Releases: mphowardlab/lammpsio
Releases · mphowardlab/lammpsio
0.9.0
Added
- Inverse map (label to type ID) in
LabelMap. - Type hints.
- Support for Python 3.14.
- Tutorial on initializing an atomistic simulation.
- Tutorial on using NumPy and Numba to accelerate calculations.
Fixed
- Jupyter notebooks for tutorials execute as scripts.
- Mass section of data file is written in the same spot as by LAMMPS.
Removed
- Support for Python 3.9.
0.8.0
Added
- Complete sphinx documentation is hosted on Read the Docs.
- Unit tests that run against LAMMPS.
- Convenience methods for working with boxes: to/from a matrix and to/from a HOOMD box definition.
Fixed
- Lists of known/unknown headers and bodies for
DataFileare now current. - Particle positions are now correctly recentered when converting a
Snapshotto a GSDFrameif theBoxis not centered at the origin.
0.7.0
Added
- Initial support for type labels of particle and topology data through the
LabelMapobject. - Interconversion of topology data with GSD
Frame. - Support for Python 3.13.
- Copying of topology data from an existing
Snapshotwhen reading a dump file.
Fixed
- Only GSD properties that have been assigned are converted to a
Snapshot. - Issues with writing topology information to data files.
- Deduction of atom type from information in
Snapshot.
Removed
- Support for Python 3.8.
0.6.1
0.6.0
0.5.0
Added
- Basic support for molecular topology data. These data are exposed as
Bonds,Angles,DihedralsandImpropersobjects that can be included in aSnapshotand read/written to aDataFile. Some features are not yet fully supported, such as conversion to/from GSD format and as an option tocopy_fromfor aDumpFile. - Testing for Python 3.12.
Changed
- Bumped license year to 2024.
0.4.1
0.4.0
Added
- Snapshot can be created from and converted to a GSD HOOMD frame.
- Package version is embedded in
__version__.
Changed
- Python 3.11 is supported and tested.
- Code style is enforced using
blackandflake8. Developers should install
requirements-dev.txtand configurepre-commit. - Classes are broken into modules for readability. The user API does not change.
- NumPy arrays use
floatandintas data types instead of specified precision.