This is a sample of what a production app might look like when using KMMBridge. It's intended to help get you started integrating a Kotlin Mutliplatform Mobile shared library into your production code.

We created sample Android and iOS projects to illustrate a simple usage of this SDK.

This is still a work in progress so some details might be changing. Feel free to open an issue if you see something that could be improved!

This repository is split into the following modules:

allshared only has iOS sources. It can include any iOS-specific API surface (e.g. callback wrappers around suspend functions) and exports analytics and breeds but not database. This is the module where KMMBridge is configured.

analytics is a module to make analytics calls. This is often a thing teams try to integrate first when introducing KMM, so it can provide a template for introducing shared analytics into your own project.

breeds is a model of a more complicated feature. It's structured around a BreedRepository which is able to list dog breeds and mark them as favorites. It also includes the storage logic. Because it lives in a separate module from allshared which is not exported, SqlDelight definitions are not unnecessarily exported to iOS.

The Android repository consumes breeds amd analytics as separate modules, while the iOS repository consumes the allshared module via the published framework.

Click “Use Template”, give your repo a name, and create it.

Add a group string to your repo. You can open gradle.properties and edit GROUP.

After the repo has been created and GROUP has been specified, go to “Actions” and run one of the available CI workflows. KMM Bridge/iOS Publish will build and publish just the iOS SDK. All Publish Will publish iOS and Android binaries.

For more detailed info and for next steps see this tutorial blog post.

Add dependencies in your build.gradle file

implementation("co.touchlab.kmmbridgekickstart:breeds:${VERSION}")
implementation("co.touchlab.kmmbridgekickstart:analytics:${VERSION}")

Initialize the SDK in onCreate of your main application by calling

val sdkHandle = startSDK(analytics = AnalyticsImpl, context = this)

with your implementation of the Analytics interface and application context, keep returned SDKHandle to have access to the CallbackBreedRepository and analytics.

(Optional) log the app start

sdkHandle.appAnalytics.appStarted()

Now you can inject the BreedRepository and Analytics classes into your ViewModel and use all its methods by using

sdkHandle.breedRepository
sdkHandle.appAnalytics
sdkHandle.breedAnalytics

Import the allshared module

import allshared

Initialize the SDK in init function in the App class

init() {
    self.handle = StartSDKKt.startSDK(analytics: AnalyticsImpl))
}

with your implementation of the Analytics interface.

The startSDK function returns SDKHandle data class that holds the CallbackBreedRepository and analytics classes, which your View Model can use to access the data and functions.

(Optional) log the app start

handle.appAnalytics().appStarted()