Structured Testing

Definition

Structured testing is a white-box unit testing technique for identifying test cases that identify defects involving the logic of a functional unit of software.

Discussion

Structured testing is the traditional primary testing technique for procedural units of software and for logically complex methods of object-oriented software. Structured testing guarantees branch coverage, and thus also guarantees statement coverage.

Objectives

The typical objectives of structured testing are to:

• Ensure adequate testing of complex operations.
• Cause failures due to logic (e.g., branching, looping) defects.

Preconditions

Structured testing can typically begin when the following preconditions hold:

• The tester(s) has been trained in structured testing.
• The class under test has been implemented.

Completion Criteria

Structured testing is typically complete if the following postconditions hold:

• Branch coverage has been achieved by creating and executing a test case for each basis path through each operation with branching and looping.

Steps

During structured testing, the unit testers typically perform the following steps:

• Identify operations with branching and looping.
• For each such operation, determine the McCabe's cyclometric complexity (i.e., the number of test cases):
  • Draw a control flow graph that shows the branches and loops of the method.
  • Calculate the cyclometric complexity by C = e - n + 2, whereby e is the number of edges and n is the number of nodes in the control flow graph.
• Identify a set of basis paths through the operation, whereby the number of basis paths equals McCabe's cyclometric complexity
• Develop a test case for each basis path by picking message parameters and states that cause the execution of the operation to follow the basis path.

Work Products

Structured testing typically results in the following work products:

• Test suites containing basis path test cases for each operation having brancing and/or looping.

Limitations

Structured testing is typically subject to the following limitations:

• This technique used to be more important when procedures contained large amounts of branching and looping; the advent of inheritance, polymorphism, and refactoring has limited its value when applied to the small, simple methods of object-oriented software.

Guidelines

• Try to reuse equivalence set test cases that traverse the basis paths.
• Basis path testing can be used to identify equivalence sets.
• Do not rely solely on structured testing, because it is no longer as important as it once was.
• Polymorphism testing replaces much of what used to be tested using structured testing.