Getting Started#


All non-optional build dependencies are installed automatically through Conan when running CMake. Conan itself can be installed manually but we recommend using the generated scipp-developer.yml for installing this and other dependencies in a conda environment (see below). Alternatively you can refer to this file for a full list of dependencies.

See Tooling for compilers and other required tools.

Getting the code, building, and installing#

You first need to clone the git repository (either via SSH or HTTPS) from GitHub. Note that this assumes you will end up with a directory structure similar to the following. If you want something different be sure to modify paths as appropriate.

|-- scipp (source code)
|   |-- build (build directory)
|   |-- install (Python library installation)
|   |-- ...
|-- ...

To build and install the library:

# Create Conda environment with dependencies and development tools
python tools/ --dir=conda --env-file=scipp-developer.yml \
  --channels=conda-forge --merge-with=developer-extra.yml
conda env create -f scipp-developer.yml
conda activate scipp-developer

# Update Git submodules
git submodule init
git submodule update

# Create build and library install directories
mkdir build
mkdir install
cd build

If you are running on Windows, you need to use a visual studio developer command prompt for the following steps. This can be opened manually from the start menu, or programatically by calling the appropriate vcvars script, for example:

"C:\Program Files\Microsoft Visual Studio\2022\Professional\VC\Auxiliary\Build\vcvars64.bat"

If you wish to build using the Visual Studio CMake generators instead, there is a windows-msbuild CMake preset for this purpose.

To build a debug version of the library:

cmake \
  -GNinja \
  -DPython_EXECUTABLE=$(command -v python3) \

# C++ unit tests
cmake --build . --target all-tests

# Benchmarks
cmake --build . --target all-benchmarks

# Install Python library
cmake --build . --target install

Alternatively, to build a release version with all optimizations enabled:

cmake \
  -GNinja \
  -DPython_EXECUTABLE=$(command -v python3) \

cmake --build . --target all-tests
cmake --build . --target all-benchmarks
cmake --build . --target install

To use the scipp Python module:

cd ../python
PYTHONPATH=$PYTHONPATH:../install python3

In Python:

import scipp as sc

Additional build options#

  1. -DDYNAMIC_LIB forces the shared libraries building, that also decreases link time.

  2. -DTHREADING enable or disable multi-threading. ON by default.

  3. -DPRECOMPILED_HEADERS toggle usage of precompiled headers. OFF by default.

  4. -DCPPCHECK toggle run of cppcheck during compilation. OFF by default.

  5. -DCTEST_DISCOVER_TESTS toggle discovery of individual tests for better (but much slower) integration with ctest. OFF by default.

Running the unit tests#

Executables for the unit tests can be found in the build directory as build/bin/scipp-XYZ-test, where XYZ is the Scipp component under test (e.g. core). all-tests can be used to build all tests at the same time. Note that simply running ctest also works, but currently it seems to have an issue with gathering templated tests, so calling the test binaries manually is recommended (and much faster).

To run the Python tests, run (in the python/ directory):

cd python
PYTHONPATH=$PYTHONPATH:./install python3 -m pytest

Building Documentation#

  • Run cmake --build . --target docs from your build directory.

  • This will build the documentation and put it on <build dir>/docs.

  • If rebuilding the documentation is slow it can be quicker to remove the docs build directory and start a fresh build.

Precommit Hooks#

If you wish, you can install precommit hooks for flake8 and yapf. In the source directory run:

pre-commit install
pre-commit run --all-files

Using scipp as a C++ library#

Using Scipp as a C++ library is not recommened at this point as the API (and ABI) is not stable and documentation is sparse. Nonetheless, it can be used as a cmake package as follows. In your CMakeLists.txt:

find_package(Scipp 0.11 REQUIRED) # replace with required version
target_link_libraries(mytarget PUBLIC scipp::dataset)

If scipp was install using conda, cmake should find it automatically. If you build and installed scipp from source use, e.g.,:

cmake -DCMAKE_PREFIX_PATH=<your_scipp_install_dir>

where <your_scipp_install_dir> should point to the CMAKE_INSTALL_PREFIX that was used when building scipp. Alternative set the Scipp_DIR or CMAKE_PREFIX_PATH (environment) variables to this path.

Generating coverage reports#

  • Run cmake with options -DCOVERAGE=On -DCMAKE_BUILD_TYPE=Debug.

  • Run cmake --build . --target coverage from your build directory.

  • Open coverage/index.html in a browser.