You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session.
Framework for unit testing Jenkins pipelines
Notifications You must be signed in to change notification settings
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Go to file
Jenkins Pipeline Unit is a testing framework for unit testing Jenkins pipelines, written in Groovy Pipeline DSL. If you use Jenkins as your CI workhorse (like us @ lesfurets.com) and you enjoy writing pipeline-as-code, you already know that pipeline code is very powerful but can get pretty complex. This testing framework lets you write unit tests on the configuration and conditional logic of the pipeline code, by providing a mock execution of the pipeline. You can mock built-in Jenkins commands, job configurations, see the stacktrace of the whole execution and even track regressions.
JenkinsPipelineUnit requires Java 11, since this is also the minimum version required by Jenkins. Also note that JenkinsPipelineUnit is not currently compatible with Groovy 4, please see this issue for more details.
Note: Starting from version 1.2 , artifacts are published to https://repo.jenkins-ci.org/releases .
repositories> repository> id>jenkins-ci-releasesid> url>https://repo.jenkins-ci.org/releases/url> repository> . repositories> dependencies> dependency> groupId>com.lesfuretsgroupId> artifactId>jenkins-pipeline-unitartifactId> version>1.9version> scope>testscope> dependency> . dependencies>
repositories < maven < url 'https://repo.jenkins-ci.org/releases/' > ... > dependencies < testImplementation "com.lesfurets:jenkins-pipeline-unit:1.9" ... >
You can write your tests in Groovy or Java, using the test framework you prefer. The easiest entry point is extending the abstract class BasePipelineTest , which initializes the framework with JUnit.
Let's say you wrote this awesome pipeline script, which builds and tests your project:
def execute() < node() < String utils = load 'src/test/jenkins/lib/utils.jenkins' String revision = stage('Checkout') < checkout scm return utils.currentRevision() > gitlabBuilds(builds: ['build', 'test']) < stage('build') < gitlabCommitStatus('build') < sh "mvn clean package -DskipTests -DgitRevision=$revision" > > stage('test') < gitlabCommitStatus('test') < sh "mvn verify -DgitRevision=$revision" > > > > > return this
Now using the Jenkins Pipeline Unit you can write a unit test to see if it does the job:
import com.lesfurets.jenkins.unit.BasePipelineTest class TestExampleJob extends BasePipelineTest < @Test void shouldExecuteWithoutErrors() < loadScript('job/exampleJob.jenkins').execute() printCallStack() > >
This test will print the call stack of the execution, which should look like so:
exampleJob.run() exampleJob.execute() exampleJob.node(groovy.lang.Closure) exampleJob.load(src/test/jenkins/lib/utils.jenkins) utils.run() exampleJob.stage(Checkout, groovy.lang.Closure) exampleJob.checkout(], extensions=[], userRemoteConfigs=[]>) utils.currentRevision() utils.sh() exampleJob.gitlabBuilds(, groovy.lang.Closure) exampleJob.stage(build, groovy.lang.Closure) exampleJob.gitlabCommitStatus(build, groovy.lang.Closure) exampleJob.sh(mvn clean package -DskipTests -DgitRevision=bcc19744) exampleJob.stage(test, groovy.lang.Closure) exampleJob.gitlabCommitStatus(test, groovy.lang.Closure) exampleJob.sh(mvn verify -DgitRevision=bcc19744)
You can define both environment variables and job execution parameters.
import com.lesfurets.jenkins.unit.BasePipelineTest class TestExampleJob extends BasePipelineTest < @Override @BeforeEach void setUp() < super.setUp() // Assigns false to a job parameter ENABLE_TEST_STAGE addParam('ENABLE_TEST_STAGE', 'false') // Assigns 1.0.0-rc.1 to the environment variable TAG_NAME addEnvVar('TAG_NAME', '1.0.0-rc.1') // Defines the previous execution status binding.getVariable('currentBuild').previousBuild = [result: 'UNSTABLE'] > @Test void verifyParam() < assertEquals('false', binding.getVariable('params')['ENABLE_TEST_STAGE']) > >
After calling super.setUp() , the test helper instance is available, as well as many helper methods. The test helper already provides basic variables such as a very simple currentBuild definition. You can redefine them as you wish.
Note that super.setUp() must be called prior to using most features. This is commonly done using your own setUp method, decorated with @Override and @BeforeEach .
Parameters added via addParam are immutable, which reflects the same behavior in Jenkins. Attempting to modify the params map in the binding will result in an error.
You can register interceptors to mock pipeline methods, including Jenkins commands, which may or may not return a result.
import com.lesfurets.jenkins.unit.BasePipelineTest class TestExampleJob extends BasePipelineTest < @Override @BeforeEach void setUp() < super.setUp() helper.registerAllowedMethod('sh', [Map]) < args -> return 'bcc19744' > helper.registerAllowedMethod('timeout', [Map, Closure], null) helper.registerAllowedMethod('timestamps', []) < println 'Printing timestamp' > helper.registerAllowedMethod('myMethod', [String, int]) < String s, int i -> println "Executing myMethod mock with args: '$s>', '$i>'" > > >
The test helper already includes mocks for all base pipeline steps as well as a steps from a few widely-used plugins. You need to register allowed methods if you want to override these mocks and add others. Note that you need to provide a method signature and a callback (closure or lambda) in order to allow a method. Any method call which is not recognized will throw an exception.
Please refer to the BasePipelineTest class for the list of currently supported mocks.
Some tricky methods such as load and parallel are implemented directly in the helper. If you want to override those, make sure that you extend the PipelineTestHelper class.
The readFile and fileExists steps can be mocked to return a specific result for a given file name. This can be useful for testing pipelines for which file operations can influence subsequent steps. An example of such a testing scenario follows:
// Jenkinsfile node < stage('Process output') < if (fileExists('output') && readFile('output') == 'FAILED. ') < currentBuild.result = 'FAILURE' error 'Build failed' > > >
@Test void exampleReadFileTest() < helper.addFileExistsMock('output', true) helper.addReadFileMock('output', 'FAILED. ') runScript('Jenkinsfile') assertJobStatusFailure() >
The shell steps ( sh , bat , etc) are used by many pipelines for a variety of tasks. They can be mocked to either (a) statically return:
Or (b), to execute a closure that returns a Map (with stdout and exitValue entries). The closure will be executed when the shell is called, allowing for dynamic behavior.
Here is a sample pipeline and corresponding unit tests for each of these variants.
// Jenkinsfile node < stage('Mock build') < String systemType = sh(returnStdout: true, script: 'uname') if (systemType == 'Debian') < sh './build.sh --release' int status = sh(returnStatus: true, script: './test.sh') if (status > 0) < currentBuild.result = 'UNSTABLE' > else < def result = sh( returnStdout: true, script: './processTestResults.sh --platform debian', ) if (!result.endsWith('SUCCESS')) < currentBuild.result = 'FAILURE' error 'Build failed!' > > > > >
@Test void debianBuildSuccess() < helper.addShMock('uname', 'Debian', 0) helper.addShMock('./build.sh --release', '', 0) helper.addShMock('./test.sh', '', 0) // Have the sh mock execute the closure when the corresponding script is run: helper.addShMock('./processTestResults.sh --platform debian') < script -> // Do something "dynamically" first. return [stdout: "Executing $script>: SUCCESS", exitValue: 0] > runScript("Jenkinsfile") assertJobStatusSuccess() > @Test void debianBuildUnstable() < helper.addShMock('uname', 'Debian', 0) helper.addShMock('./build.sh --release', '', 0) helper.addShMock('./test.sh', '', 1) runScript('Jenkinsfile') assertJobStatusUnstable() >
Note that in all cases, the script executed by sh must exactly match the string passed to helper.addShMock , including the script arguments, whitespace, etc. For more flexible matching, you can use a pattern (regular expression) and even capture groups:
helper.addShMock(~/.\/build.sh\s--(.*)/) < String script, String arg -> assert (arg == 'debug') || (arg == 'release') return [stdout: '', exitValue: 2] >
Also, mocks are stacked, so if two mocks match a call, the last one wins. Combined with a match-everything mock, you can tighten your tests a bit:
@BeforeEach void setUp() < super.setUp() helper = new PipelineTestHelper() // Basic `sh` mock setup: // - generate an error on unexpected calls // - ignore any echo (debug) outputs, they are not relevant // - all further shell mocks are configured in the test helper.addShMock() < throw new Exception('Unexpected sh call') > helper.addShMock(~/echo\s.*/, '', 0) >
The helper registers every method call to provide a stacktrace of the mock execution.
@Test void shouldExecuteWithoutErrors() < runScript('Jenkinsfile') assertJobStatusSuccess() assertThat(helper.callStack.findAll < call -> call.methodName == 'sh' >.any < call -> callArgsToString(call).contains('mvn verify') >).isTrue() >
This will also check that mvn verify was called during the job execution.
Let's say you have a simple script, and you'd like to check its behavior if a step fails.
// Jenkinsfile node() < git 'some_repo_url' sh 'make' >
You can mock the sh step to just update the pipeline status to FAILURE . To verify that your pipeline is failing, you need to check the status with BasePipelineTest.assertJobStatusFailure() .
@Test void checkBuildStatus() < helper.registerAllowedMethod('sh', [String]) < cmd -> if (cmd == 'make') < binding.getVariable('currentBuild').result = 'FAILURE' > > runScript('Jenkinsfile') assertJobStatusFailure() >
Sometimes it is useful to verify that a specific exception was thrown during the pipeline run. JUnit 4 and 5 have slightly different mechanisms for doing this.
For both examples below, assume that the following pipeline is being tested:
To do so you can use org.junit.rules.ExpectedException
// Jenkinsfile node < throw new IllegalArgumentException('oh no!') >
class TestCase extends BasePipelineTest < @Test(expected = IllegalArgumentException) void verifyException() < runScript('Jenkinsfile') > >
import static org.junit.jupiter.api.Assertions.assertThrows class TestCase extends BasePipelineTest < @Test void verifyException() < assertThrows(IllegalArgumentException) < runScript('Jenkinsfile') > > >
One other use of the callstacks is to check your pipeline executions for possible regressions. You have a dedicated method you can call if you extend BaseRegressionTest :
@Test void testPipelineNonRegression() < loadScript('job/exampleJob.jenkins').execute() super.testNonRegression('example') >
This will compare the current callstack of the job to the one you have in a text callstack reference file. To update this file with new callstack, just set this JVM argument when running your tests: -Dpipeline.stack.write=true . You then can go ahead and commit this change in your SCM to check in the change.
The default behavior of the callstack capture is to clone each call's arguments to preserve their values at time of the call should those arguments mutate downstream. That is a good guard when your scripts are passing ordinary mutable variables as arguments.
However, argument types that are not Cloneable are captured as String values. Most of the time this is a perfect fallback. But for some complex types, or for types that don't implement toString() , it can be tricky or impossible to validate the call values in a test.
Take the following simple example:
Map pretendArgsFromFarUpstream = [ foo: 'bar', foo2: 'more bar please', aNestedMap: [aa: 1, bb: 2], plusAList: [1, 2, 3, 4], ].asImmutable() node()pretendArgsFromFarUpstream is an immutable map and will be recorded as a String in the callstack. Your test may want to perform fine-grained validations via map key referencing instead of pattern matching or similar parsing. For example:
assertEquals(2, arg.aNestedMap.bb)
You may want to perform this kind of validation, particularly if your pipelines pass final and/or immutable variables as arguments. You can retain the direct reference to the variable in the callstack by setting this switch in your test setup:
helper.cloneArgsOnMethodCallRegistration = false
In case you want to have some script executed directly within a test case rather than creating a resource file for it, loadInlineScript and runInlineScript can be used.
@Test void testSomeScript() < Object script = loadInlineScript(''' node stage('Build') sh 'make' > > ''') script.execute() printCallStack() assertJobStatusSuccess() >
Note that inline scripts cannot be debugged via breakpoints as there is no file to attach to!
The abstract class BasePipelineTest configures the helper with useful conventions:
Overriding these default values is easy:
class TestExampleJob extends BasePipelineTest < @Override @BeforeEach void setUp() < baseScriptRoot = 'jenkinsJobs' scriptRoots += 'src/main/groovy' scriptExtension = 'pipeline' super.setUp() > >
This will work fine for such a project structure:
jenkinsJobs └── src ├── main │ └── groovy │ └── ExampleJob.pipeline └── test └── groovy └── TestExampleJob.groovy
To test a declarative pipeline, you'll need to subclass the DeclarativePipelineTest class instead of BasePipelineTest
// Jenkinsfile pipeline < agent none stages < stage('Example Build') < agent < docker 'maven:3-alpine' > steps < echo 'Hello, Maven' sh 'mvn --version' > > stage('Example Test') < agent < docker 'openjdk:8-jre' > steps < echo 'Hello, JDK' sh 'java -version' > > > >
import com.lesfurets.jenkins.unit.declarative.* class TestExampleDeclarativeJob extends DeclarativePipelineTest < @Test void shouldExecuteWithoutErrors() < runScript("Jenkinsfile") assertJobStatusSuccess() printCallStack() > >
The DeclarativePipelineTest class extends BasePipelineTest , so you can verify your declarative job the same way as scripted pipelines.
With Shared Libraries, Jenkins lets you share common code from pipelines across different repositories. Shared libraries are configured in the Jenkins settings and imported with @Library annotation or the library step.
Testing pipeline scripts using external libraries is not trivial because the shared library code is another repository. JenkinsPipelineUnit lets you test shared libraries and pipelines that depend on these libraries.
Here is an example pipeline using a shared library:
@Library('commons') import net.courtanet.jenkins.Utils sayHello 'World' node() < stage ('Checkout') < def utils = new Utils() checkout "$utils.gitTools()>" > stage ('Build') < sh './gradlew build' > stage ('Post Build') < String json = libraryResource 'net/courtanet/jenkins/request.json' sh "curl -H 'Content-Type: application/json' -X POST -d '$json' $acme.url>" > >
This pipeline is using a shared library called commons . Now let's test it:
// You need to import the class first import static com.lesfurets.jenkins.unit.global.lib.LibraryConfiguration.library class TestCase extends BasePipelineTest < @Test void testLibrary() < Object library = library() .name('commons') .retriever(gitSource('git@example.com:libs/commons.git')) .targetPath('path/to/clone') .defaultVersion("master") .allowOverride(true) .implicit(false) .build() helper.registerSharedLibrary(library) runScript('job/library/exampleJob.jenkins') printCallStack() > >
Notice how the shared library is defined and registered to the helper. The library definition is done via a fluent API which lets you set the same configurations as in Jenkins Global Pipeline Libraries.
The retriever and targetPath fields tell the framework how to fetch the sources of the library, and to which local path. The framework comes with two naive but useful retrievers, gitSource and localSource . You can write your own retriever by implementing the SourceRetriever interface.
Note that properties defaultVersion , allowOverride and implicit are optional with default values master , true and false .
Now if we execute this test, the framework will fetch the sources from the Git repository and load classes, scripts, global variables and resources found in the library. The callstack of this execution will look like the following:
Loading shared library commons with version master libraryJob.run() libraryJob.sayHello(World) sayHello.echo(Hello, World.) libraryJob.node(groovy.lang.Closure) libraryJob.stage(Checkout, groovy.lang.Closure) Utils.gitTools() libraryJob.checkout() libraryJob.stage(Build, groovy.lang.Closure) libraryJob.sh(./gradlew build) libraryJob.stage(Post Build, groovy.lang.Closure) libraryJob.libraryResource(net/courtanet/jenkins/request.json) libraryJob.sh(curl -H 'Content-Type: application/json' -X POST -d '' http://acme.com) There are a few types of SourceRetriever implementations in addition to the previously described GitSource one.
The ProjectSource retriever is useful if you write tests for the library itself. So it lets you load the library files directly from the project root folder (where the src and vars folders are located).
Then you can use projectSource to point to the location of the library files. Calling projectSource() with no arguments will look for files in the project root. With .defaultVersion('') , you can load it in pipelines using commons@master or commons@features which would use the same repository.
import static com.lesfurets.jenkins.unit.global.lib.ProjectSource.projectSource class TestCase extends BasePipelineTest < @Override @BeforeEach void setUp() < super.setUp() Object library = library() .name('commons') .defaultVersion('') .allowOverride(true) .implicit(true) .targetPath('') .retriever(projectSource()) .build() helper.registerSharedLibrary(library) > >
The LocalSource retriever is useful if you want to verify how well your library integrates with the pipelines. For example you may use pre-copied library files with different versions.
import static com.lesfurets.jenkins.unit.global.lib.LocalSource.localSource class TestCase extends BasePipelineTest < @Override @BeforeEach void setUp() < super.setUp() Object library = library() .name('commons') .defaultVersion('master') .allowOverride(true) .implicit(false) .targetPath('') .retriever(localSource('/var/tmp/')) .build() helper.registerSharedLibrary(library) > >
In the above example, the retriever would assume that the library files are located at /var/tmp/commons@master .
There is partial support for loading dynamic libraries. It doesn't implement all the features, but it could be useful sometimes.
Object commonsLib = library 'commons' // Assume that `sayHello` is a singleton in the `commons` library sayHello 'World' // Create an instance of a class in the `commons` library Object utils = net.courtanet.jenkins.Utils.new()
Test class example:
@Test void testDynamicLibrary() < Object library = library() .name('commons') .retriever(gitSource('git@example.com:libs/commons.git')) .targetPath('path/to/clone') .defaultVersion('master') .allowOverride(true) .implicit(false) .build() helper.registerSharedLibrary(library) // Registration for pipeline method 'library' must be made after registering the // shared library. Unfortunately, this cannot be moved to the super class. helper.registerAllowedMethod('library', [String], < String name -> helper.getLibLoader().loadLibrary(name) println helper.getLibLoader().libRecords return new LibClassLoader(helper, null) >) loadScript('job/library/exampleJob.jenkins') printCallStack() >
You might have a library that defines global variables with library class instances as arguments. For example, consider the following library class and global variable:
// src/com/example/Monster.groovy package com.example class Monster < String moniker Monster(String moniker) < this.moniker = moniker > >
// vars/monster.groovy import com.example.Monster void call(Monster monster) < println "$monster.moniker> is always very scary" >
Your pipeline uses both as follows:
Monster vampire = new Monster('Dracula') monster(vampire) // Should print "Dracula is always very scary"
If this does not yield the expected output but instead throws a MissingMethodException with the cause No signature of method: Jenkinsfile.monster() is applicable for argument types: (org.test.Monster) values: [com.example.Monster1@d34db33f] you may need to disable library class preloading in your test setup, which you do with the following switch.
helper.libLoader.preloadLibraryClasses = false
You may need to do this on a test-by-test basis, as disabling class preloading can cause problems in other cases. For example, when you have library classes that require access to the env global.
We recommend the following best-practices for organizing pipeline code:
In general, it's better to avoid having complex build logic inside of build pipelines. Although tools like JenkinsPipelineUnit are useful in testing pipelines, it's much easier to run build scripts locally (meaning, outside of a Jenkins environment). Languages like Python have much more sophisticated linting and testing tools than Groovy does.
That said, CodeNarc can be used to lint Groovy code, including Jenkinsfile files.
We recommend organizing pipeline libraries such that the bulk of the logic is organized into classes, and the singletons being thin wrappers around these classes. This approach has several advantages:
Let's say we have a library responsible for a very complex operation, in this case, adding two numbers together. 😄 Here's what that library (let's call it HardMath ) might look like:
// src/com/example/HardMath.groovy package com.example class HardMath implements Serializable < // Jenkinsfile script context, note that all pipeline steps must use this context Object script = null int complexOperation(int a, int b) < // Note the script context is required for `echo`, as it is a pipeline step script.echo "Adding $a> to $b>" return a + b > >
// vars/hardmath.groovy import com.example.HardMath int complexOperation(int a, int b) < return new HardMath(script: this).complexOperation(a, b) >
// test/com/example/HardMathTest.groovy package com.example import static org.junit.jupiter.api.Assertions.assertEquals import com.lesfurets.jenkins.unit.BasePipelineTest import org.junit.jupiter.api.BeforeEach import org.junit.jupiter.api.Test class HardMathTest extends BasePipelineTest < Object script = null @Override @BeforeEach void setUp() < super.setUp() this.script = loadScript('test/resources/EmptyPipeline.groovy') > @Test void testComplexOperation() < int result = new HardMath(script: script).complexOperation(1, 3) assertEquals(4, result) > >
// test/resources/EmptyPipeline.groovy return this
And finally, in some other project's Jenkinsfile :
// Jenkinsfile @Library('hardmath') node < stage('Hard Math') < int result = hardmath.complexOperation(5, 6) echo "The result is $result>" > >
For a larger real-world example of a pipeline library organized like the above and tested with JenkinsPipelineUnit, have a look at python-pipeline-utils .
If you already fiddled with Jenkins pipeline DSL, you may have experienced strange errors during execution on Jenkins. This is because Jenkins does not directly execute your pipeline in Groovy, but transforms the pipeline code into an intermediate format in order to run Groovy code in Continuation Passing Style (CPS).
The usual errors are partly due to the the sandboxing Jenkins applies for security reasons, and partly due to the serializability Jenkins imposes.
Jenkins requires that at each execution step, the whole script context is serializable, in order to stop and resume the job execution. To simulate this aspect, CPS versions of the helpers transform your scripts into the CPS format and check if at each step your script context is serializable.
To use this experimental feature, you can use the abstract class BasePipelineTestCPS instead of BasePipelineTest . You may see some changes in the callstacks that the helper registers. Note also that the serialization used to test is not the same as what Jenkins uses. You may find some incoherence in that respect.
JenkinsPipelineUnit aims to help developers code and test Jenkins pipelines with a shorter development cycle. It addresses some of the requirements traced in JENKINS-33925. If you are willing to contribute please don't hesitate to discuss in issues and open a pull-request.
| URL | Frameworks and Tools | Test Subject | Test Layers |
|---|---|---|---|
| https://github.com/macg33zr/pipelineUnit | Spock, Gradle(Groovy) | Jenkinsfile, scripted pipeline, SharedLibrary | UnitTest |
| https://github.com/mkobit/jenkins-pipeline-shared-library-example | Spock, Gradle (Kotlin), Junit | SharedLibrary | Integration, Unit |
| https://github.com/stchar/pipeline-sharedlib-testharness | Junit, Gradle(Groovy) | SharedLibrary | Integration, Unit |
| https://github.com/stchar/pipeline-dsl-seed | Junit, Spock, Gradle(Groovy) | scripted pipeline | Integration(jobdsl), Unit |
| https://github.com/SpencerMalone/JenkinsPipelineIntegration | Spock, Gradle(Groovy) | SharedLibrary | Integration |
| https://github.com/venosov/jenkins-pipeline-shared-library-example-victor | Junit, Gradle(Kotlin) | SharedLibrary | Unit |
