IMAS-Core is a lowlevel library with Python bindings for reading and writing fusion experiment data in standard IMAS format.
It provides easy access to data stored in HDF5, MDSplus, and other formats, making it simple to work with fusion physics data across platforms (Linux, macOS, Windows).
This repository contains the Lowlevel components of the Access Layer:
- C lowlevel interface used by the various High Level Interfaces
- Python bindings to the lowlevel interface
- Backends for reading and writing IMAS data
- MDS+ model logic for creating the models required by the MDS+ backend
Install IMAS-Core with a single command:
pip install imas-core
python -c "import imas_core"That's it! No need to compile or configure anything.
- ✅ Easy to Install - Single
pip installcommand - ✅ Multiple Formats - HDF5, MDSplus, UDA, in-memory, and more
- ✅ Cross-Platform - Works on Linux, macOS, and Windows
- ✅ IMAS Standard - Access standardized fusion data structures
- ✅ Read & Write - Both data access and creation supported
# Simple install from PyPI for Python applications
pip install imas-core
# Verify installation
python -c "import imas; print(imas.__version__)"See Building from Source for detailed build instructions.
When IMAS-Core is built and installed via CMake, it creates a complete runtime environment with:
- C/C++ Libraries (
libal.so) with full headers - Python Bindings (
imas_corePython package) - Fortran Support via pkg-config configuration
- Java Support via
imas.jar - MATLAB Support via MEX bindings
- Models for MDSplus backend access
After installation, configure your environment:
export PATH="/path/to/install/bin:$PATH"
export LD_LIBRARY_PATH="/path/to/install/lib:$LD_LIBRARY_PATH"
export PKG_CONFIG_PATH="/path/to/install/lib/pkgconfig:$PKG_CONFIG_PATH"
export HDF5_USE_FILE_LOCKING=FALSE
export PYTHONPATH="/path/to/install/lib/pythonX.X/site-packages:$PYTHONPATH"Then use IMAS-Core from your preferred language:
- Python:
import imas(see examples above) - C/C++: Link against
libal.sowith provided headers - Fortran: Use pkg-config to get compiler flags
- Java: Use
imas.jarin your classpath - MATLAB: Add MEX directory to MATLAB path
- User Guide - Complete user documentation
- Installation Guide - Installation instructions
- Backends Guide - Available data backends
- URIs Guide - Data entry URI documentation
- Configuration - Configuration options
- FAQ - Frequently asked questions
- Troubleshooting - Common issues & solutions
- Python: 3.8 or newer
- OS: Linux, macOS, or Windows
- pip: 19.0 or newer
IMAS-Core supports multiple data storage formats:
| Backend | Use Case | Remote | File-based |
|---|---|---|---|
| HDF5 | Default, local storage | No | Yes |
| MDSplus | ITER experiments | Yes | No |
| UDA | Distributed access | Yes | No |
| Memory | Testing, IPC | No | No |
| FlexBuffers | Message passing | No | Yes |
| ASCII | Debugging | No | Yes |
Can't find file?
# Check file path
import os
print(os.path.exists('/path/to/file.h5'))Need help?
- See Troubleshooting Guide
- Check FAQ
- Open an Issue on GitHub
IMAS-Core provides:
- Python API - Full Python bindings with NumPy support
- Multiple Backends - HDF5, MDSplus, UDA, and more
- Data Creation - Create and populate IMAS IDS structures
- Data Access - Read from multiple sources transparently
- Standard Format - IMAS standardized data structures
To build IMAS-Core from source:
git clone https://github.com/iterorganization/IMAS-Core.git
cd IMAS-Core
cmake -Bbuild -GNinja -DAL_PYTHON_BINDINGS=ON -DCMAKE_INSTALL_PREFIX="$(pwd)/test-install"
cmake --build build --target installSee Developer Guide for detailed instructions.
- Homepage:
- GitHub: https://github.com/iterorganization/IMAS-Core
- PyPI: https://pypi.org/project/imas-core/
- Data Dictionary: https://imas-data-dictionary.readthedocs.io/
- Issue Tracker: https://github.com/iterorganization/IMAS-Core/issues
IMAS-Core is released under the LGPL-3.0 License
- Email: imas-support@iter.org
- Documentation: https://imas-core.readthedocs.io/
- Issues: https://github.com/iterorganization/IMAS-Core/issues
