Indy Jenkins Pipeline Library

The repository contains a library of reusable Jenkins Pipeline steps and functions for that are used in Hyperledger Indy projects' CI/CD pipelines.

Indy Pipeline Steps

Indy Pipeline Steps

indyAutoMergePR

merges PR if changes consists only of updates for dependencies versions
makes sense for PR jobs only

    indyAutoMergePR {
        node = 'node-label-with-linux-docker'
                // default: indyConfig.indyNodeLabel
        prDetails = [
            owner: 'github-repo-owner',
                // default: parsed from env.CHANGE_URL or env.ghprbPullLink
            repo: 'github-repo-name',
                // default: parsed from env.CHANGE_URL or env.ghprbPullLink
            number: 1, // PR number
                // default: env.CHANGE_ID or env.ghprbPullId
        ]
        credentialsId = 'GitHub-token-with-PRs-write-permissions'
                // default: indyConfig.credentials.gitHubToken
        approveMessage = 'comment-in-PR'
                // default: 'Approved'
        statusState = 'pending|success|error|failure'
                // default: 'success'
        statusUrl = 'status-url'
                // default: env.RUN_DISPLAY_URL
        statusDescr = 'status-description'
                // default: 'Success: This commit looks good for auto merge'
        statusContext = 'status-context'
                // default: indyConfig.prStatusContext
    }

indyCleanWs

wrapper for cleanWS step
doesn't fail if ws-clenup plugin is not installed

    indyCleanWs()

indyConfig

sets indy pipelines environment
most of fields could be conigured by environment variables
returns a Map with the following fields
- rsBasePath: classpath to resources, org/hyperledger/pipeline/indy, non-configurable
- steps: a list of Maps, where key is name of one of indy steps and values are:
  - active: is step active of not, depends on env.$stepName (default: true)
  - ref: groovy closure reference to the step, depends on indyStagesInit (default: indy step closure)
- failFast: turns on/off failFast mode for parallel steps, depends on env.indyFailFast (default: false)
- dryRun: turns on/off dry run mode, depends on env.indyDryRun (default: false)
- logLevel: sets logging level, depends on env.indyLogLevel (default: 'DEBUG')
- indyNodeLabel: node label where linux docker container could be run, depends on env.indyNodeLabel (default: null)
- credentials:
  - gitHubToken: GitHub token with rights to add comments and set statuses for PRs and merge them, depends on env.indyGitHubTokenCredId (default: null)
- prStatusContext: PR status set for the PR by indyAutoMergePR, depends on env.indyPRStatusContext (default: null)

indyIsTested

check if current commit has been already tested in scope of any PR
makes sense for branch jobs only

    indyIsTested {
        node = 'node-label-with-linux-docker'
                // default: indyConfig.indyNodeLabel
        branch = 'PRs-target-branch-filter'
                // default: env.$BRANCH_NAME
        contexts = ['PRs-context-filters', ...] //default: ['continuous-integration/jenkins/pr-merge']
        age = 5      // filter PRs by number of days from the last update, default: 7
                // default: 7
    }

indyJNode

wrapper for node step
guarantees that indyCleanWs will be called after node is finished

    indyJNode(nodeLabel)

indyLogger

defines a set of log levels and provides API for each
log levels: TRACE, DEBUG, INFO, WARNING, ERROR
methods:
- trace(message, prefix=null)
- debug(message, prefix=null)
- info(message, prefix=null)
- warning(message, prefix=null)
- error(message, prefix=null)

indyNotify

provides API to send notifications

email

    indyNotify.email message

Please check docs for details about message attributes.

Default message:

    [
        body: '$DEFAULT_CONTENT',
        replyTo: '$DEFAULT_REPLYTO',
        subject: '$DEFAULT_SUBJECT',
        to: '$DEFAULT_RECIPIENTS'
    ]

slack

      indyNotify.slack message

Please check docs for details about message attributes.

indyPackaging

performs packaging calling package builders
each builder should return volumeOrPath - docker volume name or host system absolute path to the directory where package(s) is(are) placed

    indyPackaging {
        node = 'node-label-with-linux-docker'
                // default: indyConfig.indyNodeLabel
        builders = [deb: debBuilderClosure]
                // default: not-defined
        version = '1.2.3'
                // default: not-defined
    }

indyPipeline

calls steps in order defined via parameters
optionally limit execution time
optionally call failure and/or success closure callbacks

    indyPipeline {
        :
        timeout = 90 // minutes
                // default: not defined
        onFail = onFailCallback
                // default: not defined
        onSuccess = onSuccessCallback
                // default: not defined
        // format = stageName, stageConfigClosure],
        // config closures are empty by default
        stages = [
            ['indyVerifyStatic', {}],
            ['indyVerify', {
                ...
            }],
            ['indyAutoMergePR', {}],
            ['indyPackaging', {
                version = "1.2.3"
                builders.deb = buildDebUbuntu
            }]
        ]
                // default: all steps from indyConfig with dummy closures {}
    }

indyPublish

dummy step, just a definition of API
expected that it is implemented in another shared library
please, use indyStagesInit to pass the reference to the actual implementation

    indyPublish {
        packageName = 'myPackage'
            // default: not defined
        releaseVersion = '1.2.3'
            // default: not defined
        projectName = 'myProject'
            // default: not defined
        builders = [deb: debBuilderClosure]
            // default: not defined
    }

indyVerify

runs unit/integration tests

    indyVerify {
        labels = ['linux', 'windows']
                // default: not defined
        tests = [
            common: [
                resFile: { "test-result-filename.${NODE_NAME}.xml" },
                    // default: 'test-result.txt'
                testDir: 'indy_common',
                    // default: '.'
                python: 'python-executable-to-use',
                    // default: 'python'
                useRunner: true,    // run or not the custom runner for tests
                    // default: false
                docker: 'ubuntu' // reference to item in 'dockers' 
                    // default: not defined
            ],
            // ...
        ]
                // default: not defined
        dockers = [
            ubuntu: [
                imgName: 'ci-image-name',
                    // default: not defined
                dockerfile: "path-to-dockerfile",
                    // default: not defined
                contextDir: "path-to-docker-build-context-dir"
                    // default: not defined
            ]
        ]
                // default: not defined

    }

indyVerifyStatic

runs flake8 based static code verification

    indyVerifyStatic {
        node = 'node-label-with-linux-docker'
                // default: indyConfig.indyNodeLabel
        dockerEnv = [
            imgName: 'ci-image-name',
                // default: 'code-validation'
            dockerfile: "path-to-dockerfile",
                // default: 'ci/code-validation.dockerfile'
            contextDir: "path-to-docker-build-context-dir"
                // default: 'ci'
        ]
    }

indyStagesInit

virtual step: could be used by other shared libraries to set (replace) references for standard steps
should return a Map:
- key - indy step
- value - step closure reference

e.g.

    def call() {
        return [
            indyPublish: myLibraryPublishStep
        ]
    }

hyperledger-archives / indy-jenkins-pipeline-lib Goto Github PK

indy-jenkins-pipeline-lib's Introduction