A lightweight framework on top of the Qt scene graph and only few classes from Qt/Quick. It is usable from C++ and/or QML.
Go to file
Alexander Kavon 0670623fbd build functions and QskQmlExportTargets.cmake file 2023-12-29 01:53:56 -05:00
.github/workflows disable Ninja multi config 2023-12-22 15:26:58 +01:00
cmake build functions and QskQmlExportTargets.cmake file 2023-12-29 01:53:56 -05:00
doc update README to include cmake dep 2023-12-28 14:30:14 -05:00
examples add missing install variables 2023-12-21 18:02:50 +01:00
inputcontext don't use QSK_SOURCE_DIR 2023-12-21 14:05:29 +01:00
playground QskBox namespace renamed to QskBoxRenderer to avoid name clashes with 2023-12-17 17:18:35 +01:00
qmlexport build functions and QskQmlExportTargets.cmake file 2023-12-29 01:53:56 -05:00
skins QskFocusIndicator using enabled/disabled isntead of Exposed/Concealed 2023-12-02 12:05:44 +01:00
src build functions and QskQmlExportTargets.cmake file 2023-12-29 01:53:56 -05:00
support don't use QSK_SOURCE_DIR 2023-12-21 14:05:29 +01:00
tools build functions and QskQmlExportTargets.cmake file 2023-12-29 01:53:56 -05:00
.clang-format fix lower case 2023-08-09 08:30:56 +02:00
.gitignore updated README.md and chapter 3 docs to remove mention of qmake 2023-12-28 14:20:27 -05:00
.qmake.conf project file include mechanism based on QMAKEFEATURES 2019-03-31 13:05:25 +02:00
.uncrustify upgraded to uncrustify 0.76.0 - still not happy with the C++ beautifier 2023-02-28 15:49:42 +01:00
CMakeLists.txt build functions and QskQmlExportTargets.cmake file 2023-12-29 01:53:56 -05:00
COPYING BSD-3-Clause license 2023-04-06 09:23:37 +02:00
README.md update README to include cmake dep 2023-12-28 14:30:14 -05:00

README.md

QSkinny

CMake Build Matrix

The (Q)Skinny library is a framework built on top of the Qt scene graph and very few core classes from Qt/Quick. It offers a set of lightweight controls, that can be used from C++ and/or QML.

Doing the implementation in C++ allows to make use of the "scene graph" classes. Building controls from scene graph nodes allows for a lighter implementation than found with stacking "heavier" objects like QObject or QQuickItem. Offering a full featured C++ API allows the application code to benefit from following the same strategies.

Being "skinny" also means a design that separates concerns between the API and logic of the controls themselves, the styling of these controls, and the delegated rendering of the controls to the screen.

The code already provides a solid fundament for an automotive GUI with currently ~300K lines of pure C++ code. As expected it results in a good startup performance and a low memory footprint.

Nontheless QSkinny is lacking in areas like documentation or appealing default skins. The QML API has not been completed after reaching a proof of concept state. Furthermore the current selection of the implemented controls is limited to the needs of the driving projects.

QSkinny is supposed to run on all platforms being supported by Qt/Quick. But so far only Linux is actively tested. It might support all versions Qt >= 5.15, but you can rely on:

  • Qt 5.15
  • current long term supported ( LTS ) version of Qt ( at the moment Qt 6.5.x )
  • current version of Qt

On debian bullseye these packages need to be installed for Qt5: build-essential cmake qtbase5-dev qtbase5-private-dev qtdeclarative5-dev qtdeclarative5-private-dev libqt5svg5-dev. For Qt6 you need the corresponding packages.

Optional: When enabling the hunspell feature the following package needs to be installed: libhunspell-dev

Optional: When enabling the pinyin feature the following package needs to be installed: libimepinyin-dev

If you want to know more about QSkinny - or even like to give it a specific direction - please contact support@qskinny.org.

IOT dashboard

Contributing to QSkinny

QSkinny is licensed under the BSD 3 Clause License ( https://opensource.org/license/bsd-3-clause ). However we want to be able to offer more licenses on request.

If you are fine with these terms, you are welcome to contribute to QSkinny by completing the following steps:

  1. Agree to the terms of the Contributor Agreement. You can create a pull request first, then the licensing bot will check automatically whether you already signed the Agreement or not. If you have not signed it yet, it will take you to the Agreement to agree.
  2. Submit your pull request for review.