Android NDK Binder Naming Conventions - android-ndk

I am a newbie to this Android Native code development. I have some questions to get my head wrapped around some of the terms used in Android Native libraries.
What is the meaning of ABpBinder, BpBinder, IBinder, AIBinder, IBinderInterface, ABbBinder, and BbBinder?
I had a tough time finding documentation that explains these terms.
Thanks in Advance!

Related

Alloy and Alloy*

I have read the paper about Alloy* last week and it seemed very interesting to me and the fact of integrating higher-order quantifiers will help me with my work. But I want to know is it is still an on going project and if I could keep on working on there version of the Alloy Analyser UI and is it stable ?
The Alloy* project is currently not under active development. It is, however, stable, and other researchers are using it to carry out their own research.
You can get it (including sources) from http://alloy.mit.edu/alloy/hola/.
Feel free to contact me directly if you need further help with compiling/using it.
In the mean time ... Alloy (not Alloy*) is now an open source project on Github: https://github.com/AlloyTools We are discussing also including Alloy*

Using haxe as a transpiler between c++ and JavaScript [closed]

Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 6 years ago.
Improve this question
Firstly I apologise for the long post. I have been asking a few questions about haxe and its suitability as a tool to solve a problem I am facing. I have very limited knowledge of haxe but what I do know is my ideal role for haxe in my project is not the normal use of haxe. So what I'm trying to avoid is wasting time I don't have learning this language (despite it looking really fun and powerful) only to find that it isn't fit for purpose. So I am hoping to get some advice from you haxe seasoned vets on whether it is fit for purpose. I truly hope that it is.
Here is what I am trying to achieve:
I work in the games industry and I need to create a game in both C++ and JS. C++ for an embedded system and JS for online and mobile. I am very experienced in both C++ and JS.
The games I develop I only want to write the once. Now I could use something like emscripten for going from C++ to JS but this will result in compiled JS code that cant be debugged with chrome dev tools easily. I'm converned that I will run into edge case bugs with the compiled JS that I cant easily tace back to the original C++. Plus some platforms for the game would require fairness laboratories to view the source code which would be an issue for compiled JS.
What I really need is a source to source compiler that produces native and human readable c++ and JS code that I can then work with and modify in its native form if necessary, hence haxe. I've looked at the code produced by haxe for C++ and JS. JS looks perfectly easy to understand and work with. C++ not so much but still just about acceptable. I can even stop haxe from compiling and linking C++ which I dont need. I only want the source code.
So far so good.
Now I have a game framework in c++ that uses oxygine 2d engine. It is capable of showing sprites etc as well as a framework I have created for message buses and finite state machines (loads more useful classes too). I also have a similar framework in JS which uses Pixijs for its 2d engine and has its own message bus etc just like the C++ engine.
Now what I want to be able to do is write modules in haxe that when I transpile to both C++ and JS that the code can be included as part of the framework and work with its respected language engine. Each object I create in haxe will be very encapsulated and will just need to subscribe to the message bus, handle messages and send messages back. It may also need to know how to use its engines state machine. So I have no idea if this is even possible with haxe as the message bus (along with other objects) will not be written in haxe but will be supplied to the module after it has been transpiled and built in its native project. Maybe I can write my own haxe library that knows the syntax for my two game engines and can transpile depending on its target language? Not sure it that is possible.
Thanks for taking the time to read, and any advice you can give.
You can use
#if cpp
// c++ implementation
#elseif js
// javascript implementation
#end
to allow some different implementations for different targets, this can go anywhere in your code, so you can pass a Haxe value to different functions for different targets.
To use extern definitions:
http://haxe.org/manual/lf-externs.html
http://old.haxe.org/doc/js/extern_libraries
With c++ specifically it might be more complex you need to look at CFFI or Linc
https://snowkit.github.io/linc/
The complexity is probably getting the types across. Remember Haxe c++ has managed memory your engine might do things differently. Also HL is coming soon and may have some advantages.
Looking at oxygine2d it seems a bit like the Flash API? And I know pixijs is based roughly off the Flash API. I believe that OpenFL now uses pixijs for WebGL rendering. You might want to also look at NME ( has nearly same interface for C++ as OpenFL, but NME has stuff like Cppia setup and is sometimes more stable ). If the JS or C++ is too slow you need to look into shaders and Luxe/Kha for render and you might want to also try Heaps it uses Lime ( openfl ) and some haxe js webgl I think.
If you need 2D physics then use Nape it can be used with any of the Haxe toolkits (Luxe, OpenFL, Kha, Flambe), there is physaxe but that's not used as much. Kha and Nape info eg:
https://github.com/lewislepton/kha-examples/blob/master/NAPE/Sources/Project.hx
I really suspect that your making your life hard by wanting to use different engines for different platforms when probably HaxeFlixel or Punk could give you all you need running with Nape on one of the cross target toolkits, but it's very feasible to do it your way.
Maybe look at Tink or Thx for signals event buses and the like, macros (like used in Tink) allow a lot of structures to be built at compile and so can reduce much of the overhead by doing it before run-time, so tweens can be pre-calculated by the compiler. There is a hxcpp debugger and if you run Haxe in chrome it should give you the Haxe line number via js source mapping.
Hopefully I have covered some aspects of your large question.
Haxe allows you to mix and match with native, but I think to use Pixijs and oxygine 2d is probably not ideal as there is bound to be enough good game tools in pure Haxe with target optimizations built in, you just need to go onto Kha IRC, Luxe gitter, and Haxe IRC and OpenFL forum and ask a few questions.

Can we use #MagicConstant annotation in our code?

When im developing Android Applications i come to realize that the usual is to avoid enums due to performance.
So i would like to know, can we use the #MagicConstant in our own code ?
Found what i was looking for Android Studio since 19.01 of the support library has the annotations library which is handy for catching errors
http://tools.android.com/tech-docs/support-annotations
The #IntDef was what i was looking for

Example usage of Graphics.Win32.Dialogue (Haskell library)

I was browsing the Haskell Platform documentation and found this library.
It has only one line of explanation: A collection of FFI declarations for interfacing with Win32.
Is this library intended for building UI on windows ?
If so could anyone show a short example ?
TL;DR: Yes, but don't. Ugliness alert.
It is indeed a set of bindings directly onto the win32 API, which means you can use it to make a UI, but you essentially have to write like a C programmer who doesn't have a toolkit.
It's not pretty, and I'd like to strongly recommend you use a toolkit like GTK or WX, or better still a Functional Reactional Programming library like reactive banana. Those libraries will give you much more idiomatic and eassy-to-understand Haskell code, and portability comes for free.
Occasionally some library you use doesn't feature something you need, whereupon you might want to delve into the windows API.
If you're determined to use this, you need a good Win32 API tutorial to learn from, together with a good reference whilst actually coding. There are loads out there if you google, and plenty of books, but none of them fit into a stackoverflow answer. Whilst I don't know of any Win32 API tutorials written in Haskell, using the bindings provided in Graphics.Win32 means all the function names match up with those in the online documentation, so you should be able to translate.

Is document view architecture unique to MFC

Is there any other class library like MFC which provides Document/View architecture or is it totally unique concept of MFC?
I realize that this is an old question, but for the benefit of those seeing this question via Google searches, the wxWidgets cross-platform GUI API contains a document-view framework as well. I never used it before, so I cannot tell how it comparess with the MFC implementation.
No, it is not, Apple used it for iPhone programming, at least until iOS 5 and if you take a look at the Stanford Universirty iOS Programming courses on iTunesU, you'll see that they'll teach it too. Moreover I've studied it in my OOP course here in Italy, so I can guess it's a worldwide accepted way to organize your code.
Certainly the concept can be found in other languages/frameworks. But, if you compare MFC with WinForms and WPF, then yes, this technique is unique.

Resources