Programing Style
Download as PPT, PDF1 like2,314 views
This document discusses programming style for object-oriented programming. It covers principles like reusability, extensibility, and robustness. For reusability, it recommends keeping methods coherent, small, consistent, and separating policy from implementation. For extensibility, it suggests encapsulating classes, hiding data structures, and avoiding type casts. For robustness, it advises protecting against errors, validating arguments, and avoiding predefined limits. Overall, it emphasizes readability, understandability, and extensibility of code through object-oriented design.
Programing Style
- 1. Programming Style Chapter 14 Part 3: Implementation Object-Oriented Modeling and Design Byung-Hyun Ha bhha@pusan.ac.kr Sumber dari: ike.ie.pusan.ac.kr/w/images/c/cc/08sal_ch014.ppt
- 2. Lecture Outline Introduction Object-Oriented Style Reusability Extensibility Robustness Programming-in-the-large
- 3. Introduction The experienced programmer follows principles to make readable programs that live beyond the immediate need Good style is important in all programming, but it is even more important in OO design and programming because much of the benefit of the OO approach is predicated on producing reusable, extensible, understandable programs
- 4. Object-Oriented Style Reusability Extensibility Robustness Programming-in-the-large
- 5. Reusability Kind of reusability Sharing of newly-written code within a project Reuse of previously-written code on new projects Style rules for reusability Keep methods coherent Keep methods small Keep methods consistent Separate policy and implementation Provide uniform coverage Broaden the method as much as possible Avoid global information Avoid modes
- 6. Reusability Using inheritance Subroutines Factoring Delegation Encapsulate external code
- 7. Extensibility OO principles for extensibility Encapsulate classes Hide data structures Avoid traversing multiple links or methods Avoid cast statements on object type Distinguish public and private operations
- 8. Robustness Guidelines for robustness Protect against errors • User errors and low-level system errors • Programming bugs Optimize after the program runs Validate arguments Avoid predefined limits Instrument the program for debugging and performance monitoring
- 9. Programming-in-the-large Guidelines Do not prematurely begin programming Keep method understandable Make methods readable Use exactly the same names as in the object model Choose name carefully Use programming guidelines Package into modules Document classes and methods Publish the specification
- 10. Appendix: OO and Programming We already discussed… Farm toString() and priority queue Window programming They cannot be possible without inheritance and polymorphism If you want to prepare those functionalities in the examples, you should consider the use of OO concepts Otherwise, your program will never be understandable (even for yourself), extensible, and reusable
- 11. Appendix: OO and Programming Inventory example Inventory simulation Assumptions • Two types of suppliers • Three types of demands • Two types of policies • Two ways of displaying results If you want to carry out simulation for every possible combination of settings, you have to write 24 (= 2x3x2x2) programs • Too complex to write and manage Let’s make them using one program (Inv_manage1.java) • How about it? • Could you understand? • Do you think you can easily extend the program?
- 12. Appendix: OO and Programming Inventory example (cont’) Let’s use OO concepts (Inv_manage2.java, …) First of all, easy to understand Coherent, small, clear, consistent, extendible, reusable, … Inv_manage Supplier Demand Policy Display put_order today_deliever today_demand today_order show
- 13. Appendix: OO and Programming Tree example Tree traversal with different purposes Structural way (Tree1.java) • We should implement traversal algorithm every time we need OO way (Tree2.java) • Don’t care about how to traverse, only need to write what to do 4 5 6 2 3 1 2 5 7 1 32