odoyle

O'Doyle Rules!

Odoyle is the core of Rules Engine. It's sole purpose is to execute your defined rules against a given set data and return if the given rules are satisfied or not.

Trivia This package is named Family O'Doyle (bullies) from the movie Billy Madison (1995) where would say O'Doyle Rules after pulling a "prank" on Billy.

Installation

npm install --save odoyle

Usage

The Odoyle class accepts a single, optional, argument:

• ops An Object that contains all operators that your rules can use. See Custom operators for more detailed information.

const Odoyle = require('odoyle');
const odoyle = new Odoyle();

odoyle.rules

The rules method executes the given rules, against the given dataset and returns a boolean that indicates if the rules pass or not. The method accepts the following arguments:

• data An Object that contains data that the rules can reference using the key property in the rules definition.
• ruleset An Array or Object that represents the rule(s) that needs to be executed against the dataset. See Defining rules to learn how to create the rules.

const data = { foo: 90 };
const rules = { key: 'foo', op: '>=', value: 30 };

odoyle.rules(data, rules); // True as 90 >= 30

Defining rules

A rule is represented as single Object that has the following keys:

• key Which data it should use as input value. The key can use a dot notation e.g. foo.bar to reach into deeply nested structures of the dataset to select the data point.
• value Expected value used by the operator.
• op The operator that should be ran against the input and value.

If we want to create a rule that only triggers when clicks >= 20 we would describe it in such a manner:

{
  key: 'clicks',
  op: '>=',
  value: 20
}

When rules are wrapped in an Array we automatically assume that every rule in the array needs to be pass. You can alter this behavior by using an object that specifies the in which condition the rule needs to be executed. We support the following conditions:

• every All rules should be satisfied.
• some Only one rule needs to be satisfied.
• not None of the rules should be satisfied.

{
  every: [
    {
      some: [
        { key: 'clicks', op: '=', value: 300 },
        {
          every: [
            { key: 'clicks', op: '=', value: 19 },
            { key: 'another', op: '=', value: 77 }
          ]
        }
      ]
    }
    { key: 'clicks', op: '<', value: 20 },
    { key: 'clicks', op: '>', value: 10 }
  ],
  some: [
    { key: 'another', op: '>', value: 10 },
    { key: 'another', op: '<', value: 10 }
  ]
}

Note that you can nest conditions as deeply as you want and combine it with other conditions even to create complex rules that might need to react to multiple data points.

Operators

There are various of operators that your rule can use to determine if it satisfied the rule or not. Note that all >/< based operators only function correctly when it's given a numeric value, while the =/~/$/^ function

• = input == value
• != input !== value
• > input > value
• >= input >= value
• < input < value
• <= input <= value
• ~ input.contains(value)
• ^ input.startsWith(value)
• $ input.endsWith(value)

We've opted to use symbols/operators instead of fully named words such as lessThanInclusive to reduce the potential size of your rulesets if you where to store them as JSON objects in databases etc.

Custom operators

In addition to the default operators, you can create your own operators. The first argument of the Odoyle class accepts an Object where the key is the operator, and the value a function that executes the requested operation. Thee function receives 2 arguments:

• input The value that is read from the given dataset.
• value The value that is defined in the rule.

const odoyle = new Odoyle({
  '=': (input, value) => input === value,
  '>=': (input, value) => input >== value,
})

In addition to overriding the default set, you can also directly manipulate the ops property of the class to add, remove, or override operators. The ops property is the Map() instance that stores the defined operators.

const odoyle = new Odoyle();

odoyle.ops.set('===', (input, value) => input === value);

Note that we still assume the operator as first argument, and the function that executes the operation as second argument.

Debugging

The project uses diagnostics to add debugging information to the project. Start your application with:

DEBUG=odoyle

To get information why certain rules failed. E.g. you supplied a key but we couldn't extract the data, or the provided op doesn't exist. Maybe a JS was caused during the execution.

License

MIT