def parse_ecp(string, symb=None):
if symb is not None:
symb = _std_symbol(symb)
raw_data = string.splitlines()
for i, dat in enumerate(raw_data):
dat0 = dat.split(None, 1)
if dat0 and dat0[0] == symb:
break
if i + 1 == len(raw_data):
raise BasisNotFoundError('ECP not found for %s' % symb)
seg = []
for dat in raw_data[i:]:
dat = dat.strip().upper()
if dat: # remove empty lines
if ((dat[0].isalpha()
and dat.split(None, 1)[0] != symb.upper())):
break
else:
seg.append(dat)
else:
seg = string.splitlines()
ecptxt = []
for dat in seg:
x = dat.split('#')[0].strip().upper()
if (x and not x.startswith('END') and not x.startswith('ECP')):
ecptxt.append(x)
return _parse_ecp(ecptxt)
def _parse(raw_basis, optimize=True):
basis_add = []
for line in raw_basis:
dat = line.strip()
if not dat or dat.startswith('#'):
continue
elif dat[0].isalpha():
key = dat.split()[1]
if key == 'SP':
basis_add.append([0])
basis_add.append([1])
else:
basis_add.append([MAPSPDF[key]])
else:
try:
line = [float(x) for x in dat.replace('D', 'e').split()]
except Exception as e:
raise BasisNotFoundError(
'\n' + str(e) +
'\nor the required basis file not existed.')
if key == 'SP':
basis_add[-2].append([line[0], line[1]])
basis_add[-1].append([line[0], line[2]])
else:
basis_add[-1].append(line)
basis_sorted = []
for l in range(MAXL):
basis_sorted.extend([b for b in basis_add if b[0] == l])
if optimize:
basis_sorted = optimize_contraction(basis_sorted)
basis_sorted = remove_zero(basis_sorted)
return basis_sorted
def parse(string, symb=None, optimize=True):
'''Parse the basis text which is in NWChem format. Return an internal
basis format which can be assigned to attribute :attr:`Mole.basis`
Empty lines, or the lines started with #, or the lines of "BASIS SET" and
"END" will be ignored are ignored.
Args:
string : A string in NWChem basis format. Empty links and the lines of
"BASIS SET" and "END" will be ignored
Kwargs:
optimize : Optimize basis contraction. Convert the segment contracted
basis to the general contracted basis.
Examples:
>>> mol = gto.Mole()
>>> mol.basis = {'O': gto.basis.parse("""
... #BASIS SET: (6s,3p) -> [2s,1p]
... C S
... 71.6168370 0.15432897
... 13.0450960 0.53532814
... 3.5305122 0.44463454
... C SP
... 2.9412494 -0.09996723 0.15591627
... 0.6834831 0.39951283 0.60768372
... 0.2222899 0.70011547 0.39195739
... """)}
>>> gto.basis.parse("""
... He S
... 13.6267000 0.1752300
... 1.9993500 0.8934830
... 0.3829930 0.0000000
... He S
... 13.6267000 0.0000000
... 1.9993500 0.0000000
... 0.3829930 1.0000000
... """, optimize=True)
[[0, [13.6267, 0.17523, 0.0], [1.99935, 0.893483, 0.0], [0.382993, 0.0, 1.0]]]
'''
if symb is not None:
symb = _std_symbol(symb)
string = _search_basis_block(re.split(BASIS_SET_DELIMITER, string),
symb)
if not string:
raise BasisNotFoundError('Basis not found for %s' % symb)
raw_basis = []
for dat in string.splitlines():
dat = dat.split('#')[0].strip() # Use # to start comments
dat_upper = dat.upper()
if (dat and not dat_upper.startswith('END')
and not dat_upper.startswith('BASIS')):
raw_basis.append(dat)
return _parse(raw_basis, optimize)
def search_seg(basisfile, symb):
with open(basisfile, 'r') as fin:
# ignore head
dat = fin.readline().lstrip(' ')
while dat.startswith('#'):
dat = fin.readline().lstrip(' ')
# searching
while dat:
if dat[:-1].rstrip(' ') == symb:
seg = []
fin.readline() # pass the "back to top" mark
while (dat and not dat.startswith('back to top')):
seg.append(dat)
dat = fin.readline().lstrip(' ')
return seg[:-1]
dat = fin.readline().lstrip(' ')
raise BasisNotFoundError('Basis not found for %s in %s' %
(symb, basisfile))
def _parse(raw_basis, optimize=True):
basis_parsed = [[] for l in range(MAXL)]
key = None
for line in raw_basis:
dat = line.strip()
if not dat or dat.startswith('#'):
continue
elif dat[0].isalpha():
key = dat.split()[1].upper()
if key == 'SP':
basis_parsed[0].append([0])
basis_parsed[1].append([1])
else:
l = MAPSPDF[key]
current_basis = [l]
basis_parsed[l].append(current_basis)
else:
dat = dat.replace('D','e').split()
try:
dat = [float(x) for x in dat]
except ValueError:
if DISABLE_EVAL:
raise ValueError('Failed to parse basis %s' % line)
else:
dat = list(eval(','.join(dat)))
except Exception as e:
raise BasisNotFoundError('\n' + str(e) +
'\nor the required basis file not existed.')
if key is None:
raise RuntimeError('Failed to parse basis')
elif key == 'SP':
basis_parsed[0][-1].append([dat[0], dat[1]])
basis_parsed[1][-1].append([dat[0], dat[2]])
else:
current_basis.append(dat)
basis_sorted = [b for bs in basis_parsed for b in bs]
if optimize:
basis_sorted = optimize_contraction(basis_sorted)
basis_sorted = remove_zero(basis_sorted)
return basis_sorted
def load(filename_or_basisname, symb, optimize=OPTIMIZE_CONTRACTION):
'''Convert the basis of the given symbol to internal format
Args:
filename_or_basisname : str
Case insensitive basis set name. Special characters will be removed.
or a string of "path/to/file" which stores the basis functions
symb : str
Atomic symbol, Special characters will be removed.
Examples:
Load STO 3G basis of carbon to oxygen atom
>>> mol = gto.Mole()
>>> mol.basis = {'O': load('sto-3g', 'C')}
'''
symb = ''.join([i for i in symb if i.isalpha()])
if os.path.isfile(filename_or_basisname):
# read basis from given file
try:
return parse_nwchem.load(filename_or_basisname, symb, optimize)
except BasisNotFoundError:
with open(filename_or_basisname, 'r') as fin:
return parse_nwchem.parse(fin.read(), symb)
name = _format_basis_name(filename_or_basisname)
if '@' in name:
split_name = name.split('@')
assert len(split_name) == 2
name = split_name[0]
contr_scheme = _convert_contraction(split_name[1])
else:
contr_scheme = 'Full'
if not (name in ALIAS or _is_pople_basis(name)):
try:
return parse_nwchem.parse(filename_or_basisname, symb)
except BasisNotFoundError:
try:
return parse_nwchem.parse(filename_or_basisname)
except IndexError:
raise BasisNotFoundError('Invalid basis name %s' %
filename_or_basisname)
except IndexError:
raise BasisNotFoundError(filename_or_basisname)
if name in ALIAS:
basmod = ALIAS[name]
elif _is_pople_basis(name):
basmod = _parse_pople_basis(name, symb)
else:
raise BasisNotFoundError(filename_or_basisname)
if 'dat' in basmod:
b = parse_nwchem.load(join(_BASIS_DIR, basmod), symb, optimize)
elif isinstance(basmod, (tuple, list)) and isinstance(basmod[0], str):
b = []
for f in basmod:
b += parse_nwchem.load(join(_BASIS_DIR, f), symb, optimize)
else:
if sys.version_info < (2, 7):
fp, pathname, description = imp.find_module(basmod, __path__)
mod = imp.load_module(name, fp, pathname, description)
b = mod.__getattribute__(symb)
fp.close()
else:
mod = importlib.import_module('.' + basmod, __package__)
b = mod.__getattribute__(symb)
if contr_scheme != 'Full':
b = _truncate(b, contr_scheme, symb, split_name)
return b