def handle_unparsablespecies(cif_string):
"""
Handles CIF parsing errors arising from unrecognizable species
:param cif_string: (str) cif file
:return: pymatgen structure object with appended unparsable species
"""
cif_string_new = ''
symbols = []
coords = []
occupancies = []
cif = CifFile.from_string(cif_string).data
for block in cif:
if 'standardized' in block:
cif_stdblock = cif[block]
break
for i, sym in enumerate(cif_stdblock['_atom_site_type_symbol']):
if 'OH' in sym:
symbols.append(sym)
coords.append([float(cif_stdblock['_atom_site_fract_x'][i]), float(cif_stdblock['_atom_site_fract_y'][i]),
float(cif_stdblock['_atom_site_fract_z'][i])])
occupancies.append(float(cif_stdblock['_atom_site_occupancy'][i]))
for key in cif:
cif_string_new += str(cif[key]) + '\n'
cif_string_new += '\n'
new_struct = CifParser.from_string(cif_string_new).get_structures()[0]
for specie_no in range(len(symbols)):
new_struct.append({DummySpecie('X'): occupancies[specie_no]}, coords[specie_no],
properties={"molecule": [symbols[specie_no]]})
return new_struct
def get_pmg_dict(cifstring: str):
"""
use pmg dict to parse cifstring, only deal with one structure per file
:param cifstring:
:return:
"""
cifdata = CifFile.from_string(cifstring).data
idnetifiers = list(cifdata.keys())
if len(idnetifiers) > 1:
warnings.warn('W: find more than 1 structures in this cif file!')
elif len(idnetifiers) == 0:
warnings.warn('W: no structure found by pymatgen parser!')
try:
identifier = idnetifiers[0]
except IndexError:
raise CifFileError('no identifier found in the ciffile!')
pymatgen_dict = list(cifdata.items())[0][1].data
# jmol writes '_atom_site_type_symbol', but not '_atom_site_label'
if '_atom_site_label' not in pymatgen_dict.keys():
warnings.warn('W: _atom_site_label not found in parsed dict')
atom_site_label = []
symbols = pymatgen_dict['_atom_site_type_symbol']
for i in range(len(symbols)):
s = symbols[i]
atom_site_label.append('{}{}'.format(s, i))
pymatgen_dict['_atom_site_label'] = atom_site_label
return identifier, pymatgen_dict
def handle_unparsablespecies(cif_string):
"""
Handles CIF parsing errors arising from unrecognizable species
:param cif_string: (str) cif file
:return: pymatgen structure object with appended unparsable species
"""
cif_string_new = ''
symbols = []
coords = []
occupancies = []
cif = CifFile.from_string(cif_string).data
for block in cif:
if 'standardized' in block:
cif_stdblock = cif[block]
break
for i, sym in enumerate(cif_stdblock['_atom_site_type_symbol']):
if 'OH' in sym:
symbols.append(sym)
coords.append([
float(cif_stdblock['_atom_site_fract_x'][i]),
float(cif_stdblock['_atom_site_fract_y'][i]),
float(cif_stdblock['_atom_site_fract_z'][i])
])
occupancies.append(float(cif_stdblock['_atom_site_occupancy'][i]))
for key in cif:
cif_string_new += str(cif[key]) + '\n'
cif_string_new += '\n'
new_struct = CifParser.from_string(cif_string_new).get_structures()[0]
for specie_no in range(len(symbols)):
new_struct.append({DummySpecie('X'): occupancies[specie_no]},
coords[specie_no],
properties={"molecule": [symbols[specie_no]]})
return new_struct
def __init__(self, filename, occupancy_tolerance=1., site_tolerance=1e-4):
self._occupancy_tolerance = occupancy_tolerance
self._site_tolerance = site_tolerance
if isinstance(filename, six.string_types):
self._cif = CifFile.from_file(filename)
else:
self._cif = CifFile.from_string(filename.read())
# store if CIF contains features from non-core CIF dictionaries
# e.g. magCIF
self.feature_flags = {}
self.errors = []
def is_magcif():
"""
Checks to see if file appears to be a magCIF file (heuristic).
"""
# Doesn't seem to be a canonical way to test if file is magCIF or
# not, so instead check for magnetic symmetry datanames
prefixes = [
'_space_group_magn', '_atom_site_moment',
'_space_group_symop_magn'
]
for d in self._cif.data.values():
for k in d.data.keys():
for prefix in prefixes:
if prefix in k:
return True
return False
self.feature_flags['magcif'] = is_magcif()
def is_magcif_incommensurate():
"""
Checks to see if file contains an incommensurate magnetic
structure (heuristic).
"""
# Doesn't seem to be a canonical way to test if magCIF file
# describes incommensurate strucure or not, so instead check
# for common datanames
if not self.feature_flags["magcif"]:
return False
prefixes = ['_cell_modulation_dimension', '_cell_wave_vector']
for d in self._cif.data.values():
for k in d.data.keys():
for prefix in prefixes:
if prefix in k:
return True
return False
self.feature_flags['magcif_incommensurate'] = is_magcif_incommensurate(
)
for k in self._cif.data.keys():
# pass individual CifBlocks to _sanitize_data
self._cif.data[k] = self._sanitize_data(self._cif.data[k])
def get_pmg_dict(cifstring):
cifdata = CifFile.from_string(cifstring).data
idnetifiers = list(cifdata.keys())
if len(idnetifiers) > 1:
warnings.warn('W: find more than 1 structures in this cif file!')
elif len(idnetifiers) == 0:
warnings.warn('W: no structure found by pymatgen parser!')
identifier = idnetifiers[0]
pymatgen_dict = list(cifdata.items())[0][1].data
return identifier, pymatgen_dict
def fix_incorrectlyparsedstructures_symbols(cif_string):
"""
Fixes already parsed CIF files with structures that have data with either incorrect labels. This function will
replace the labels with symbols.
:param cif_string: (str) cif file
:return: corrected cif string
"""
cif_string_new = ''
cif = CifFile.from_string(cif_string).data
for block in cif:
if 'standardized' in block:
cif_stdblock = cif[block]
break
for i, sym in enumerate(cif_stdblock['_atom_site_type_symbol']):
if sym != cif_stdblock['_atom_site_label'][i]:
cif_stdblock['_atom_site_label'][i] = sym
for key in cif:
cif_string_new += str(cif[key]) + '\n'
cif_string_new += '\n'
return cif_string_new
def fix_incorrectlyparsedstructures_symbols(cif_string):
"""
Fixes already parsed CIF files with structures that have data with either incorrect labels. This function will
replace the labels with symbols.
:param cif_string: (str) cif file
:return: corrected cif string
"""
cif_string_new = ''
cif = CifFile.from_string(cif_string).data
for block in cif:
if 'standardized' in block:
cif_stdblock = cif[block]
break
for i, sym in enumerate(cif_stdblock['_atom_site_type_symbol']):
if sym != cif_stdblock['_atom_site_label'][i]:
cif_stdblock['_atom_site_label'][i] = sym
for key in cif:
cif_string_new += str(cif[key]) + '\n'
cif_string_new += '\n'
return cif_string_new
def fix_incorrectlyparsedstructures_manually(cif_string):
"""
Fixes already parsed CIF files with random errors in them (used for last 8 incorrectly parsed structures).
:param cif_string: (str) cif file
:return: corrected cif string
"""
cif_string_new = ''
cif = CifFile.from_string(cif_string).data
for block in cif:
if 'standardized' in block:
cif_stdblock = cif[block]
break
for i, sym in enumerate(cif_stdblock['_atom_site_type_symbol']):
if sym == 'Mn':
cif_stdblock['_atom_site_occupancy'][i] = 0.76
elif sym == 'Ti':
cif_stdblock['_atom_site_occupancy'][i] = 0.12
elif sym == 'V':
cif_stdblock['_atom_site_occupancy'][i] = 0.12
for key in cif:
cif_string_new += str(cif[key]) + '\n'
cif_string_new += '\n'
return cif_string_new
def fix_incorrectlyparsedstructures_manually(cif_string):
"""
Fixes already parsed CIF files with random errors in them (used for last 8 incorrectly parsed structures).
:param cif_string: (str) cif file
:return: corrected cif string
"""
cif_string_new = ''
cif = CifFile.from_string(cif_string).data
for block in cif:
if 'standardized' in block:
cif_stdblock = cif[block]
break
for i, sym in enumerate(cif_stdblock['_atom_site_type_symbol']):
if sym == 'Mn':
cif_stdblock['_atom_site_occupancy'][i] = 0.76
elif sym == 'Ti':
cif_stdblock['_atom_site_occupancy'][i] = 0.12
elif sym == 'V':
cif_stdblock['_atom_site_occupancy'][i] = 0.12
for key in cif:
cif_string_new += str(cif[key]) + '\n'
cif_string_new += '\n'
return cif_string_new
db = client.springer
coll = db['pauling_file_unique_Parse']
newcoll = db['incorrect_labels']
if __name__ == '__main__':
d = 0
remove_keys = []
for doc in coll.find({'key': 'sd_1903187'}).batch_size(75).sort('_id', pymongo.ASCENDING).skip(d).limit(500):
d += 1
print '#########################'
print 'On record # {} and key {}'.format(d, doc['key'])
# new_cif_string = fix_incorrectlyparsedstructures_symbols(doc['cif_string'])
if 'structure' in doc:
print Structure.from_dict(doc['structure']).composition
cif = CifFile.from_string(doc['cif_string']).data
for block in cif:
if 'standardized' in block:
cif_stdblock = cif[block]
break
# print cif_stdblock['_atom_site_label']
# print cif_stdblock['_atom_site_type_symbol']
incorrect_symbol = False
for i, sym in enumerate(cif_stdblock['_atom_site_type_symbol']):
if sym not in cif_stdblock['_atom_site_label'][i] and ' + ' not in sym:
# print sym, cif_stdblock['_atom_site_label'][i]
cif_stdblock['_atom_site_label'][i] = sym
incorrect_symbol = True
if incorrect_symbol:
cif_string_new = ''
for key in cif: