Batches

The basic operation of the program is focused on input data structured as a batch (e.g., a batch of scanned documents). The batch serves only as a logical bracket holding the documents together. This is particularly relevant when the processing status is being monitored.

Input data might also come in the form of an individual document (e.g., an incoming email). For this input data, the program artificially generates a parent batch that contains only this single document. The data in the internal management structures thus always retains the same structure, regardless of whether the source system provides the data as a batch or as individual documents.

In the program, the individual documents contained in a batch are the main object type, and most processing steps work with it. Data is also always output to the target system by document. The batch structure of the input data is therefore not maintained.

The Document

A document consists of index data and one or more file attachments. File attachments are not mandatory. Thus, depending on the type of input data, a document might consist only of pure data fields.

Index data is comprised of two categories: header data and item data. Only one of each header data fields is always contained in a document. Item data, on the other hand, might occur multiple times as tabular data in a corresponding number of table rows. Technically speaking, it is possible for one document to contain any number of different tables. In practice, however, the number of tables contained in a document is usually just one. In the case of invoice data, a table such as this would be one for invoice items.

Theoretically, tabular data could be broken down further, into subtables. However, in practical use, barely any external system can send or receive data with this sort of structure.