http://download.qt.io/official_releases/online_installers/qt-unified-windows-x86-online.exe
Install pkgconfig, sourcetree, python3 and jre.
pkg-config |
pkg-config is a helper tool used when compiling applications and libraries.
|
Sourcetree |
Sourcetree is a free git client for windows and mac.
|
python3 |
Python is a widely used high-level programming language for general-purpose programming. Python3 is requiredby SolAR in order to create new components, based on templates.
|
jre |
Java Runtime Environment, or JRE, is required in order to download third parties libraries based on your environment.
|
You can use or QT creator, or Visual.
QT creator is recommended. |
Install QT community by downloading and launching the following file:
http://download.qt.io/official_releases/online_installers/qt-unified-windows-x86-online.exe
During installation choose a minimal QT 5.9.0 with msvc2015 64bits, and QT Tools→QT creator (default selection) only
Open the "tools/options" window and section "build and run"; please check compiler and debugger is installed. Every tab should be filled in.
Compiler
Debugger
If not, make sure your Visual installation is complete (see sections just above).
If Qt does not autodetect the windows debugger, you might need to manually install the debugging tools for Windows, as stated on [http://doc.qt.io/qtcreator/creator-debugger-engines.html this Qt Page]. You can get the Windows Driver Kit that contains the debugging tools from [https://developer.microsoft.com/en-us/windows/hardware/windows-driver-kit this page].
You can either install Visual Community 2017 (adapted for OpenSource) or install Visual Studio 2017.
Create a new C++ project and follow steps advised by the prompt.
If the compiler is not installed, you will have an option to install it right from the new project window.
You should verify you are able to run a console application "hello world" in C++.
If you have a problem when running Qmake, this is probably due a mismatck between your Qt installation and your Visual Studio version.
do this :
* cd C:\Qt\{your qt version -for instance 5.11.1} * dir
You should see a directory named msvc2017_64 if you ave installed VS2017, and msvc2015_64 if you installed VS2015. I this is missing, please do this :
. cd c:\\MaintenanceTool.exe . follow the installation step to indicate the MSVC version missing, as you can see in this screenshot :
This section helps you to install all code needed for contribution to SolAR framework. It helps you do the same as the automatic script "fromscratch" in the getting started page, but doing it by yourself, in case the script does not work on your computer. Please feel free to give your feedbacks ! |
To download SolAR for contribution, please first use the setup tool in the "use it" section of this web site setup
You will then use the SolARFramework directory created to download the SolAR Source Code.
Check you have a SolAR directory
This will be used for your development. All code concerning SolARFramework will be there.
check you have a BCOMDEVROOT env variable on your local hard drive
This will be used for your compilation, to know where to find third parties.
Please download following projects, using or a GIT bash, or a graphic GIT tool as sourcetree, or by downloading on the github SolarFramework site:
Name |
GitHub url |
|
Framework |
SolARframework |
|
Modules |
SolARModuleOpenCV |
|
SolARModuleTools |
||
Third parties |
Compiled third parties |
Please refer to the Use it/ setup section ../use_it/setup |
you have to choose to checkout on the master or on the develop branch. If you want to contribute, prefer the "develop" branch. |
You have to compile those projects. Please open each of them in QT Creator (is recommended) and do this:
click on "run qmake"
configure the build option with the option "make" + install in parameters (it will install the project in the BCOMDEVROOT directory)
click on "build"
Name |
Projects |
|
Framework |
SolARframework |
SolARFramework/SolARFramework.pro |
Modules |
SolARModuleOpenCV |
Modules/SolARModuleOpenCV/SolARModuleOpenCV.pro |
SolARModuleTools |
Modules/SolARModuleTools/SolARModuleTools.pro |
Some other projects are interesting to build and test, if you want to learn more about SolAR.
Name |
Test projects |
|
Test projects |
ImageConvertor |
Modules/SolARModuleOpenCV/tests/SolARImageConvertor/dynamic/SolARImageConvertorOpencvTest.pro |
ImageLoader |
Modules/SolARModuleOpenCV/tests/SolARImageLoader/dynamic/SolARImageOpenCVDynTest.pro |
|
CameraCalibration |
Modules/SolARModuleOpenCV/tests/SolARCameraCalibration/static/SolARCameraCalibration.pro |
You can see you have installed:
the Core framework ins SolARFramework directory
a module based on OpenCV .
a module based on SolAR developments
The OpenCV container is a first implementation based on SolAR API. in this directory, you will see some features and tests.
You can for example open the SolARDescriptorMatcher program. For this program, and for each test, you can choose a dynamic or a static version. Let’s see with the static version here.
Open the SolARContainerOpenCV\tests\SolARDescriptorMatcher\static folder.
There are source files, and a packagedepencies.txt files, precizing which third parties are used.
packagedependencies.txt is the file used by the pipeline manager to load third parties. |
Just have a look at packagedepencies.txt, it contains the definition of third parties used by this program.
Just remind that the third parties shoule be available in your SolARFramework\SolARLibraries\thirdParties repository.
Open the pro file in your favorite IDE.
Please use your git tool to update the source code.
git fetch
git pull
Rebuild and explained previously