Coverage Testing

This lesson is about coverage testing and how to use it.

Learning Objectives

In this section you will learn about the limitations of using structural tools for black box testing and the importance of coverage measures such as device compatibility, input file format, and output file format. The section also highlights the drawbacks of measuring code coverage as a way to determine progress and drive project completion. Additionally, it introduces the concept of using code coverage to identify untested areas and design tests accordingly.

 

Core Topics and Takeaways

  • Coverage testing

  • Device compatibility

  • Input/Output file compatibility

Video Highlights

Topics

Key Concepts

Video Location

Topics

Key Concepts

Video Location

Structural tools are not helpful for black box testers, who focus on coverage measures such as device compatibility, input/output file format, and size coverage.

  • The VP of development was disappointed when the tester couldn't provide code coverage information.

  • The tester emphasized the importance of passing compatibility testing with printers and testing compatibility with various word processors and graphics programs.

  • Coverage measures for the tester included device compatibility coverage, input file format coverage, and output file format and size coverage.

  • Brian Merrick is a well-known author of code coverage tools and wrote about the misuse of code coverage in a classic paper.

00:01

Code coverage is not a reliable measure of the quality of tests or the progress of testing.

  • Increasing code coverage does not necessarily result in better tests.

  • Code coverage only measures how many statements have been tested, not the effectiveness of the tests.

  • Pressuring people to increase their code coverage numbers is not an effective way to complete a project.

  • Code coverage can be useful for identifying areas of the program that haven't been tested yet.

01:52

Learning Highlights

The importance of not relying solely on structural tools for black-box testing, suggesting alternative coverage measures for better evaluation. Focusing on numbers doesn't necessarily lead to effective tests. The lecture advocates for using coverage to identify untested areas for targeted testing rather than measuring completeness.