Software Testing Techniques

Software Testing Techniques

• White-box testing and black-box testing are two software testing techniques.

Two Unit Testing Techniques

• Black-box testing: Focuses on the function of a product, not its internal structure. Tests if the specified function is fully operational and error-free. Includes tests at the software interface. Also known as functional testing or behavioral testing.

• White-box testing: Examines the internal workings of a product. Ensures that internal operations match specifications, and all internal components are exercised. Focuses on procedural details, logical paths, and specific conditions/loops. Also called structural testing or glass-box testing.

White-box Testing

• Uses the control structure of component-level design to create test cases.
• Guarantees all independent paths in a module are exercised at least once.
• Exercises every logical decision (true and false sides).
• Executes all loops at their boundaries.
• Tests internal data structures for validity.

Basis Path Testing

• A white-box testing technique by Tom McCabe.
• Derives a logical complexity measure from a procedural design.
• Uses this measure as a guide for defining a basis set of execution paths.
• Test cases derived to exercise the basis set execute every statement in the program at least once.

Flow Graph Notation

• A circle in a graph is a node representing a sequence of one or more procedural statements.
• A node with a conditional expression is a predicate node.
• Each compound condition (multiple Boolean operators) is a separate predicate node.
• A predicate node has two edges (True and False).
• An edge (link) shows flow of control.
• Edges do not intersect or cross other edges.
• Areas bounded by edges/nodes are regions.

Independent Program Paths

• A path from the start node to the end node that introduces at least one new set of statements/conditions.
• Must traverse at least one new edge compared to previous paths.

Cyclomatic Complexity

• A quantitative measure of the logical complexity of a program.
• Defines the number of independent paths in the basis set.
• Provides an upper bound for the required test cases.
• Calculated in three ways:
  • Number of regions.
  • V(G) = E - N + 2 where E = edges, and N = nodes.
  • V(G) = P + 1 where P = predicate nodes.

Deriving the Basis Set and Test Cases

• Use the program design or code to draw a flow graph.
• Determine the cyclomatic complexity of the resultant flow graph.
• Create a basis set of linearly independent paths.
• Create test cases that force each path to execute.

Black-box Testing

• Complements white-box testing by uncovering different classes of errors.
• Focuses on functional requirements and the information domain of a software program.
• Utilizes in latter testing stages, after white-box testing.
• Identifies input conditions to fully exercise all functional requirements.

Black-box Testing Categories

• Incorrect or missing functions.
• Interface errors.
• Errors in data structures or external data access.
• Behavior or performance errors.
• Initialization and termination errors.

Equivalence Partitioning

• A black-box testing technique dividing input domain into classes.
• Ideal for reducing total test cases.
• Based on equivalence classes for valid and invalid input conditions.
• Selects test cases from each equivalence class to exercise the largest number of attributes.

Guidelines for Defining Equivalence Classes

• Range input conditions have one valid range and two invalid ranges.
• Specific value input conditions have one valid and two invalid values.
• Sets of values have one valid set and one invalid set.
• Boolean values have one valid and one invalid Boolean condition.

Boundary Value Analysis

• A black-box testing technique focusing on boundaries of input domain.
• Complements equivalence partitioning.
• Selecting test cases at the edges of classes, from both input and output domains.

Guidelines for Boundary Value Analysis

• Test cases for input ranges include values at the boundaries and just above/below these boundaries.
• Test cases for input values include minimum, maximum values and just above and below the values.
• Test output values analogous to input values.
• Test to stress data structures, by testing at minimum and maximum boundaries.

Walkthrough, Inspection & Peer Review

• Walkthrough: Study of a document for completeness/correctness, from beginning to end.
• Inspection: Focus on a specific aspect within a document.
• Peer Review: Comparison of two similar documents for completeness/correctness.

Error, Defect & Bug

• Error: A coding mistake found by the developer.
• Defect: A mismatch between expected and actual software behaviour, found by the tester (also called issue or flaw).
• Bug: A problem in software during use, found by customers or users.