What's New in Rulex 4.4

Workflow Review Tool

NEW

The Workflow Review Tool is used to check the validity of your workflow. Constraints and best practices are defined in an excel file, which can be configured as required.

This tool is essential to ensure that your workflows satisfy compliance constraints to avoid generating errors in production, and is particularly useful when new constraints are identified, as the updated excel file can be used to check multiple workflows and spot where the new constraint is not respected quickly and easily.

Best practices can also be identified to improve the quality of your workflows and avoid warnings.

Import/Export XML

NEW

Two new tasks have been added to Rulex, to make it possibly to directly import single and multiple XML files into a workflow, and to export these same files.

Export Tasks can export relevances

CHANGE

Export tasks in Rulex now have the possibility to export relevances. This was previously possible only for Excel export tasks.

This option is also available when converting structures to datasets.

Http API remote connector added to export tasks

CHANGE

The tasks which deal with exporting data to XML, Json and Text Files can now use Http API as a remote connector.

Export to Json file structure

CHANGE

The Export to Json task has two new options which allows you to decide whether to export data in a flat or nested format, and remove missing records from the exported file.

Import From DB - Oracle Access

CHANGE

The Import From DB task has been modified in the way it connects to the Oracle DB. Connections are now made using the ODBC driver present on the machine instead of the OCILIB library. This change will improve performance and increase resilience of the connection. 

The filename added to the Import from Excel task as a new column on the imported dataset presented errors in some cases. In particular an extra "/" was added in bulk import operations or a URL encoded version of the name was displayed in SharePoint imports.

The filename is now correctly displayed in the final imported dataset without any extra characters.

Export to Excel on SharePoint blocks the computation, failing to correctly export the required dataset.

A series of extra debug print logs has been added to monitor the creation of the exported file. Extra error checks have been added in the XLSX creation to prevent any crashes due to low disk resources. The creation procedure of the exported XLSX file has been changed to lower effective disk usage. A timeout limit of 20 minutes has been added to SharePoint upload operations to highlight possible hung status of the SharePoint server.

The amount of stack memory used in Rulex threads was too high and this limited the possible number of parallel computations on a single VM. 

The amount of stack memory required by a single thread of computation has been decreased by a factor of 10. Consequently the maximum number of parallel computations on a single VM of the same factor has increased.

Imports operations followed by a Concat task did not correctly manage dataset metadata for nominal types for in-memory computations. This occasionally lead to memory crashes in subsequent tasks.

This issue has now been solved.

The versioning system in Rulex system did not manage conflicts correctly.

Conflict management has been completely revised to deal with limitations of the used external Git library. This revision means Rulex can identify real line conflicts inside a wider conflict shown by the Git library.

Hard Reset operations did not cancel dataset tables on working databases. They were effectively erased only by subsequent computations.

Now Hard Reset operations correctly erase all data on the working database.

PRCX export operations also exported data presented in uncomputed tasks,

Data remaining in uncomputed tasks are now ignored by PRCX export operations.

Occasionally the exit code was not correctly caught in PowerShell by the launching application.

An extra print in the log has been added in the logout procedure to check the exit code returned by Rulex and eventually debug the possible issue on the launching application.

The primary key in Export to DB tasks was created with the name __KEY__ when incremental index primary keys were not used.

The user-defined name is now used to create the incremental index primary key.

The format of the query was not managed correctly if insert/update options were selected in the Export to DB task

The query is now formatted correctly.