JenkinsPipelineUnit Testing Framework

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.

Table of Contents

1. Usage
2. Configuration
3. Declarative Pipeline
4. Testing Shared Libraries
5. Writing Testable Libraries
6. Note On CPS
7. Contributing
8. Demos and Examples

Usage

Add to Your Project as Test Dependency

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.

Maven

<repositories>
    <repository>
    <id>jenkins-ci-releases</id>
    <url>https://repo.jenkins-ci.org/releases/</url>
    </repository>
    ...
</repositories>

<dependencies>
    <dependency>
        <groupId>com.lesfurets</groupId>
        <artifactId>jenkins-pipeline-unit</artifactId>
        <version>1.9</version>
        <scope>test</scope>
    </dependency>
    ...
</dependencies>

Gradle

repositories {
    maven { url 'https://repo.jenkins-ci.org/releases/' }
    ...
}

dependencies {
    testImplementation "com.lesfurets:jenkins-pipeline-unit:1.9"
    ...
}

Start Writing Tests

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({$class=GitSCM, branches=[{name=feature_test}], extensions=[], userRemoteConfigs=[{credentialsId=gitlab_git_ssh, url=github.com/lesfurets/JenkinsPipelineUnit.git}]})
            utils.currentRevision()
               utils.sh({returnStdout=true, script=git rev-parse HEAD})
         exampleJob.gitlabBuilds({builds=[build, test]}, 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)

Mocking Jenkins Variables

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.

Mocking Jenkins Commands

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.

Mocking readFile and fileExists

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()
}

Mocking Shell Steps

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:

• A string for standard output
• A return code

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)
}

Analyzing the Mock Execution

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.

Checking Pipeline Status

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()
}

Checking Pipeline Exceptions

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!')
}

JUnit 4

class TestCase extends BasePipelineTest {
    @Test(expected = IllegalArgumentException)
    void verifyException() {
        runScript('Jenkinsfile')
    }
}

JUnit 5

import static org.junit.jupiter.api.Assertions.assertThrows

class TestCase extends BasePipelineTest {
    @Test
    void verifyException() {
        assertThrows(IllegalArgumentException) { runScript('Jenkinsfile') }
    }
}

Compare the Callstack with a Previous Implementation

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.

Preserve Original Callstack Argument References

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() {
    doSomethingWithThis(pretendArgsFromFarUpstream)
}

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

Running Inline Scripts

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!

Configuration

The abstract class BasePipelineTest configures the helper with useful conventions:

• It looks for pipeline scripts in your project in root (./.) and src/main/jenkins paths.
• Jenkins pipelines let you load other scripts from a parent script with load command. However load takes the full path relative to the project root. The test helper mock successfully the load command to load the scripts. To make relative paths work, you need to configure the path of the project where your