how it works

The pyBPPIBridge manages the Data source and the BPPI connection and build a Data pipeline between the both of us. Of course in between the two connections, the bridges can also prepare the data and also make some data transformations depending of the needs and the configuration.

Currently the bridges loads the data by using the repository first. So after gathering the data source dataset all of those data are pushed into a table into the BPPI Repository. It's possible to add another step to load automatically load these data into a BPPI project by creating a To Do into BPPI. This To Do can be executed by just specifying an option in the bridge configuration.

This way (managing a intermediate step with the BPPI Repository) enables the BPPI administrator to create and add other data operations and also make himself the BPPI mapping (Timeline iD, Event Id and Timestamp).

Data Pipeline steps

The Data pipeline follows the ETL pattern, that means it first Extract (E) the data from the data source, make some Transformations (T) if needed and lastly Load (L) the result into the BPPI Repository.

These are the different steps when running the bridge:

graph TD;
    Access-to-the-data-source-->Read-the-data;
    Read-the-data-->Prepare-data;
    Prepare-data-->Connect-to-bppi;
    Connect-to-bppi-->Load-data-in-a-Repository-Table;
    Load-data-in-a-Repository-Table-->Execute-To-Do

Loading

Note: the last step is optional.

• The first two steps are specifics to the Data source
• The Bridge gather the data from the data source and then manage to do some transformations (filtering, format change, etc.)
• The connection to BPPI is done via the Native API interfaces
• The data are loaded into a Repository table
• (Optional) A BPPI To Do is executed. For example to make some transformations and load the Repository table into a BPPI Project

In the example below, the BPPI To Do execute different BPPI Data operations before loading the result into a BPPI Project: