The Imixs-Data project provides modules to import, export, extract, group and view data related to the Imixs-Workflow engine. All modules are based on Jakarta EE and integrate seamlessly into the Imixs-Workflow processing life cycle via CDI events and Signal Adapters.
| Module | Description |
|---|---|
| imixs-data-documents | OCR text extraction from document attachments using Apache Tika |
| imixs-data-groups | Group and organize related workitems under a master process |
| imixs-data-views | Select, view and export data views |
| imixs-data-importer | Import documents from external sources (FTP, IMAP, CSV) |
Imixs-Data-Documents provides Plugins and Adapter classes to extract textual information from attached documents — including Optical character recognition (OCR) — during the processing life cycle of a workitem. The extracted text can be used for further processing, AI prompt enrichment or full-text search.
The text extraction is based on Apache Tika and is controlled via BPMN model configuration. Two extraction modes are supported:
TIKA — standard mode, sends the document to the Tika /tika endpoint and returns the full extracted text as a single text block.
RMETA — recursive metadata mode, uses the Tika /rmeta/text endpoint to extract text from container files (e.g. .eml, .msg) with fine-grained control over which embedded documents (PDFs, ZIPs, etc.) are included. Technical or binary files like CAD files (.step) can be cleanly excluded even when nested inside ZIP archives.
Imixs-Data-Groups provides services and adapters to group and organize related workitems under a master process. A data group is a business process that is referenced by other processes within the same process instance via the item $uniqueIdRef.
Typical use cases are consolidating payment transactions of a customer in a "Statement of Account", or grouping invoices that need to be exported into another IT system under an "Export process".
The DataGroupAdapter supports the following operations:
- ADD — add the current workitem to a data group (creates one if none exists)
- REMOVE — remove the current workitem from a data group
- EXECUTE — process all referred workitems with a given event in the same transaction
- EXPORT — export the data of a data group into a CSV or Excel file
Imixs-Data-Views provides methods to define, select and export data views. A data view is a reusable query and column definition that can be used across modules. The DataViewExportAdapter exports a data view directly based on its own query, while the DataGroupExportAdapter (see Imixs-Data-Groups) uses a data view definition to export the workitems referenced by a data group.
Imixs-Data-Importer provides a generic scheduler-based import service to import documents from various external sources into the Imixs-Workflow processing life cycle. The service is based on the Imixs-Scheduler API and uses CDI events, making it highly extensible for custom import sources.
The following standard import sources are supported out of the box:
- IMAP — import documents from an email mailbox
- FTP — import documents from an FTP server
- CSV — import datasets from CSV files located on an FTP server
Each module can be included independently:
<!-- OCR / Document Text Extraction -->
<dependency>
<groupId>org.imixs.workflow</groupId>
<artifactId>imixs-data-documents</artifactId>
<scope>compile</scope>
</dependency>
<!-- Data Groups -->
<dependency>
<groupId>org.imixs.workflow</groupId>
<artifactId>imixs-data-groups</artifactId>
<scope>compile</scope>
</dependency>
<!-- Data Views -->
<dependency>
<groupId>org.imixs.workflow</groupId>
<artifactId>imixs-data-views</artifactId>
<scope>compile</scope>
</dependency>
<!-- Data Importer -->
<dependency>
<groupId>org.imixs.workflow</groupId>
<artifactId>imixs-data-importer</artifactId>
<scope>compile</scope>
</dependency>