Plants

RIDDL supports the definition of complex data streaming models. There are three basic definitions involved in setting up a pipeline: a plant, a pipe, and a processor. There are several kinds of pipes and processors, and they can be combined to form entire systems of data processing known as a plant. These metaphors come from chemical processing concepts which you can visualize as something like this:

All you have to do is remember the 3 P’s:

• Pipes - Pipes are conduits for reliably transmitting messages of a particular type of data between processors that are connected to the pipe.
• Processors - While pipes reliably transport data from its producers to its consumers, processors are the producers, consumers, and transformers of data. Notably, entities are processors of pipes too.
• Plants - Plants combine pipes and processors together with support for rich semantics so that arbitrarily complex streams can be modeled in RIDDL.

A RIDDL plant is a definition that combines pipes with processors to specify a model of how data should flow. You may define as many plants as needed but each plant is a closed system without the ability of the RIDDL model to express the sharing of data between the plants. This is done deliberately to prevent unintentional contamination of data in large models.

The purpose of a plant definition is to provide the blueprint for how a set of pipes, processors, and entities are joined together so that data may flow end-to-end. This is done by using:

• the names and types of inlets in processors
• the names and types of outlets in processors
• the names and content types of pipes
• the definition of a joint to connect pipes and processors

Plants are entirely type safe. This means the data type that a pipe transmits must match the data type of the publishing processors (outlets) and the data types of the consuming processors (inlets). When riddlc processes a plant specification, it ensures that all the inlet and outlet data types match the data types of the connected pipes.

An entity may also be used as a processor under some special circumstances:

• as a source - An entity may be used as a source of events if a command handler is defined for the entity.
• as a sink - An entity may be used as a sink for events if a reaction handler is defined for the entity.
• as a flow - An entity may be used as a flow to convert commands into events

Not Implemented

This section has not been completed.

• Domains
• Contexts

• Processors
• Pipes
• Joints