Practical UML Statecharts in C/C++

Dr. Miro Samek is the founder and president of Quantum Leaps, an open source company providing lightweight, state machine-based, event-driven application frameworks for embedded systems. He is the author of Practical Statecharts in C/C++ (CMP Books, 2002), has written numerous articles for magazines, including a column for C/C++ Users Journal, is a regular speaker at the Embedded Systems Conferences, and serves on the editorial review board of the Embedded Systems Design magazine. For a number of years, he worked in various Silicon Valley companies as an embedded software architect and before that he worked as an embedded software engineer at GE Medical Systems (now GE Healthcare). Dr. Samek earned his Ph.D. in nuclear physics at GSI (Darmstadt, Germany).

Preface
PART I STATECHARTS

Chapter 1 Whirlwind Tour of Programming with Statecharts
1.1 Why Bother?
1.2 The Traditional Event-Action Paradigm
1.3 State Machines ? A Better Way of Programming
1.3.1 The Time Bomb Example
1.3.2 The Calculator Example
1.5 Object-Oriented Analogy
1.6 The Event-driven Framework
1.6 Summary

Chapter 2 A Crash Course in Statecharts
2.1 The Essence of Finite State Machines
2.2 The Essence of UML Statecharts
2.3 Examples of State Models
2.4 Summary

Chapter 3 Standard State Machine Implementations
3.1 State Machine Interface
3.2 Nested switch Statement
3.3 State Table
3.4 State Design Pattern
3.5 Optimal FSM Implementation
3.6 State Machines and C++ Exception Handling
3.7 Role of Pointer-to-Member Functions
3.8 Implementing Guards, Junctions, and Choice Points
3.9 Implementing Entry and Exit Actions
3.10 Dealing with State Hierarchy
3.11 Summary

Chapter 4 QEP: A Minimal Hierarchical Event Processor
4.1 General Structure of the QEP Event Processor
4.2 An Annotated Example (QHsm)
4.3 QEP Structure
4.3.1 QEP Source Code Structure
4.3.2 Internal Representation of a State Machine
4.3.3 Initialization of a State Machine
4.3.4 Dispatching Events to a FSM
4.3.5 Executing a Transition in a FSM
4.3.6 Dispatching Events to a HSM
4.3.7 Executing a Transition in a HSM
4.3.8 Static Transition Optimization in a HSM
4.4 Porting and Configuring QEP
4.5 Caveats
4.6 Summary

Chapter 5 Implementing State Machines with QEP
5.1 Implementing a HSM with QEP
5.1.1 Step 1: Enumerating Signals
5.1.2 Step 2: Defining Events
5.1.3 Step 3: Defining the QCalc State Machine
5.1.4 Step 4: Declaring the QCalc States
5.1.5 Step 5: Initializing the HSM
5.1.6 Step 6: Implementing the State Handler Functions
5.2 Implementing a FSM with QEP
2.5 Pitfalls to Avoid While Coding State Machines with QEP
2.5.1 Incomplete State Handlers2-37
2.5.2 Confusing Statecharts with Flowcharts2-38
2.5.3 Ill-Formed State Handlers2-39
2.5.4 Suboptimal Signal Granularity2-42
2.5.5 Violating the Run To Completion Semantics2-42
4.6 Summary

Chapter 6 State Patterns
6.1 Ultimate Hook
6.2 Reminder
6.3 Deferred Event
6.4 Orthogonal Component
6.5 Transition to History
6.6 Summary


PART II EVENT-DRIVEN FRAMEWORK

Chapter 7 QF: A Minimal Event-Driven Embedded Framework
7.1 Conventional Approach to Multithreading
7.2 Computing Model of QF
7.3 Annotated Example
7.3.1 The ?Airplane in the Tunnel? Game
7.3.2 The Active Object Design
7.3.3 The Implementation
7.3.4 The Port for ARM Cortex-M3
7.3.5 Testing
7.4 Summary

Chapter 8 Design of QF
8.1 Handling Errors and Exceptional Conditions
8.2 Memory Management
8.3 Mutual Exclusion and Blocking
8.4 Active Objects
8.5 Event Management in QF
8.6 Event Delivery Mechanisms in QF
8.9 Deferring and Recalling Events in QF
8.7 Time Events
8.8 Summary

Chapter 9 Implementation of QF
9.1 Code Organization
9.2 Critical Section in QF
9.3 General QF Policies Enforced by Assertions
9.4 Active Object class
9.5 Native QF Event Queue
9.6 Native QF Memory Pool
9.7 Native QF Priority Set
9.8 Native QF Scheduler

Chapter 10 Porting QF
10.1 QF Porting Guide
10.2 QF on Bare-Metal Targets (the Vanilla Port)
10.3 Using QF with a preemptive Real-Time Kernel (C/OS-II)
10.4 QF port to a POSIX-Compliant OS (Linux)
10.5 Summary

Chapter 11 Conclusion
11.2 Rules for Developing Event-Driven Embedded Applications
11.3 Heuristics
11.4 Sizing Event Queues and Event Pools
11.5 System Integration
11.6 Summary of Key Elements
11.7 An Invitation


Appendix A QK: A Single-Stack Preemptive Kernel
A.2 Run-to-Completion Processing
A.3 Synchronous and Asynchronous Preemptions
A.4 Stack Utilization
A.4 Comparison with a Traditional RTOS
A.5 Summary

Appendix B QS: Software Tracing for Event Driven Systems
B.1 Software Tracing Concepts
B.2 Structure of QS Trace Records
B.3 QS Filters
B.4 QS Data Protocol
B.5 QS Trace Buffer
B.6 Configuring and Porting QS
B.7 Summary

Appendix C Inheriting Entire State Models in C++
C.1 Statechart Refinement Example in C++
C.3 Caveats
C.4 Summary

Appendix D Guide to Notation
D.1 Class Diagrams
D.2 Statechart Diagrams
D.3 Sequence Diagrams
D.4 Timing Diagrams

Appendix E CD-ROM
E.1 Source Code Structure
E.2 Installation
E.3 Licensing
E.4 Answers to the Exercises
E.5 QP Reference Guide (Doxygen)
E.6 Resources

Bibliography
Index