here.

18.2 XML Query

XML parsers (e.g., DOM) are able to access and manipulate XML data using program- ming languages. XQuery is another W3C standard, developed to query XML documents. As a point of reference, XQuery is to XML as SQL is to databases. It is designed to replace proprietary middleware languages as well as Web application development languages. XQuery is built on XPath expressions. Indeed, XQuery 1.0 and XPath 2.0 share the same data model and support the same functions and operators. In this chapter, we list some

TABLE 18.8

Example DOM (VB.NET)

1. Imports System.Xml 2. Imports System.IO 3. Public Class Form1 4. Private Sub Button1_Click(ByVal sender As System.Object, ByVal e As System.

EventArgs) Handles Button1.Click 5. Dim xmldoc As New XmlDataDocument() 6. Dim xmlnode As XmlNodeList 7. Dim i As Integer 8. Dim str As String 9. Dim fs As New FileStream("book.xml", FileMode.Open, FileAccess.Read) 10. xmldoc.Load(fs) 11. xmlnode=xmldoc.GetElementsByTagName("Part") 12. For i=0 To xmlnode.Count − 1 13. xmlnode(i).ChildNodes.Item(0).InnerText.Trim() 14. str=xmlnode(i).ChildNodes.Item(0).InnerText.Trim() 15. MsgBox(str) 16. Next 17. End Sub 18. End Class

371Working with XML (II)

important XPath operators (Table 18.9). For a comprehensive list of XPath functions, please visit the W3C website: http://www.w3.org/2005/xpath-functions (Table 18.10).

Please note that XQuery supports the use of namespaces and is case sensitive. In XQuery, all elements, attributes, and variables must be valid XML names. An XQuery variable is defined with a $ followed by a name (e.g., $x), and an XQuery comment is delimited by “(:” and “:)” (e.g., (: comments :)). The basic expression in XQuery is FLWOR where F refers to for, L refers to let, W refers to where, O refers to order by, and R refers to return.

Using an XML example (Table 18.11), some basics of XQuery are illustrated. This XML con- tains the information about best-selling single-volume books. For each book, the title, authors, language, first publication year, sales volume, price, and the types of the book are presented.

Example 18.3

An XQuery example on the XML from Table 18.11 is as follows:

for $x in doc ("books.xml")/Best-selling-single-volume-books//Book where $x/Price >10.99 order by $x/title let $x/Price: =10.99*0.80 return $x/Title, $x/Price

TABLE 18.9

XPath (XQuery) Location Operators

Operator Description

. The current node

.. The parent node / The root element // All descendants @ Attribute identifier * All child nodes

TABLE 18.10

List of Operators Used in XPath (XQuery)

Operator Description

| Computes two node sets, for example: //Book | //Part will return a node set with all Book and Part elements

+ Addition − Subtraction * Multiplication div Division = Equal != Not equal < Less than <= Less than or equal to > Greater than >= Greater than or equal to or Or and And mod Modulus (division remainder)

372 Developing Windows-Based and Web-Enabled Information Systems

TABLE 18.11

Example XML Document on Best-Selling Single-Volume Books (books.xml) <?xml version="1.0" encoding="UTF-8"?> <Best-selling-single-volume-books>

<Book Type="Kindle"> <Title>A Tale of Two Cities</Title> <Authors>

<Author> <Lastname>Dickens</Lastname> <Firstname>Charles</Firstname>

</Author> </Authors> <Language>English</Language> <FirstPublished>1859</FirstPublished> <Sales>Over 200 Million</Sales> <Price>3.50</Price>

</Book> <Book Type="Kindle">

<Title>The Lord of Rings</Title> <Authors>

<Author> <Lastname>Tolkien</Lastname> <Firstname>J.R.R.</Firstname>

</Author> </Authors> <Language>English</Language> <FirstPublished>1954–1955</FirstPublished> <Sales>150 million</Sales> <Price>3.10</Price>

</Book> <Book Type="Kindle">

<Title>The Little Prince</Title> <Authors>

<Author> <Lastname>Saint-Exupéry</Lastname> <Firstname>Antoine</Firstname>

</Author> </Authors> <Language>French</Language> <FirstPublished>1937</FirstPublished> <Sales>Over 100 million</Sales> <Price>8.74</Price>

</Book> <Book Type="eBook">

<Title>Dream of the Red Chamber</Title> <Authors>

<Author> <Lastname>Cao</Lastname> <Firstname>Xueqing</Firstname>

</Author> </Authors> <Language>Chinese</Language> <FirstPublished>1754–1791</FirstPublished> <Sales>100 million</Sales> <Price>18.59</Price>

</Book> </Best-selling-single-volume-books>

373Working with XML (II)

The for clause selects all Book elements within the Books element and assigns it to a variable x ($x). The where clause selects only the Book elements that has a Price sub- element, and the value of the Price sub-element is greater than 10.99. Additionally, the order by clause defines the sort order. By default, the ascending order is used. The let clause sets the discounted prices (80%) for the book, while the return clause returns the titles of the books satisfying the condition and the new prices. The result of the XQuery example is:

Dream of the Red Chamber, 14.82

The purpose of querying an XML document is to retrieve the related data within the document as well as to transmit it to other applications. Transforming the data into other formats (e.g., HTML, other XML formats) is often desired, which leads to another W3C standard, extensible stylesheet language (XSL).

18.3 XML Transformation

Cascading style sheets (CSS) is a commonly used tool for data transformation and pre- sentation. However, similar to DTD, CSS does not support namespaces and suffers from extensibility, which, in turn, limits its application to XML. On the other hand, these limitations are addressed by XSL, which is a W3C standard. Together with XPath, XSL Transformation (XSLT) is able to create different presentations of XML content. This is done by the use of an XSL stylesheet, which explicitly defines how to transform the values of a queried XML element, attribute, and text into a desired output format. Note that an XSL stylesheet in itself is a well-formed XML and that it supports namespaces. Therefore, syntax rules on XML elements, attributes, and text discussed in Chapter 17 will also apply to the stylesheet. The vocabularies of XSLT though are mostly made up of XML elements. A partial list of XSLT elements are summarized in Table 18.12.

Example 18.4

Let us use an example on the best-selling book XML (Table 18.11) to illustrate the appli- cation of XSLT. In this example, the transformed output is in HTML format (Figure 18.2). To connect the XSLT file to an XML document, the <?xml-stylesheet?> processing function is used. To do so, the following element is added at the beginning of the XML document from Table 18.11:

<?xml-stylesheet type="text/xsl" href="BestSellingBooks.xsl"?>

Here, it is assumed that the XSL file is located in the same folder as the XML document. Considering the XSLT example (BestSellingBooks.xsl) (Table 18.13):

Line 1: Mandatory XML declaration. Line 2: Declare the xsl:stylesheet element. In this document, there are four

namespaces declared specifically: an XSL transform namespace with prefix xsl, an XSL format namespace with prefix fo, an XMLschema namespace with prefix xs, and an XPath functions namespace with prefix fn.

Line 3: Declare that the output method is hypertext markup language (HTML), that is, the source XML document is to be transformed to an HTML document.

374 Developing Windows-Based and Web-Enabled Information Systems

As a result, the necessary HTML tags will be added to the formatting, includ- ing <html>, <body>, <h2>, <tr>, <th>, and <td>.

Line 4: Declare the <xsl:template> element. The stylesheet starts to search the XML document using the template element. The match attribute specifies the matching pattern. In this example, the matching pattern is “/” (root ele- ment, see Table 18.9: XPath location operator); the stylesheet will start from the root element.

Lines 5 to 15: Format the information for presentation using HTML tags. Lines 16 to 34: Instruct the XSL processor to scan through the document. For

each <Book> element under <Best-selling-single-volume-books>, the value of the <Title> element, the value of the attribute <@Type> (see Table 18.9), the value of the element <Authors/Author/Lastname>, the value of the <Sales> element, and the value of the <Price> element are retrieved and displayed in a table format.

FIGURE 18.2 Display of XML after applying XSLT.

TABLE 18.12

XSLT Elements

Elements Description

xsl:stylesheet Defines a root element of a stylesheet. This element can be used interchangeably with transform, but most stylesheets use stylesheet as a de facto standard

xsl:template Applies rules in a match or select action xsl:value-of Retrieves a string value of a node and writes it to the output node tree xsl:for-each Iteratively processes each node in a node set defined by an XPath expression xsl:sort Defines a sort key used by apply-templates to a node set and by for-each to

specify the order of iterative processing of a node set xsl:if Conditionally applies a template if the test attribute expression evaluates to true xsl:choose Makes a choice based on multiple options; used with elements when and

otherwise xsl:apply-templates Applies templates to all children of the current node or a specified node set

using the optional select attribute. Parameters can be passed using the with-param element

xsl:element Adds an element to the output node tree. Names, namespaces, and attributes can be added with the names, namespaces, and use-attribute-sets attributes

xsl:attribute Adds an attribute to the output node tree; must be a child of an element xsl:attribute-set Adds a list of attributes to the output node tree; must be a child of an element xsl:text Adds text to the output node tree xsl:comment Adds a comment to the output node tree processing-instruction Adds a processing instruction to the output node tree

375Working with XML (II)

This example demonstrates the use of XSLT and XPath to transform an XML document to an HTML document. Indeed, an XML document can be transformed to a number of other document formats as well as generate other XML documents. For example, the devel- opers can search for the Books having an “eBook” Type and then construct a new XML document that contains only eBook information. Some other common output file formats are text and portable document format (PDF). Unlike HTML, XML, and text format, con- verting an XML document to a PDF file will require a formatting objects processor (FOP). Currently, the most popular FOP engine is the Apache FOP processor.

TABLE 18.13

XSLT Example (BestSellingBooks.xsl) 1. <?xml version="1.0" encoding="UTF-8"?> 2. <xsl:stylesheet version="2.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"

xmlns:fo="http://www.w3.org/1999/XSL/Format" xmlns:xs="http://www.w3.org/2001/ XMLSchema" xmlns:fn="http://www.w3.org/2005/xpath-functions">

3. <xsl:output method="html"/> 4. <xsl:template match="/"> 5. <html> 6. <body> 7. <h2>Best Selling Book (Single Volume)</h2> 8. <table border="1"> 9. <tr> 10. <th>Title</th> 11. <th>Type</th> 12. <th>Author</th> 13. <th>Sales</th> 14. <th>Prices</th> 15. </tr> 16. <xsl:for-each select="Best-selling-single-volume-books/Book"> 17. <tr> 18. <td> 19. <xsl:value-of select="Title"/> 20. </td> 21. <td> 22. <xsl:value-of select="@Type"/> 23. </td> 24. <td> 25. <xsl:value-of select="Authors/Author/Lastname"/> 26. </td> 27. <td> 28. <xsl:value-of select="Sales"/> 29. </td> 30. <td> 31. <xsl:value-of select="Price"/> 32. </td> 33. </tr> 34. </xsl:for-each> 35. </table> 36. </body> 37. </html> 38. </xsl:template> 39. </xsl:stylesheet>

376 Developing Windows-Based and Web-Enabled Information Systems

18.4 Interfacing XML with Database Applications

As W3C continues to improve XML-related technologies and standards, most applications, specifically database systems, are providing corresponding interfaces with XML. Database systems are well known for consistent data storage, retrieval, and manipulation. At the same time, XML has been accepted as a standard for Web-based information systems and a medium for information exchange between organizations. Since a database is essentially the backbone for any information system, the integration between XML and a relational database system (RDBS) is a must.

18.4.1 Relational Database versus XML

To understand the relationship between XML and RDBS, we will first explain the dif- ferences between XML and RDBS in terms of (1) data modeling, (2) data schema, and (3) instance as follows.

The concept of data modeling refers to the definition of the structure of data. Since an RDBS is designed to store large volumes of data and provide efficient access with ensured integrity, an RDBS requires explicit definitions on the data types. In contrast, XML is designed to exchange data embedded in a structured document.

By definition, a data schema is the utilization of data modeling concepts to structure certain domain data. For example, an RDBS schema can describe the database structure as well as its relational table structure. It has to be defined before any data can be stored in the system. On the other hand, XML is more like a semi-structured document as it allows implicit, partial, or even an incomplete schema. This is mainly because an XML document may evolve over time. When the structure schema is well defined, developers usually use a DTD or an XML schema to specify the schema. XML documents are validated through a DTD or an XML schema, as discussed in Chapter 17. Additionally, although both DTD and XML schema support multiple data types, an XML schema is relatively limited and is not as strong as an RDBS. Furthermore, a DTD or an XML schema is not mandatory to be applied on an XML document.

At the instance level, XML is self-describing, which means that parts of the schema defini- tion in terms of tags are replicated within each XML document, no matter if the schema is defined explicitly or not. For the example in Table 18.11, the Author tags are repeated for each Book element. This is in contrast to an RDBS where there is only one schema for the whole database. It is the way that the schema, along with the data, is stored that provides XML the flexibility in integrating heterogeneous data sources and allowing changes to its structure. However, the replicated schema information implies space cost for storage, time cost for retrieval, and the danger of inconsistencies in case of schema updates (Deutsch et al., 1999).

Example 18.5

Use Healthcaresystem as an example to compare relational database and XML. Consider the relational database for Healthcaresystem, which has three tables: Drugs,

Stores, and Sales. The Drugs table contains the information about the ID, name, price, and general description of the drug. The Stores table has the information on the Store_ ID, the name of the store, and the contact phone number. The Sales table summarizes the quantity sold by the store for each specific drug.

From an RDBS perspective, first of all, the data types of each should be explicitly defined. For example, “Price” is a number while “Description” is a string. Next, the relational data- base schema needs to be defined. That is, there are four columns in the “Drugs” table,

377Working with XML (II)

specifically, Drug_ID, Name, Price, and Description. Additionally, there are three col- umns in the “Stores” table, which includes Store_ID, Name, and Telephone. Lastly, there are three columns in the “Sales” table: Transaction_ID, Drug_ID, Quantity, and Store_ID. In addition, constraints should be specified when defining the schema, that is, the pri- mary key in the “Drugs” (Drug_ID) table is a foreign key in the “Sales” (Drug_ID) table; the primary key in the “Stores” (Store_ID) table is a foreign key in “Sales” (Store_ID); and the primary key in the “Sales” table is a composite key, which is composed of Drug_ID and Store_ID. Let the tables be populated with some records as follows:

Drugs

Drug_ID Name Price Description

T_001 Tylenol $17.99 100 325mg Tablets Bottle T_003 Tylenol $8.99 T_002 Tri_Vitamin $14.99 1500-400-35 Solution 50mL Bottle C_001 Claritin $35.99 10 Tablets

Stores

Store_ID Name Phone_Number

S_001 Walgreen 1-800-746-7287 S_002 CVS 1-800-925-4733 S_003 Fry’s Pharmacy 1-866-221-4141

Sales

Transaction_ID Drug_ID Quantity Store_ID

ST_001 T_001 25 S_001 ST_002 T_001 300 S_002 ST_003 C_002 150 S_001

If this information is presented in an XML document, it is

<?xml version="1.0" encoding="UTF-8"?> <Healthcaresystem> <Drugs> <Drug Drug_ID="T_001"> <Name>Tylenol</Name> <Price>$17.99</Price> <Description>100 325mg Tablets Bottle</Description> </Drug>

<Drug Drug_ID="T_003"> <Name>Tylenol</Name> <Price>$8.99</Price> <Description></Description> </Drug>

<Drug Drug_ID="T_002"> <Name>Tri_Vitamin</Name> <Price>$14.99</Price> <Description>1500-400-35 Solution 50ml Bottle</Description> </Drug> <Drug Drug_ID="C_001"> <Name>Claritin</Name> <Price>$35.99</Price> <Description>10 Tablets</Description>

378 Developing Windows-Based and Web-Enabled Information Systems

</Drug> </Drugs>

<Stores> <Store Store_ID="S_001"> <Name>Walgreen</Name> <Telephone>1-800-746-7287</Telephone> </Store>

<Store Store_ID="S_002"> <Name>CVS</Name> <Telephone>1-800-925-4733</Telephone> </Store>

<Store Store_ID="S_003"> <Name>Fry's Pharmacy</Name> <Telephone>1-866-221-4141</Telephone> </Store> </Stores>

<Sales> <Sale Drug_ID="T_001" Store_ID="S_001"> <Quantity>25</Quantity> </Sale>

<Sale Drug_ID="T_001" Store_ID="S_002"> <Quantity>300</Quantity> </Sale>

<Sale Drug_ID="C_002" Store_ID="S_001"> <Quantity>501</Quantity> </Sale> </Sales> </Healthcaresystem>

An associated DTD (Table 18.14) or an XML schema (Table 18.15) for this XML docu- ment are as follows.

TABLE 18.14

DTD Document <!ELEMENT Healthcaresystem (Drugs, Stores, Sales)*> <!ELEMENT Drugs (Drug)*> <!ELEMENT Drug (Name, Price, Description)*> <!ATTLIST Drug Drug_ID ID #REQUIRED> <!ELEMENT Name (#PCDATA)> <!ELEMENT Price (#PCDATA)> <!ELEMENT Description (#PCDATA)> <!ELEMENT Stores (Store)*> <!ELEMENT Store (Name | Telephone)*> <!ATTLIST Store Store_ID ID #REQUIRED> <!ELEMENT Name (#PCDATA)> <!ELEMENT Telephone (#PCDATA)> <!ELEMENT Sales (Sale)*> <!ELEMENT Sale (Quantity)*> <!ATTLIST Sale Transaction_ID ID #REQUIRED DrugID IDREF #REQUIRED Store_ID IDREF #REQUIRED> <!ELEMENT Quantity (#PCDATA)>

379Working with XML (II)

TABLE 18.15

XML Schema Document <?xml version="1.0" encoding="UTF-8"?> <xsd:schema elementFormDefault="qualified" xmlns:xsd="http://www.w3.org/2001/XMLSchema"> <xsd:element name="Healthcaresystem"> <xsd:complexType> <xsd:sequence minOccurs="0" maxOccurs="unbounded"> <xsd:element ref="Drugs"/> <xsd:element ref="Stores"/> <xsd:element ref="Sales"/> </xsd:sequence> </xsd:complexType> </xsd:element> <xsd:element name="Drugs"> <xsd:complexType> <xsd:sequence minOccurs="0" maxOccurs="unbounded"> <xsd:element ref="Drug"/> </xsd:sequence> </xsd:complexType> </xsd:element> <xsd:element name="Drug"> <xsd:complexType mixed="true"> <xsd:sequence minOccurs="0" maxOccurs="unbounded"> <xsd:element ref="Name"/> <xsd:element ref="Price"/> <xsd:element ref="Description"/> </xsd:sequence> <xsd:attribute name="Drug_ID" type="xsd:string" use="required"/> </xsd:complexType> </xsd:element> <xsd:element name="Name" type="xsd:string"/> <xsd:element name="Price" type="xsd:string"/> <xsd:element name="Description" type="xsd:string"/> <xsd:element name="Stores"> <xsd:complexType> <xsd:sequence minOccurs="0" maxOccurs="unbounded"> <xsd:element ref="Store"/> </xsd:sequence> </xsd:complexType> </xsd:element> <xsd:element name="Store"> <xsd:complexType mixed="true"> <xsd:sequence minOccurs="0" maxOccurs="unbounded"> <xsd:element ref="Name"/> <xsd:element ref="Telephone"/> </xsd: sequence> <xsd:attribute name="Store_ID" type="xsd:string" use="required"/> </xsd:complexType> </xsd:element> <xsd:element name="Name" type="xsd:string"/> <xsd:element name="Telephone" type="xsd:string"/> <xsd:element name="Sales"> <xsd:complexType> <xsd:sequence minOccurs="0" maxOccurs="unbounded"> <xsd:element ref="Sale"/> </xsd:sequence> </xsd:complexType> </xsd:element>

(continued)

380 Developing Windows-Based and Web-Enabled Information Systems

From the example, it is concluded that both RDBS and XML are able to store data, but in different ways. Therefore, almost all relational database applications provide the interfaces with XML for importing and exporting.

18.4.2 XML and Microsoft Access

Since Office XP, Microsoft Access has enhanced its capability to export and import data by fully embracing the XML standard and embedding XML support in its application and programmability features. In this section, we will learn about how XML data can be imported and exported by using the Access graphical user interface (GUI). We will learn how XML schema can be used to ensure data integrity for imports and exports. We will also learn how to leverage XSL to convert semi-structured XML data properly for import- ing into a format that can be directly used by Access.

18.4.2.1 Exporting Access Database Tables to XML

Let us first create three tables in an Access database (ch18_AccesstoXML.mdb)—Drugs, Stores, and Sales—following Example 18.5. The relationships between the tables are shown in Figure 18.3.

The simplest way to export XML from this Access database is to right-click on a table. An export menu appears on the pop-up menu (see Figure 18.4).

FIGURE 18.3 Access database example: three tables—Drugs, Sales, Stores and the relationship among the three.

TABLE 18.15 (Continued)

XML Schema Document <xsd:element name="Sale"> <xsd:complexType mixed="true"> <xsd:sequence minOccurs="0" maxOccurs="unbounded"> <xsd:element ref="Quantity"/> </xsd:sequence> <xsd:attribute name="Drug_ID" type="xsd:string" use="required"/> <xsd:attribute name="Store_ID" type="xsd:string" use="required"/> </xsd:complexType> </xsd:element> <xsd:element name="Quantity" type="xsd:string"/> </xsd:schema>

381Working with XML (II)

Access will then prompt for a location where it can place the XML file containing the data from the table. In addition to the tables, we can also export the data from a query, as well as the datasheet, a form, or a report into an XML file.

Before the objects are to be exported, a prompt will be shown to ask for the configura- tions (see Figure 18.5). The resulting XML file contents will depend on the options that we select.

Let us look at the generated XML file. This file contains the data contents of the object exported from Access (Figure 18.6). Table 18.16 shows the final XML from the export. We can see how the fields from the Access table are related to the elements in the XML document.

In addition, Access has exported an XSD schema to preserve the definitions that describe how the fields relate one to the other. As seen, the XSD contains information such as whether there is a key constraint on the database table, whether the key enforces unique- ness, what the field names are, as well as their corresponding data types and lengths (see Table 18.17).

18.4.2.2 Importing XML to an Access Database

As is the case with XML exports, there is the ability to import an XML document into Access through its GUI. During the importing process, the most important step is to ensure that the structure of the data in the XML file conforms to the structure of the database table. If

FIGURE 18.4 Export XML via user interface in Access.

FIGURE 18.5 Export configuration settings.

382 Developing Windows-Based and Web-Enabled Information Systems

TABLE 18.16

XML Exported from Access Database (ch18_04.xml) <?xml version="1.0" encoding="UTF-8"?> <dataroot xmlns:od="Zurn:schemas-microsoft-com:officedata" xmlns:xsi="http://www. w3.org/2001/XMLSchema-instance" xsi:noNamespaceSchemaLocation="Drugs.xsd" generated="2011-07-03T13:15:14"> <Drugs>

<Drug_ID>T_001</Drug_ID> <Name>Tylenol</Name> <Price>17.99</Price> <Description>100 325mg Tablets Bottle</Description>

<Drug_ID>T_003</Drug_ID> <Name>Tylenol</Name> <Price>8.99</Price> <Description></Description>

<Drug_ID>T_002</Drug_ID> <Name>Tri_Vitamin</Name> <Price>14.99</Price> <Description>1500-400-35 Solution 50ml Bottle</Description>

<Drug_ID>C_001</Drug_ID> <Name>Claritin</Name> <Price>35.99</Price> <Description>10 Tablets</Description>

</Drugs> </dataroot>

FIGURE 18.6 Design of the table exported from the Access database.

383Working with XML (II)

Access encounters a problem when inserting data into the database table, it will place the errors in a new table called “ImportErrors.” When importing a new XML document and the table does not exist, Access will create it; otherwise, new errors will be appended to the Access table.

To begin an import attempt (an example import file is shown in Table 18.18), the user can choose the File → Get External Data → Import menu in Access, or choose a pop-up menu when right-clicking in the pane that lists tables, queries, or other objects in Access. Either way, a dialog box is popped up to navigate to the XML file that contains the data to be imported. After selecting the file and clicking the button to import the file, another dialog box prompts to configure the import options (Figure 18.7).

The import behavior can be different according to the setting that the users choose. For example:

• The import option is “Structure Only:” This imports the table structure only. If the table does not exist, it will be created. If the table already exists, a new one will be added with a number appended to differentiate it from the pre-existing table of the same name.

• The import option is “Structure and Data:” This imports the table and the data into  the database. If the table exists, a new one will be added with a number appended. The difference between this and the previous option is that this one tells Access to import the data as well.

TABLE 18.17

XML Schema Exported from Access Database <?xml version="1.0" encoding="UTF-8"?> <xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema" xmlns:od="urn:schemas-microsoft-com:officedata"> <xsd:element name="dataroot"></xsd:element> <od:index index-name="PrimaryKey" index-key="Drug_ID" primary="yes" unique="yes" clustered="no" order="asc"/> <xsd:element name="Drugs"> <xsd:complexType> <xsd:sequence> <xsd:element name="Drug_ID" minOccurs="0" od:jetType="text" od:sqlSType="nvarchar"> </xsd:element> <xsd:element name="Name" minOccurs="0" od:jetType="text" od:sqlSType="nvarchar"> </xsd:element> <xsd:element name="Price" minOccurs="0" od:jetType="decimal" od:sqlSType="decimal"> <xsd:simpleType> <xsd:restriction base="xsd:decimal"> <xsd:totalDigits value="18"/> <xsd:fractionDigits value="2"/> </xsd:restriction> </xsd:simpleType> </xsd:element> <xsd:element name="Description" minOccurs="0" od:jetType="text" od:sqlSType="nvarchar"> </xsd:element> </xsd:sequence> </xsd:complexType> </xsd:element> </xsd:schema>

384 Developing Windows-Based and Web-Enabled Information Systems