Develop Reference

Pure functional programming on android

Are there any advancements in this area? I want to be able to write purely functional code on Android in Haskell or similar languages. I've tried some examples with Scala but it seems to be a pain to get started. Are there any other functional JVM languages which I can use to write Android applications?
Edit: functional languages that write native android applications. My mistake about the JVM.

I doubt that you can find anything mature for writing Haskell-like code for Android. You do need to implement Java abstractions which are required by Android API (implement activity, etc.).
But if you really want to write for Android in a purely functional style you can try to implement your business logic in a pure functional language that compiles to JVM and call it from your Java classes. That approach would be much simpler than trying to implement it entirely in pure functional style.
As your language choice, you can try
Frege, it even has a library for android - froid
Eta lang, it is very new and probably nobody has tried to use it for Android yet

I you want a painless solution in terms of Gradle builds etc., you have only two options: Java and Kotlin, of which of course you should choose Kotlin ;)
Kotlin has most of the things you need to write in functional style:
functions as first class citizen
higher-order functions
immutable collections
var and val like in Scala
if-else as a statement
elements of pattern matching (where statement)
tail recursion
and more...
If you also include funKTionale and kotlinx.collections.immutable, you'll have all the functional goodies like: Option, Try, currying, memoization, persistent data structures and so on...
To start with Kotlin just install the latest Android Studio 3 Preview, which already has built-in Kotlin support.
BTW, don't be so polarized into "pure" functional ;) After all, being 100% "pure" means no side-effects, which means your app can't interact with the user ;)
Hope this helps :)

I've never tried it personally, but you can do F# programming using Xamarin.Android (and, I believe, with Xamarin Forms too). You can see the guide here. (It also includes sample code).
As some background, F# is the .NET Framework's functional language. It is derived from ML; in fact, many ML scripts can be compiled almost "directly" as F# (with the caveat that you may have to do some renaming because F# has some additional keywords that ML doesn't have in order to support several .NET-specific extensions).
Xamarin allows for native development for Android, iOS, and Windows phone. Xamarin Forms allows for a single code-base for all three platforms (it's a competitor to Ionic).
One more quick point: Android does not use the JVM, even if you're writing apps in Java. (In fact, Android does not even support all Java 8 features yet). Through Android 4.4 it used Dalvik; after that, it started using Android Runtime.
You could also try using a JVM language like Scala to create a JAR file and create a bindings library for it.
Please also note that you'll end up using at least somewhat of a mixed paradigm - for example, things like Activities are objects, and the XML files used to define an Android screen is, for all practical purposes, declarative. Edit: This last point is slightly debatable - see the comments.
One final possibility: I haven't checked this out too closely, but try also this link for a site claiming you can do Scala in Android.

Between Pure functional and Java, there is a way which is IMO Pragmatic functional. For instance Redux achieves that in the React arena.
My goal is to write an app (Activity) having an immutable state that advanced as a result of interactions is functional.
In the browser you can see that done with elm (Haskel like language which is also web platform)
Since we want an Android app, I opened the Android Studio used the wizard to create and app with Navigaton Drawer Action bar (with Drawer, FloatingActionBar),
Then converted it to use Elm concepts of immutable model functional approach into a working POC based on a small ElmBase class and idioms.
The code is written in Kotlin (JetBrains tool of choice for the JVM).
You can find the app at my GitHub https://github.com/saffih/ElmDroid The sweet spot of that approach is that it leverages Kotlin being strongly typed and the editor does code completions very well,
making lot's of the code completed for me in a way I have never seen before - amazing experience (But it require using idioms like the sealed class and when properly).

Multiplatform app development with individual guis

I am currently planning the development of a multiplatform app, and I am not sure wich way to do it fits my requirements best, since all possibilties I could think of failed to satisfy me.
First I want to write an app for android, which should feel like a normal typical android app. So I want to use the standard actionbar and android design look and feel in my other gui elements. After finishing the android app I am planning on developing an ios app, which should have a different design, so i am going to redevelop the gui for this anyway.
But I don´t want to rewrite the other code wich represents the intelligence of the app, independent from the gui. I came up with the following possibilities:
1. Java GUI With native library
Here I abstract all the code of my app in a C++ library (since as far as I know ios supports the usage of c++ libraries too) and develop the gui android typically in java. The library would than have a function to start and would inform the gui about every change via callback functions.
Pro
I can reuse all the code that would be the same on both platforms. I just would implement the gui seperately
The design of the android gui is straightforward as I want it to be. It looks like typical android because it is typical android.
Cons
I dislike the usage of JNI very much. Especially the signature and names of the callback functions (calling java functions from c++) are not checked at compile time and require a lot of manual work. If I rename a function and forget to rework the native part I only notice this mistake at runtime.
2. Build the GUI on the native side
Here I had difficulties finding out what is possible, especially for 2.2
2.1 Use Qt
I have only a few first step experiences with Qt in general but as far as I understood i would have the following pros and cons:
Pro
Reuse most parts of the code for ios and Android. I would than redesign the gui for each platform to make them feel natural. I can´t evaluate how much qt may even assists me at doing that
Cons
I have to copy the android gui by using other qt widgets. This is more effort and I don´t know if one can replicate the android gui elements (like the actionbar) so that the user wouldn´t notice it.
2.2 Using the android framework from the native side
I dont know if this is possible at all, I wasn´t able to find this information. Can I use the class "NativeActivity" and use the android framework to build the gui and use e.g. the actionbar? If this is possible somehow it would have the pros from 1. and maybe wouldn´t have it cons?
Do you have any feedback to my ideas or maybe even new approaches I didn´t think of? How do other multiplatform apps like WhatsApp solve this problem? Do they have redundant code for each platform?
Thank you,
Tobi

I would say that it depends very much on your application requirements. By my opinion, a better solution is to develop a separate application for each platform using recommended SDK's for that platform, and implement in native C++ only the time consuming data processing algorithms.
Application runtime on mobile platforms is not so straightforward as on desktop platforms. You should take into account background and foreground processing, specific application life cycling, accessing system resources such as network, file system, etc. And all these issues do differ on iOS and Android.
Regarding possibilities that you listed.
Qt/QML is ok only in case two requirements are met:
1.1 Your application is a foregroud application without any background operations.
1.2 You purchase a commercial Qt license because only commercial Qt can be submitted to Apple iTunes app store (even GPL apps are under
question).
Using NDK Native Activities on Android with cross-platform C/C++ backend. Android NDK API offers much less API then Android Java SDK, so a lot of things you will have to implement or wrap manually. It is a hard road.
Mixing Java code and C/C++ using JNI gives you more of Android SDK API. But you should remember that an Android activity's life cycle is not somewhat that you're used to deal with when developing on C/C++.
Approach that we are using
We've been developing an application with a huge amount of cross platform functionality that should work on Windows/Linux/Mac OSx/Android/iOS. We're using the approach as follows.
Cross-platform core is written in pure C++.
We have adaptors to GUI interface for each platform.
On desktop we use Qt as it reaches all desktop platforms with minor adapting to each platform.
On iOS the GUI is built using iOS SDK with Objective-C and C++ core is linked as a Framework. Still, we had to patch our core in some way for iOS background requirements and so.
On Android we wrap our C++ core in a background process and build all the GUI using only Android Java SDK. Foreground GUI activities interact with the core via local sockets, so we don't need to bother with activities life cycling in our C++ core. But the adaptor is a litle bit complicated.
Nevertheless, both mobile platforms often require workarounds and adaptations in C++ core which add a number of #ifdef'ed branches in code for each platform.

Cross-platform xamarin application and native library bindings

I currently have a native Android AND iOS library that contains different "modules" composed of models (the smallest part), views and controllers (the biggest part).
I would like to start writing apps using xamarin without losing all the work I did on this library. But it seems like if the native library binding is more suited for utilities libraries, or libraries that only contains 1 part of the MVC components, not the 3 which are depending from each other.
So after reading the Xamarin documentation about cross platform development and native library binding, I think that I should rewrite all the model part of my library using a Xamarin C# library, I would drop Core Data on iOS, and ORM Lite on Android, and instead I would use the SQLite library provided. This job does not sounds to big, so it is not a problem (and on long term it will reduce the maintenance effort). But now comes the problem/question, when my models will be C# classes, how will I be able to use these models in my controllers which are still written using Obj-C and Java ? Can I obtain header files from the C# models classes that I can import in the iOS / Android code ? Or do I also need to rewrite the controllers using C# (that would be a no-go to using Xamarin for me) ?
I am open to any migration suggestion/strategy that would avoid losing multiple months of work, so feel free to answer if you have one :)

I currently have a native Android AND iOS library that contains different "modules" composed of models (the smallest part), views and controllers (the biggest part).
That probably means you have currently two libraries, one for iOS and one for Android. One written in Objective-C and one in Java.
I would like to start writing apps using xamarin without losing all the work I did on this library. But it seems like if the native library binding is more suited for utilities libraries, or libraries that only contains 1 part of the MVC components, not the 3 which are depending from each other.
What makes you think this? You can bind whatever you want to bind. And you don't need to bind everything to start with. If for now you only want to bind some parts from your native library that deals with UI, then skip the rest. You can add bindings later.
I think that I should rewrite all the model part of my library using a Xamarin C# library, I would drop Core Data on iOS, and ORM Lite on Android, and instead I would use the SQLite library provided.
That is a valid approach. However then you will not be able to continue using your native code. There is no way to talk from ObjC to C#. Only vice versa. Even if it was possible: you would have to change your controllers to talk to SQLite.Net instead CoreData, or add an abstraction layer, which in turns you will have to bind and implement on the C# side of things. All of this is probably more work than sticking with the native versions or rewriting everything in C#.
Can I obtain header files from the C# models classes that I can import in the iOS / Android code ?
No, you cannot.
Or do I also need to rewrite the controllers using C# (that would be a no-go to using Xamarin for me) ?
You have the choice:
stick with what you have and live with the limitations of maintaining two libraries (and two application code bases)
Bind the libraries and create a managed abstraction around them. Then use this abstraction from a PCL and reference that PCL from your Xamarin C# and Android application. Live with this solution, until you're unhappy enough from maintaining two code bases, and then:
Rewrite your native code and switch to Xamarin and have one code base.
Maybe one more note on bindings: bindings are not meant to exchange part of the native functionality. They transport native functionality into the managed world. They do not enable you to pick the best of both worlds and then create one jack of all trades solution.
You end up with the same result as you would if you stayed in the native world: think what you would have to do if you wanted to change from CoreData on iOS to a Json based storage for instance? You will have to touch all of your controllers unless you prepared everything to support abstraction of the data layer.

Cross Mobile Options

I created an Android app. While creating one specific app was an interesting challenge, I'm now looking into creating a group of similar apps.
I'd like to create a group of similar Android apps and then move on to creating the same on tablets and iOS... (anything mobile).
I've considered doing so with a product called PhoneGap or doing a web based mobile app. Both of these options seem less than ideal. Doing the Android app I've been frustrated by Java's lack of control and low level constructs. Moving to something like a web based app seems like the exact wrong direction.
C++ is my language of choice. It has the ability to work at a low level, is highly portable across platforms, and has significant support for generic coding which would be useful for generating a group of similar apps. However, the Android documentation suggests to not use C++ unless your goal is porting existing code or dealing with computationally heavy tasks.
I'm leaning towards using C++ anyway, but are there other options I've not considered?
Thanks

You could in theory write your logic in C++ and then have UI layers on top that make use of it. If you are really comfortable with C++ that might be the way to go.
Almost any other parts (networking, UI, animation, etc) are better off being done in the native language of the platform. Use of cross platform solutions always limits you in some way, and usually leads to an application that is not as good as it could be for any platform.

Well, Google's recommendation to not use C++ is based on the following, I believe. C++ is low level, so you can get extra performance out of it if you know what you are doing. Google makes the reasonable assumption that many programmers do not. It is easier for an inexperienced programmer to do harm in C++ then to get a performance boost.
But, if you know what you are doing, it can help you. UI elements on both iOS and Android are implemented in their main language (obj-c, and Java respectively) so there is not a great way around that, but you can write core logic and other functions in C++ and it will be portable between them (iOS can use C++ directly and Android can use it via the Native Development Kit).
There are a few other options available. The one I ended up using is Appcelerator Titanium but please stay away from it. If your project gets complicated or large at all you will hate yourself for choosing it, as I did. Another interesting one that uses C++ instead of Javascript is Marmalade. I haven't used it though, so I can't comment on it.
A non-free solution that I hear good things about is Xamarin, who have ported both environments to C# and a .NET using Mono. However, you still have to write two versions of your code for the UI as far as I can tell.

Android NativeActivity

The Android NDK has just been significantly expanded to include support for writing android applications entirely in native C/C++ code. One can now capture input events on the keyboard and touch screen using native code, and also implement the application lifecycle in C/C++ using the new NativeActivity class.
Given all the expanded native capabilities, would it be worthwhile to completely bypass Java and write Android application in native code?

The NDK is not native per-se. It is to a large extent a JNI wrapper around the Android SDK. Using NativeActivity gives you a convenient way of dealing with certain app-life cycle events, and add your own native code on top. ALooper, AInputQueue etc. are all JNI wrappers of the Java SDK counterparts, some with additional code that is private and unaccessible for real apps.
When it comes to Android development, there's no such thing as writing an application entirely in native C++ - you will (in every real App case that I can think of) always need to use the Android API:s, which are to a huge extent pure Java. Wether you use these through wrappers provided by the NDK or wrappers that you create yourself doesn't really change this.
So, to answer your question: No, it wouldn't be worthwhile, because you would end up writing JNI wrappers for SDK calls instead of writing JNI wrappers to your own Java methods that do the same thing, with less code, simpler code and faster code. For example, showing a dialog using "pure c++", involves quite many JNI calls. Just calling a Java method through JNI that does the same thing will give you faster code (one JNI call), and, arguably, code that is easier to maintain.
To fully understand what you can do, you really must examine the Android source code. Start with native_app_glue.c, which is available in the NDK, then continue with the OS implementation of AActivity, ALooper, AInputQueue etc. Google Code Search is a great help in this. :-)
If it is easy to do in Java, and includes many calls, call a method through JNI that does it all, rather than writing all the extra code to do it with multiple JNI calls. Preserve as much of your existing C++ code as is reasonable.

Not if you are just making a standard application. The Java SDK is more complete than its Native counterpart right now so you would still be making things more difficult for yourself.
If you are not doing something that requires the NDK (read: real time performance sensitive) then stick with Java.

Just some food for thought but if you have an app on iOS and Android, some C/C++ code might be shareable. Obviously the iOS Obj-C and platform specific code wouldn't work elsewhere. (Ditto for the Android specific stuff). But you might be able have some shared code that's platform neutral.

If you can, stick with the java style apps until versions of Android supporting native activities constitute a significant fraction of the installed base.
For things that were hard to do before - particularly ports of existing code - this will probably be a big help.
It's not entirely clear yet what has changed vs. just writing your own thin java wrapper. For example, is there still a copy of the dalvik VM hanging around?