Rasputin

Rasputin can convert a point set of (x, y, z) coordinates to a triangulated irregular network. Specifically, it has been developed to convert raster dems (digital elevation models) into simplified triangulated surface meshes. The rasputin_store program did this in the CGAL era and does not exist in the post-CGAL tree; the installed CLI is rasputin, which currently exposes only version while the core is rebuilt. See docs/increments/ for what has landed.

It is also possible to compute the shade cast from a given, planar sun ray vector. This shade is computed based on the cell center of the simplified surface mesh.

Implementation strategy

Rasputin is being rebuilt on a CGAL-free and GDAL-free stack: a C++20 core with async Python bindings. The C++ dependencies are header-only.

C++ core

• pybind11 generates the Python wrappers.
• A constrained Delaunay triangulation library, used exactly once to build the initial triangulation — everything downstream of it is owned in-tree. Detria (MIT, header-only, C++20) is the intended choice, with poly2tri (BSD-2) as fallback; see parallel_refinement.md for the comparison.
• Exact orient2d and incircle predicates via a filtered kernel over a vendored backend (lib/detria/, MIT, pinned). Not header-only: the backend is confined to one translation unit, src/predicates/detria_exact.cpp.
• Catch2 v3 for unit tests. Property tests use Catch2 GENERATE over a seeded range rather than rapidcheck — each property suite records why at the top of the file.

Parallelism uses the standard library by default (<thread>, <atomic>, std::execution); TBB or OpenMP can be opted in via a CMake flag.

Python layer

• Pydantic V2 for models and validation.
• Typer for the CLI.
• Shapely and PyProj for geometry and CRS handling.
• NumPy for array interchange across the binding layer.

Raster reading deliberately avoids Rasterio, which wraps GDAL. A pure-Python reader will be introduced under src_python/tin_engine/io/.

Removed

CGAL, GMP and MPFR (triangulation and simplification move to the CDT wrapper plus in-tree refinement), Armadillo, Pillow, Meshio, and Howard Hinnant's date library (superseded by C++20 <chrono>). The pre-migration tree is preserved under legacy/ and still references several of these.

Installation

The C++ dependencies are header-only, and Catch2 is fetched automatically by CMake via FetchContent, so no manual checkout is needed.

Create the Python environment:

python3 -m venv .venv
.venv/bin/pip install -e ".[dev]"

Building the Python extension needs no extra step: pybind11 is declared in [build-system].requires, and scikit-build-core invokes CMake for you. Rasputin does not aim at being backwards compatible with older compilers. The build requires C++20. Date and time handling uses the C++20 <chrono> calendar types (sys_days, year/month/day) directly, so the compiler must provide them. The following are verified to build and pass the test suite:

• AppleClang 21.0.0
• g++ 16.2.0

All date handling is UTC; no timezone database is required, so libc++ is fine despite not yet shipping std::chrono::zoned_time.

On macOS, note that /usr/bin/c++ dispatches through xcode-select. If it resolves to an old toolchain, point it at the Command Line Tools:

sudo xcode-select --switch /Library/Developer/CommandLineTools

You can ensure that the right compiler is used for building Rasputin by setting the CXX environment variable, for example:

export CXX=/opt/homebrew/bin/g++-16

If you are using gcc, make sure that CXX points to g++ and not gcc.

Rasputin is build using CMake. On Ubuntu, CMake can be installed with the command

sudu apt-get install cmake

or on Arch,

sudo pacman -S cmake

A relatively recent version of CMake is needed; the build declares a minimum of 3.24.

Additionally, you need Python 3.12 or newer. Then, to install Rasputin, change to the Rasputin root source directory and run

pip install .

or, for a development install with the test and lint tooling, use the virtualenv shown above.

Minimal Example

The post-CGAL core is under construction: the meshing pipeline is not wired up yet, so the installable surface is currently the geometry primitives and the CLI. To check that the compiled extension imported correctly, run this in ipython:

from tin_engine import Point2, Point3, dot, cross

a = Point2(3.0, 4.0)
b = Point2(1.0, 2.0)

print("dot(a, b)  =", dot(a, b))
print("cross(a, b) =", cross(a, b))
print("cross(x, y) =", cross(Point3(1.0, 0.0, 0.0), Point3(0.0, 1.0, 0.0)))

This should print out:

dot(a, b)  = 11.0
cross(a, b) = 2.0
cross(x, y) = Point3(0, 0, 1)

The CLI is installed as rasputin:

rasputin version

The legacy CGAL-based pipeline that used to be demonstrated here is archived under legacy/ for reference during the port. It is not packaged and not importable from an installed rasputin.

Data

High quality DTM data for Norway can be downloaded from free here. Choose "Nedlasting" from the left hand side of the map, and choose "Landsdekkende", check "UTM-sone 33" and finally click DTM10. Download and unpack in, for instance, $HOME/rasputin_data/dem_archive, and export RASPUTIN_DATA_DIR=$HOME/rasputin_data.

It is possible to include land cover types in your triangulation, through the GlobCover dataset from ESA. It is a raster based 300m (approx) resolution data set that contains 23 different land cover types. Download the data set and unpack it in $RASPUTIN_DATA_DIR/globcov to access the land types using the rasputin.globcov_repository.GlobCovRepository class.

Acknowledges

The original layout of this project followed the recommendation from an excellent blog post by Benjamin R. Jack, and both the CMakeExtension and the CMakeBuild classes were taken from his blog as well. They lived in the setup.py that the post-CGAL migration replaced with pyproject.toml, so they are no longer in the tree -- but the debt stands. Thanks!

Use cases

Bhattarai, B. C., Silantyeva, O., Teweldebrhan, A. T., Helset, S., Skavhaug, O., and Burkhart, J. F.: Impact of Catchment Discretization and Imputed Radiation on Model Response: A Case Study from Central Himalayan Catchment, Water, 12, 2020b; https://doi.org/10.3390/w12092339

Silantyeva, O., Skavhaug, O., Bhattarai, B.C., Helset, S., Tallaksen, L.M., Nordaas, M., and Burkhart, J.F.: Shyft and Rasputin: a toolbox for hydrologic simulations on triangular irregular networks. https://doi.org/10.31223/X5CS95