def _get_xyz_dset(f: File) -> Tuple[np.ndarray, Dict[str, List[int]]]:
"""Return the ``"xyz"zz dataset from **f**.
Parameters
----------
f : |h5py.File|_
An opened hdf5 file.
Returns
-------
:math:`k*m*n*3` |np.ndarray|_ and |dict|_:
An array of :math:`k` MultiMolecule instances with :math:`m` molecules and
:math:`n` atoms.
"""
key = 'xyz'
# Construct a dictionary with atomic symbols and matching atomic indices
iterator = enumerate(f[key].attrs['atoms'])
idx_dict = group_by_values(iterator)
# Extract the Cartesian coordinates; sort in chronological order
i = f.attrs['sub-iteration']
j = f[key].shape[0] - i
ret: np.ndarray = np.empty_like(f[key])
ret[:j] = f[key][i:]
ret[j:] = f[key][:i]
return ret, idx_dict
def mol_from_hdf5(filename: PathType, i: int = -1, j: int = 0) -> Tuple3:
"""Read a single dataset from a (multi) .xyz.hdf5 file.
Returns values for the :class:`MultiMolecule` ``coords``, ``atoms`` and ``bonds`` parameters.
Parameters
----------
filename : :class:`str`
The path+name of the .xyz.hdf5 file.
i : :class:`int`
The (sub-)iteration number of the to-be returned trajectory.
j : :class:`int`
The index of the to-be returned dataset.
For example: :code:`j=0`is equivalent to the :code:`'xyz.0'` dataset.
Returns
-------
:class:`numpy.ndarray`, :class:`dict` [:class:`str`, :class:`list` [:class:`int`]] and :class:`numpy.ndarray`.
* A 3D array with Cartesian coordinates of :math:`m` molecules with :math:`n` atoms.
* A dictionary with atomic symbols as keys and lists of matching atomic indices as values.
* A 2D array of bonds and bond-orders.
""" # noqa
with h5py.File(filename, 'r', libver='latest') as f:
dset = f[f'xyz.{j}']
return (dset[i],
group_by_values(
enumerate(dset.attrs['atoms'].astype(str).tolist())),
dset.attrs['bonds'])
def get_indices(mol: Molecule) -> Dict[str, np.ndarray]:
"""Construct a dictionary with atomic symbols as keys and arrays of indices as values."""
elements = [at.symbol for at in mol.atoms]
symbol_enumerator = enumerate(elements)
idx_dict = group_by_values(symbol_enumerator)
for k, v in idx_dict.items():
idx_dict[k] = np.fromiter(v, dtype=int, count=len(v))
return idx_dict
def _get_idx_dict(kf: KFReader) -> IdxDict:
"""Extract atomic symbols and matching atomic indices from **kf**.
Parameters
----------
kf : |plams.KFReader|_
A KFReader instance constructed from a KF binary file.
Returns
-------
|dict|_ [|str|_, |list|_ [|int|_]]:
A dictionary with atomic symbols and a list of matching atomic indices.
"""
at_list = kf.read('Molecule', 'AtomSymbols').split()
return group_by_values(enumerate(at_list))
def _get_idx_dict(f: Iterable[str], atom_count: int) -> Dict[str, List[int]]:
"""Extract atomic symbols and matching atomic indices from **f**.
Parameters
----------
f : |io.TextIOWrapper|_
An opened .xyz file.
atom_count : int
The number of atoms per molecule.
Returns
-------
|dict|_ [|str|_, |list|_ [|int|_]]:
A dictionary with atomic symbols and a list of matching atomic indices.
"""
stop = 1 + atom_count
atom_list = [at.split(maxsplit=1)[0].capitalize() for at in islice(f, 1, stop)]
return group_by_values(enumerate(atom_list))