def parse_dipole_moment(path: str) -> Optional[float]:
"""
Parse the dipole moment in Debye from an opt job output file.
Args:
path: The ESS log file.
Returns: Optional[float]
The dipole moment in Debye.
"""
lines = _get_lines_from_file(path)
log = ess_factory(path)
dipole_moment = None
if isinstance(log, GaussianLog):
# example:
# Dipole moment (field-independent basis, Debye):
# X= -0.0000 Y= -0.0000 Z= -1.8320 Tot= 1.8320
read = False
for line in lines:
if 'dipole moment' in line.lower() and 'debye' in line.lower():
read = True
elif read:
dipole_moment = float(line.split()[-1])
read = False
elif isinstance(log, MolproLog):
# example: ' Dipole moment /Debye 2.96069859 0.00000000 0.00000000'
for line in lines:
if 'dipole moment' in line.lower() and '/debye' in line.lower():
splits = line.split()
dm_x, dm_y, dm_z = float(splits[-3]), float(splits[-2]), float(
splits[-1])
dipole_moment = (dm_x**2 + dm_y**2 + dm_z**2)**0.5
elif isinstance(log, OrcaLog):
# example: 'Magnitude (Debye) : 2.11328'
for line in lines:
if 'Magnitude (Debye)' in line:
dipole_moment = float(line.split()[-1])
elif isinstance(log, QChemLog):
# example:
# Dipole Moment (Debye)
# X 0.0000 Y 0.0000 Z 2.0726
# Tot 2.0726
skip = False
read = False
for line in lines:
if 'dipole moment' in line.lower() and 'debye' in line.lower():
skip = True
elif skip:
skip = False
read = True
elif read:
dipole_moment = float(line.split()[-1])
read = False
elif isinstance(log, TeraChemLog):
# example: 'DIPOLE MOMENT: {-0.000178, -0.000003, -0.000019} (|D| = 0.000179) DEBYE'
for line in lines:
if 'dipole moment' in line.lower() and 'debye' in line.lower():
splits = line.split('{')[1].split('}')[0].replace(',',
'').split()
dm_x, dm_y, dm_z = float(splits[0]), float(splits[1]), float(
splits[2])
dipole_moment = (dm_x**2 + dm_y**2 + dm_z**2)**0.5
else:
raise ParserError(
'Currently dipole moments can only be parsed from either Gaussian, Molpro, Orca, QChem, '
'or TeraChem optimization output files')
if dipole_moment is None:
raise ParserError('Could not parse the dipole moment')
return dipole_moment
def parse_xyz_from_file(path: str) -> Optional[Dict[str, tuple]]:
"""
Parse xyz coordinated from:
- .xyz: XYZ file
- .gjf: Gaussian input file
- .out or .log: ESS output file (Gaussian, Molpro, Orca, QChem, TeraChem) - calls parse_geometry()
- other: Molpro or QChem input file
Args:
path (str): The file path.
Raises:
ParserError: If the coordinates could not be parsed.
Returns: Optional[Dict[str, tuple]]
The parsed cartesian coordinates.
"""
lines = _get_lines_from_file(path)
file_extension = os.path.splitext(path)[1]
xyz = None
relevant_lines = list()
if file_extension == '.xyz':
for i, line in enumerate(reversed(lines)):
splits = line.strip().split()
if len(splits) == 1 and all([c.isdigit() for c in splits[0]]):
# this is the last number of atoms line (important when parsing trajectories)
num_of_atoms = int(splits[0])
break
else:
raise ParserError(
f'Could not identify the number of atoms line in the xyz file {path}'
)
index = len(lines) - i - 1
relevant_lines = lines[index + 2:index + 2 + num_of_atoms]
elif file_extension == '.gjf':
start_parsing = False
for line in lines:
if start_parsing and line and line != '\n' and line != '\r\n':
relevant_lines.append(line)
elif start_parsing:
break
else:
splits = line.split()
if len(splits) == 2 and all([s.isdigit() for s in splits]):
start_parsing = True
elif 'out' in file_extension or 'log' in file_extension:
xyz = parse_geometry(path)
else:
record = False
for line in lines:
if '$end' in line or '}' in line:
break
if record and len(line.split()) == 4:
relevant_lines.append(line)
elif '$molecule' in line:
record = True
elif 'geometry={' in line:
record = True
if not relevant_lines:
raise ParserError(
f'Could not parse xyz coordinates from file {path}')
if xyz is None and relevant_lines:
xyz = str_to_xyz(''.join([line for line in relevant_lines if line]))
return xyz
def parse_trajectory(path: str) -> List[Dict[str, tuple]]:
"""
Parse all geometries from an xyz trajectory file or an ESS output file.
Args:
path (str): The file path.
Raises:
ParserError: If the trajectory could not be read.
Returns: List[Dict[str, tuple]]
Entries are xyz's on the trajectory.
"""
lines = _get_lines_from_file(path)
ess_file = False
if path.split('.')[-1] != 'xyz':
try:
log = ess_factory(fullpath=path)
ess_file = True
except InputError:
ess_file = False
if ess_file:
if not isinstance(log, GaussianLog):
raise NotImplementedError(
f'Currently parse_trajectory only supports Gaussian files, got {type(log)}'
)
traj = list()
done = False
i = 0
while not done:
if i >= len(lines) or 'Normal termination of Gaussian' in lines[
i] or 'Error termination via' in lines[i]:
done = True
elif 'Input orientation:' in lines[i]:
i += 5
xyz_str = ''
while len(
lines
) and '--------------------------------------------' not in lines[
i]:
splits = lines[i].split()
xyz_str += f'{qcel.periodictable.to_E(int(splits[1]))} {splits[3]} {splits[4]} {splits[5]}\n'
i += 1
traj.append(str_to_xyz(xyz_str))
i += 1
else:
# this is not an ESS output file, probably an XYZ format file with several Cartesian coordinates
skip_line = False
num_of_atoms = 0
traj, xyz_lines = list(), list()
for line in lines:
splits = line.strip().split()
if len(splits) == 1 and all([c.isdigit() for c in splits[0]]):
if len(xyz_lines):
if len(xyz_lines) != num_of_atoms:
raise ParserError(
f'Could not parse trajectory, expected {num_of_atoms} atoms, '
f'but got {len(xyz_lines)} for point {len(traj) + 1} in the trajectory.'
)
traj.append(
str_to_xyz(''.join(
[xyz_line for xyz_line in xyz_lines])))
num_of_atoms = int(splits[0])
skip_line = True
xyz_lines = list()
elif skip_line:
# skip the comment line
skip_line = False
continue
else:
xyz_lines.append(line)
if len(xyz_lines):
# add the last point in the trajectory
if len(xyz_lines) != num_of_atoms:
raise ParserError(
f'Could not parse trajectory, expected {num_of_atoms} atoms, '
f'but got {len(xyz_lines)} for point {len(traj) + 1} in the trajectory.'
)
traj.append(
str_to_xyz(''.join([xyz_line for xyz_line in xyz_lines])))
if not len(traj):
raise ParserError(f'Could not parse trajectory from {path}')
return traj
def parse_frequencies(
path: str,
software: str,
) -> np.ndarray:
"""
Parse the frequencies from a freq job output file.
Args:
path (str): The log file path.
software (str): The ESS.
Returns: np.ndarray
The parsed frequencies (in cm^-1).
"""
lines = _get_lines_from_file(path)
freqs = np.array([], np.float64)
if software.lower() == 'qchem':
for line in lines:
if ' Frequency:' in line:
items = line.split()
for i, item in enumerate(items):
if i:
freqs = np.append(freqs, [(float(item))])
elif software.lower() == 'gaussian':
with open(path, 'r') as f:
line = f.readline()
while line != '':
# this line intends to only capture the last occurrence of the frequencies
if 'and normal coordinates' in line:
freqs = np.array([], np.float64)
if 'Frequencies --' in line:
freqs = np.append(freqs,
[float(frq) for frq in line.split()[2:]])
line = f.readline()
elif software.lower() == 'molpro':
read = False
for line in lines:
if 'Nr' in line and '[1/cm]' in line:
continue
if read:
if line == os.linesep:
read = False
continue
freqs = np.append(freqs, [float(line.split()[-1])])
if 'Low' not in line and 'Vibration' in line and 'Wavenumber' in line:
read = True
elif software.lower() == 'orca':
with open(path, 'r') as f:
line = f.readline()
read = True
while line:
if 'VIBRATIONAL FREQUENCIES' in line:
while read:
if not line.strip():
line = f.readline()
elif not line.split()[0] == '0:':
line = f.readline()
else:
read = False
while line.strip():
if float(line.split()[1]) != 0.0:
freqs = np.append(freqs, [float(line.split()[1])])
line = f.readline()
break
else:
line = f.readline()
elif software.lower() == 'terachem':
read_output = False
for line in lines:
if '=== Mode' in line:
# example: '=== Mode 1: 1198.526 cm^-1 ==='
freqs = np.append(freqs, [float(line.split()[3])])
elif 'Vibrational Frequencies/Thermochemical Analysis After Removing Rotation and Translation' in line:
read_output = True
continue
elif read_output:
if 'Temperature (Kelvin):' in line or 'Frequency(cm-1)' in line:
continue
if not line.strip():
break
# example:
# 'Mode Eigenvalue(AU) Frequency(cm-1) Intensity(km/mol) Vib.Temp(K) ZPE(AU) ...'
# ' 1 0.0331810528 170.5666870932 52.2294230772 245.3982965841 0.0003885795 ...'
freqs = np.append(freqs, [float(line.split()[2])])
else:
raise ParserError(
f'parse_frequencies() can currently only parse Gaussian, Molpro, Orca, QChem and TeraChem '
f'files, got {software}')
logger.debug(
f'Using parser.parse_frequencies(). Determined frequencies are: {freqs}'
)
return freqs