1. Object Oriented Analysis And Design-IT0207 iiI Semester
Unit-V -SOFTWARE QUALITY

2. Introduction
To develop and deliver robust system, we need a high level of confidence that
Each component will behave correctly
Collective behavior is correct
No incorrect collective behavior will be produced
Debugging and testing
Quality assurance test, testing strategies, OO on Software quality, test plan and test cases.

3. Quality Assurance Tests
Reason why quality assurance is needed,
Computers are infamous for doing what we tell them to do, not necessarily what we want them to do.
To close this gap, code must be free of errors or bugs that cause unexpected results

4. Debugging
Software debugging system can provide tools for finding errors in programs and correcting them
Different kinds of errors are
Syntax errors
Run-Time errors
Logic errors

5. Quality Assurance Tests
Quality assurance testing can be divided into two major categories
Error based testing (testing the boundary condition)
Scenario based testing (Usage-based testing)
Error based Testing
Search a given class’s method for particular clues of interests, then describe how these clues should be tested
Scenario based Testing
Concentrates on what the user does, not what the product does

6. Testing Strategies
Extend of testing a system is controlled by many factors, such as the risks involved, limitations on resources and deadlines
There are many test strategies, but most testing uses a combination of them
Black box testing
White box testing
Top-Down testing
Bottom-Up testing

7. Testing Strategies Black Box Testing
Used to represent a system whose inside workings are not available for inspection.
Test items are tested as “Black”, since its logic is unknown, all that is known is what goes in and what comes out, i.e. input and output
Works nicely in testing objects in OO environment

8. Testing Strategies
Black Box Testing
Input Output

9. Testing Strategies White Box Testing
Assumes that the specific logic is important and must be tested to guarantee the system’s proper functioning.
Main use is in error-based testing, when we already have tested all objects of an application and all external or public methods of an object that are of greater importance
One form of white box testing, called path testing, makes certain that each path in a object’s method is executed at least once during testing.

10. Testing Strategies Two types of path testing are
Statement based testing
Branch based testing

11. Testing Strategies Top-Down Testing
It assumes that the main logic or object interactions and systems message of the application need more testing than an individual object’s method or supporting logic
It can detect the design flows early in the implementation
It supports testing the user interface and event-driven systems

12. Testing Strategies Top-Down Testing
It can test the navigation through screens and verify that it matches the requirement, users can see, at an early stage, how the final application will look and feel.
Useful for Scenario-based testing.
Useful to test subsystem and system integration.

13. Testing Strategies Bottom-Up testing
Starts with the details of the system and proceeds to higher levels by a progressive aggregation of details until they collectively for the requirements for the system.
This approach is more appropriate for testing the individual object in a system.
Test each object, combine them and test their interaction and the message passed among object by utilizing the top-down approach.

14. Testing Strategies Bottom-Up testing
In bottom-up, start with methods and classes that call or rely on no others, then progress to next level up. Those methods and classes that use only the bottom level ones already tested
It leads to integration testing which leads to system testing

15. Test Cases
To have a comprehensive testing scheme, the test must cover all methods or a good majority of them.
All services of the system should be checked atleast once.
To test a system, given input and received output checked with expected output

16. Test Cases The main objectives of testing are:
Testing is the process of executing a program with the intent of finding errors.
A good test case is the one that has a high probability of detecting an as-yet undiscovered error.

17. Guidelines for Developing Quality Assurance Test Cases
A test case is a set of what-if questions
Thomas has developed guidelines that have been adapted for the UA
Describe which feature or service your test attempts to cover
If the test case is based on a use-case, it is good idea to refer to the use-case name. Use- cases are the source of test cases.
Specify what you are testing and which particular feature
Test the normal use of the object’s methods.

18. Guidelines for Developing Quality Assurance Test Cases
Test the abnormal but reasonable use of object’s method.
Test the boundary condition
Test object’s interactions and message sent among them
The internal quality of the software, such as reusability and extendibility should be assessed as well.

19. Test Plan
Test plan is developed to detect and identify potential problems before delivering the software to its users
It provides the roadmap for testing activities, whether usability, user satisfaction or quality assurance tests.

20. Test Plan Steps needed to create a test plan:
Objectives of the test
Development of a test case
Test analysis
All passed tests should be reported with the revised program called regression testing, which can discover errors introduced during the debugging process

21. Who should do testing?
For a small application, the designer or the design team usually will develop the test plan and test cases and in some situations, actually will perform the tests.
Many organizations have a separate team, such as a quality assurance group, that works closely with the design team and is responsible for these activities

22. Who should do testing?
Most software companies also use beta testing a popular inexpensive and effective way to test software on a select group of the actual users of the system
In contrast, alpha testing is done by in- house testers, such as programmers, software engineering and internal users.

23. Guidelines for Developing Test Plans
You may have requirements that dictate a specific appearance or format for your test plan.
Test plan should contain a schedule and a list of required resources
After you have determined what types of testing are necessary, you need to document specifically what you are going to do

24. Guidelines for Developing Test Plans
A configuration control system provides a way of tracking the changes to the code.
A well thought out design tends to produce better code and result in more complete testing, so it is a good idea to buy to keep the plan up to date.
At the end of each month, take time to complete routine updates.

25. Continuous Testing
Software is tested to determine whether it conforms to the specifications of requirements.
Software is maintained when errors are found and corrected and software is extended when new functionality is added to an already existing program.
There are different reasons for testing such as to test for potential problems in a proposed design or to compare two or more designs to determine which is better, given a specific task or set of tasks

26. Continuous Testing
Testing the whole system and detecting bugs is more difficult than testing smaller pieces of the application as it is being developed.
Testing must take place on a continuous basis and this refining cycle must continue through out the development process until we are satisfied with the results.

27. Continuous Testing The steps to successful testing:
Understand and communicate the business case for improved testing
Develop an internal infrastructure to support continuous testing.
Look for leaders who will commit to and own the process
Measures and document your findings in a defect recording system
Publicize improvements as they are made and let people know what they are doing better

28. Debugging Principles Bug Locating Principle: Think
If you reach an impasse, sleep on it.
If the impasse remains, describe the problem to someone else
Use debugging tools
Experimentation should be done as a last

29. Debugging Principles Debugging Principle:
Where there is a one bug, there is likely to be another
Fix the error, not just the symptom of it
The probability of the solution being correct drops as the size of the program increases
Because of the possibility that an error correction will create a new error

30. Review Questions
What is testing ? List out the different testing strategies.
What is black box testing?
When would you use white box testing?
If you don’t know what is inside the box, how would you test it?
What is the importance of developing a test case?
What is white box testing?
What is path testing?
What is statement testing coverage?
What is branch testing coverage?
What is regression testing?
What is a test plan? What steps are followed in developing a test plan?
What is top-down testing?
What is bottom -up testing?
What are guidelines for developing test plans?
What do you mean by continuous testing?
What are the Myer’s debugging principles?
What is debugging?
What is error based testing? Give an example.
What is scenario -based testing?
What are two types of path testing?
Differentiate beta testing and alpha testing?
What are the steps involved in successful testing?
Define Unit Testing?

31. Bibiliography Object oriented system development by Ali Brahami
Object oriented methodology by Booch
A text book on UML by Srimathi