Data-CAT 0.7.2

What's Changed

• REL: Update master branch after release (0.7.1) by @BvB93 in #63
• TST: Enable tests for Python 3.10 by @BvB93 in #64
• DEP: Replace pd.DataFrame.append with pd.concat by @BvB93 in #65
• BUG: Fixed size-1 molecular dataframes not being correctly updated by @BvB93 in #66

Full Changelog: 0.7.1...0.7.2

Data-CAT 0.7.1

What's Changed

• DEP: Deprecate usage of Molecule.get_formula and modernize CI by @BvB93 in #62

Full Changelog: 0.7.0...0.7.1

Data-CAT 0.7.0

What's Changed

• ENH: Add a new PDBContainer method for creating rdkit molecules by @BvB93 in #55
• ENH: Expose the index_to_pandas function by @BvB93 in #56
• BUG: Fixed an issue where properties weren't properly assigned to RDKit atoms by @BvB93 in #57
• BUG: Triage a bug wherein the hdf5 index would be of the wrong dtype by @BvB93 in #58
• MAINT: Store whether a Molecule is optimized when retrieving it from .hdf5 by @BvB93 in #59
• BLD: Bump the PLAMS minimum version to >= 1.5.1 by @BvB93 in #60
• REL: Release Data-CAT 0.7.0 by @BvB93 in #61

Full Changelog: 0.6.0...0.7.0

Data-CAT 0.6.0

Merge pull request #54 from nlesc-nano/dtype

BUG: Fixed an issue where property-dataset generation could fail for string arrays

Data-CAT 0.5.1

• Fixed an issue where not all properties were copied from the .csv file to the .hdf5 file.

Data-CAT 0.5.0

See https://github.com/nlesc-nano/data-CAT/milestone/5:

• Removed the .yaml job recipe storage component of the Database class.
• Added a function for validating property Groups.
• Cleaned up the Database class.
• Ensure that all content written to the .csv files is also written to the .hdf5 file.
• Pass the "logger" group directly to the hdf5_log functions.
• Added prop_to_dataframe(), a function for converting property Datasets into DataFrames.
• Added set-based methods to PDBContainer: .union(), .intersection(), .difference() & .symmetric_difference().
• Added PDBContainer.concatenate().

Data-CAT 0.4.0

See https://github.com/nlesc-nano/data-CAT/milestone/3:

• Updated the hdf5 logging system.
• Moved all dtype-related functions and data to dataCAT.dtype.
• Cleaned up the global namespace.
• Added the PDBContainer.index attribute.
• Create the index Dataset when calling PDBContainer.create_hdf5_group(). Aforementioned Dataset is used as dimensional scale.
• Provide users more control over the keyword arguments used for creating datasets.
• Fixed an issue wherePDBContainer.to_molecules() would set values as numpy scalars rather than using its builtin counterparts.
• Added an API for creating and managing datasets for containing quantum-mechanical properties in hdf5 format.
• Cleaned up the backwards-compatibility checks in dataCAT.create_database.