Complete Overview of Unit Testing Techniques and Best Practices

November 24, 2018 

Complete Overview of Unit Testing Techniques and Best Practices

What is Unit Testing?

It is a type of testing in which individual units or functions of software testing. Its primary purpose is to test each unit or function. A unit is the smallest testable part of an application. It mainly has one or a few inputs and produces a single output. In procedural programming, a unit referred as an individual program, while object-oriented programming languages include Base/Superclass, abstract class, Derived/Child class takes place. Unit test frameworks, drivers, stubs and mocks /fake objects used in Unit Testing. It works on the basis of a White box technique.


How Does Unit Testing Work?

To do Unit Testing mocks required. Is there a need of mocks to make testing on functions? Yes, without creating mocks functions cannot be unit tested. Unit testing works on the basis of mock objects. Mock objects work to fill in for missing parts of a program. For example, there might be a function that needs variables or objects that not created yet. To test function, mock objects created. In such conditions, mock objects fill missing parts.

Techniques within Unit testing

White-Box testing

It's referred to as a glass box testing/transparent testing. In this type of testing, the tester is aware of an internal functionality. The internal structure of an item or function to be tested is unknown.

Black-Box testing

It is a type of testing, tester not aware of the internal functionality of a system. The internal structure of function to be tested is unknown.

Gray-Box testing

It's referred to as semi-transparent testing. It is a combination of a Black Box and White Box testing. It is the type of testing in which tester aware with internal functionality of a method or unit but not in a more deep level like white box testing. In this, the user partially aware of the internal functionality of a system.

Different type of testing covered under a Gray box testing is mentioned as following -

  • Matrix testing.
  • Pattern Testing.
  • Orthogonal Pattern testing.
  • Regression Testing.

Benefits of Unit Testing

Unit Tests reveal a basic understanding of API units or functions to understand the functionality of the code.

It is a way to write test cases for all functions and methods so that whenever a change causes a fault then in that case bug quickly identified and fixed.

A modular approach followed in Unit testing, single-single functionality or part of the code tested without waiting for other of code to be completed.


Why Unit Testing Matters?

It is a type of testing that increases the speed of testing.

Unit testing helps to maintain and change the code. If good unit test cases are written for a code and run every time code changes, any defect revealed after code changes.

Codes should be in the more reusable way, in a modular way more unit testing will be possible.

Writing test cases takes time but, this thing can be compensated by less amount of time it takes to run the tests.

Unit tests are more reliable, and development will be faster in the long run too.

It makes debugging easy. In this when a test fails, then only the latest changes need to be debugged. With testing at a higher level or higher phases, changes made over many days/ weeks/months need to be scanned or detected.

The cost to fix detected bugs during Unit Testing lesser in comparison to the defects at higher levels. Compare the cost to fix a defect when a bug detected in a phase of unit testing with the same bug detected in Acceptance Testing.

The effort required to find and fix defects found during Unit Testing is less in comparison to the effort required to fix defects found during System Testing or Acceptance Testing.


How to Adopt Unit Testing

To do Unit Testing, unit test cases created. Unit test cases automated but still created manually. To do Unit Testing with an automated approach following steps need to be considered -

Write another section of code in an application to test a function. Later on, this section of code finally removed to test code when an application is complete or when an application is in working state.

Isolation of a code - Isolate function to test it more rigorously. Isolate code to do Automated Unit Testing in a better way. Isolating functions/code helps to do testing in a good way. It helps to reveal dependencies between functions of code.

Use the Unit Test framework for developing automated test cases. During the execution of test cases, framework display logs for failed test cases.

Nowadays, many frameworks automatically flag and report in summary for failed test cases. Depending upon failure, failure may halt subsequent testing.


Best Practises of Unit Testing

There is no need to create test cases for every condition. Instead of that focus on tests that impact the behavior of the entire system. Before fixing a bug, write a test the expose the defects.

  • The bug will re-occur if not appropriately cached.
  • Test suites should be more comprehensive.
  • Write test cases before fixing the defect.
  • Write test cases independent of each other. For example, if a class of code depends upon a database, do not write a case that interacts with a database to test a class. Instead of that create an abstract interface around that database connection and after that implement an interface with a mock object.
  • The most important thing for Unit Testing is to cover all the paths, it is required to pay attention to loop conditions also.
  • Write cases to verify a behavior, also write test cases to ensure the performance of the code.
  • Execute test cases continuously and frequently.

Unit Testing Tools

Nowadays, there are many tools available that assists for Unit Testing. Few tools to perform Unit Testing -

Jtest - Parasoft Jtest is an IDE plugin that leverages an open source framework (JUnit,mockito, PowerMock, and Spring)with guided and also provides one-click actions to create, scale and maintain unit tests. By automating the time-consuming aspects of Unit Testing, it focuses on business logic and creates more meaningful test suits.

JUnit - It is a tool that is free to use for Java programming language. It also provides support for assertions to identify test method. It is a type of tool that tests data first then inserted into a piece of code.

NUnit - It's used for Unit Testing framework used, and this tool used for all.Net languages. It is an open source tool to write scripts manually. It supports data-driven tests run in parallel.

JMockit - It is an open source unit testing tool. It is a code coverage tool. It helps in mocking of API's with recording and verification of syntax. It is a type of tool that offers line coverage, path coverage and data coverage.

EMMA - It is an open source toolkit for analyzing and reporting code written in a Java language.EMMA support coverage types like method, line, basic block. It is a Java-based tool.

PHPUnit - It is a Unit Testing tool for PHP language. It takes small portions of code called as units and tests each unit separately. This testing tool uses pre-defined assertion methods to assert that the system should behave in a specific manner.


Concluding Unit Testing

In conclusion, Unit testing only required to test each function independently. This type of testing saves time and cost. If the same issue faced at Acceptance Testing then, in that case, the price will be more as compared to the cost of Unit Testing.

Sometimes to do Unit Testing, there is a requirement to create mock objects, to satisfy or fulfill the dependency of a method or a function. If a function is dependent on other function and other function still not created, then the mock function object used.