Mapping

Mapping is walkerOS's code-free way to reshape events as they move through the flow. You describe the transformation as data (JSON or object literal), and the collector applies it at two stations: source push (clean and normalize incoming data) and destination transform (adapt to each tool's expected format).

Two concepts

Everything in walkerOS mapping is built from two building blocks:

Mapping.Value: the polymorphic primitive. Whenever a config field accepts a value (data, map.*, policy.*, set[], …), you write a Mapping.Value. There are eight forms (path, constant, map, loop, set, fn, condition, consent).
Mapping.Rule: the object on a destination's mapping[entity][action]. It composes values into a complete rule (name, data, condition, consent, policy, batch, ignore, skip, settings).

When a destination doc shows data: any, the value you put there is a Mapping.Value. The whole rule around it is a Mapping.Rule.

Minimal example

// Destination rule: transform "product view" for GA4
{
  product: {
    view: {
      name: 'view_item',
      data: {
        map: {
          item_id: 'data.id',
          value: 'data.price',
          currency: { value: 'USD' },
        },
      },
    },
  },
}

product.view is the entity-action key. name renames the event. data is a Mapping.Value (specifically the map form) that builds the destination payload.

Try it

Map events by their entity-action structure, live:

Configuration

Result

Where mapping runs

Source mapping cleans, filters, or renames incoming events before they reach the collector:

Destination mapping adapts events to each tool's API shape:

Both mappings are independent. The same event can be transformed differently at each stage.

Two concepts​

Minimal example​

Try it​

Where mapping runs​

See also​

Two concepts

Minimal example

Try it

Where mapping runs

See also