| Part of a series on |
| Software development |
|---|
Paradigms and models |
Standards and bodies of knowledge |
Unit testing,a.k.a.component ormodule testing, is a form ofsoftware testing by which isolatedsource code is tested to validate expected behavior.[1]
Unit testing describes tests that are run at the unit-level to contrast testing at theintegration orsystem level.[2]
Unit testing, as a principle for testing separately smaller parts of large software systems, dates back to the early days of software engineering. In June 1956 at US Navy's Symposium on Advanced Programming Methods for Digital Computers, H.D. Benington presented theSAGE project. It featured a specification-based approach where the coding phase was followed by "parameter testing" to validate component subprograms against their specification, followed then by an "assembly testing" for parts put together.[3][4]
In 1964, a similar approach is described for the software of theMercury project, where individual units developed by different programmes underwent "unit tests" before being integrated together.[5] In 1969, testing methodologies appear more structured, with unit tests, component tests and integration tests collectively validating individual parts written separately and their progressive assembly into larger blocks.[6] Some public standards adopted in the late 1960s, such as MIL-STD-483[7] and MIL-STD-490, contributed further to a wide acceptance of unit testing in large projects.
Unit testing was in those times interactive[4] or automated,[8] using either coded tests or capture and replay testing tools. In 1989,Kent Beck described a testing framework forSmalltalk (later calledSUnit) in "Simple Smalltalk Testing: With Patterns". In 1997,Kent Beck andErich Gamma developed and releasedJUnit, a unit test framework that became popular withJava developers.[9]Google embraced automated testing around 2005–2006.[10]
A unit is defined as a single behaviour exhibited by the system under test (SUT), usually corresponding to a requirement[definition needed]. While a unit may correspond to a single function or module (inprocedural programming) or a single method or class (inobject-oriented programming), functions/methods and modules/classes do not necessarily correspond to units. From the system requirements perspective only the perimeter of the system is relevant, thus only entry points to externally visible system behaviours define units.[clarification needed][11]
Unit tests can be performed manually or viaautomated test execution. Automated tests include benefits such as: running tests often, running tests without staffing cost, and consistent and repeatable testing.
Testing is often performed by the programmer who writes and modifies the code under test.Unit testing may be viewed as part of the process of writing code.
 | This sectionneeds additional citations forverification. Please helpimprove this article byadding citations to reliable sources in this section. Unsourced material may be challenged and removed.(September 2019) (Learn how and when to remove this message) |
During development, a programmer may code criteria, or results that are known to be good, into the test to verify the unit's correctness.
During test execution, frameworkslog tests that fail any criterion and report them in a summary.
For this, the most commonly used approach is test - function - expected value.
Aparameterized test is a test that accepts a set of values that can be used to enable the test to run with multiple, different input values. A testing framework that supports parametrized tests supports a way to encode parameter sets and to run the test with each set.
Use of parametrized tests can reduce test code duplication.
Parameterized tests are supported byTestNG,JUnit,[14]XUnit andNUnit, as well as in various JavaScript test frameworks.[citation needed]
Parameters for the unit tests may be coded manually or in some cases are automatically generated by the test framework. In recent years support was added for writing more powerful (unit) tests, leveraging the concept of theories, test cases that execute the same steps, but using test data generated at runtime, unlike regular parameterized tests that use the same execution steps with input sets that are pre-defined.[citation needed]
Test code needs access to the code it is testing, but testing should not compromise normal design goals such asinformation hiding, encapsulation and theseparation of concerns. To enable access to code not exposed in the external API, unit tests can be located in the same project ormodule as the code being tested.
Inobject oriented design this still may not provide access to private data and methods. Therefore, extra work may be necessary for unit tests. InJava and other languages, a developer can usereflection to access private fields and methods.[15] Alternatively, aninner class can be used to hold the unit tests so they have visibility of the enclosing class's members and attributes. In the.NET Framework and some other programming languages,partial classes may be used to expose private methods and data for the tests to access.
It is important that code solely for accommodating tests does not remain in the production code. InC and other languages,compiler directives such as#if DEBUG ... #endif can be placed around such additional classes and indeed all other test-related code to prevent them being compiled into the released code. This means the released code is not exactly the same as what was unit tested. The regular running of fewer but more comprehensive, end-to-end, integration tests on the final release build can ensure (among other things) that no production code exists that subtly relies on aspects of the test harness.
There is some debate among developers, as to whether it is wise to test private methods and data anyway. Some argue that private members are a mere implementation detail that may change, and should be allowed to do so without breaking numbers of tests. Thus it should be sufficient to test any class through its public interface or through its subclass interface, which some languages call the "protected" interface.[16] Others say that crucial aspects of functionality may be implemented in private methods and testing them directly offers advantage of smaller and more direct unit tests.[17][18]
Sometimes, in the agile software development, unit testing is done peruser story and comes in the later half of the sprint after requirements gathering and development are complete. Typically, the developers or other members from the development team, such asconsultants, will write step-by-step 'test scripts' for the developers to execute in the tool. Test scripts are generally written to prove the effective and technical operation of specific developed features in the tool, as opposed to full fledged business processes that would be interfaced by theend user, which is typically done duringuser acceptance testing. If the test-script can be fully executed from start to finish without incident, the unit test is considered to have "passed", otherwise errors are noted and the user story is moved back to development in an 'in-progress' state. User stories that successfully pass unit tests are moved on to the final steps of the sprint - Code review, peer review, and then lastly a 'show-back' session demonstrating the developed tool to stakeholders.
In test-driven development (TDD), unit tests are written before the related production code is written. Starting with a failing test, then addsjust enough production code to make the test pass, then refactors the code as makes sense and then repeats by adding another failing test.
Unit testing is intended to ensure that the units meet theirdesign and behave as intended.[19]
By writing tests first for the smallest testable units, then the compound behaviors between those, one can build up comprehensive tests for complex applications.[19]
One goal of unit testing is to isolate each part of the program and show that the individual parts are correct.[1] A unit test provides a strict, writtencontract that the piece of code must satisfy.
Unit testing finds problems early in thedevelopment cycle. This includes both bugs in the programmer's implementation and flaws or missing parts of the specification for the unit. The process of writing a thorough set of tests forces the author to think through inputs, outputs, and error conditions, and thus more crisply define the unit's desired behavior.[citation needed]
The cost of finding a bug before coding begins or when the code is first written is considerably lower than the cost of detecting, identifying, and correcting the bug later. Bugs in released code may also cause costly problems for the end-users of the software.[20][21][22] Code can be impossible or difficult to unit test if poorly written, thus unit testing can force developers to structure functions and objects in better ways.
Unit testing enables more frequent releases in software development. By testing individual components in isolation, developers can quickly identify and address issues, leading to faster iteration and release cycles.[23]
Unit testing allows the programmer torefactor code or upgrade system libraries at a later date, and make sure the module still works correctly (e.g., inregression testing). The procedure is to write test cases for allfunctions andmethods so that whenever a change causes a fault, it can be identified quickly.
Unit tests detect changes which may break adesign contract.
Unit testing may reduce uncertainty in the units themselves and can be used in abottom-up testing style approach. By testing the parts of a program first and then testing the sum of its parts,integration testing becomes much easier.[citation needed]
Some programmers contend that unit tests provide a form of documentation of the code. Developers wanting to learn what functionality is provided by a unit, and how to use it, can review the unit tests to gain an understanding of it.[citation needed]
Test cases can embody characteristics that are critical to the success of the unit. These characteristics can indicate appropriate/inappropriate use of a unit as well as negative behaviors that are to be trapped by the unit. A test case documents these critical characteristics, although many software development environments do not rely solely upon code to document the product in development.[citation needed]
In some processes, the act of writing tests and the code under test, plus associated refactoring, may take the place of formal design. Each unit test can be seen as a design element specifying classes, methods, and observable behavior.[citation needed]
Testing will not catch every error in the program, because it cannot evaluate every execution path in any but the most trivial programs. Thisproblem is a superset of thehalting problem, which isundecidable. The same is true for unit testing. Additionally, unit testing by definition only tests the functionality of the units themselves. Therefore, it will not catch integration errors or broader system-level errors (such as functions performed across multiple units, or non-functional test areas such asperformance). Unit testing should be done in conjunction with othersoftware testing activities, as they can only show the presence or absence of particular errors; they cannot prove a complete absence of errors. To guarantee correct behavior for every execution path and every possible input, and ensure the absence of errors, other techniques are required, namely the application offormal methods to prove that a software component has no unexpected behavior.[citation needed]
An elaborate hierarchy of unit tests does not equal integration testing. Integration with peripheral units should be included in integration tests, but not in unit tests.[citation needed] Integration testing typically still relies heavily on humanstesting manually; high-level or global-scope testing can be difficult to automate, such that manual testing often appears faster and cheaper.[citation needed]
Software testing is a combinatorial problem. For example, every Boolean decision statement requires at least two tests: one with an outcome of "true" and one with an outcome of "false". As a result, for every line of code written, programmers often need 3 to 5 lines of test code.[citation needed] This obviously takes time and its investment may not be worth the effort. There are problems that cannot easily be tested at all – for example those that arenondeterministic or involve multiplethreads. In addition, code for a unit test is as likely to be buggy as the code it is testing.Fred Brooks inThe Mythical Man-Month quotes: "Never go to sea with two chronometers; take one or three."[24] Meaning, if twochronometers contradict, how do you know which one is correct?
Another challenge related to writing the unit tests is the difficulty of setting up realistic and useful tests. It is necessary to create relevant initial conditions so the part of the application being tested behaves like part of the complete system. If these initial conditions are not set correctly, the test will not be exercising the code in a realistic context, which diminishes the value and accuracy of unit test results.[citation needed]
To obtain the intended benefits from unit testing, rigorous discipline is needed throughout the software development process.
It is essential to keep careful records not only of the tests that have been performed, but also of all changes that have been made to the source code of this or any other unit in the software. Use of aversion control system is essential. If a later version of the unit fails a particular test that it had previously passed, the version-control software can provide a list of the source code changes (if any) that have been applied to the unit since that time.[citation needed]
It is also essential to implement a sustainable process for ensuring that test case failures are reviewed regularly and addressed immediately.[25] If such a process is not implemented and ingrained into the team's workflow, the application will evolve out of sync with the unit test suite, increasing false positives and reducing the effectiveness of the test suite.
Unit testing embedded system software presents a unique challenge: Because the software is being developed on a different platform than the one it will eventually run on, you cannot readily run a test program in the actual deployment environment, as is possible with desktop programs.[26]
Unit tests tend to be easiest when a method has input parameters and some output. It is not as easy to create unit tests when a major function of the method is to interact with something external to the application. For example, a method that will work with a database might require a mock up of database interactions to be created, which probably won't be as comprehensive as the real database interactions.[27][better source needed]
Below is an example of a JUnit test suite. It focuses on theAdder class.
classAdder{publicintadd(inta,intb){returna+b;}}
The test suite usesassert statements to verify the expected result of various input values to thesum method.
import staticorg.junit.Assert.assertEquals;importorg.junit.Test;publicclassAdderUnitTest{@TestpublicvoidsumReturnsZeroForZeroInput(){Adderadder=newAdder();assertEquals(0,adder.add(0,0));}@TestpublicvoidsumReturnsSumOfTwoPositiveNumbers(){Adderadder=newAdder();assertEquals(3,adder.add(1,2));}@TestpublicvoidsumReturnsSumOfTwoNegativeNumbers(){Adderadder=newAdder();assertEquals(-3,adder.add(-1,-2));}@TestpublicvoidsumReturnsSumOfLargeNumbers(){Adderadder=newAdder();assertEquals(2222,adder.add(1234,988));}}
| This sectiondoes notcite anysources. Please helpimprove this section byadding citations to reliable sources. Unsourced material may be challenged andremoved.(September 2019) (Learn how and when to remove this message) |
Using unit-tests as a design specification has one significant advantage over other design methods: The design document (the unit-tests themselves) can itself be used to verify the implementation. The tests will never pass unless the developer implements a solution according to the design.
Unit testing lacks some of the accessibility of a diagrammatic specification such as aUML diagram, but they may be generated from the unit test using automated tools. Most modern languages have free tools (usually available as extensions toIDEs). Free tools, like those based on thexUnit framework, outsource to another system the graphical rendering of a view for human consumption.[28]
Unit testing is the cornerstone ofextreme programming, which relies on an automatedunit testing framework. This automated unit testing framework can be either third party, e.g.,xUnit, or created within the development group.
Extreme programming uses the creation of unit tests fortest-driven development. The developer writes a unit test that exposes either a software requirement or a defect. This test will fail because either the requirement isn't implemented yet, or because it intentionally exposes a defect in the existing code. Then, the developer writes the simplest code to make the test, along with other tests, pass.
Most code in a system is unit tested, but not necessarily all paths through the code. Extreme programming mandates a "test everything that can possibly break" strategy, over the traditional "test every execution path" method. This leads developers to develop fewer tests than classical methods, but this isn't really a problem, more a restatement of fact, as classical methods have rarely ever been followed methodically enough for all execution paths to have been thoroughly tested.[citation needed] Extreme programming simply recognizes that testing is rarely exhaustive (because it is often too expensive and time-consuming to be economically viable) and provides guidance on how to effectively focus limited resources.
Crucially, the test code is considered a first class project artifact in that it is maintained at the same quality as the implementation code, with all duplication removed. Developers release unit testing code to the code repository in conjunction with the code it tests. Extreme programming's thorough unit testing allows the benefits mentioned above, such as simpler and more confident code development andrefactoring, simplified code integration, accurate documentation, and more modular designs. These unit tests are also constantly run as a form ofregression test.
Unit testing is also critical to the concept ofEmergent Design. As emergent design is heavily dependent upon refactoring, unit tests are an integral component.[citation needed]
An automated testing framework provides features for automating test execution and can accelerate writing and running tests. Frameworks have been developed fora wide variety of programming languages.
Generally, frameworks arethird-party; not distributed with a compiler orintegrated development environment (IDE).
Tests can be written without using a framework to exercise the code under test usingassertions,exception handling, and othercontrol flow mechanisms to verify behavior and report failure. Some note that testing without a framework is valuable since there is abarrier to entry for the adoption of a framework; that having some tests is better than none, but once a framework is in place, adding tests can be easier.[29]
In some frameworks advanced test features are missing and must be hand-coded.
Some programming languages directly support unit testing. Their grammar allows the direct declaration of unit tests without importing a library (whether third party or standard). Additionally, the Boolean conditions of the unit tests can be expressed in the same syntax as Boolean expressions used in non-unit test code, such as what is used forif andwhile statements.
Languages with built-in unit testing support include:
Languages with standard unit testing framework support include:
Some languages do not have built-in unit-testing support but have established unit testing libraries or frameworks. These languages include:
{{cite book}}:ISBN / Date incompatibility (help)The contractor shall then code and test software Units, and enter the source and object code, and associated listings of each successfully tested Unit into the Developmental Configuration