New features in 7.2

Clover AI Assistant in Wrangler

Clover AI Assistant offers productivity boost to all users working CloverDX Wrangler. It allows you to create your Wrangler transformations by simply asking the Assistant what you’d like to do with your data. It will then figure out the steps needed to get the results you are looking for. It will also automatically generate block descriptions helping you create clean and understandable jobs.

The Assistant can also generate job summaries or even answer questions about your data sets, steps usage and more.

With the introduction of the AI Assistant, we’ve updated the structure of Wrangler jobs and made various smaller improvement to transformation steps. Wrangler jobs now contain blocks which allow you to lead conversation with the Assistant and come back to it at any time to continue building your job or simply review your past chat.

Working with transactional data sets in Wrangler

You can now read and write data from/to transactional data sets using new data sources and data targets that will be generated for you automatically when you create new data sets.

All transactional data sets will have two entries in the Data Catalog – one entry for data source and one for data target. The sources and targets will respect your permissions and will not show if you do not have any access to the data set.

With this feature, you will be able to easily ingest data into your Data Manager for further processing. All data quality errors that are flagged by Wrangler when the job is running, will be automatically pushed to the Data Manager where the users will see the error detail and will have the chance to fix the data.

The targets will also offer mappings to ensure that the data layout you are writing is the same as needed by each data set.

Data Manager improvements

This release brings many improvements in Data Manager that will help you deliver more complete experience to Data Manager users.

You can now configure columns to be of a URL data type. URL-typed columns allow you add a web address as the value and will be shown as clickable links in the Data Manager’s data grid.

This functionality will allow you to, for example, show a picture of an invoice or link to an external system providing additional information about the row.

You can now also use Data Services to provide real time suggestions in Data Manager. Instead of submitting static suggestions as part of your data validation logic, you can use URL to a Data Service which will be called whenever user clicks on given cell in Data Manager. This allows you to provide up-to-date value recommendations instead of relying on static values.

To make sure you can easily use Data Services in Data Manager, we have introduced a new “ds” protocol to link to Data Services simply by their name rather than knowing the full URL. This will make it easier to build data quality rules or custom actions for your data sets.

To make working with larger data sets easier, we implemented a Search and Replace function in the data editor. This functionality allows you to replace many values quickly without having to change each of them separately. The changes made like this are fully audited so you always know who made the change in your data.

We’ve extended the API for Data Manager to cover additional operations. The new API allows you to create and edit data sets as well as work with data within them providing complete funcitonality to integrate Data Manager into external applications.

CloverDX Server support for Java 21

Starting with this release, CloverDX Server supports deployments with Java 21 for both Tomcat and Red Hat Web Server stacks. The Server now supports both Java 17 as well as Java 21 with Java 17 being the default. Next version of CloverDX – CloverDX 7.3.0 – will introduce support for Java 21 into Designer and will make Java 21 the default for deployments.

Other changes

Customizable error handling for OpenAIClient component

OpenAIClient component now includes new interface functions that allow you to customize the behavior of the component when an error occurs. The new functions newChatOnError, sendRequestOnError, prepareQueryOnError and processResponseOnError match the approach of onError handlers as used in other CTL-enabled components.

More detailed CloverDX security configuration

We’ve added the ability for more detailed configuration of how password and accounts behave on CloverDX Server. You can now configure additional password complexity rules, password expiration policies, session timeouts, etc.

Deprecated TableauWriter removed

We removed deprecated component TableauWriter. There is no direct replacement for this component. We recommend writing data to a different target (e.g., a database) and then connecting your Tableau instance to it rather than working with Tableau files directly.