JNA VS Introducing .NET Multi-platform App UI (MAUI)

As far as I understand it, with JNA, all calls into C code go through libffi: https://github.com/java-native-access/jna/blob/master/www/Fu...

This means that every call sets up some libffi data structures and libffi uses this information to perform the native call. Likewise in the other direction for return values. With JNI (and Panama), Hotspot can directly emit the argument/return code a the call, not too dissimilar from what a C or C++ compiler would do. There is still some overhead from maintaining JVM invariants. For example, I think a thread blocked in an FFI call can still participate in a safepoint. But that applies to JNI as well.

If you were prepared to go to the JVM you might try JNA. https://github.com/java-native-access/jna

Other good example, but for Java platform is JNA library. Do not mix it with Java's JNI, which is a bad example of how it could be done.

In case you haven't used it (assuming that you already know about JNI, for comparison) - https://github.com/java-native-access/jna is about as easy as it gets for native interop. Also, as mentioned in another comment, with the Java FFM (Foreign Function and Memory) API already in preview mode, pretty soon, there will be no external dependencies at all, and Java should be able to interop with any language that can talk to C.

The complexity of JNI has given rise to some community-driven libraries that make it simpler to do FFI in Java. Java Native Access (JNA) is one of them. It's built on top of JNI and at least makes FFI easier to use, especially as it removes the need to write any C binding code manually and reduces the chances of memory safety issues. Still, it has some of the disadvantages of being JNI-based and is slightly slower than JNI in many cases. However, JNA is widely used and battle-tested, so definitely a better option than using JNI directly.

This is about calling into any native operating system APIs, as long as they are callable via C or C++ (which these days means "all" operating system APIs).

JNI is somewhat harder to use, because you need custom glue on both sides of the border: Some custom classes in Java and some custom code on the C (and C++) side.

This proposal would remove the need for the glue on the C side and would allow a pure java solution.

Something like this has existed in third-party form for a while as JNA (https://github.com/java-native-access/jna), but now it's going to be built into the JRE itself (if the proposal passes through review)

I use JNA https://github.com/java-native-access/jna , as you can write the entire interface in Java faster as well as easier without the need of messing with the complexities of JNI.

I am migrating XF app into MAUI and writing a simple Navigation framework because Prism doesn't work well and I didn't use anything advanced anyway. So, I am surfing the code of MAUI to intercept all the back buttons, etc. I haven't found a single mention of apis related to predictive back "RegisterOnBackInvokedCallback", "OnBackInvokedDispatcher", "OnBackPressedDispatcher", "AddCallback", "android:enableOnBackInvokedCallback" Also I don't see any issue on github that would say "Support Android Predictive back". Only one kinda related https://github.com/dotnet/maui/issues/8680

Below issue is still reproducing in Maui .net7.0 version also. #10163

Maui is Open Source, MIT License

https://github.com/dotnet/maui

.NET is Open Source

https://dotnet.microsoft.com/en-us/platform/open-source

I do share your skepticism of Microsoft, but it looks like the economics and cash flow dynamics have changed drastically after the advent of the cloud.

Microsoft is more focused on getting developers onto its ecosystem and help them with open source projects with the hope that they will use its Azure cloud services and bring in the money.

My skepticism is a bit relaxed now and I have no qualms using .NET.

I hope I am not wrong.

It's a bit of a hit and miss as of today. CLI, back-end and natively compiled libraries (think dll/so/dylib or even .lib/.a - you can statically link NAOT binaries into other "unmanaged" code) work best, GUI - requires more work.

Avalonia[0] and MAUI[1] have known working templates with it, but YMMV.

[0] https://github.com/lixinyang123/AvaloniaAOT / https://github.com/AvaloniaUI/Avalonia/ / honorable mention https://github.com/VincentH-Net/CSharpForMarkup

[1] https://github.com/dotnet/maui (try out with just true in csproj - it is known to work e.g. on iOS)

While this is the quite endorsed by the community: https://github.com/dotnet/maui/discussions/339

I think the fundamental issue is that desktop Linux is way too fragmented. Not only just GTK2/3 and Qt but you have GNOME, KDE, XFCE, Cinnamon and then you have X11, Xorg, Wayland...

To be honest, all those craps are why desktop Linux never took off. I'm very safe to say MAUI for Linux will eventually renders components off its own using framebuffer and hardware acceleration APIs such as OpenGL or Vulkan just because of the market fragmentations...

If desktop Linux truly wants to get the attention, it will need to unify. Fixing dependency hell using Flatpak is the right direction.

There is an existing old fork of MAUI for Linux that uses GTK: https://github.com/jsuarezruiz/maui-linux

First I came across this Github issue: https://github.com/dotnet/maui/issues/9687 - According to this issue, this is a known bug for MAUI iOS, but it works OK for MAUI Android. As I said, I target Android only and I have the exact same issue. It's apparantly fixed with some of the latest versions for MAUI but the problem still occurs to me even with MAUI version:

I saw this and tried to imitate (ofc my lack of experience wouldn't allow me to do it in the exact way). Already found some documentation that allowed to understand better. Thanks for the insigh.

I implemented the Auth controller following this sample code from Microsoft.