def parse_multi_xyz(filename):
"""
Extract multiple molecules from an XYZ file
Note: This file will fail if not given a valid XYZ file
TODO: Do some more elegant parsing to ensure that the xyz file is valid
Args:
filename (str): The multi-XYZ file to be parsed.
Returns:
molecules (list of Molecule objects)
"""
molecules = list()
with open(filename) as molfile:
text = molfile.readlines()
linenum = 0
while linenum < len(text):
try:
num_atoms = int(text[linenum].strip())
mol = Molecule.from_str(
"".join(text[linenum:linenum + num_atoms + 2]), "xyz")
molecules.append(mol)
linenum += num_atoms + 2
except ValueError:
break
return molecules
def test_main(self):
o = Molecule.from_str(rhb18xyz, "xyz")
o.set_charge_and_spin(-1, 3)
task = AdfTask("optimize", **rhb18)
inp = AdfInput(task)
inp.write_file(o, self.tempfile)
s = readfile(join(test_dir, "adf", "RhB18_adf.inp"))
self.assertEqual(readfile(self.tempfile), s)
def test_main(self):
o = Molecule.from_str(rhb18xyz, "xyz")
o.set_charge_and_spin(-1, 3)
task = AdfTask("optimize", **rhb18)
inp = AdfInput(task)
inp.write_file(o, self.tempfile)
s = readfile(join(test_dir, "adf", "RhB18_adf.inp"))
self.assertEqual(readfile(self.tempfile), s)
def test_to_from_file_string(self):
for fmt in ["xyz", "json", "g03"]:
s = self.mol.to(fmt=fmt)
self.assertIsNotNone(s)
m = Molecule.from_str(s, fmt=fmt)
self.assertEqual(m, self.mol)
self.assertIsInstance(m, Molecule)
self.mol.to(filename="CH4_testing.xyz")
self.assertTrue(os.path.exists("CH4_testing.xyz"))
os.remove("CH4_testing.xyz")
def test_to_from_file_string(self):
for fmt in ["xyz", "json", "g03"]:
s = self.mol.to(fmt=fmt)
self.assertIsNotNone(s)
m = Molecule.from_str(s, fmt=fmt)
self.assertEqual(m, self.mol)
self.assertIsInstance(m, Molecule)
self.mol.to(filename="CH4_testing.xyz")
self.assertTrue(os.path.exists("CH4_testing.xyz"))
os.remove("CH4_testing.xyz")
def mol_from_string(string, fmt):
"""
Reads a string into a pymatgen Molecule. Uses the pymatgen IMolecule method from_str.
Args:
string: a string containing the molecular data
fmt: the conversion format to use
Returns:
a pymatgen Molecule object
"""
try:
return Molecule.from_str(string, fmt)
except:
printx("Error: could not convert string '"+str(fmt)+"' to Molecule.\n"
+"Default supported formats are xyz, gaussian and pymatgen JSON molecules.\n"
+"Installing openbabel allows for more extensions.", priority=1)
return
def test_get_smiles(self):
single_molecule = get_smiles(join(files_dir, "molecules"), ["1453094"])
self.assertEqual(single_molecule[0], 'c12c(cc(c(=O)n1C)C=O)cccc2')
all_mols = [
"1453094", "1738108", "1873402", "2045554", "21925165", "22125071",
"28599994", "31695576", "5078635", "6657763"
]
for mol in all_mols:
smiles = get_smiles(join(files_dir, "molecules"), [mol])[0]
file = join(files_dir, "molecules", mol, "{}.mol".format(mol))
mol_smiles = Molecule.from_str(smiles, "smi")
mol_file = Molecule.from_file(file)
smiles_species = sorted(
[str(e) for e in mol_smiles.species if str(e) != "H"])
file_species = sorted(
[str(e) for e in mol_file.species if str(e) != "H"])
self.assertSequenceEqual(smiles_species, file_species)
def test_atom_block_key(self):
block = AdfKey("atoms")
o = Molecule.from_str(h2oxyz, "xyz")
for site in o:
block.add_subkey(AdfKey(str(site.specie), list(site.coords)))
self.assertEqual(str(block), atoms_string)
def test_atom_block_key(self):
block = AdfKey("atoms")
o = Molecule.from_str(h2oxyz, "xyz")
for site in o:
block.add_subkey(AdfKey(str(site.specie), list(site.coords)))
self.assertEqual(str(block), atoms_string)
def from_string(cls, string, fmt="coord"):
"""
Creates an instance from a string. Could be the string of a coord file
or any format supported by pymatgen Molecule.
Args:
string (str): the string with the data.
fmt (str): the format of the data. could be "coord" for Turbomole
coord file or any format supported in pymatgen Molecule.
Returns:
An instance of MoleculeSystem.
"""
if fmt == "coord":
dg = DataGroups(string=string)
coordinates_str = dg.sdg("$coord", strict=True)
if not coordinates_str:
raise ValueError("The string does not contain $coord!")
mol, fi = get_mol_and_indices_frozen(coordinates_str)
int_def_str = dg.sdg("$intdef", strict=True)
int_def = []
if int_def_str:
lines = []
# remove empty lines and comments
for l in int_def_str.splitlines():
lstrip = l.strip()
if lstrip and not lstrip.startswith("#"):
lines.append(l)
int_def_str = "\n".join(lines)
# split based on the presence of the index plus the status.
# In a case like this:
# 1 k 1.0000000000000 stre 4 1 val= 1.80084
# 2 k 1.0000000000000 bend 4 3 1 val= 106.27756
# 1.0000000000000 bend 3 2 1
# 1.0000000000000 bend 2 4 1
# 3 f 1.0000000000000 tors 1 2 3 4
# will split in 3 groups based on the presence of the digit plus k, f, d or i
# at the beginning of the line.
r = r"^\s*\d+\s+[kfdi]\s+.*?(?=\s*\d+\s+[kfdi]\s+|\Z)"
for group in re.findall(r, int_def_str,
re.DOTALL | re.MULTILINE):
int_def.append(InternalDefinition.from_string(group))
user_def_bonds_str = dg.sdg("$user-defined bonds", strict=True)
user_def_bonds = set()
if user_def_bonds_str:
# parses a line of this form:
# 1-2, 3-4, 5|6
# splitting first on "," and then on "-" and "|"
for l in user_def_bonds_str.splitlines():
l = l.strip()
if not l or l.startswith("#"):
continue
for bond in l.split(","):
for separator in ("-", "|"):
if separator in bond:
bond_indices = bond.split(separator)
if len(bond_indices) != 2:
raise ValueError(
"Cannot parse user-defined bonds for line: {}"
.format(l))
index_1 = int(bond_indices[0]) - 1
index_2 = int(bond_indices[1]) - 1
user_def_bonds.add(
(index_1, separator, index_2))
break
else:
raise ValueError(
"Cannot parse user-defined bonds for line: {}".
format(l))
return cls(mol,
int_def=int_def,
frozen_indices=fi,
user_defined_bonds=user_def_bonds)
else:
return cls(Molecule.from_str(string, fmt))