Fixes #4981

Description of changes:
- add a quickstart section in the readme and Sphinx docs
- make Ubuntu installation instructions more fine-grained
- update macOS installation instructions

Co-authored-by: PythonFZ <fabian.zills@web.de>

Fixes #5030

Description of changes:
- new feature: generate non-bonded tabulated potentials from sympy expressions

Lukas Dorschel authored Mar 06, 2025 and Jean-Noël Grad committed Mar 06, 2025

Co-authored-by: Jean-Noël Grad <jgrad@icp.uni-stuttgart.de>

* adds a function `add_bond(bond_id, particle_ids)`
* use it (rather than storage-aware code) in method "add_bond" for the Particle Handle script interface
* use it in the BindCenters collision detection scheme. Here, we check via a flag, that one-sided bond storage is used, because the current implementation doesn't guarantee that cross-mpi-rank pairs are handled on both involved ranks

This reverts commit 7db93800.
Branch protection rules require extra Docker flags.

Description of changes:
- bump waLBerla and Stokesian Dynamics releases
- update version requirements of C++ and Python libraries
- address deprecation warnings from CMake 3.31 and 4.0
- replace deprecated `except:` and `only:` by `rules:` in GitLab-CI
- refactor CMake targets and fixtures in the testsuite

Simplify target dependencies in the CMake testsuite and remove
unused CMake variables. Use the modern version of `add_test()`.
Bring tutorial/sample/benchmark tests one level higher in the
testsuite tree. Increase verbosity of the CI shell script.

Description of changes:
- `observables.PairwiseDistances`: an observable that tracks pairwise distances between two sets of particles
- `accumulators.ContactTimes`: an accumulator that tracks contact times when particles are within some distance
- `samples/contact_time.py`: a sample script showcasing how to use these new feature

The contact time is defined as:

$$\tau = t_f - t_0$$

where  $t_f$ is the last measured time at which those particles were in contact and  $t_0$ the first measured time at which those particles were in contact. For example, if two particles have been only in contact during two consecutive configurations then their  contact time is $\tau = \Delta t$, where $\Delta t$ is the time step. If they are only in contact for a single configuration then  their contact time is $\tau = 0$.

blancoapa authored Feb 21, 2025 and Jean-Noël Grad committed Feb 21, 2025

Add a pairwise distance observable and an accumulator to track
particle contact times.

Co-authored-by: Jean-Noël Grad <jgrad@icp.uni-stuttgart.de>

Fixes #4665

Somesh Kurahatti authored Feb 19, 2025 and Jean-Noël Grad committed Feb 19, 2025

Co-authored-by: Rudolf Weeber <weeber@icp.uni-stuttgart.de>
Co-authored-by: Jean-Noël Grad <jgrad@icp.uni-stuttgart.de>

Fixes #1400 

Description of changes:
- Splits the kinetic energy returned by `espressomd.analyze.Analysis.energy()` into contributions due to translation and rotation

David Beyer authored Feb 18, 2025 and Jean-Noël Grad committed Feb 18, 2025

Co-authored-by: Jean-Noël Grad <jgrad@icp.uni-stuttgart.de>

Use plain LbmPy bounce back boundary, but with streaming_pattern=pull, as we call it before the stream sweep. The patched boundary additional data handling is still needed. This accounts for the fact that the C++ boundary implementation queries the fluid node (not the boundary node) for the boundary velocity.

via `system.analysis.particle_bond_energy(particle,bond)`.

This also makes sure `_bond_id` is set on `BondedInteraction` instances once the bond has been added to the system, and for instances returned by `particle.bonds`.

Description of changes:
- bump all compiler version requirements
   - C++20 core features are now almost fully supported
- remove now obsolete re-implementations of C++20 algorithms
- massively reduce the complexity of the CMake build system for tutorials
- fix deprecation warnings from recent versions of NumPy, SciPy, JupyterLab, and the Python standard library
- improve code coverage and remove unreachable code

Fix dangling references and MPI deadlocks.
Simplify management of object lists and maps.

Replace Utils functionality by equivalent C++20 features.
Remove unused or duplicated code. Improve code coverage.

Use standard CMake targets to configure tutorials. Mark figure files
as dependencies of the Jupyter notebooks that actually use them.
Make python and html targets depend on the original tutorial file in
the source directory. Convert solution cells to hidden solutions.