This is a desktop application for manipulating images using a graphical technique called "PixelSorting".

Written in C++14 using Qt and cmake, Chromatic Sort contains a high-performance multithreaded core with minimal memory requirements.

Available soon on Windows, macOS, and Linux operating systems. See the "Download" section below.

Standalone downloads for Windows, macOS, and Linux are available at the Github Releases page for this project. Alternatively, the project can be built from source on Windows, macOS, or Linux. See the 'Building' section below.

• Provides an interface for two common pixelsort idioms: Asendorf sort and block sort
• Fine grained control of colors and shapes being sorted
• Built-in tools for repeating pixelsort operations through an image
• The app contains both a GUI and a command-line interface for easy scripting
• Image processing back-end is configurable between QImage, Magick::Image, and stb_image
• Written in a cross-platform subset of standard C++14 and compiled using -Wall -Werror / -W4
• Cross-platform cmake project which uses the Qt GUI toolkit

This requires a C++14 compatible compiler. Visual C++ 2017 for Windows, Clang for macOS and g++ for Linux are recommended. There are three additional dependencies required: Qt, CMake, and boost:

1. Download and install the Qt5 libraries for your system.
2. Download and install cmake for your system.
3. There is currently an additional dependency on boost which we're trying to phase-out. Download and install boost for your system. Follow the getting started steps to compile the Program Options library.

Download or clone this project's source code using

git clone https://github.com/rdmello/Chromatic-Sort.git --recursive

The --recursive flag is required since this project includes google test as a Git submodule which is downloaded and compiled to run tests.

An out-of-tree build is recommended with CMake. Here are the steps for this:

1. Create a new subdirectory in the top-level Chromatic Sort directory called build and change working directory to this new subdirectory:

mkdir build
cd build

2. Create Makefiles using cmake. On Unix (macOS or Linux) platforms, this should be sufficient:

cmake ..

On Windows, you may need to specify the version of Visual Studio and the project bitness:

cmake .. -G "Visual Studio 15 2017 Win64"

This cmake command is essential for correct operation of the executable. If cmake shows any errors, you should investigate it and try to fix it. In particular, you may need to set the CMAKE_PREFIX_PATH on Windows to help cmake find your Qt installation.

3. Building the project using a C++ compiler. On Unix systems, automake should just work:

make
make test
./pixelsort

On Windows, cmake should create a Visual Studio solution file (.sln). Open this .sln file in the version of Visual Studio used earlier in the cmake command. Then build the project and execute it from the GUI. For optimal performance, build in Release x64 mode.

1. Add another statusbar entry for image name, dimensions
2. Modify sort rotation selector to allow for more angles
3. Add Zoom and pan controls to the image display
4. Maybe move the GUI toolkit to something non-Qt so that we can release this under a more permissive license?

The PixelSort application should be accessible from a command-line interface. Type pixelsort --help for more information on the options for this.