All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
0.6.4 - 2024-10-01
- Encoders and parsers for CREST to support
energy
,gradient
,hessian
, andoptimization
calculations.
0.6.3 - 2024-09-12
parse_optimization_dir(...) -> OptimizationResults
for TeraChem.
0.6.2 - 2024-08-13
CREST
encoder and directory parser for conformer search output directories.
0.6.1 - 2024-08-08
- Program version parser for
CREST
stdout.
0.6.0 - 2024-06-10
- Updated to
qcio 0.10.0
to use newStructure
rather thanMolecule
nomenclature.
0.5.3 - 2024-04-04
- TeraChem
parse_git_commit
updated toparse_version_control_details
to accommodate older versions of TeraChem compiled from theHg
days.
- TeraChem
parse_version
parser. It was unused given the switch to only parsingSinglePointResults
objects and notProvenance
objects as well. Theparse_version_string
function is used byqcop
to get the version of the program. We do not need to set the version of the program atSinglePointResults.extras.program_version
anymore.
0.5.2 - 2023-09-27
- All input parsing details from the library.
encode
top level function and encoder for TeraChem input files.
- Added
FileType.stdout
as defaultfiletype
argument toparse
decorator to reduce boilerplate in parsers.
0.5.1 - 2023-09-19
- Dropped Python dependency from
^3.8.1
to^3.8
. Can't remember what older dependency required3.8.1
but it's not needed anymore.
0.5.0 - 2023-08-31
- Updated
pydantic
fromv1
->v2
.
0.4.1 - 2023-07-19
- Updated
qcio
from0.3.0
->0.4.0
.
0.4.0 - 2023-07-17
- Updated to used
qcio>=0.3.0
flattened models and theSinglePointResults
object.
0.3.2 - 2023-06-29
- Updated package description in pyproject.toml from TeraChem specific parsing and MolSSI to all QC packages and qcio.
0.3.1 - 2023-06-29
- Git commit parsing for TeraChem as part of version parsing
qcio
>=0.1.0
->>=0.2.0
0.3.0 - 2023-06-28
- Dropped support for
QCSchema
models and changed to qcio data models. parse
function now raisesNotImplementedError
and the default use case is to useparse_computed_prop
instead and ignore inputs and provenance data. This is the minimum spec since QC programs can be powered using structured inputs and qcop. I may go back to parsing entireSinglePointSuccess/FailedOutput
objects if use cases arise.
0.2.1 - 2023-03-25
- Generalized
CUDA error:
regex to catch all CUDA errors.
0.2.0 - 2023-03-24
cli
interface for parsing TeraChem outputs from the command line.parse_natoms
,parse_nmo
,parse_total_charge
,parse_spin_multiplicity
- Removed Hessian matrix dimension checking from
parse_hessian
. Dimension validation is already done viapydantic
on theAtomicResult
object.
0.1.0 - 2023-03-23
- Basic parsers for energy, gradient, Hessian (frequencies) calculations.
- Can return either
AtomicResult
orFailedOperation
objects depending on whether calculation succeeded or failed. - Tests for all parsers and the main
parse
function.