Connor Tumbleson

Thanks @REAndroid - makes sense. The application ended in `com.something.courier` which matches ARSCLib output. Makes sense how you got to it now. I assume you did that instead of hard-coding...

Sounds like you've done what I was going to ask and isolated the disassembly to the resources. Skipping disassembly of resources and launching resulted in no issue. Odd however the...

 I disassembled, rebuilt, disassembled and then compared. Source changes are the lost of default values, but I'm guessing the jvm/something knows default values of unset scalars....

I have not. If you preform the same test I did in [this comment](https://github.com/iBotPeaches/Apktool/issues/3489#issuecomment-1890999605) you can see if its just as confusing or provides additional context.

Not positive on path forward on this. So to be blunt - it may sit for quite a while. We adjusted disassembly process to just handle the "2" part of...

If you try with no sources (-s) and then again except with no resources (-r) - how does your test change?

We've had a pattern over the last decade of adding apache compress, ripping it back out and here we are adding it back. I do like this idea of further...

I don't think a bug here. If you force manifest decoding (ie you don't decode `resources.arsc`) we cannot resolve those references. How it actually built is the odd thing. I'll...

This is so interesting to me that we built an obviously invalid Manifest and aapt2 was fine with it. My guess is all the protections we've ripped out over the...

Am I correct this is just the compression of the outer zip (ie apk)? Or am I misunderstanding? We have plenty of logic at the moment that reviews all the...