def parse_ecp(string, symb=None):
from pyscf.gto.mole import _std_symbol
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[0].isalpha() and dat.split(None, 1)[0] != symb.upper())):
break
elif dat: # remove blank lines
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 format_basis(basis_tab):
'''Convert the input :attr:`Cell.basis` to the internal data format::
{ atom: (l, kappa, ((-exp, c_1, c_2, ..), nprim, nctr, ptr-exps, ptr-contraction-coeff)), ... }
Args:
basis_tab : dict
Similar to :attr:`Cell.basis`, it **cannot** be a str
Returns:
Formated :attr:`~Cell.basis`
Examples:
>>> pbc.format_basis({'H':'gth-szv'})
{'H': [[0,
(8.3744350009, -0.0283380461),
(1.8058681460, -0.1333810052),
(0.4852528328, -0.3995676063),
(0.1658236932, -0.5531027541)]]}
'''
fmt_basis = {}
for atom in basis_tab.keys():
atom_basis = basis_tab[atom]
if isinstance(atom_basis, str) and 'gth' in atom_basis:
fmt_basis[atom] = basis.load(atom_basis, _std_symbol(atom))
else:
fmt_basis[atom] = atom_basis
return mole.format_basis(fmt_basis)
def format_basis(basis_tab):
'''Convert the input :attr:`Cell.basis` to the internal data format.
``{ atom: (l, kappa, ((-exp, c_1, c_2, ..), nprim, nctr, ptr-exps, ptr-contraction-coeff)), ... }``
Args:
basis_tab : dict
Similar to :attr:`Cell.basis`, it **cannot** be a str
Returns:
Formated :attr:`~Cell.basis`
Examples:
>>> pbc.format_basis({'H':'gth-szv'})
{'H': [[0,
(8.3744350009, -0.0283380461),
(1.8058681460, -0.1333810052),
(0.4852528328, -0.3995676063),
(0.1658236932, -0.5531027541)]]}
'''
fmt_basis = {}
for atom in basis_tab.keys():
atom_basis = basis_tab[atom]
if isinstance(atom_basis, str) and 'gth' in atom_basis:
fmt_basis[atom] = basis.load(atom_basis, _std_symbol(atom))
else:
fmt_basis[atom] = atom_basis
return mole.format_basis(fmt_basis)
def convert_basis_to_nwchem(symb, basis):
'''Convert the internal basis format to NWChem format string'''
from pyscf.gto.mole import _std_symbol
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 format_basis(basis_tab):
"""Convert the input :attr:`Cell.basis` to the internal data format.
``{ atom: (l, kappa, ((-exp, c_1, c_2, ..), nprim, nctr, ptr-exps, ptr-contraction-coeff)), ... }``
Args:
basis_tab : list
Similar to :attr:`Cell.basis`, it **cannot** be a str
Returns:
Formated :attr:`~Cell.basis`
Examples:
>>> pbc.format_basis({'H':'gth-szv'})
{'H': [[0,
(8.3744350009, -0.0283380461),
(1.8058681460, -0.1333810052),
(0.4852528328, -0.3995676063),
(0.1658236932, -0.5531027541)]]}
"""
fmt_basis = {}
for atom in basis_tab.keys():
symb = _symbol(atom)
rawsymb = _rm_digit(symb)
stdsymb = _std_symbol(rawsymb)
symb = symb.replace(rawsymb, stdsymb)
if isinstance(basis_tab[atom], str):
fmt_basis[symb] = basis.load(basis_tab[atom], stdsymb)
else:
fmt_basis[symb] = basis_tab[atom]
return fmt_basis
def convert_basis_to_nwchem(symb, basis):
'''Convert the internal basis format to NWChem format string'''
from pyscf.gto.mole import _std_symbol
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):
from pyscf.gto.mole import _std_symbol
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 format_basis(basis_tab):
'''Convert the input :attr:`Cell.basis` to the internal data format::
{ atom: (l, kappa, ((-exp, c_1, c_2, ..), nprim, nctr, ptr-exps, ptr-contraction-coeff)), ... }
Args:
basis_tab : dict
Similar to :attr:`Cell.basis`, it **cannot** be a str
Returns:
Formated :attr:`~Cell.basis`
Examples:
>>> pbc.format_basis({'H':'gth-szv'})
{'H': [[0,
(8.3744350009, -0.0283380461),
(1.8058681460, -0.1333810052),
(0.4852528328, -0.3995676063),
(0.1658236932, -0.5531027541)]]}
'''
def convert(basis_name, symb):
if basis_name.lower().startswith('unc'):
return uncontract(basis.load(basis_name[3:], symb))
else:
return basis.load(basis_name, symb)
fmt_basis = {}
for atom in basis_tab.keys():
symb = _atom_symbol(atom)
stdsymb = _std_symbol(symb)
if stdsymb.startswith('GHOST-'):
stdsymb = stdsymb[6:]
atom_basis = basis_tab[atom]
if isinstance(atom_basis, (str, unicode)):
if 'gth' in atom_basis:
bset = convert(str(atom_basis), symb)
else:
bset = atom_basis
else:
bset = []
for rawb in atom_basis:
if isinstance(rawb, (str, unicode)) and 'gth' in rawb:
bset.append(convert(str(rawb), stdsymb))
else:
bset.append(rawb)
fmt_basis[symb] = bset
return mole.format_basis(fmt_basis)
def convert_ecp_to_nwchem(symb, ecp):
'''Convert the internal ecp format to NWChem format string'''
from pyscf.gto.mole import _std_symbol
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 convert_ecp_to_nwchem(symb, ecp):
'''Convert the internal ecp format to NWChem format string'''
from pyscf.gto.mole import _std_symbol
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 search_ecp(basisfile, symb):
from pyscf.gto.mole import _std_symbol
symb = _std_symbol(symb)
with open(basisfile, 'r') as fin:
fdata = re.split(ECP_DELIMITER, fin.read())
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[0].isalpha() and dat.split(None, 1)[0] != symb.upper())):
return seg
elif dat: # remove blank lines
seg.append(dat)
return []
def parse(string, symb=None):
'''Parse the basis text which is in NWChem format, return an internal
basis format which can be assigned to :attr:`Mole.basis`
Lines started with # are ignored.
'''
from pyscf.gto.mole import _std_symbol
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)
def format_pseudo(pseudo_tab):
r'''Convert the input :attr:`Cell.pseudo` (dict) to the internal data format::
{ atom: ( (nelec_s, nele_p, nelec_d, ...),
rloc, nexp, (cexp_1, cexp_2, ..., cexp_nexp),
nproj_types,
(r1, nproj1, ( (hproj1[1,1], hproj1[1,2], ..., hproj1[1,nproj1]),
(hproj1[2,1], hproj1[2,2], ..., hproj1[2,nproj1]),
...
(hproj1[nproj1,1], hproj1[nproj1,2], ... ) )),
(r2, nproj2, ( (hproj2[1,1], hproj2[1,2], ..., hproj2[1,nproj1]),
... ) )
)
... }
Args:
pseudo_tab : dict
Similar to :attr:`Cell.pseudo` (a dict), it **cannot** be a str
Returns:
Formatted :attr:`~Cell.pseudo`
Examples:
>>> pbc.format_pseudo({'H':'gth-blyp', 'He': 'gth-pade'})
{'H': [[1],
0.2, 2, [-4.19596147, 0.73049821], 0],
'He': [[2],
0.2, 2, [-9.1120234, 1.69836797], 0]}
'''
fmt_pseudo = {}
for atom in pseudo_tab:
symb = _symbol(atom)
rawsymb = _rm_digit(symb)
stdsymb = _std_symbol(rawsymb)
symb = symb.replace(rawsymb, stdsymb)
if isinstance(pseudo_tab[atom], str):
fmt_pseudo[symb] = pseudo.load(pseudo_tab[atom], stdsymb)
else:
fmt_pseudo[symb] = pseudo_tab[atom]
return fmt_pseudo
def format_pseudo(pseudo_tab):
'''Convert the input :attr:`Cell.pseudo` (dict) to the internal data format.
``{ atom: ( (nelec_s, nele_p, nelec_d, ...),
rloc, nexp, (cexp_1, cexp_2, ..., cexp_nexp),
nproj_types,
(r1, nproj1, ( (hproj1[1,1], hproj1[1,2], ..., hproj1[1,nproj1]),
(hproj1[2,1], hproj1[2,2], ..., hproj1[2,nproj1]),
...
(hproj1[nproj1,1], hproj1[nproj1,2], ... ) )),
(r2, nproj2, ( (hproj2[1,1], hproj2[1,2], ..., hproj2[1,nproj1]),
... ) )
)
... }``
Args:
pseudo_tab : dict
Similar to :attr:`Cell.pseudo` (a dict), it **cannot** be a str
Returns:
Formatted :attr:`~Cell.pseudo`
Examples:
>>> pbc.format_pseudo({'H':'gth-blyp', 'He': 'gth-pade'})
{'H': [[1],
0.2, 2, [-4.19596147, 0.73049821], 0],
'He': [[2],
0.2, 2, [-9.1120234, 1.69836797], 0]}
'''
fmt_pseudo = {}
for atom in pseudo_tab.keys():
symb = _symbol(atom)
rawsymb = _rm_digit(symb)
stdsymb = _std_symbol(rawsymb)
symb = symb.replace(rawsymb, stdsymb)
if isinstance(pseudo_tab[atom], str):
fmt_pseudo[symb] = pseudo.load(pseudo_tab[atom], stdsymb)
else:
fmt_pseudo[symb] = pseudo_tab[atom]
return fmt_pseudo
def convert_basis_to_psi4(symb, basis):
from pyscf.gto.mole import _std_symbol
'''Convert pyscf internal basis format to Gaussian format string
Psi4 uses Gaussian 94 format '''
res = []
symb = _std_symbol(symb)
SPDF = ('S', 'P', 'D', 'F', 'G', 'H', 'I', 'J')
MAXL = 8
MAPSPDF = {'S': 0, 'P': 1, 'D': 2, 'F': 3, 'G': 4, 'H': 5, 'I': 6, 'J': 7}
# element name
res.append('%-2s 0' % (symb))
# gaussian formatting
for bas in basis:
for i in range(len(bas[1]) - 1):
res.append('%s %s 1.00' % (SPDF[bas[0]], len(bas[1:])))
for dat in bas[1:]:
res.append('%15.9f %15.9f ' % (dat[0], dat[i + 1]))
#if len(bas[1]) > 2:
# for i in range(len(bas[1])-1):
# res.append('%s %s 1.00' % (SPDF[bas[0]], len(bas[1:])))
# for dat in bas[1:]:
# res.append('%15.9f %15.9f ' % (dat[0], dat[i+1]))
#elif len(bas[1]) == 2:
# res.append('%s %s 1.00' % (SPDF[bas[0]], len(bas[1:])))
# for dat in bas[1:]:
# res.append('%15.9f %15.9f ' % (dat[0], dat[1]))
#
#else:
# raise RuntimeError(
# 'Warning! Please manually check basis set format!')
# closing
res.append('****')
return '\n'.join(res)