mod-connect

mod-connect is a CLI tool used to create ingestion pipelines that will regularly build and publish LST artifacts to your artifact repository. This will enable the Moderne agent to pick up the LSTs so that they can be used in the Moderne platform to run recipes against.

Setting up these pipelines will not affect or require changes to your existing workflows or pipelines.

Prerequisites

• You will need to have Java 8 or higher installed to build or run this tool.

Installation instructions

1. Go to the releases section of this repository and download the mod-connect.zip file found under the latest release.

2. Unzip the file by running:

   unzip mod-connect.zip

3. Add the binary to your $PATH:

   export PATH=$PATH:$(pwd)/mod-connect/bin

4. You should now have a mod-connect tool you can run use to run various CLI commands. Use the jenkins or github subcommand to configure your ingestion pipeline. For more information about the commands you can run the following command:

   mod-connect help

Usage instructions

mod-connect jenkins

This command will create a Jenkins Job for each configured repository that will build and publish LST artifacts to your artifact repository on a regular basis.

Before you can run this command, you will need to prepare a repos.csv file that follows this structure:

scmHost, repoName, branch, mavenTool, gradleTool, jdkTool, repoStyle, repoBuildAction, repoSkip, skipReason

Column	Required	Notes
scmHost	Optional	The URL of the source code management tool where the repository is hosted. Defaults to github.com.
repoName	Required	The repository that should be ingested. Follows the format of: organization/repository (e.g., google/guava).
branch	Optional	The branch of the above repository that should be ingested.
mavenTool	Optional	The name of the Maven tool that should be used to run Maven jobs. Specified in the Jenkins Global Tool Configuration page: {controllerUrl}/manage/configureTools/
gradleTool	Optional	The name of the Gradle tool that should be used to run Gradle jobs. Specified in the Jenkins Global Tool Configuration page: {controllerUrl}/manage/configureTools/
jdkTool	Optional	The name of the JDK tool that should be used to run JDK jobs. Specified in the Jenkins Global Tool Configuration page: {controllerUrl}/manage/configureTools/
repoStyle	Optional	The OpenRewrite style name to apply during ingestion.
repoBuildAction	Optional	Additional arguments that are added to the Maven or Gradle build command.
skip	Optional	If set to true, this repository will be skipped and not ingested.
skipReason	Optional	The context for why the repo is being skipped.

Once you've created the repos.csv file, you can set up the ingestion pipeline by running:

mod-connect jenkins --fromCsv repos.csv \
  --jenkinsUser $JENKINS_USER \
  --jenkinsPwd $JENKINS_PWD \
  --publishUrl $ARTIFACTORY_REPO_URL \
  --publishCredsId artifactCreds \
  --gitCredsId myGitCreds \
  --controllerUrl $JENKINS_URL

Note: You can specify defaults for the repositories in your repos.csv file by using the --defaultMaven, --defaultGradle, and --defaultJdk options. If you include those in your command, the defaults you specify for those parameters will be used when a row in your repos.csv does not include mavenTool, gradleTool, or jdkTool respectively.

mod-connect github

This command will create a GitHub workflow that builds and publishes LST artifacts to your artifact repository on a regular basis. A workflow can be created for ingesting a single repository (by specifying the --path parameter or by manually adding the moderne-publish-action to your repository) or a workflow can be created for ingesting a mass number of repositories (by specifying the --fromCsv parameter).

If you want to ingest a single repository

Please run the following command:

mod-connect github --path $my-repo-folder \
 --publishUserSecretName $ghSecretUserToPublish \
 --publishPwdSecretName $ghSecretPwdToPublish \
 --publishUrl $urlToPublish` 

If you want to ingest a mass number of repositories

1. Create a repos.csv file with the following columns:

repoName, branch, javaVersion, style, repoBuildAction, skip, skipReason

Column	Required	Notes
repoName	Required	The repository that should be ingested. Follows the format of organization/repositoryRepository (e.g., google/guava).
branch	Optional	The branch of the above repository that should be ingested.
javaVersion	Optional	The Java version used to compile this repository.
style	Optional	The OpenRewrite style name to apply during ingestion.
repoBuildAction	Optional	Additional arguments that are added to the Maven or Gradle build command.
skip	Optional	If set to true, this repository will be skipped and not ingested.
skipReason	Optional	The context for why the repo is being skipped.

2. Create and initialize a GitHub repository where this ingestion pipeline will be set up (e.g., acme/moderne-ingest).
3. Add the following secrets to the above repository:

Secret	Notes
PUBLISH_AST_USER	The username to publish into your Artifactory or Nexus.
PUBLISH_AST_PWD	The password to publish into your Artifactory or Nexus.
GH_PAT	The secret with read access to clone the repository.
GH_DISPATCH_PAT	The classic secret with write access restricted to the new repository to dispatch workflows.

4. Create a Personal Access Token (WORKFLOW_PAT) with write and workflow access to the new repository (or reuse the GH_DISPATCH_PAT) to automatically commit and run the workflows in the new repository.

5. Run the following command:

mod-connect github --fromCsv $my-csv \ 
  --publishUserSecretName PUBLISH_AST_USER \
  --publishPwdSecretName PUBLISH_AST_PWD \
  --artifactoRepoUrl $urlToPublish \
  --dispatchSecretName GH_DISPATCH_PAT \
  --repoReadSecretName GH_PAT \
  --repo $ingestRepo \
  --accessToken $WORKFLOW_PAT

mod-connect gitlab

This command will create a GitLab pipeline job for each configured repository to build and publish LST artifacts to your artifact repository.

Before you can run this command, you will need to prepare a repositories.csv file that follows this structure:

repoName, branch, repoStyle, repoBuildAction, repoSkip, skipReason

Column	Required	Notes
repoName	Required	The repository that should be ingested. Follows the format of: organization/repository (e.g., google/guava).
branch	Optional	The branch of the above repository that should be ingested.
repoStyle	Optional	The OpenRewrite style name to apply during ingestion.
additionalBuildArgs	Optional	Additional arguments that are added to the mod build command.
skip	Optional	If set to true, this repository will be skipped and not ingested.
skipReason	Optional	The context for why the repo is being skipped.

Once you've created the repos.csv file, you can set up the ingestion pipeline by running:

mod-connect gitlab \
  --fromCsv repos.csv \
  --publishUrl $PUBLISH_URL \
  --publishUserSecretName $PUBLISH_USER_SECRET \
  --publishPwdSecretName $PUBLISH_PASSWORD_SECRET \
  --repositoryAccessUserSecretName $GITLAB_ACCESS_USER_SECRET \
  --repositoryAccessTokenSecretName $GITLAB_ACCESS_TOKEN_SECRET \
  --dockerImageBuildJob $DOCKER_BUILD_IMAGE \
  --downloadCLIUrl $DOWNLOAD_CLI_URL

Note: This command should typically be run from (a fork of) moderne-gitlab-ingest

Development

If you wish to contribute to this project, please follow these instructions.

Prerequisites

In order to build this project you will need to have:

• Java 8 or higher installed on your machine
• Docker running (for the tests)

How to build and run mod-connect locally

1. Clone the project to your machine:

git clone git@github.com:moderneinc/mod-connect.git

2. Run the following command to build the project. This will compile the project, run the tests, and create a zip file in the build/distributions directory:

./gradlew build

3. With the zip file created, unzip it by running:

unzip build/distributions/mod-connect.zip

4. You should now have the latest build of the mod-connect tool locally!

cd mod-connect/bin
./mod-connect help