Today, we’ve shipped Jetpack annotations, collecti...
# multiplatform
y
Today, we’ve shipped Jetpack annotations, collections and datastore with their KMP versions as alpha versions. https://developer.android.com/jetpack/androidx/versions/all-channel Some docs pages are delayed due to a problem but we’ll try to get them updated asap. Most APIs are common so hopefully missing KMP docs won’t be a big problem. Let us know if you get a chance to try them out.
😛 1
👀 6
🎉 9
r
Is this a one-off release again, or will all future releases of these libs now have all their KMP targets?
y
All future releases are expected to have their iOS counterparts. The reason why we didn't have dev02 is because around the same time, we were focusing on multiprocess support in datastore and wanted to ship it. So we really didn't have anything usable until we did the refector to merge single and multi process into 1 implementation (and multi process is only on Android for now)
Also, if you look at dev01, it is pretty much a complete implementation of data store 1.0, except for another major change to unbundle from File APIs. That will ideally allow us to adopt kotlinx.io in the future
We called them dev instead of alpha because we are still building the infra for these. So we were extra careful(and still are) not to over promise on the KMP related things
There is still a long way to go before we can scale the number of KMP supported libraries but I'm very bullish on this effort. As the ecosystem moves forward, i think KMP has a very bright future
r
Thanks! Really exciting to see the Android team investing more here.
Ran into this issue FYI that I hadn’t seen in dev01: https://issuetracker.google.com/issues/274786186
y
Thanks for reporting it. We are having a similar problem with the sample app where the compiler SOs 🤦‍♂️. We will be looking into it.
quickfix is live as a new version. it is a bug in the kotlin gradle plugin but the fix wont be there until 1.9; so we decided to ship new artifacts w/o the dependency constraints.
r
Thanks! It's really cool to see that a native-only issue is now important enough to merit a quick-fix release :)
y
lets not make that an expectation for now 🙂 but in this case it was pretty bad since we didn’t have any workaround for it