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 KeyError('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 convert_basis_to_nwchem(symb, basis):
'''Convert the internal basis format to NWChem format string'''
res = []
symb = _std_symbol(symb)
# pass 1: comment line
ls = [b[0] for b in basis]
nprims = [len(b[1:]) for b in basis]
nctrs = [len(b[1]) - 1 for b in basis]
prim_to_ctr = {}
for i, l in enumerate(ls):
if l in prim_to_ctr:
prim_to_ctr[l][0] += nprims[i]
prim_to_ctr[l][1] += nctrs[i]
else:
prim_to_ctr[l] = [nprims[i], nctrs[i]]
nprims = []
nctrs = []
for l in set(ls):
nprims.append(str(prim_to_ctr[l][0]) + SPDF[l].lower())
nctrs.append(str(prim_to_ctr[l][1]) + SPDF[l].lower())
res.append('#BASIS SET: (%s) -> [%s]' %
(','.join(nprims), ','.join(nctrs)))
# pass 2: basis data
for bas in basis:
res.append('%-2s %s' % (symb, SPDF[bas[0]]))
for dat in bas[1:]:
res.append(' '.join('%15.9f' % x for x in dat))
return '\n'.join(res)
def search_seg(basisfile, symb):
symb = _std_symbol(symb)
with open(basisfile, 'r') as fin:
fdata = re.split(BASIS_SET_DELIMITER, fin.read())
dat = _search_seg(fdata, symb)
if dat is None:
return []
else:
return [x.upper() for x in dat.splitlines() if x and 'END' not in x]
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 convert_ecp_to_nwchem(symb, ecp):
'''Convert the internal ecp format to NWChem format string'''
symb = _std_symbol(symb)
res = ['%-2s nelec %d' % (symb, ecp[0])]
for ecp_block in ecp[1]:
l = ecp_block[0]
if l == -1:
res.append('%-2s ul' % symb)
else:
res.append('%-2s %s' % (symb, SPDF[l].lower()))
for r_order, dat in enumerate(ecp_block[1]):
for e, c in dat:
res.append('%d %15.9f %15.9f' % (r_order, e, c))
return '\n'.join(res)
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
... """)}
'''
if symb is not None:
symb = _std_symbol(symb)
string = _search_seg(re.split(BASIS_SET_DELIMITER, string), symb)
if string is None:
raise KeyError('Basis not found for %s' % symb)
bastxt = []
for dat in string.splitlines():
x = dat.split('#')[0].strip().upper() # Use # to start comments
if (x and not x.startswith('END') and not x.startswith('BASIS')):
bastxt.append(x)
return _parse(bastxt, optimize)
def search_ecp(basisfile, symb):
symb = _std_symbol(symb)
with open(basisfile, 'r') as fin:
fdata = re.split(ECP_DELIMITER, fin.read())
if len(fdata) <= 1:
return []
fdata = fdata[1].splitlines()
for i, dat in enumerate(fdata):
dat0 = dat.split(None, 1)
if dat0 and dat0[0] == symb:
break
seg = []
for dat in fdata[i:]:
dat = dat.strip().upper()
if dat: # remove empty lines
if ((dat[0].isalpha() and dat.split(None, 1)[0] != symb.upper())):
return seg
else:
seg.append(dat)
return []
def search_seg(basisfile, symb):
symb = _std_symbol(symb)
with open(basisfile, 'r') as fin:
fdata = re.split(BASIS_SET_DELIMITER, fin.read())
raw_basis = _search_basis_block(fdata, symb)
return [x.upper() for x in raw_basis.splitlines() if x and 'END' not in x]