This release contains the following new features and enhancements:
When importing from SQL, you can now define a filtering condition (predicate). The SQL query automatically includes the initial part of the SELECT statement, such as SELECT * FROM <table/view_name> WHERE. Users only need to provide the predicate (the actual condition to evaluate), such as id > 10 or name = 'Alice'. All standard SQL predicates are supported. Additionally, administrators can define a default predicate in the SQL default import options. This default predicate is automatically pre-filled when users run the Import SQL service.
When exporting SQL, you can now export to a view. Previously, you could only export to a table. Please note that due to a limitation, this is not available for PostgreSQL databases. Additionally, you cannot export to a view that includes join tables.
An administrator can now configure the add-on's import, export, and transfer services to run as background tasks. This allows users to access and work in other areas of EBX® during longer running services that move larger amounts of data.
The add-on now includes the Digital asset name lookup transformation function. You can use this function to export digital asset file names instead of UUIDs. This simplifies the process of using the Digital asset UUID lookup function to automatically attach digital assets to records during import.
The TransformationDefinition interface is now available in the public API to use when coding transformation mappings.
When using the add-on's services in an EBX® workflow, the Accept button now only displays after the service finishes.
New options were added when importing or transferring data that control how the system handles errors caused by invalid records. For constraints defined in the data model, options determine whether constraints are enforced, invalid records are skipped, or the operation stops when an error is encountered. For transformation errors, options determine whether invalid records are skipped, or the operation stops when an error is encountered.
When exporting related data, the display of related tables in the mapping screen is no longer affected by dataspace, or dataset permissions. If the user initiating the export has sufficient permission to view a related table at the data model level, it will display in the UI. However, if the user does not have sufficient permission at the dataspace or dataset level, an empty file/sheet is exported for the related tables.
The Apache Commons IO library was updated to version 2.18.0.
This release contains the following closed issues:
[DINT-3860] An import simulation should not return a Validation category error for null primary keys when Check empty or null primary key is disabled.
[DINT-3939] When running an import simulation where there are white spaces in primary keys, the simulation result should only return a Business category error.
[DINT-4719] When importing a CSV file that has an empty first line and the option to specify that the first row contains a header is set to Yes, an error occurs in the mapping screen.
[DINT-4730] When transferring data between datasets, and a template is loaded, an error occurs.
[DINT-4740] The mapping screen does not display if it's loaded from a template that contains missing transformations.
[DINT-4905] The statistics contained in the log after a simulation are not correct.
[DINT-4975] The tooltip text for transformations continues to display even after users are no longer mousing over it.
[DINT-5042] When importing CSV at the dataset level, an error can occur when clicking on the Simulation button.
[DINT-5079] A mapping is not automatically created between a single-valued source field and multi-valued target field even though both fields have the same label and datatype.
[DINT-5092] When importing CSV or Excel to a table and cancelling the operation, or navigating away from the screen, a confirmation pop-up displays even when no changes are made to the settings.
[DINT-5093] Loading an existing mapping from a template throws a IndexOutOfBoundsException when a constant transformation function has no parameters.
[DINT-5143] Unchecking a tab when importing at the dataset level removes any saved mappings.
[DINT-5153] A ParsingException occurs for an unexpected character when exporting a table.
[DINT-5164] When running an import simulation, an error occurs when the import source includes empty columns.
[DINT-5172] An issue occurs when cancelling a CSV or Excel export and processing is still in progress.
[DINT-5174] Under certain conditions export to Excel is not working.
[DINT-5272] An error occurs when using the API to convert data using a mapping constant of the Date type.
[DINT-5311] An issue occurs in acquiring a lock when transferring data using a workflow script.
This release contains the following known issues:
[DINT-4339] A JNDI data source cannot be configured when EBX® is deployed on a WebLogic, or JBoss server.
[DINT-5113] When using a cross reference function during import, and the data source's configured lookup field contains missing or empty values, the add-on creates duplicate records.
Fonts are no longer shipped with the JDK 11. Font rendering in Java 11 JVMs now uses pre-installed operating system fonts. This can cause issues in Linux environments when exporting Excel files. As a workaround, you can install the fontconfig package.
When using the SQL export service and exporting to a PostgreSQL database, the option to export to a view is not supported.
When exporting to SQL with TDV, if any error occurs, the operation is cancelled.
Due to a limitation in Excel, long fields with the number data type display all numbers past the 16th place as a 0.
Importing and exporting SQL: Read and write is not possible with the Timestamp data type with MS SQL Server.
Configuring a schema using the JNDI configuration in your web server is not supported.
When creating a custom transformation function, you must use a predefined data type for its input and output fields.