Software Engineering Exam 2 Flashcards
Module
A lexically continguous sequence of program statements, bounded by elements and identified as an aggregate.
Cohesion
The degree of interaction within a module.
Coupling
The degree of interaction between modules.
Data Encapsulation
Integrating data and operations on the data within the same module helps isolate changes and improve maintainability.
Abstract Data Types
Types of data with associated operations but without public data attributes.
Information Hiding
Concealing details that might change within a module to ensure modularity.
Object-Oriented Paradigm
Emphasis on objects, inheritance, polymoprhism, and encapsulation to improve software structure and reusability.
Modules should be designed for?
High cohesion and low coupling .
Coincidental Cohesion
Performs unrelated tasks, leading to poor maintainability and reuse.
Logical Cohesion
Performs related tasks chosen by the caller, like all input/output functions grouped together.
Temporal Cohesion
Groups actions related by time. These actions are weakly related functionally.
Procedural Cohesion
Actions are related by the sequence in which they occur.
Communicational Cohesion
Actions use the same data set.
Functional Cohesion
The highest level, where each module performs exactly one taks.
Informational Cohesion
Actions operate on a shared data structure, each action independent with its entry point.
Which cohesion is ideal and why?
Functional cohesion. Highly reusable, easily maintainable, and easier to debug since it performs a single task.
Content Coupling
One module directly uses or modifies another’s content, leading to high interdependency.
Common Coupling
Multiple modules share access to global data, reducing readability and maintainability.
Control Coupling
Modules control each other’s flow by passing control information.
Stamp Coupling
Modules share complex data structures but only use parts of them, reducing reusability.
Data Coupling
Modules interact via simple data parameters, which is ideal as it minimizes dependencies.
Objects and Classes
An object is an instantiation of a class, which is an abstract data type that supports inheritance.
Inheritance
A mechanism where a subclass inherits properties and behaviors from a parent class, promoting code reuse.
Polymorphism
A key concept in OOP that allows objects of different classes to be treated as objects of a common superclass. It enables s single interface to represent different underlying forms (data types).
Dynamic Binding
Method calls are resolved at runtime, which can enhance flexibility but complicate maintenance.
Advantages of polymorphism and inheritance
Flexibility, code reuse, and adaptability of polymorphic objects.
Advantages of Object-Oriented Design
- Balances focus on data and actions, improving modularity.
- Encapsulates object details, supporting high cohesion and low coupling.
Disadvantages of polymorphism and inheritance
Increases complexity and potential for maintenance difficulties due to intertwined dependencies.
Challenges of Object-Oriented Design
- Requires careful design to avoid complexity, especially with inheritance and dynamic binding.
- First projects often take more effort due to the learning curve and petentially larger project scopes.
Reuse
Involves using parts of one product to assist in developing another product with different functionality.
Opportunistic Reuse
Reusable parts are added to a database after a product is built.
Systematic Reuse
Reusable components are created first, then used to build products.
Benefits of reuse
Accelerated product development. No need to redesign or retest reused components.
Common impediments of reuse
- Not-Invented-Here Syndrome: Hesitation to use external code.
- Fault concerns: Worries about bugs in reused code.
- Storage and retrieval issues: Challenges in organizing reusable components.
- Cost: Expenses related to making items reusable and creating reuse process.
Libraries/Toolkits
Sets of reusable routines where control logic is managed by the user.
Application Frameworks
Structures that include control logic, allowing users to customize specific sections.
Design Patterns
Standard solutions to recurring design problems, often consisting of customizable classes.
Wrapper (Adapter Pattern)
Used to resolve interface mismatches between classes.
Bridge Pattern
- Decouples abstraction from implementation to allow independent modification.
- Commonly used in hardware-dependent systems, such as printer drivers, to isolate changes to a single side.
Iterator Pattern
Enables traversal of elements in a collection without exposing the collection’s structure.
Abstract Factory Pattern
Provides an interface to create families of related objects without specifying their conrete classes.
Architecture Reuse
Large-scale reuse through frameworks and toolkits.
Creational Patterns
Focus on object creation.
Structural Patterns
Deal with relationships between entities.
Behavioral Patterns
Address communication between objects.
Maintenance Impact
Reuse can reduce maintenance effort as code changes in one location propagate to reused instances.
Portability
Portability refers to a product’s adaptability across different platforms.
Data and Actions
A product must consider both the data it manages and the actions performed on this data.
Operation-Oriented Design
Focuses on actions.
Data-Oriented Design
Emphasizes data structure.
Hybrid (Object-Oriented Design)
Combines data and actions within objects.
Architectural Design
- High-level design based on specifications.
- Focuses on breaking down the system into modules.
Detailed Design
Elaborates on specific algorithms and data structures within each module.
Data Flow Analysis
Determines points of abstraction for input and output data and decomposes into modules.
Transaction Analysis
More suitable for transaction-heavy applications. Focuses on reusability by creating generic, reusable modules.
Data-Oriented Design Principle
The product structure should reflect the structure of its data.
Data-Oriented Design Popularity
Less common than operation-oriented design and has been largely replaced by object-oriented methods.
Object-Oriented Design’s Aim
Design a system in terms of classes derived during Object-Oriented Analysis.
Two Main Steps of Object-Oriented Design
- Complete the Class Diagram: Define attribute formats and assign methods to either a class or a client.
- Detailed design: Refine the class and sequence of operations for each method.
The Design Workflow
Transform analysis artifacts into usable blueprints for programmers. Split large workflows into smaller subsystems for parallel implementation.
Role of Software Architect
Balances functional and nonfunctional requirements. Makes tradeoffs within budget and time constraint.
Design Review
Checks that the design aligns with specifications and ensures its correctness.
Design Inspection
Verifies each aspect of the design, identifying potential faults before implementation.
