Generative Programming

Methods, Tools, and Applications

Krzysztof Czarnecki, Ulrich Eisenecker

832 pages, parution le 22/06/2000

Ajouter à une liste

Indisponible

Résumé

Generative Programming (GP) offers great promise to application developers. It makes the idea of moving from one-of-a-kind software systems to the semi-automated manufacture of wide varieties of software quite real. In short, GP is about recognizing the benefits of automation in software development. Generative Programming covers methods and tools that will help you design and implement the right components for a system family and automate component assembly. The methods presented here are applicable for all commercial development--from "programming in the small," at the level of classes and procedures--to "programming in the large," or developing families of large systems.

Generative Programming is your complete guide and reference to this emerging discipline. It provides in-depth treatment of critical technologies and topics including:

Domain Engineering
Feature Modeling
Generic Programming
Aspect-Oriented Programming
Template Metaprogramming in C++
Generators
Microsoft's Intentional Programming

Using this book you will learn how these techniques fit together and, more importantly, how to apply them in practice. The text contains three comprehensive case studies in three different domains: programming domain (container data structures), business domain (banking), and scientific computing (matrix computations).

Table of contents

Chapter 1 What Is This Book About?: 1.1 From Handcrafting to Automated Assembly Lines; 1.2 Generative Programming; 1.3 Benefits and Applicability
Part I Analysis and Design Methods and Techniques
Chapter 2 Domain Engineering: 2.1 Why Is This Chapter Worth Reading?; 2.2 What Is Domain Engineering?; 2.3 Domain Analysis; 2.4 Domain Design and Domain Implementation; 2.5 Application Engineering; 2.6 Product-Line Practices; 2.7 Key Domain Engineering Concepts; 2.7.1 Domain; 2.7.2 Domain Scope and Scoping; 2.7.3 Relationships between Domains; 2.7.4 Features and Feature Models; 2.7.5 Method Tailoring and Specialization; 2.8 Survey of Domain Analysis and Domain Engineering Methods; 2.8.1 Feature-Oriented Domain Analysis (FODA); 2.8.1.1 FODA Process; 2.8.2 Organization Domain Modeling (ODM); 2.8.2.1 The ODM Process; 2.8.3 Draco; 2.8.4 Capture; 2.8.5 Domain Analysis and Reuse Environment (DARE); 2.8.6 Domain-Specific Software Architecture (DSSA) Approach; 2.8.7 Algebraic Approach; 2.8.8 Other Approaches; 2.9 Domain Engineering and Related Approaches; 2.10 Historical Notes; 2.11 Summary
Chapter 3 Domain Engineering and Object-Oriented Analysis and Design: 3.1 Why Is This Chapter Worth Reading?; 3.2 OO Technology and Reuse; 3.2.1 Solution Space; 3.2.2 Problem Space; 3.3 Relationship between Domain Engineering and Object-Oriented Analysis and Design (OOA/D) Methods; 3.4 Aspects of Integrating Domain Engineering and OOA/D Methods; 3.5 Horizontal versus Vertical Methods; 3.6 Selected Methods; 3.6.1 Rational Unified Process 5.0; 3.6.2 00ram; 3.6.3 Reuse-driven Software Engineering Business (RSEB); 3.6.4 FeatuRSEB; 3.6.5 Domain Engineering Method for Reusable Algorithmic Libraries (DEMRAL)
Chapter 4 Feature Modeling: 4.1 Why Is This Chapter Worth Reading?; 4.2 Features Revisited; 4.3 Feature Modeling; 4.4 Feature Models; 4.4.1 Feature Diagrams; 4.4.1.1 Mandatory Features; 4.4.1.2 Optional Features; 4.4.1.3 Alternative Features; 4.4.1.4 Or-Features; 4.4.1.5 Normalized Feature Diagrams; 4.4.1.6 Expressing Commonality in Feature Diagrams; 4.4.1.7 Expressing Variability in Feature Diagrams; 4.4.2 Other Infon-nation Associated with Feature Diagrams in a Feature Model; 4.4.3 Assigning Priorities to Variable Features; 4.4.4 Availability Sites, Binding Sites, and Binding Modes; 4.4.4.1 Sites; 4.4.4.2 Availability Sites; 4.4.4.3 Binding Sites and Binding Modes; 4.4.4.4 Relationship between Optimizations and Availability Sites, Binding Sites, and Binding Modes; 4.5 Relationship between Feature Diagrams and Other Modeling Notations and Implementation Techniques; 4.5.1 Single Inheritance; 4.5.2 Multiple Inheritance; 4.5.3 Parameterized Inheritance; 4.5.4 Static Parameterization; 4.5.5 Dynamic Parameterization; 4.6 Implementing Constraints; 4.7 Tool Support for Feature Models; 4.8 Frequently Asked Questions about Feature Diagrams; 4.9 Feature Modeling Process; 4.9.1 How to Find Features?; 4.9.2 Role of Variability in Modeling; 4.9.2.1 Separation of Concerns and Perspectives; 4.9.2.2 Decomposition Techniques; 4.9.2.3 Variability in Modeling
Chapter 5 The Process of Generative Programming: 5.1 Why Is This Chapter Worth Reading?; 5.2 Generative Domain Models; 5.3 Main Development Steps in Generative Programming; 5.4 Adapting Domain Engineering for Generative Programming; 5.5 Domain-Specific Languages; 5.6 DEMRAL: Example of a Domain Engineering Method for Generative Programming; 5.7 Outline of DEMRAL; 5.8 Domain Analysis; 5.8.1 Domain Definition; 5.8.2 Domain Modeling; 5.8.2.1 Identification of Key Concepts; 5.8.2.2 Feature Modeling; 5.8.2.2.1 Feature Starter Set for ADTs; 5.8.2.2.2 Feature Starter Set for Algorithms; 5.9 Domain Design; 5.9.1 Scope Domain Model for Implementation; 5.9.2 Identify Packages; 5.9.3 Develop Target Architectures and Identify the Implementation Components; 5.9.4 Identify User DSLs; 5.9.5 Identify Interactions between DSLs; 5.9.6 Specify DSLs and Their Translation; 5.9.7 Configuration DSLs; 5.9.8 Expression DSLs; 5.10 Domain Implementation
Part II Implementation Technologies
Chapter 6 Generic Programming: 6.1 Why Is This Chapter Worth Reading?; 6.2 What Is Generic Programming?; 6.3 Generic versus Generative Programming; 6.4 Generic Parameters; 6.5 Parametric versus Subtype Polymorphism; 6.5.1 Genericity in Java; 6.6 Bounded versus Unbounded Polymorphism; 6.7 A Fresh Look at Polymorphism; 6.8 Parameterized Components; 6.9 Parameterized Programming; 6.9.1 Types, Interfaces, and Specifications; 6.9.2 Adapters; 6.9.3 Vertical and Horizontal Parameters; 6.9.4 Module Expressions; 6.10 C++ Standard Template Library; 6.10.1 Iterators; 6.10.2 Freestanding Functions versus Member Functions; 6.11 Generic Methodology; 6.12 Historical Notes
Chapter 7 Component-Oriented Template-Based C++ Programming Techniques: 7.1 Why Is This Chapter Worth Reading?; 7.2 Types of System Configuration; 7.3 C++ Support for Dynamic Configuration; .4 C++ Support for Static Configuration; 7.4.1 Static Typing; 7.4.2 Static Binding; 7.4.3 Inlining; 7.4.4 Templates; 7.4.5 Parameterized Inheritance; 7.4.6 typedefs; 7.4.7 Member Types; 7.4.8 Nested Classes; 7.5 Prohibiting Certain Template Instantiations; 7.6 Static versus Dynamic Parameterization; 7.7 Wrappers Based on Parameterized Inheritance; 7.8 Template Method Based on Parameterized Inheritance; 7.9 Parameterizing Binding Mode; 7.10 Consistent Parameterization of Multiple Components; 7.11 Static Interactions between Components; 7.11.1 Components with Influence; 7.11.2 Components under Influence; 7.11.3 Structured Configurations; 7.11.4 Recursive Components; 7.11.5 Intelligent Configuration
Chapter 8 Aspect-Oriented Decomposition and Composition: 8.1 Why Is This Chapter Worth Reading?; 8.2 What Is Aspect-Oriented Programming?; 8.3 Aspect-Oriented Decomposition Approaches; 8.3.1 Subject-Oriented Programming; 8.3.2 Composition Filters; 8.3.3 Demeter / Adaptive Programming; .3.4 Aspect-Oriented Decomposition and Domain Engineering; 8.4 How Aspects Arise; 8.5 Composition Mechanisms; 8.5.1 Requirements on Composition Mechanisms; 8.5.1.1 Minimal Coupling; 8.5.1.2 Different Binding Times and Modes; 8.5.1.3 Noninvasive Adaptability; 8.5.2 Example: Synchronizing a Bounded Buffer; 8.5.3 ?TangledO Synchronized Stack; 8.5.4 Separating Synchronization Using Design Patterns; 8.5.5 Separating Synchronization Using SOP; 8.5.6 Some Problems with Design Patterns and Some Solutions; 8.5.6.1 Object Schizophrenia; 8.5.6.2 Preplarming Problem; 8.5.6.3 Traceability Problem; 8.5.7 Implementing Noninvasive, Dynamic Composition in Smalltalk; 8.5.7.1 Model of the Composition; 8.5.7.2 Composition API; 8.5.7.3 Instance-Specific Extension Protocol; 8.5.7.3.1 Defining Methods in Instances; 8.5.7.3.2 Adding Instance Variables to Instances; 8.5.8 Kinds of Crosscutting; 8.6 How to Express Aspects in Programming Languages; 8.6.1 Separating Synchronization Using AspectJ Cool; 8.6.2 Implementing Dynamic Cool in Smalltalk; 8.6.2.1 Example: Synchronizing an Assembly System; 8.6.2.2 Architecture of the Smalltalk Implementation of Dynamic Cool; 8.7 Implementation Technologies for Aspect-Oriented Programming; 8.7.1 Technologies for Implementing Aspect-Specific Abstractions; 8.7.2 Technologies for Implementing Weaving; 8.7.3 AOP and Specialized Language Extensions; 8.7.4 AOP and Active Libraries; 8.8 Final Remarks
Chapter 9 Generators: 9.1 Why Is This Chapter Worth Reading?; 9.2 What Are Generators?; 9.3 Transformational Model of Software Development; 9.4 Technologies for Building Generators; 9.5 Compositional versus Transformational Generators; 9.6 Kinds of Transformations; 9.6.1 Compiler Transformations; 9.6.1.1 Refinements; 9.6.1.2 Optimizations; 9.6.2 Source-to-Source Transformations; 9.7 Transformation Systems; 9.7.1 Scheduling Transformations; 9.7.2 Existing Transformation Systems and Their Applications; 9.8 Selected Approaches to Generation; 9.8.1 Draco; 9.8.2 GenVoca; 9.8.2.1 Example: Applying GenVoca to the Domain of Data Containers; 9.8.2.2 GenVoca Model; 9.8.2.3 Defining Realms and Layers in P++; 9.8.2.4 Implementing GenVoca Layers in C++; 9.8.2.5 Composition Validation; 9.8.2.6 Transformations and GenVoca; 9.8.2.7 Frameworks and GenVoca; 9.8.3 Approaches Based on Algebraic Specifications
Chapter 10 Static Metaprogramming in C++: 10.1 Why Is This Chapter Worth Reading?; 10.2 What Is Metaprogramming?; 10.3 A Quick Tour of Metaprogramming; 10.4 Static Metaprogramming; 10.5 C++ As a Two-Level Language; 10.6 Functional Flavor of the Static Level; 10.6.1 Class Templates As Functions; 10.6.2 Integers and Types As Data; 10.6.3 Symbolic Names Instead of Variables; 10.6.4 Constant Initialization and typedef-Statements Instead of Assiginnent; 10.6.5 Template Recursion Instead of Loops; 10.6.6 Conditional Operator and Template Specialization As Conditional Constructs; 10.7 Template Metaprogramming; 10.8 Template Metafunctions; 10.9 Metaftinctions As Arguments and Return Values of Other Metaftinctions; 10.10 Representing Metainformation; 10.10.1 Member Traits; 10.10.2 Traits Classes; 10.10.3 Traits Templates; 10.10.4 Example: Using Template Metafunctions and Traits Templates to Implement Type Promotions; 10.10.5 Compile-Time Lists and Trees As Nested Templates; 10.11 Compile-Time Control Structures; 10.11.1 Explicit Selection Constructs (? :, IF<>, and SWITCH<>); 10.11.1.1 Implementing IF<> without Partial Template Specialization; 10.11.1.2 Switch Construct (SWITCH<>); 10.11.1.3 Taking Advantage of Lazy Behavior; 10.11.1.4 SWITCH<> without Partial Template Specialization; 10.11.2 Template Recursion As a Looping Construct; 10.11.3 Explicit Looping Constructs (WHILE<>, DO<>, and FOR<>); 10.11.3.1 While-Loop (WHILE<>); 10.11.3.2 Do-Loop (DO<>); 10.11.3.3 For-Loop (FOR<>); 10.12 Code Generation; 10.12.1 Simple Code Selection; 10.12.2 Composing Templates; 10.12.3 Generators Based on Expression Templates; 10.12.4 Recursive Code Expansion; 10.12.5 Explicit Loops for Generating Code (EWHILE<>, EDO<>, and EFOR<>); 10.12.5.1 EWHILE<>; 10.12.5.2 EDO<>; 10.12.5.3 EFOR<>; 10.12.5.4 Nesting Static E-Loops; 10.13 Example: Using Static Execute Loops to Test Metafunctions; 10.14 Partial Evaluation in C++; 10.15 Workarounds for Partial Template Specialization; 10.16 Problems of Template Metaprogramming; 10.17 Historical Notes
Chapter 11 Intentional Programming: 11.1 Why Is This Chapter Worth Reading?; 11.2 What Is Intentional Programming?; 11.3 Technology behind IP; 11.3.1 System Architecture; 11.3.2 Representing Programs in IP: The Source Graph; 11.3.2.1 Treelike Structures; 11.3.2.2 Graphlike Structures; 11.3.2.3 Source Graphs: The Big Picture; 11.3.2.4 The Essence of Source Graphs: Abstraction Sharing and Parameterization; 11.3.3 Source Graph + Methods = Active Source; 11.3.3.1 Kinds of Methods; 11.4 Working with the IP Programming Environment; 11.4.1 Editing; 11.4.2 Further Capabilities of the IP Editor; 11.4.3 Extending the IP System with New Intentions; 11.5 Advanced Topics; 11.5.1 Questions, Methods, and a Frameworklike Organization; 11.5.2 Source-Pattem-Based Polymorphism; 11.5.3 Methods As Visitors; 11.5.4 Asking Questions Synchronously and Asynchronously; 11.5.5 Reduction; 11.6 The Philosophy behind IP; 11.6.1 Why Do We Need Extendible Programming Environments? or What Is the Problem with Fixed Programming Languages?; 11.6.1.1 Problems with General-Purpose Languages and Conventional Libraries; 11.6.1.2 Problems with Comprehensive Application-Specific Languages; 11.6.1.3 The IP Solution: An Extendible Programming Environment; 11.6.2 Moving Focus from Fixed Languages to Language Features and the Emergence of an Intention Market; 11.6.3 Intentional Programming and Component-Based Development; 11.6.4 Frequently Asked Questions; 11.7 Summary
Part III Application Examples
Chapter 12 List Container: 12.1 Why Is This Chapter Worth Reading?; 12.2 Overview; 12.3 Domain Analysis; 12.4 Domain Design; 12.5 Implementation Components; 12.6 Manual Assembly; 12.7 Specifying Lists; 12.8 The Generator; 12.9 Extensions
Chapter 13 Bank Account: 13.1 Why Is This Chapter Worth Reading?; 13.2 The Successful Programming Shop; 13.3 Design Pattems, Frameworks, and Components; 13.4 Domain Engineering and Generative Programming; 13.5 Feature Modeling; 13.6 Architecture Design; 13.7 Implementation Components; 13.8 Configurable Class Hierarchies; 13.9 Designing a Domain-Specific Language; 13.10 Bank Account Generator; 13.11 Testing Generators and Their Products
Chapter 14 Generative Matrix Computation Library (GMCL): 14.1 Why Is This Chapter Worth Reading?; 14.2 Why Matrix Computations?; 14.3 Domain Analysis; 14.3.1 Domain Definition; 14.3.2 Domain Modeling; 14.4 Domain Design and Implementation; 14.4.1 Matrix Type Generation; 14.4.1.1 Target Architecture and Matrix Implementation Components; 14.4.1.2 Matrix Configuration DSL; 14.4.1.3 Matrix Configuration Generator; 14.4.1.3.1 Configuration DSL Parser; 14.4.1.3.2 Assigning Defaults to DSL Features; 14.4.1.3.3 Matrix Component Assembler; 14.4.2 Generating Code for Matrix Expressions; 14.4.2.1 Evaluating Matrix Expressions; 14.4.2.2 Modeling the Elementwise Expression Evaluation Using Objects; 14.4.2.3 Overview of the Implementation; 14.4.2.4 Matrix Operator Templates; 14.4.2.5 Matrix Expression Templates; 14.4.2.6 Matrix Cache; 14.4.2.7 Generating getElement; 14.4.2.8 Metafunctions for Computing Result Types of Expressions; 14.4.2.9 Generating Matrix Assignment; 14.4.2.10 Lessons Learned; 14.4.2.11 Problems with Template Metaprograrnming; 14.4.3 Implementing the Matrix Component in IP
Appendices: Appendix A Conceptual Modeling; A.1 What Are Concepts?; A.2 Theories of Concepts; A.2.1 Basic Terminology; A.2.2 The Classical View; A.2.3 The Probabilistic View; A.2.4 The Exemplar View; A.2.5 Summary of the Three Views; A.3 Important Issues Concerning Concepts; A.3.1 Stability of Concepts; A.3.2 Concept Core; A.3.3 Informational Contents of Features; A.3.4 Feature Composition and Relationships between Features; A.3.5 Quality of Features; A.3.6 Abstraction and Generalization; A.4 Conceptual Modeling, Object-Orientation, and Software Reuse; Appendix B Instance-Specific Extension Protocol for Smalltalk; Appendix C Protocol for Attaching Listener Objects in Smalltalk; Appendix D Glossary of Matrix Computation Terms; Appendix E Metafunction for Evaluating Dependency Tables
Glossary of Generative Programming Terms
References
Index

L'auteur - Krzysztof Czarnecki

Krzysztof Czarnecki

is a researcher and consultant with the Software Engineering Lab at DaimlerChrysler Research and Technology in Ulm, Germany. He gained firsthand experience with Aspect-Oriented Programming and Intentional Programming during research visits at the Xerox Palo Alto Research Center and the Microsoft Research in Redmond, Washington. He received an M.S. degree in computer science from California State University at Sacramento, and M.S. and Ph.D. degrees in computer science from the University of Ilmenau in Germany.

L'auteur - Ulrich Eisenecker

Ulrich Eisenecker

is a professor of computer science at the University of Applied Sciences Kaiserslautern at Zweibrocken, where he chairs the department for componentware and windows interfaces. Prior to his university career he spent nearly a decade in industry. He is an editor of a special supplement on component software of the German IT magazine OBJEKTspektrum, for which he also writes a C++ column. He has published more than a hundred articles, and frequently speaks at national and international IT conferences.

Caractéristiques techniques

	PAPIER
Éditeur(s)	Addison Wesley
Auteur(s)	Krzysztof Czarnecki, Ulrich Eisenecker
Parution	22/06/2000
Nb. de pages	832
Format	18,6 x 23,3
Couverture	Broché
Poids	1269g
Intérieur	Noir et Blanc
EAN13	9780201309775
ISBN13	978-0-201-30977-5