3.14.0

This release of ml-gradle includes significant changes within ml-app-deployer 3.14.0 for how databases are deployed. These changes were made for supporting resource files in bundles (see below) as well as laying the groundwork for further usage of the MarkLogic Configuration Management API for speeding up the deployment of applications by including as many resources as possible in a single request. The latter support should be available in the 3.15.0 releases of ml-gradle and ml-app-deployer.

All issues addressed by this release:

• #449 ServerEvalTask now has a client property to allow for customizing the DatabaseClient that it uses
• #451 Added mlNewTrigger for generating a new trigger file
• #453 mlInstallAdmin now accepts a parameter for the realm
• #454 Bundles now support an ml-config directory for resources. See Bundles for more information. Note that this includes support for merging certain resource types as well.
• #456 Multiple instances of DeleteCollectionsTask can now be invoked at once

3.13.0

Highlights from the full set of changes:

• #420 mlRestApi is now named mlBundle; see the docs on bundles for more information
• #446 Loading data is now part of a deployment, and data can also be loaded from a bundle
• #447 Installing system plugins is now part of a deployment, and plugins can also be loaded from a bundle
• #448 Schemas can now be loaded from a bundle
• And see the list of changes in ml-app-deployer 3.13.0

3.12.0

This release addresses the following items:

• #428 ml-gradle now resolves file paths correctly when run using Java 11 with the Gradle daemon enabled. See "mlIgnoreProjectDir" in the Property Reference if you need to disable this functionality because you're using absolute paths for property values.
• #423 A database and all of its forests can now be deleted via mlDeleteDatabase (which requires -Pconfirm=true as a safety measure)
• #421 MlcpTask now allows for a custom DatabaseClient to be used for logging output
• #419 DMSDK jobs that accept a "transform" parameter can now accept a transform name plus parameter names and values
• #418 CopySearchOptionsTask simplifies copying search options so that they're available via multiple REST servers
• #416 mlReloadSchemas was fixed so that it does not clear a content database
• Also includes enhancements from ml-app-deployer 3.12.0 and marklogic-data-movement-components 1.2.0

3.11.0

Summary of all issues:

• #415 Can now preview a deployment
• Support for faster deployment times via new properties for using the Configuration Management API (CMA) to deploy forests, amps, and privileges
• #400 Can now customize the DatabaseClient used by any DataMovementTask
• And see the release notes for ml-app-deployer 3.11.0 as well

3.10.2

This release addresses one small item:

• #409 WatchTask now has an extension point for DHF to use

3.10.1

This release contains a couple small enhancements in support of the upcoming DHF 4.1 release:

• #407 mlLoadSchemas now runs all commands in the "mlSchemaCommands" list
• #408 WatchTask now includes a method that can be overridden by a subclass which is called on every iteration of the loop
• Also includes ml-app-deployer 3.10.1, which is likewise in support of the upcoming DHF 4.1 release

3.10.0

Addresses 2 issues, along with a lot of changes in ml-app-deployer 3.10.0:

• #386 New mlDeleteResourceTimestampsFile task that is related to incremental deployments - see https://github.com/marklogic-community/ml-gradle/wiki/Incremental-deployment
• #394 The mlUnitTest task now depends on version 0.12.0 of marklogic-unit-test
• Initial support for incremental deployments to speed up deploying an application)
• And be sure to see the release notes for ml-app-deployer 3.10.0 which has some significant changes

3.9.0

Highlights from the full set of issues:

• #126 mlReloadSchemas no longer clears the schemas database; mlDeleteUserSchemas is instead used so that schema documents created via the deployment of resources such as temporal collections and view schemas are not deleted.
• #371 mlExportToFile can now write JSON documents to a valid JSON array, thanks to improvements in the marklogic-data-movement-components 1.1 release
• #372 mlNewProject no longer erroneously creates a REST API configuration file when no REST port is specified
• Improvements from ml-app-deployer 3.9.0

3.8.3

Bugfix release that provides a few fixes from ml-app-deployer and one small new feature in ml-gradle:

• #368 mlWatch now supports a callback for after modules are loaded
• Includes bug fixes from ml-app-deployer 3.8.3

3.8.2

This release has no changes in ml-gradle itself; it just depends on version 3.8.2 of ml-app-deployer which has a fix for allowing Data Hub Framework users to depend on a version of ml-app-deployer/ml-gradle higher than 3.6.3.