Scripts and Rules

Execution logic in neuron-js is defined via an ExecutionScript.

ExecutionScript Structure

A script is essentially a collection of Rules. When a script is executed, the engine iterates through these rules in the order they are defined.

The Anatomy of a Rule

A Rule is a self-contained logical unit. It consists of two main parts:

1. Conditions: A list of checks that must pass for the rule to fire. If this list is empty, the rule is considered an "Always" rule and will always execute its actions.
2. Actions: A list of operations to perform if the conditions pass. If this list is empty, the rule will "Do Nothing"—it will evaluate conditions but trigger no side effects.

{
  "id": "HighValueCustomerRule",
  "conditions": [...],
  "actions": [...]
}

Condition Logic

Conditions are evaluated sequentially and follow a Sum of Products logic (ANDs grouped by ORs).

AND Logic

By default, all conditions in the list are joined by AND. Every condition must evaluate to true for the actions to trigger.

OR Grouping

You can create an OR relationship by setting orCondition: true in a condition's options. This starts a new block of conditions.

• Each block is evaluated as an AND group.
• All blocks are joined by OR.
• If any block evaluates to true, the rule fires.

Example result: (Block 1 AND Block 1) OR (Block 2 AND Block 2)

Advanced Features:

• Inversion: Set inverted: true to negate a condition (NOT logic).
• Disabling: Rules or individual conditions can be disabled via options without removing them from the script.

Sequential Execution

If a rule's conditions are met:

1. The actions within that rule are executed one by one.
2. Each action can modify the ExecutionContext.
3. The modified context is passed to the next action, and eventually to the next rule in the script.

This sequential nature allows for "pipelining" logic, where earlier rules prepare data or set flags for later rules to consume.