@vallemar that was fast. You don't need changes in core/application. Could you please remove them from that PR?

@farfromrefug if it refers to this file packages/core/application/index.android.ts if I don't put the changes the application gives an error when opening, I haven't tried one by one. I will find which one is necessary and update the branch

I have seen that they have added this to 8.4 amazing! I'm sure this needs some change, there were times it worked for me and other times it didn't, but I think it was because of how I have my local environment. I'm not going to be able to do anything for the whole weekend, I just took the pc to the technician because I think the graphics card is broken

Nx Cloud Report

CI is running for commit 9753520.

📂 Click to track the progress, see the status, the terminal output, and the build insights.

Sent with 💌 from NxCloud.

Guys, I'm testing @farfromrefug fork that has these changes and I have to say that I think the difference in Android when browsing is quite a bit, with the current ns-core package I lose the animation because it gets stuck when entering a page with a lot of components and with the fork the animation makes it perfect and it doesn't seem to get stuck

@vallemar it is quite a big change. We need to test that quite a bit. Not sure what the best approach would be to ensure there is no regression. Firs nx report seems to report a tsc issue on this. @NathanWalker I dont see what command line it is failing on to reproduce locally. Can you tell me ?
@vallemar did you try that PR does it works fine for you?

@farfromrefug I had it working but there were times that it had some failures to start, I don't know if it was because I changed the core in node_modules or because. On top of that, another commit has slipped in, sorry. I would have to try it again but when I try to run the core on my pc my pc crashes

@vallemar OK if you can force push without the unwanted commit that would be great.
About your case where it fails if you can create a reproducible example I could fix it

@farfromrefug Using this version my application is working perfectly and without blinking in the navigation. The only thing is that I have these two warnings and I don't know why, since the BackstackEntry type is being exported in the d.ts

WARNING in ../../../../git/NativeScript/packages/core/ui/frame/index.android.ts 516:57-71
export 'BackstackEntry' (imported as 'BackstackEntry') was not found in '.' (possible exports: Frame, FrameBase, NavigationType, _stack, actionBarVisibilityProperty, attachStateChangeListener, defaultPage, getFrameById, goBack, moduleLoaded, reloadPage, setActivityCallbacks, setFragmentCallbacks, setFragmentClass, topmost)

WARNING in ../../../../git/NativeScript/packages/core/ui/frame/index.android.ts 516:91-105
export 'BackstackEntry' (imported as 'BackstackEntry') was not found in '.' (possible exports: Frame, FrameBase, NavigationType, _stack, actionBarVisibilityProperty, attachStateChangeListener, defaultPage, getFrameById, goBack, moduleLoaded, reloadPage, setActivityCallbacks, setFragmentCallbacks, setFragmentClass, topmost)

@vallemar can you try to replace import { ExpandedEntry } from './fragment.transitions.android'; in with import type { ExpandedEntry } from './fragment.transitions.android'; in index.android.ts

@farfromrefug this does not solve the problem, the error comes from BackstackEntry not from ExpandedEntry

Yes but I thought it would be this. Strange I don't have the error in my fork. Need to look at this PR locally

@vallemar i fixed the warning but i cant commit in your branch. You need to change the first line of core/ui/frame/index.android.ts with this

import { AndroidFrame as AndroidFrameDefinition, AndroidActivityCallbacks, AndroidFragmentCallbacks, NavigationTransition } from '.';
import type { BackstackEntry } from '.';

@farfromrefug done! Thanks!

I am using @farfromrefug fork, I am noticing that in the navigation the lottie components take a while to load (I think the lists also take longer to display). I'm not sure if it's because of these changes, because of the component itself or because of something else, I'm just leaving it here as a note to check before continuing with this

@vallemar you need to give more context. What kind of navigation ?

@farfromrefug it's a simple forward navigation, using the Fade transition, here's a video showing the delay

in the En ruta button it seems to be more noticeable because it makes the button grow and with the background it is more noticeable (it is the first one that I show in the video)

@vallemar ok would be good to create comparable example. Forward navigation should make no difference. Backward navigation should be very different though

I think this should be taken up again, I am having navigation delays when using NS but I don't see them if I use the fork...

I think this should be taken up again, I am having navigation delays when using NS but I don't see them if I use the fork...

I noticed there's a long discussion regarding this PR.
What's the current state of core regarding flashes and delays?

@farfromrefug's fork works very well.

Maybe reopen this PR?

Maybe implement this fix in the upcoming new version 9?

@farfromrefug's fork works very well.

Maybe reopen this PR?

Maybe implement this fix in the upcoming new version 9?

This comment has no specific context regarding the problem and I noticed you have already commented in another old closed PR.
Please avoid spam and open an issue describing your problem.

@CatchABus I think i know the context. every user using google maps, mapbox, maplibre... will have issue with current N on android. opening a page on top of a map view then going back will end up with a black view.
This was the reason why i rewrote the fragment handling in this PR.
some users ask me about that issue and i pointed them to this PR.
it is true that this PR should be reopened. we never really new why it was reverted after being merged. thus leaving no chance to fix the issues preventing this from being merged