Mapping
Mapping transforms your events - either as they come from sources or before they go to destinations. Use the same mapping syntax in both places.
Why Mapping?
For Sources: Clean up messy input data, filter unwanted events, normalize formats
For Destinations: Transform events to match what each tool expects (GA4, Meta, etc.)
When to Use
Source Mapping:
- Filter test/debug events before they reach your collector
- Rename inconsistent event names from different sources
- Validate or normalize data before processing
Destination Mapping:
- Transform event names to match destination requirements (e.g.,
product view
→view_item
for GA4) - Reshape data to fit destination APIs
- Add required fields like currency codes
How It Works
Map events by their entity-action structure:
// Source: Clean up what comes in
sources: {
browser: {
config: {
mapping: {
product: {
click: { name: 'product view' } // Standardize clicks to views
},
test: {
'*': { ignore: true } // Filter out test events
}
}
}
}
}
// Destination: Format for specific tools
destinations: {
gtag: {
config: {
mapping: {
product: {
view: {
name: 'view_item', // GA4 expects this name
data: {
map: {
item_id: 'data.id',
value: 'data.price'
}
}
}
}
}
}
}
}
Both mappings are independent - one event can be transformed differently at each stage.
What You Can Do
- Rename events to match your needs or destination requirements
- Filter events by ignoring unwanted ones
- Reshape data to match destination formats
- Add static values like currency codes
- Validate data before sending
- Require consent to respect user privacy
Event Mapping with getMappingEvent
getMappingEvent(event: WalkerOS.PartialEvent, mapping?: Mapping.Rules): Promise<Mapping.Result>
This function finds the appropriate mapping configuration for an event based on its entity and action.
Basic Event Mapping
Map specific entity-action combinations to custom event names:
Wildcard Mappings
Use wildcards (*
) to match multiple entities or actions:
Conditional Mappings
Use conditions to apply different mappings based on event properties:
Ignoring Events
Skip processing certain events by setting ignore: true
:
Value Mapping with getMappingValue
getMappingValue(value: unknown, mapping: Mapping.Data, options?: Mapping.Options): Promise<WalkerOS.Property | undefined>
This function transforms values using various mapping strategies.
String Key Mapping
Use a string to extract a value by its property path:
Array Access
Access array elements using dot notation:
Static Values
Return static values using the value
property:
Custom Functions
Transform values using custom functions:
Object Mapping
Create new objects by mapping properties:
Array Processing with Loop
Process arrays and transform each item:
Validation
Validate values and return undefined if validation fails:
Consent-Based Mapping
Only return values when required consent is granted:
Usage Examples
Source Mapping
Normalize events before they reach the collector:
await startFlow({
sources: {
browser: {
code: sourceBrowser,
config: {
mapping: {
// Rename DOM clicks to views
product: { click: { name: 'product view' } },
// Ignore test events
test: { '*': { ignore: true } },
},
policy: {
// Normalize user email
'user.email': { fn: (e) => e.user?.email?.toLowerCase() }
}
}
}
}
});
Destination Mapping
Transform for specific destination APIs:
await startFlow({
destinations: {
gtag: {
code: destinationGtag,
config: {
mapping: {
product: {
view: {
name: 'view_item',
data: {
map: {
item_id: 'data.id',
value: 'data.price',
currency: { value: 'USD' }
}
}
}
}
}
}
}
}
});
Combined Flow
Event processed twice with different configs:
// 1. Browser sends: "product click"
// 2. Source mapping: "product click" → "product view"
// 3. Destination mapping: "product view" → "view_item"
// 4. GA4 receives: "view_item"
Best Practices
- Source mapping: Normalize, filter, validate incoming events
- Destination mapping: Transform to destination-specific formats
- Use specific mappings over wildcards for better performance
- Validate critical data before sending to destinations
- Respect consent by using consent-based mappings
- Keep transformations simple - complex logic in custom functions