Generate SPL Example Project

Download SPL Core

You can either install the SPL Core from the Python Package Index (PyPI)

pip install spl-core

or clone the repository from GitHub and install its dependencies:

git clone https://github.com/avengineers/spl-core
cd spl-core
.\\build.ps1 -install

SPL Core has a command-line interface called please.

Note

If you have cloned the SPL Core repository you find the please script in the .venv/Scripts directory.

Make sure the installation was successful by running the following command:

please --help

Initialize a new SPL Project

Generate the example project:

please init --project-dir C:\\tmp\\MyProject

This will create a new SPL project in the directory C:\\tmp\\MyProject.

Before you can build the project, you need to install its dependencies. Change to the project directory and run the following command:

.\\build.ps1 -install

Now you can build the project:

.\\build.ps1 -build

Select one of the variants to build, e.g., EnglishVariant. The created executable can be found in the directory build\\EnglishVariant\\prod\\my_main.exe.

Run the executable in the terminal will output:

Hello, World!

Generated Project

PowerShell Build Script

The build.ps1 PowerShell script acts as a centralized tool to streamline several tasks related to the project, such as installing dependencies, running, and testing the project. It can be viewed as a build automation script that ensures consistency in development and deployment activities.

You can provide several parameters to customize the script’s behavior:

• install : Install all dependencies required to build.

• build : Build the target.

• clean : Clean build, wipe out all build artifacts.

• buildKit : Build kit to be used.

• target : Target to be built.

• variants : Variants (of the product) to be built (List of strings, leave empty to be asked or “all” for automatic build of all variants)

• filter : filter for selftests; define in pytest syntax: https://docs.pytest.org/en/6.2.x/usage.html; e.g. “EnglishVariant or test_EnglishVariant.py”

• ninjaArgs : Additional build arguments for Ninja (e.g., “-d explain -d keepdepfile” for debugging purposes)

• reconfigure : Delete CMake cache and reconfigure.

The following is a flowchart describing the script’s operation:

        flowchart TD

    Start(Start Script)
    Install{Install?}
    Build{Build?}
    Clean{Clean?}
    HandleClean[Remove All Build Artifacts]
    End(End Script)
    PressKey[Press Any Key to Continue]

    CheckTargetSelftests{Target 'selftests'?}
    HandleSelfTests[Execute Selftests]
    HandleVariants[Detect Variants]
    CleanVariant{Clean?}
    HandleVariantClean[Remove Variant Build Artifacts]
    ReconfigureCheck{Reconfigure?}
    HandleReconfigure[Remove Variant CMake Files]
    CMakeConfigure[Configure & Generate CMake]
    CMakeBuild[Build with CMake]

    Start --> Install
    Install -->|Yes| Bootstrap(Invoke Bootstrap)

    subgraph "BOOTSTRAP"
        Bootstrap-->GitConfig(Git-Config)
    end
    GitConfig-->Build

    Install -->|No| Build
    Build -->|No| PressKey
    Build -->|Yes| CheckTargetSelftests

    subgraph "BUILD"
        CheckTargetSelftests -->|Yes| Clean
        Clean -->|Yes| HandleClean
        Clean -->|No| HandleSelfTests
        HandleClean --> HandleSelfTests
        CheckTargetSelftests -->|No| HandleVariants
        HandleVariants --> CleanVariant
        CleanVariant -->|Yes| HandleVariantClean
        CleanVariant -->|No| ReconfigureCheck
        HandleVariantClean --> ReconfigureCheck
        ReconfigureCheck -->|Yes| HandleReconfigure
        ReconfigureCheck -->|No| CMakeConfigure
        HandleReconfigure --> CMakeConfigure
        CMakeConfigure --> CMakeBuild
    end

    HandleSelfTests --> PressKey
    CMakeBuild --> PressKey
    PressKey --> End
    

Python Dependencies

There are some SPL Core features which require additional Python packages (the list is not exhaustive):

• kconfiglib : used to generate the configuration header file (autoconf.h) from the Kconfig files.

• hammocking : used to generate the mockups for the unit tests.

• sphinx : used to generate the documentation.

These Python dependencies are defined in the Pipfile and will be automatically installed when running the build.ps1 script with the -install parameter. There will be a Python virtual environment created in the .venv directory.

Build Tools Dependencies

The build tools dependencies (like CMake, Ninja, Compiler etc.) are defined in the scoopfile.json file and will be automatically installed when running the build.ps1 script with the -install parameter. The tools are installed in the user directory under scoop.

C:/Users/<username>/scoop/apps

Note

For more information about scoop and how to import dependencies from a scoopfile.json file, please refer to the scoop documentation.