This project provides a set of Ada bindings (aka language projection) to the Microsoft Windows WinRT API and is targeted at two distinct use cases

• Windows Store applications
• Windows native Desktop applications

These bindings have been generated from the metadata describing the WinRT api provided in the Microsoft Windows 10 SDK 20H1 (build 10.0.19041)

v 1.0 Released

You will need a working gcc/gnat x64 Windows build environment. This release has NOT yet been tested against 32 bit environments. You may need to build gcc/gnat x64 from source if required. Details/scripts on how to do this can be found on the mingw64 website, alternatively install using MSYS2

Winrt is dependent upon Winrt_Runtime project. Please clone and build/install this project before attempting to use this project.

The Winrt.gpr needs to know where to install the project files/artifacts, this is specified by the variable named Base_Installation_Dir within the gpr file. You will need to change this value to suit your environment.

For example

• MSYS2 (gcc version 8.3) - Base_Installation_Dir := "lib\gcc\x86_64-w64-mingw32\8.3.1/";
• MSYS2 (gcc version 9.1) - Base_Installation_Dir := "lib\gcc\x86_64-w64-mingw32\9.1.1/";
• AdaCore CE 2018 (gcc version 7.3) - Base_Installation_Dir := "lib\gcc\x86_64-pc-mingw32\7.3.1/";
• AdaCore CE 2019 (gcc version ?.?) - Base_Installation_Dir := "lib\gcc\x86_64-pc-mingw32\?.?.?/";

From a command prompt that has your gcc/gnat x64 distribution/build environments bin directory in its path, simply run the following commands.

• .\gprbuild -p -P Winrt.gpr
• .\gprinstall -f -p -P Winrt.gpr

Alternatively, their is a build.cmd script as part of the distribution that runs the above two commands.

If you have previously built this runtime for a different target/version. Then please clean your previous env by executing the following command

• .\gprclean -P Winnrt.gpr

Please see the appropriate readme within the appropiate directory. The following samples are provided

• Desktop Examples

  • Ocr - demonstrates how to use the Optical Character Recognition API (via the Storage/Streams/Bitmap API's) Also demonstrates a possible approach to handling async functions/procedures from within Ada (aka C#'s async/await) The other alternative approach is to use delegates/callbacks, which is demonstrated in the Network example.

  • Network - demonstrates the use of the tcp/ip Steam socket to connect to a server

• Store Examples

  tba

Welcome