1 Abstract Data Types (ADT) 4 Why use the objects? The need for - PDF document

1 Last update: 16 April 2004 Programming in the large Bertrand Meyer Chair of Software Engineering Programming in the large - Lecture 3 Reading assignment for next week 2 � OOSC2: � Chapter 3: Modularity � Chapter 6: Abstract Data Types Chair of Software Engineering Programming in the large - Lecture 3 3 Lecture 3: Abstract Data Types Chair of Software Engineering Programming in the large - Lecture 3 1

Abstract Data Types (ADT) 4 � Why use the objects? � The need for data abstraction � Moving away from the physical representation � Abstract data type specifications � Applications to software design Chair of Software Engineering Programming in the large - Lecture 3 The first step 5 � A system performs certain actions on certain data. � Basic duality: � Functions [or: Operations, Actions] � Objects [or: Data] Action Object Processor Chair of Software Engineering Programming in the large - Lecture 3 Finding the structure 6 � The structure of the system may be deduced from an analysis of the functions (1) or the objects (2). � Resulting analysis and design method: � Process-based decomposition: classical (routines) � Object-oriented decomposition Chair of Software Engineering Programming in the large - Lecture 3 2

Arguments for using objects 7 � Reusability: Need to reuse whole data structures, not just operations � Extendibility, Continuity: Objects remain more stable over time. Employee information Produce Paychecks Paychecks Hours worked Chair of Software Engineering Programming in the large - Lecture 3 Object technology: A first definition 8 � Object-oriented software construction is the approach to system structuring that bases the architecture of software systems on the types of objects they manipulate — not on “the” function they achieve. Chair of Software Engineering Programming in the large - Lecture 3 The O-O designer’s motto 9 � Ask NOT first WHAT the system does: Ask WHAT it does it TO! Chair of Software Engineering Programming in the large - Lecture 3 3

Issues of object-oriented design 10 � How to find the object types. � How to describe the object types. � How to describe the relations and commonalities between object types. � How to use object types to structure programs. Chair of Software Engineering Programming in the large - Lecture 3 Description of objects 11 � Consider not a single object but a type of objects with similar properties. � Define each type of objects not by the objects’ physical representation but by their behavior: the services (FEATURES) they offer to the rest of the world. � External, not internal view: ABSTRACT DATA TYPES Chair of Software Engineering Programming in the large - Lecture 3 The theoretical basis 12 � The main issue: How to describe program objects (data structures): � Completely � Unambiguously � Without overspecifying? (Remember information hiding) Chair of Software Engineering Programming in the large - Lecture 3 4

A stack, concrete object 13 capacity x “Push” x on stack representation : x representation [ count ] := x count (array_up) count := count + 1 1 representation Chair of Software Engineering Programming in the large - Lecture 3 A stack, concrete object 14 capacity “Push” x on stack representation : count x representation [ count ] := x (array_up) count := count + 1 1 representation x “Push” x on stack representation : representation [ free ] := x (array_down) free := free - 1 free x 1 representation Chair of Software Engineering Programming in the large - Lecture 3 A stack, concrete object 15 capacity “Push” x on stack representation : representation [ count ] := x count (array_up) count := count + 1 1 representation “Push” x on stack representation : representation [ free ] := x (array_down) free := free - 1 x free 1 representation “Push” operation: new new ( n ) item previous n n . item := x item previous n . previous := last (linked) previous head := n item Chair of Software Engineering Programming in the large - Lecture 3 5

Stack: An Abstract Data Type (ADT) 16 � Types: STACK [ G ] -- G : Formal generic parameter � Functions (Operations): put : STACK [ G ] × G → STACK [ G ] remove : STACK [ G ] → STACK [ G ] item : STACK [ G ] → G empty : STACK [ G ] → BOOLEAN new : STACK [ G ] Chair of Software Engineering Programming in the large - Lecture 3 Using functions to model operations 17 ( ) , put = s x s’ Chair of Software Engineering Programming in the large - Lecture 3 Reminder: Partial functions 18 � A partial function, identified here by → , is a function that may not be defined for all possible arguments. � Example from elementary mathematics: � inverse: ℜ → ℜ , such that inverse ( x ) = 1 / x Chair of Software Engineering Programming in the large - Lecture 3 6

The STACK ADT (continued) 19 � Preconditions: remove ( s : STACK [ G ]) require not empty ( s ) item ( s : STACK [ G ]) require not empty ( s ) � Axioms: For all x : G , s : STACK [ G ] item ( put ( s, x )) = x remove ( put ( s, x )) = s put ( , ) = empty ( new ) (or: empty ( new ) = True ) s x s’ not empty ( put ( s , x )) (or: empty ( put ( s , x )) = False ) Chair of Software Engineering Programming in the large - Lecture 3 Exercises 20 � Adapt the preceding specification of stacks (LIFO, Last-In First-Out) to describe queues instead (FIFO). � Adapt the preceding specification of stacks to account for bounded stacks, of maximum size capacity. � Hint: put becomes a partial function. Chair of Software Engineering Programming in the large - Lecture 3 21 End of lecture 3 Chair of Software Engineering Programming in the large - Lecture 3 7