This repo contains the official Swift SDK for accessing feature flags and for event tracking with Confidence.
It also contains the Confidence OpenFeature Provider, to be used in conjunction with the OpenFeature SDK.
For documentation related to flags management and event tracking in Confidence, refer to the Confidence documentation website.
Functionalities:
- Managed integration with the Confidence backend
- Prefetch and cache flag evaluations, for fast value reads even when the application is offline
- Automatic data collection about which flags have been accessed by the application
- Event tracking for instrumenting your application
In the dependencies section of Package.swift add:
.package(url: "[email protected]:spotify/confidence-sdk-swift.git", from: "1.2.0")
and in the target dependencies section add:
.product(name: "Confidence", package: "confidence-sdk-swift"),
You have two options, both start from File > Add Packages... in the code menu.
First, ensure you have your GitHub account added as an option (+ > Add Source Control Account...). You will need to create a Personal Access Token with the permissions defined in the Xcode interface.
- Add as a remote repository
- Search for
[email protected]:spotify/confidence-sdk-swift.git
and click "Add Package"
- Search for
- Clone the repository locally (only recommended if you are making changes to the SDK)
- Clone locally using your preferred method
- Use the "Add Local..." button to select the local folder
If your app is using some of the features of Swift 6, we recommend setting the Strict Concurrency Checking to Minimal.
import Confidence
let confidence = Confidence
.Builder(clientSecret: "mysecret", loggerLevel: .NONE)
.withContext(context: ["user_id": ConfidenceValue(string: "user_1")])
.build()
await confidence.fetchAndActivate()
- The
clientSecret
for your application can be generated in the Confidence portal. - The
loggerLevel
sets the verbosity level for logging to console. This can be useful while testing your integration with the Confidence SDK. withContext()
sets the initial context. The context is a key-value map used for sampling and for targeting, so it determines how flags are evaluated by the Confidence backend.
Note: the Confidence SDK has been intended to work as a single instance in your Application. Creating multiple instances in the same runtime could lead to unexpected behaviours.
After creating the confidence instance, you can choose between different strategies to initialize the SDK:
-
await confidence.fetchAndActivate()
: async function that fetches the flags from the Confidence backend according to the current context, stores the result in storage, and make the same data ready for the Application to be consumed. -
confidence.activate()
: this loads fetched flags data from storage and makes that available for the Application to consume right away.
If you wish to avoid waiting on backend calls when the Application starts, the suggested approach is to call
confidence.activate()
and then call confidence.asyncFetch()
to update the flag values in storage to be used on a future activate()
.
Important: confidence.activate()
ignores the current context: even if the current context has changed since the last fetch, flag values from the last fetch will be exposed to the Application.
The context is set when instantiating the Confidence instance, but it can be updated at runtime:
await confidence.putContext(context: ["key": ConfidenceValue(string: "value")])
await confidence.putContext(key: "key", value: ConfidenceValue(string: "value"))
await confidence.removeContext(key: "key")
These functions are async functions, because the flag values are fetched from the backend for the new context, put in storage and then exposed to the Application.
Note: Changing the context could cause a change in the flag values.
Note: When a context change is performed and the SDK is fetching the new values for it, the old values are still available for the Application to consume but marked with evaluation reason STALE
.
When integrating the SDK in your Application, it's important to understand the implications of changing the context at runtime:
- You might want to keep the flag values unchanged within a certain session
- You might want to show a loading UI while re-fetching all flag values
- You might want the UI to dynamically adapt to underlying changes in flag values
You can find examples on how to implement these different scenarios in the Demo Application project within this repo.
Once the Confidence instance is activated, you can access the flag values using the
getValue
method or the getEvaluation
functions.
Both functions use generics to return a type defined by the default value type.
The method getEvaluation
returns an Evaluation
object that contains the value
of the flag, the reason
for the value returned and the variant
selected.
The method getValue
will simply return the assigned value or the default.
In the case of an error, the default value will be returned and the Evaluation
contains information about the error.
let message: String = confidence.getValue(key: "flag-name.message", defaultValue: "default message")
let messageFlag: Evaluation<String> = confidence.getEvaluation(key: "flag-name.message", defaultValue: "default message")
let messageValue = messageFlag.value
// message and messageValue are the same
The Confidence instance offers APIs to track events, which are uploaded to the Confidence backend:
try confidence.track(eventName: "MyEvent", data: ["field": ConfidenceValue(string("value"))])
The SDK takes care of storing events in case of offline and retries in case of transient failures.
Note that the data struct can't contain the key context
, as that is reserved for entries set via putContext
(see below):
violating this rule will cause the track function to throw an error.
To set context data to be appended to all tracked events, here is an example:
confidence.putContext(context: ["os_version": ConfidenceValue(string: "17.0")])
If you want to use OpenFeature, an OpenFeature Provider for the OpenFeature SDK is also available. See the dedicated Provider Readme.