Destinations
WalkerOS is meant to be vendor-agnostic. While clients create events, destinations are used to manage how events are processed and sent to various analytics or data storage tools. The purpose of using destinations in walkerOS is to ensure that data captured from your website or application is best-organized to easily integrated with different tools if proper consent was granted. This helps in maintaining data quality and simplifying the setup of new tools by simply mapping the events to the desired format.
Destinations initialize and process events only if a user granted consent.
How to use
Destinations are added to a client
(see web
or node). Before
receiving events from the client, the proper consent states are checked each
time automatically. Destinations receive events through the push
interface. Each destination can have its own configuration, which is set up in
the config
object. This configuration includes general
settings for the destination and individual event settings. The optional
init
function in a destination gets called before actually pushing
events. This function must return true upon successful initialization for the
events to be processed.
Configuration
The configuration of a destination is set up in the config
. All properties are
optional. A complete destination configuration might look like this:
{
id: "demo",
custom: { foo: "bar" },
consent: { demo: true },
init: false,
loadScript: false,
mapping: {
// Read more in the mapping section
page: {
view: { name: "pageview" },
},
"*": {
visible: { batch: 2000 },
},
},
policy: {
'data.gclid': { consent: { marketing: true } },
},
queue: true,
verbose: false,
onError: (error) => console.error("demo error", error),
onLog: (message) => console.log("demo log", message),
on: {
// Client-related on-events
consent: [{ marketing: console.log }],
ready: [console.log],
run: [console.log],
session: [console.log],
},
};
Overview of all properties:
Property | Value | Description |
---|---|---|
id | string | A unique key for the destination |
consent | object | Required consent states to init and push events |
custom | any | Used for a destinations individual settings |
init | boolean | If the destination has been initialized by calling the init method |
loadScript | boolean | If an additional script to work should be loaded |
mapping | object | Configuration how to transform events (see Mapping) |
policy | boolean | Enriches, validates, and redacts properties from the event (see MappingValue) |
queue | boolean | Disable processing of previously pushed events |
verbose | boolean | Enable verbose logging |
onError | function | Custom error handler |
onLog | function | Custom log handler |
on | On.Config | Rules for on-functions that gets triggered on specific events |
Call elb('walker destination', { push: console.log }, config);
to add the
destination to a client. The destination will log all events straight to the
console. Edit a destinations configuration at runtime by accessing
walkerjs.destinations.<id>
.
To grant required consent call
elb('walker consent', { demo: true });
.
Methods
A client communicates with a destination through the methods. It's also the
clients job to check for proper consent, calling the init
method or batching
events. The only required method is push
to send events to the destination.
init
The init
method is optional and gets called before pushing events. It's used
to eventually load additional scripts or set up the environment. To interrupt
further processing, the method must return false
.
The walker.js checks the config.init
value to see if a destination has been
initialized, or not. This way you can add a destination that has already been
initialized.
// Optional init function
const init = (config) => {
if (config.verbose) config.onLog(config.custom.foo);
config.custom.count = 0;
};
push
The push
method gets called to send events to the destination, along with the
config
and eventually matching mapping
.
// push function
const push = (event, config, mapping) => {
config.custom.count++;
event.data.count = config.custom.count;
console.log('demo push', { event, config, mapping });
};
// elb("page view");
// Output with the mapping above
// demo log bar
// demo push { event: { data: { count: 1 }, event: 'pageview' }, config: { ... }, mapping: { ... } }
pushBatch
The pushBatch
method is optional and gets called if the event mapping
is
configured with a batch
number to bundle multiple events before sending them
to the destination.
const pushBatch = (batch, config, instance) => {
console.log('demo pushBatch', { batch, config });
};
The batch
is an object with the key
of the used mapping, the events
array
and the mapping
object itself.
// Configuring a destinations mapping with
const mapping = {
'*': {
visible: { batch: 2000 },
},
};
// Calling a matching event two times
elb('product visible');
elb('promotion visible');
// The destinations pushBatch receives the following batch after 2 seconds
batch = {
key: '* visible',
events: [{ event: 'product visible' }, { event: 'promotion visible' }],
mapping: { batch: 2000 },
};
on
The on
method is used to set up rules for on-functions that get triggered by
the client (individually for walker.js and node).
const on = {
consent: [{ marketing: console.log }],
ready: [console.log],
run: [console.log],
session: [console.log], // walker.js only
};
Once a client triggers an event, the destinations on
method gets called.
Replace the console.log
with your custom function.
Available Destinations
📄️ Mapping
How to transform events into another required format.
🗃️ Node
3 items
🗃️ Web
7 items