Automate everything involving MarkLogic with Gradle

ml-gradle is a Gradle plugin that can automate everything you do with MarkLogic. Deploy an application, add a host, backup a database, stub out a new project, load modules as you modify them, run an MLCP or CoRB or Data Movement job - if it involves MarkLogic, ml-gradle either automates it already or can be extended to do so.

You can use ml-gradle right away with the simple tutorial below, or learn more:

• The ml-gradle Wiki guides you through all of the ml-gradle documentation
• Read the Getting Started guide for more details on setting up a new project
• Browse the example projects for working examples of different ml-gradle features

Don't want to use Gradle? Then consider using the ml-app-deployer Java library on which ml-gradle depends. ml-app-deployer provides most of the functionality within ml-gradle without having any dependency on Gradle or Groovy - ml-gradle is then a fairly thin wrapper around ml-app-deployer to expose its functionality within a Gradle environment.

Compatibility guidelines

If you are using ml-gradle 4.6.0 or higher, you must use Gradle 7.6 or higher.

If you wish to use Gradle 7 or higher, you must use ml-gradle 4.3.0 or higher.

For older versions of ml-gradle, Gradle 6 or higher is recommended.

ml-gradle has been tested and used extensively on multiple versions of Gradle 7 and 8. As described below, both ml-gradle and the MarkLogic Java Client have been tested on Java 8, 11, and 17.

Start using ml-gradle

ml-gradle depends on at least Java 8 and MarkLogic 9 or higher, so if you have those installed, you're just a few minutes away from using ml-gradle to start a new project and deploy an application from it. Note that in addition to Java 8, you can use Java 11 or Java 17, but please see the MarkLogic Java Client documentation for information on dependencies you may need when using Java 11 or Java 17.

First, install Gradle.

Then, in an empty directory, create a file named "build.gradle" with your favorite text editor and enter the following:

plugins { id "com.marklogic.ml-gradle" version "4.8.0" }

Then run:

gradle mlNewProject

(If you are trying ml-gradle 4.5.0, you will need to do gradle -PmlUsername= -PmlPassword= mlNewProject due to a bug that has been fixed in ml-gradle 4.5.1).

This starts a project wizard to stub out files for your new application. You can accept all the defaults, but be sure to enter a valid port number for the "REST API port" question. ml-gradle will then print the following logging:

Updating build.gradle so that the Gradle properties plugin can be applied
Writing: build.gradle
Writing: gradle.properties
Writing: gradle-dev.properties
Writing: gradle-local.properties
Writing: gradle-qa.properties
Writing: gradle-prod.properties
Making directory: src/main/ml-config
Making directory: src/main/ml-modules
Writing project scaffolding files

You now have an ml-gradle project stubbed out with support for deploying to multiple environments via the Gradle properties plugin.

Now deploy it!

gradle mlDeploy

And you should see more ml-gradle logging like this:

:mlDeleteModuleTimestampsFile
:mlPrepareBundles
:mlDeployApp
:mlPostDeploy UP-TO-DATE
:mlDeploy
BUILD SUCCESSFUL

And once that's complete, you can go to the MarkLogic Admin UI on port 8001 to see the resources that have been created (the names of these resources start with the application name you selected in the project wizard, which defaults to myApp):

• Under App Servers, a new REST server named myApp on the port you chose
• Under Databases, a new content datase named myApp-content and a new modules database named myApp-modules
• Under Forests, 3 new forests for myApp-content and 1 new forest for myApp-modules
• Under Security/Users, 3 new users, each prefixed with myApp
• Under Security/Roles, 5 new roles, each prefixed with myApp

Congratulations! You've used ml-gradle to stub out a new project and deploy its application to MarkLogic. You're now ready to start adding more resources and modules to your project. See the links above this tutorial to learn more about using ml-gradle.