sorry for the direct @ but I was hoping to get this on your radar.
Is there any chance to get more frequent updates? I am not speaking about big stuff, but specially some very powerful packages like react-native-screens. They are fixing critical bugs very frequently (specially for android) and we’re stuck on 2.0.0.alpha.12 for now with EXPO SDK36. Specially alpha.13 fixes something very important stuff regarding nested createNativeStackContainer.
I know the package is alpha, but its evolving fast and we have to hack around currently to run our stuff on iOS and Android. Ejecting is not a option for me right now, as I am relying too heavy on Expo and Expo push service.
hi @hirbod! this would be really nice to be able to do but the overhead of doing a new release is very steep - each release involves:
extensive manual smoke testing, because we can’t catch everything with our automated tests just yet
doing a new simulator/emulator build for ios/android
submitting a new build to apple and google. this is can be time consuming and stressful for us due to the sometimes inconsistent and arbitrary nature of policies.
dealing with support requests caused by discrepancies in libraries and client versions for mid-cycle releases.
another concern is that once we update to 2.0.0-alpha.13, another user will request that we update to 2.0.0-alpha.14 because it fixes an important bug that they encounter.
we’re working on making the bare workflow support for notifications and other services so in the future it will be easier to eject and customize as needed if this kind of issue arises. for 2020 we plan to do one release per quarter, and only patch those releases in the middle of a release cycle if there is an urgent bug that breaks existing or promised functionality.
Thank you for your detailed explanation. Under these aspects I understand the effort which is behind a release and I understand that updates are always a bit more complicated. I thank you very much for your commitment and the good work. And thanks to the whole team.
I don’t want to sound selfish, but does this mean that SDK 37 is not expected before March? (or 36.0.1)
In my case it’s actually about react-native-screens (already on alpha.23) which won’t work on Android without the updates. Waiting three months would be a pretty long time.
I would be happy if there would be a release in the meantime. The best is of course, when all Unimodules are finished, then everybody can decide for himself, if he wants to wait or if he wants to take care of the solutions himself.
If there is anything I could do to help, shout out. I’d love to give something back to the community.
Well, I hope not, considering that there are several keyboard and statusbar issues with android, keyboardavoingview and native-screens
I have high hopes for a faster release. Can’t wait for Expo push land to bare, this will be a huge step into the right direction. Had a test eject today, wasn’t that smooth but finally got it working to test some things. react-native-screens alpha.22 might fix my nesting issue, but introduces ton of other issues. So, its so heavily alpha currently that it’s not worth ejecting for me anyway currently.
Looking forward for future updates. Meanwhile, I’ll try to give something back on supporting questions I know answers for.
the issues you mention can all be worked around by users for a couple of months until the next release. again, if we were to fix the couple of issues that are important for you specifically, there would still be other issues that are important to other users and they could rightfully ask why we don’t release a new version with their specific improvements. if we can be shown that there are serious issues that didn’t exist in a previous release and users cannot work around then we will do our best to get fixes as soon as possible. but when it comes to adding new features or fixing long-standing issues we don’t rush to minor releases.
