PyHOPE (Python High-Order Preprocessing Environment) is an open-source Python framework for the generation of three-dimensional unstructured high-order meshes. These meshes are needed by high-order numerical methods like Discontinuous Galerkin, Spectral Element Methods, or pFEM, in order to retain their accuracy if the computational domain includes curved boundaries.
PyHOPE has been developed by the Numerics Research Group (NRG) lead by Prof. Andrea Beck at the Institute of Aerodynamics and Gas Dynamics at the University of Stuttgart, Germany.
PyHOPE is heavily inspired by HOPR (High Order Preprocessor) and shares the same input/output format. Furthermore, PyHOPE utilizes Gmsh for the initial mesh generation and conversion before converting it to its internal representation. The internal representation is loosely based on meshio but augmented with additional information required for high-order meshes.
Installation
PyHOPE is built using standard Python packages. You can install PyHOPE by following these steps.
- Optional: Create and activate a virtual environment
Creating a virtual environment is a recommended practice to manage project dependencies. It isolates the packages required for PyHOPE and prevents potential conflicts between different package versions. To create and activate a virtual environment namedvenv, use the following commands.python -m venv venv source venv/bin/activate
If you choose not to use a virtual environment, skip this step and proceed directly to the installation of PyHOPE.
[!IMPORTANT]
For new shell sessions, the virtual environment must be re-sourced usingsource venv/bin/activatebefore usingpyhopecommands.
- Install PyHOPE
PyHOPE is installed usingpip, the Python package installer. This command fetches the PyHOPE package and its dependencies from PyPI (Python Package Index) and installs them.python -m pip install pyhope
[!NOTE]
PyHOPE can optionally use Numba to accelerate some compute-heavy routines. If Numba is not available in your environment, PyHOPE will fall back to pure NumPy implementations.python -m pip install pyhope[numba]
Testing
PyHOPE features internal health checks to verify that everything works as expected. The checks can be invoked directly from the terminal.
pyhope --verify [tutorials] # Run all health checks
pyhope --verify-health # Run Python health checks
pyhope --verify-install [tutorials] # Run PyHOPE mesh generation checks
pyhope --verify-unittest # Run PyHOPE unit tests
[!NOTE]
By default, PyHOPE looks for thetutorialsdirectory relative to the current working directory or the git root. If neither exists, PyHOPE downloads the tests from GitHub while using available authentication methods.
[!IMPORTANT]
Be aware that the PyHOPE repository uses Git Large File Storage (LFS) for some of its larger mesh files.
Getting Help
PyHOPE help output is formatted to serve as self-hosting INI format. A list of all options and the default values can be accessed by running the following command.
pyhope --help
Documentation
Refer to the PyHOPE documentation for the getting started guide, examples, and usage instructions. PyHOPE is heavily inspired by HOPR (High Order Preprocessor) and shares the same input/output format. For technical details and the reference mesh format specification, see the HOPR documentation.
Usage
PyHOPE can either be invoked directly from the command line or used as a Python library.
Command Line Usage
PyHOPE can be invoked from the command line. After installation, its functionalities can be accessed directly from the terminal by passing a valid configuration file.
pyhope [parameter.ini]
The following output is obtained when running the example configuration file tutorials/1-01-cartbox/parameter.ini.
$ pyhope tutorials/1-01-cartbox/parameter.ini
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
┃ P y H O P E — Python High-Order Preprocessing Environment
┃ PyHOPE version x.x.x
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
│ INIT PROGRAM...
│ nThreads │ nProcs-2 │ DEFAULT │
├─────────────────────────────────────────────
│ INIT OUTPUT...
│ ProjectName │ 1-01-cartbox │ *CUSTOM │
│ OutputFormat │ 0 [HDF5] │ *CUSTOM │
│ OutputBytes │ 0 [int32] │ DEFAULT │
│ DebugMesh │ T │ *CUSTOM │
│ DebugVisu │ F │ *CUSTOM │
├─────────────────────────────────────────────
│ INIT MESH...
│ Mode │ 1 [Internal] │ *CUSTOM │
│ NGeo │ 9 │ *CUSTOM │
├─────────────────────────────────────────────
│ GENERATE MESH...
├────
│ nZones │ 1 │ *CUSTOM │
├── Generating zone 1
│ Corner │ (/0.,0.,0.,,1.,0.,0.,,1.,1.,... │ *CUSTOM │
│ nElems │ (/8,8,8/) │ *CUSTOM │
│ ElemType │ 108 [hexahedron] │ *CUSTOM │
│ StretchType │ (/0,0,0/) │ DEFAULT │
│ BCIndex │ (/1,2,3,4,5,6/) │ *CUSTOM │
├────
├── Setting boundary conditions
├────
│ BoundaryName │ BC_zminus │ *CUSTOM │
│ BoundaryType │ (/4,0,0,0/) │ *CUSTOM │
│ BoundaryName │ BC_yminus │ *CUSTOM │
│ BoundaryType │ (/2,0,0,0/) │ *CUSTOM │
│ BoundaryName │ BC_xplus │ *CUSTOM │
│ BoundaryType │ (/2,0,0,0/) │ *CUSTOM │
│ BoundaryName │ BC_yplus │ *CUSTOM │
│ BoundaryType │ (/2,0,0,0/) │ *CUSTOM │
│ BoundaryName │ BC_xminus │ *CUSTOM │
│ BoundaryType │ (/2,0,0,0/) │ *CUSTOM │
│ BoundaryName │ BC_zplus │ *CUSTOM │
│ BoundaryType │ (/9,0,0,0/) │ *CUSTOM │
├────
│ vv │ (/1.,0.,0./) │ *CUSTOM │
│ vv │ (/0.,1.,0./) │ *CUSTOM │
│ vv │ (/0.,0.,1./) │ *CUSTOM │
├────
├── Generated mesh with 512 cells
├─────────────────────────────────────────────
├── BUILD DATA STRUCTURE...
├────
├── Removing duplicate points
├────
├── Generating sides
├─────────────────────────────────────────────
│ SORT MESH...
├────
│ MeshSorting │ 1 [SFC] │ DEFAULT │
│ MeshSortingSFC │ 0 [default] │ DEFAULT │
├────
├── Sorting elements along space-filling curve
├─────────────────────────────────────────────
│ CHECK NORMALS POINTING OUTWARDS...
├────
│ CheckSurfaceNormals │ True │ DEFAULT │
│ Processing Elements |█████████████████████████████████| 512/512 [100%] in 0.0s (24000.00/s)
├─────────────────────────────────────────────
│ CONNECT MESH...
├────
│ doPeriodicCorrect │ False │ DEFAULT │
│ doMortars │ True │ DEFAULT │
│ doMortarRebuild │ 1 [auto] │ DEFAULT │
│ Processing Sides |█████████████████████████████████| 3072/3072 [100%] in 0.0s (24000.00/s)
├────
│ Number of sides : 3072
│ Number of inner sides : 2688
│ Number of mortar sides (big) : 0
│ Number of mortar sides (small) : 0
│ Number of boundary sides : 384
│ Number of periodic sides : 0
├─────────────────────────────────────────────
│ CHECK CONNECTIVITY...
├────
│ CheckConnectivity │ True │ DEFAULT │
│ Processing Elements |█████████████████████████████████| 512/512 [100%] in 0.0s (24000.00/s)
├─────────────────────────────────────────────
│ CHECK WATERTIGHTNESS...
├────
│ CheckWatertightness │ True │ DEFAULT │
│ Processing Elements |█████████████████████████████████| 512/512 [100%] in 0.0s (24000.00/s)
├─────────────────────────────────────────────
│ CHECK JACOBIANS...
├────
│ CheckElemJacobians │ True │ DEFAULT │
│ Processing Elements |█████████████████████████████████| 512/512 [100%] in 0.0s (24000.00/s)
├────
│ Scaled Jacobians
├─────────────────
│<0.0 │ 0.00
│ 0.0-0.1 │ 0.00
│ 0.1-0.2 │ 0.00
│ 0.2-0.3 │ 0.00
│ 0.3-0.4 │ 0.00
│ 0.4-0.5 │ 0.00
│ 0.5-0.6 │ 0.00
│ 0.6-0.7 │ 0.00
│ 0.7-0.8 │ 0.00
│ 0.8-0.9 │ 0.00
│>0.9-1.0 │ ▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 512.00
├─────────────────
├─────────────────────────────────────────────
│ OUTPUT MESH...
├────
│ Curved Hexahedra : 512
├────
├── Writing HDF5 mesh to "1-01-cartbox_mesh.h5"
├── Writing XDMF mesh to "1-01-cartbox_DebugMesh.xdmf"
┢━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
┃ PyHOPE completed in [0.25 sec]
┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Python Library Usage
PyHOPE can be included in other Python libraries. PyHOPE exposes its functionally via runtime contexts defined by Context Managers. The following Python code loads a HOPR HDF5 mesh and derived quantities. For a complete list of currently implemented functions, see the PyHOPE documentation.
from pyhope import Basis, Mesh
with Mesh('1-01-cartbox_mesh.h5') as m:
elems = m.elems
lobatto_nodes = Basis.legendre_gauss_lobatto_nodes(order=m.nGeo)
Cite
This is a scientific project. If you use PyHOPE for publications or presentations in science, please support the project by citing the following article.
@article{kopper2025pyhope:joss,
title = {{PyHOPE}: A Python Toolkit for Three-Dimensional Unstructured High-Order Meshes},
author = {Kopper, Patrick and Blind, Marcel P. and Schwarz, Anna and Kurz, Marius and Rodach, Felix and Copplestone, Stephen M. and Beck, Andrea D.},
journal = {Journal of Open Source Software},
year = {2025},
volume = {10},
number = {115},
pages = {8769},
publisher = {The Open Journal},
doi = {10.21105/joss.08769}
}
In addition, you can also directly refer to this repository as
@misc{kopper2025pyhope:repo,
title = {{PyHOPE}: A Python Toolkit for Three-Dimensional Unstructured High-Order Meshes},
author = {Kopper, Patrick and Blind, Marcel P. and Schwarz, Anna and Kurz, Marius and Rodach, Felix and Copplestone, Stephen M. and Beck, Andrea D.},
year = {2025},
howpublished = {\url{https://github.com/hopr-framework/PyHOPE}},
doi = {10.5281/zenodo.17414843}
}
Metadata
Release files for PyHOPE 1.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| pyhope-1.1.0.tar.gz | 202.6 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| pyhope-1.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 474.6 kB
Release files / pyhope-1.1.0.tar.gz
| Download URL | pyhope-1.1.0.tar.gz |
|---|---|
| Size | 202.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
b08885850f7bbb628eb6a6bb799d86aa24ea4bfa6db74026dcde57eea71773e5
|
|
BLAKE2b-256 checksum How to use checksums |
c8c2de4d615e51e0384505a1cc73531d31008a32dc85e6d94554e92dedfa011b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 7, 2026.
Transparency logRelease files / pyhope-1.1.0-py3-none-any.whl
| Download URL | pyhope-1.1.0-py3-none-any.whl |
|---|---|
| Size | 272.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
d3f004c8c0953ec0e26e6b0e02d815a12f40a35bb1ebf197757c7be7f6412819
|
|
BLAKE2b-256 checksum How to use checksums |
2ffd257e625ac8d5ce9f22b71459f7d49f7aaab8a0907b311a34cc8475b1f12b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 7, 2026.
Transparency log