Skip to content

Jenkins Multiple GitHub Checkout#

Introduction#

Jenkins Multiple GitHub Checkout provides a Groovy Library for use in your Jenkins pipelines for cloning and checking out multiple Git repositories into your workspace.

Requirements#

  • Setup a jenkins instance
  • Create a new github git repo holding a shared library.
  • Use the shared library from a Jenkins pipeline with one stage
  • The shared lib functionality: clone from GitHub

Its arguments/capabilities:

  • Clone from private repo using creds given by the Jenkins pipeline in a secured way

  • Clone list of repos – list of URLs.

    Only in case of a list with more than 1 repo, place each one in a dedicated folder. In this case the folder name is the folder repo name.

Jenkins Configuration#

Manage Jenkins -> Global Trusted Pipeline Libraries

Jenkins SSH Credential

Usage#

See the groovydoc for API details.

Sample Jenkins Pipeline using Jenkins Multiple GitHub Checkout

@Library('JenkinsMultipleGitHubCheckout') _

pipeline {
    agent any

    stages {
        stage('MultiClone') {
            steps {
                 multiClone(
                    [credentialsId: "JenkinsTest",
                     repoUrls: ['https://github.com/DaverSomethingSomethingOrg/conan-system-packaging.git',
                               'https://github.com/DaverSomethingSomethingOrg/conan-github-workflows.git'
                     ]])
                 multiClone(
                    [credentialsId: "JenkinsTest",
                     repoUrls: ['https://github.com/DaverSomethingSomethingOrg/conan-build-container.git'
                     ]])
            }
        }
    }
}

Documentation#

DevContainer#

An Ubuntu-based DevContainer configuration is provided loaded with groovy, groovydoc, and mkdocs. The Docker file is in .devcontainer/Dockerfile.

MkDocs and groovydoc#

This Documentation site is generated using MkDocs.

MkDocs currently doesn't have any ability to publish Groovy docstrings directly, and it looks like there are no plans to provide such.

For now I'm simply using groovydoc and sending the output into the MkDocs tree directly without formatting.

The provided script generateDocsRelease.sh takes care of regenerating the groovydoc site under docs/ prior to deploying the MkDocs site.