def parser(name, data):
""" Parse Empire specific xyz file """
tmol = Molecule(name)
nat = int(data[0])
tmol.comment = data[1].strip()
tmol.newAtoms(nat)
for j in range(nat):
line = data[j + 2].split()
tmol.setAtom(j, line[0], line[1:4])
tmol.setFmt("angstrom", scale=True)
vec = [0, 0, 0]
for i in range(3):
vec[i] = [float(x) for x in data[nat + 3 + i].split()]
tmol.setVec(vec)
tmol.setCellDim(1, fmt="angstrom")
return tmol, None
def parser(name, data):
""" Parse Gaussian Cube file """
tmol = Molecule(name)
fmt = "bohr"
# two lines of comments, combine
tmol.comment = data[0] + ";" + data[1]
# nat, origin[3]
nat = int(data[2].split()[0])
origin = [float(data[2].split()[i]) for i in [1, 2, 3]]
# n of datapoints(dir), cell_vec[3]
nvol = [0, 0, 0]
tvec = [0, 0, 0]
for i in [0, 1, 2]:
line = data[i + 3].split()
nvol[i] = int(line[0])
if nvol[i] < 0:
nvol[i] = -nvol[i]
fmt = "angstrom"
tvec[i] = [float(line[j]) * nvol[i] for j in [1, 2, 3]]
tmol.setVec(tvec)
pse = list(glob_pse.keys())
tmol.newAtoms(nat)
for i in range(nat):
# line = Z, charge, coord(x, y, z)
line = data[i + 6].split()
tmol.setAtom(i, pse[int(line[0])], line[2:5], charge=float(line[1]))
tmol.setFmt(fmt, scale=False)
tmol.setCellDim(1, fmt=fmt)
# rest of file has datagrid, x is outer loop, z inner
vol = [[[0] * nvol[2] for i in range(nvol[1])] for j in range(nvol[0])]
i = 6 + nat
line = data[i].split()
for x in range(nvol[0]):
for y in range(nvol[1]):
for z in range(nvol[2]):
while not line and i < (len(data) - 1):
i += 1
line = data[i].split()
vol[x][y][z] = float(line.pop(0))
tmol.setVol(vol, origin)
return tmol, None
def parser(name, data):
""" Parse xyz files (angstrom) """
# create list of mol, trajectory support
tmol = Molecule(name, steps=0)
i = 0
while i < len(data):
# handle empty lines at eof or between molecules
if not data[i].strip().isdigit():
i += 1
continue
# fixed format nat and comment
tmol.newStep()
nat = int(data[i])
tmol.comment = data[i + 1].strip()
# read coordinates and types
tmol.newAtoms(nat)
for j in range(nat):
line = data[j + i + 2].split()
tmol.setAtom(j, line[0], line[1:4])
tmol.setFmt('angstrom', scale=True)
i += nat + 2
return tmol, None
def parser(name, data):
""" Parse Lammps custom dump files
Preliminary implementation!
Needs to be 'custom' format with either:
'id element xs ys yz'
or
'id element x y z'
Only orthogonal cells for now
Assumes angstrom
"""
tmol = Molecule(name, steps=0)
i = 0
while i < len(data):
line = data[i]
if 'ITEM' in line:
if 'TIMESTEP' in line:
i += 2
tmol.newStep()
elif 'NUMBER OF ATOMS' in line:
nat = int(data[i + 1])
i += 2
elif 'BOX BOUNDS' in line:
tvec = [[0, 0, 0], [0, 0, 0], [0, 0, 0]]
tvec[0][0] = float(data[i + 1].split()[1]) -\
float(data[i + 1].split()[0])
tvec[1][1] = float(data[i + 2].split()[1]) -\
float(data[i + 2].split()[0])
tvec[2][2] = float(data[i + 3].split()[1]) -\
float(data[i + 3].split()[0])
tmol.setVec(tvec)
tmol.setCellDim(1, fmt='angstrom')
i += 4
elif 'ATOMS' in line:
line = line.split()
if 'id' not in line or 'element' not in line or\
('xs' not in line and 'x' not in line):
raise NotImplementedError("Lammps dump in not (yet) "
"recognized format")
# ididx = line.index('id') - 2
elidx = line.index('element') - 2
if 'xs' in line:
xidx = line.index('xs') - 2
yidx = line.index('ys') - 2
zidx = line.index('zs') - 2
fmt = 'crystal'
else:
xidx = line.index('x') - 2
yidx = line.index('y') - 2
zidx = line.index('z') - 2
fmt = 'angstrom'
if 'q' in line:
qidx = line.index('q') - 2
else:
qidx = False
tmol.newAtoms(nat)
for j in range(nat):
at = data[j + 1 + i].split()
if qidx:
tmol.setAtom(j, at[elidx],
[at[xidx], at[yidx], at[zidx]],
charge=float(at[qidx]))
else:
tmol.setAtom(j, at[elidx],
[at[xidx], at[yidx], at[zidx]])
tmol.setFmt(fmt, scale=True)
i += nat + 1
else:
i += 1
return tmol, None
def parser(name, data):
""" Parse Lammps data file
Element parsed via comment in 'Masses' part
Assumes angstrom, atom_style in [angle,atomic,bond,charge,full,molecular]
"""
tmol = Molecule(name)
i = 0
tvec = [[0, 0, 0], [0, 0, 0], [0, 0, 0]]
while i < len(data):
line = data[i].strip()
if 'atoms' in line:
nat = int(line.split()[0])
elif 'atom types' in line:
types = [0] * int(line.split()[0])
elif 'xlo xhi' in line:
tvec[0][0] = float(line.split()[1]) - float(line.split()[0])
elif 'ylo yhi' in line:
tvec[1][1] = float(line.split()[1]) - float(line.split()[0])
elif 'zlo zhi' in line:
tvec[2][2] = float(line.split()[1]) - float(line.split()[0])
elif 'xy xz yz' in line:
tvec[1][0], tvec[2][0], tvec[2][1] = line.split()[:3]
elif 'Masses' in line:
for j in range(i + 2, i + 2 + len(types)):
if '#' in data[j]:
el = data[j].split('#')
t = el[1].strip()
types[int(el[0].split()[0]) - 1] = t
tmol.pse[t]['m'] = float(el[0].split()[1])
else:
raise NotImplementedError("Can't deduce elements")
i += len(types) + 1
elif 'Atoms' in line:
atomlist = []
for j in range(i + 2, i + 2 + nat):
atomlist.append(data[j].strip().split())
if '#' in line and line.split('#')[1].strip() in lammps_atom_style:
style = lammps_atom_style[line.split('#')[1].
strip()].split('} ')
atypepos = style.index('{2:d')
cpos = style.index('{3: .4f')
try:
qpos = style.index('{6: .4f')
except:
qpos = None
else:
# make a guess for column-content:
nargs = len(atomlist[0])
atypepos = 1
cpos = -3
testimageflags = 0
for j in range(nargs, 1, -1):
try:
test = [int(k[j]) for k in atomlist]
except:
pass
else:
# check for imageflags (3 integer columns)
if nargs > 7 and j > 3:
testimageflags += 1
if testimageflags == 3:
cpos = -6
# check if one of the first two entries
# and not an empty charge column
elif j == 2 and set(test) != set([0]):
atypepos = j
break
# guess where charge is found:
if nargs + cpos - 1 > atypepos:
qpos = atypepos + 1
else:
qpos = False
# gotta load 'em all!
tmol.newAtoms(nat)
for j, at in enumerate(atomlist):
tmol.setAtom(j, types[int(at[atypepos]) - 1],
at[cpos:cpos + 3 if cpos + 3 else None],
float(at[qpos]) if qpos else None)
tmol.setFmt('angstrom', scale=True)
tmol.setVec(tvec)
tmol.setCellDim(1, fmt='angstrom')
i += 1
return tmol, None
def parser(name, data):
""" Parse PWScf output to trajectory """
tmol = Molecule(name, steps=0)
i = 0
vec = [[0, 0, 0], [0, 0, 0], [0, 0, 0]]
gamma = False
while i < len(data):
line = data[i].split()
# ignore empty lines
if not line:
pass
# read number of atoms
elif line[0:3] == ['number', 'of', 'atoms/cell']:
nat = int(line[4])
# read cell dimension
elif line[0] == 'celldm(1)=':
celldm = float(line[1])
# read initial cell vectors
elif line[0:2] == ['crystal', 'axes:']:
for j in [0, 1, 2]:
temp = data[i + 1 + j].split()
vec[j] = [float(x) for x in temp[3:6]]
# read initial positions:
elif line[0] == 'site':
tmol.newStep()
tmol.setCellDim(celldm)
tmol.setVec(vec)
tmol.newAtoms(nat)
for j in range(nat):
atom = data[j + i + 1].split()
tmol.setAtom(j, atom[1], atom[6:9])
tmol.setFmt('alat', scale=True)
i += nat
# read k-points:
elif line[0] == 'gamma-point':
gamma = True
elif line[0:3] == ['number', 'of', 'k'] and not gamma:
nk = int(line[4])
if nk < 100:
kpoints = []
for j in range(i + 2, i + nk + 2):
kp = data[j].split()
kpoints.append([kp[4], kp[5], kp[6].strip('), '), kp[9]])
tmol.setKpoints('discrete', kpoints)
tmol.setKpoints('active', 'discrete')
i += nk
# read step-vectors if cell is variable
elif line[0] == 'CELL_PARAMETERS':
for j in [0, 1, 2]:
temp = data[i + 1 + j].split()
vec[j] = [float(x) for x in temp[0:3]]
# read step-coordinates
elif line[0] == 'ATOMIC_POSITIONS':
tmol.newStep()
tmol.setCellDim(celldm)
tmol.setVec(vec)
tmol.newAtoms(nat)
for j in range(nat):
atom = data[j + i + 1].split()
tmol.setAtom(j, atom[0], atom[1:4],
fix=[not int(x) for x in atom[4:]])
tmol.setFmt(line[1].strip('()'), scale=True)
i += nat
# break on reaching final coordinates (duplicate)
elif line[0] == 'Begin':
break
# ignore everything else
else:
pass
i += 1
return tmol, None
def parser(name, data):
""" Parse (animated) XCrysDen structure files
Optionally contains datagrids/animations/forces
Parses multiple datagrids into multiple steps
"""
animated = False
periodic = False
pbcformats = ["MOLECULE", "POLYMER", "SLAB", "CRYSTAL"]
i = 0
while i < len(data):
if data[i][0] == '#':
pass
elif "ANIMSTEPS" in data[i]:
tmol = Molecule(name, steps=int(data[i].strip("ANIMSTEPS \r\n")))
animated = True
elif data[i].strip() in pbcformats:
periodic = True
# Parse coordinates of isolated molecule
elif not periodic and "ATOMS" in data[i]:
if animated:
tmol.changeStep(int(data[i].strip("ATOMS \r\n")) - 1)
else:
tmol = Molecule(name, steps=1)
j = 1
while i + j < len(data) and len(data[i + j].split()) in [4, 7]:
line = data[i + j].split()
tmol.newAtom(line[0], line[1:4])
j += 1
tmol.setFmt("angstrom", scale=True)
i += j - 1
# Parse periodic structure containing cell and coordinates
elif periodic and "PRIMVEC" in data[i]:
vec = [data[i + j].split() for j in range(1, 4)]
if animated and data[i].strip("PRIMVEC \r\n"):
tmol.changeStep(int(data[i].strip("PRIMVEC ")) - 1)
tmol.setVec(vec)
tmol.setCellDim(1, fmt='angstrom')
elif animated:
tmol.setVecAll(vec)
tmol.setCellDimAll(1, fmt='angstrom')
else:
tmol = Molecule(name, steps=1)
tmol.setVec(vec)
tmol.setCellDim(1, fmt='angstrom')
i += 3
elif periodic and "PRIMCOORD" in data[i]:
if animated:
tmol.changeStep(int(data[i].strip("PRIMCOORD \r\n")) - 1)
nat = int(data[i + 1].split()[0])
tmol.newAtoms(nat)
for j in range(nat):
line = data[j + 2 + i].split()
tmol.setAtom(j, line[0], line[1:4])
tmol.setFmt("angstrom", scale=True)
i += nat + 1
elif periodic and "CONVVEC" in data[i]:
# could be used, but not atm
pass
elif periodic and "CONVCOORD" in data[i]: # pragma: no cover
# should not even be there
pass
elif "BEGIN_" in data[i]:
datatype = data[i].strip("BEGIN_BLOCK_ \r\n")
endblock = "END_BLOCK_" + datatype
endgrid = "END_" + datatype
j = 1
grids = []
# only parse 3D datagrids
while endblock not in data[i + j]:
if datatype in data[i + j] and datatype == "DATAGRID_3D":
nvol = list(map(int, data[i + j + 1].split()))
vol = [[[0] * nvol[2] for y in range(nvol[1])]
for z in range(nvol[0])]
origin = list(map(float, data[i + j + 2].split()))
# assume spanning space equals cell
j += 6
line = data[i + j].split()
for z in range(nvol[2]):
for y in range(nvol[1]):
for x in range(nvol[0]):
while not (line or endgrid in data[i + j + 1]):
j += 1
line = data[i + j].split()
vol[x][y][z] = float(line.pop(0))
del vol[-1]
for n in vol:
del n[-1]
for m in n:
del m[-1]
grids.append((vol, origin))
j += 1
j += 1
if grids:
tmol.changeStep(0)
tmol.setVol(*grids[0])
if len(grids) > 1:
for k in range(1, len(grids)):
tmol.copyStep()
tmol.setVol(*grids[k])
i += j
i += 1
return tmol, None
def parser(name, data):
""" Parse PWScf input files
Namelists will be parsed and saved in PW parameter set
Supported Cards:
- ATOMIC_SPECIES
- ATOMIC_POSITIONS
- K_POINTS
- CELL_PARAMETERS
Not supported:
- CONSTRAINTS
- OCCUPATIONS
- ATOMIC_FORCES (PWSCFv5)
"""
tmol = Molecule(name)
tparam = {"type": "pwi", "name": name}
tvec = []
# parse data and create tparam
while data:
header = data.pop(0).strip().split()
# ignore empty lines
if not header:
pass
# parse namelists
elif header[0][0] == '&':
tnl = OrderedDict()
# parse entries
line = regex(', \s*(?![^()]*\))', data.pop(0).strip())
while line[0] != '/':
for j in range(len(line)):
if line[j]:
tnl[line[j].split('=')[0].strip()] =\
line[j].split('=')[1].strip()
line = regex(', \s*(?![^()]*\))', data.pop(0).strip())
tparam[header[0].lower()] = tnl
# parse card
elif header[0][0].isupper():
# ATOMIC_SPECIES:
# Name Weight PP-file
if header[0] == 'ATOMIC_SPECIES':
for i in range(int(tparam['&system']['ntyp'])):
line = data.pop(0).strip().split()
tmol.pse[line[0]]['m'] = float(line[1])
tmol.pse[line[0]]['PWPP'] = line[2]
# ATOMIC_POSITIONS fmt
# Name x y z
elif header[0] == 'ATOMIC_POSITIONS':
fmt = header[1].strip('{()}')
tmol.newAtoms(int(tparam['&system']['nat']))
for i in range(int(tparam['&system']['nat'])):
# support empty lines
temp = data.pop(0).split()
while not temp:
temp = data.pop(0).split()
tmol.setAtom(i, temp[0], temp[1:4],
fix=[not int(x) for x in temp[4:]])
# K_POINTS fmt
elif header[0] == 'K_POINTS':
active = header[1].strip('{()}')
# Gamma point only
if active == 'gamma':
tmol.setKpoints('active', 'gamma')
# Monkhorst Pack Grid:
# x y z offset
elif active == 'automatic':
line = data.pop(0).split()
tmol.setKpoints('mpg', line)
tmol.setKpoints('active', 'mpg')
# else:
# number of kpoints
# x y z weight
else:
nk = int(data.pop(0).split()[0])
kpoints = []
for i in range(nk):
kpoints.append(data.pop(0).split())
tmol.setKpoints('discrete', kpoints)
tmol.setKpoints('active', 'discrete')
if active == 'tpiba':
pass
elif active == 'tpiba_b':
tmol.setKpoints('options', {'bands': True,
'crystal': False})
elif active == 'crystal':
tmol.setKpoints('options', {'bands': False,
'crystal': True})
elif active == 'crystal_b':
tmol.setKpoints('options', {'bands': True,
'crystal': True})
# CELL_PARAMETERS
# only needed if ibrav=0
# tbd changed between pw4 and pw5, ignored for now
elif header[0] == 'CELL_PARAMETERS':
for i in [0, 1, 2]:
line = data.pop(0).strip().split()
tvec.append([float(x) for x in line])
else:
pass
# Identify cell parameter representation, parse it
ibrav = tparam['&system']['ibrav']
tparam['&system']['ibrav'] = '0'
# Get cell values
sys = tparam['&system']
a = b = c = cosab = cosac = cosbc = None
if 'celldm(1)' in sys:
a = float(sys['celldm(1)'])
del sys['celldm(1)']
tmol.setCellDim(a, fmt='bohr')
if 'celldm(2)' in sys:
b = float(sys['celldm(2)'])
del sys['celldm(2)']
if 'celldm(3)' in sys:
c = float(sys['celldm(3)'])
del sys['celldm(3)']
if 'celldm(4)' in sys:
cosab = float(sys['celldm(4)'])
del sys['celldm(4)']
if 'celldm(5)' in sys:
cosac = float(sys['celldm(5)'])
del sys['celldm(5)']
if 'celldm(6)' in sys:
cosbc = float(sys['celldm(6)'])
del sys['celldm(6)']
if ibrav == '14':
cosab, cosbc = cosbc, cosab
elif 'A' in sys:
a = float(sys['A'])
del sys['A']
tmol.setCellDim(a, fmt='angstrom')
if 'B' in sys:
b = float(sys['B']) / a
del sys['B']
if 'C' in sys:
c = float(sys['C']) / a
del sys['C']
if 'cosAB' in sys:
cosab = float(sys['cosAB'])
del sys['cosAB']
if 'cosAC' in sys:
cosac = float(sys['cosAC'])
del sys['cosAC']
if 'cosBC' in sys:
cosbc = float(sys['cosBC'])
del sys['cosBC']
elif ibrav != '0':
raise ValueError('Neither celldm(1) nor A specified')
def checkCellVal(v, n):
if v is None:
raise ValueError(n + ' is needed, but was not specified')
if ibrav == '0':
# check if CELL_PARAMETERS card has been read
# if not present, throw error
if not tvec:
raise ValueError('ibrav=0, but CELL_PARAMETERS missing')
else:
tmol.setVec(tvec)
elif ibrav == '1':
# simple cubic
pass
elif ibrav == '2':
# face centered cubic
tmol.setVec([[-0.5, 0, 0.5], [0, 0.5, 0.5], [-0.5, 0.5, 0]])
elif ibrav == '3':
# body centered cubic
tmol.setVec([[0.5, 0.5, 0.5], [-0.5, 0.5, 0.5], [-0.5, -0.5, 0.5]])
elif ibrav == '4':
# hexagonal
checkCellVal(c, 'c')
tmol.setVec([[1, 0, 0], [-0.5, sqrt(3) * 0.5, 0], [0, 0, c]])
elif ibrav == '5':
# trigonal
checkCellVal(cosab, 'cosab')
tx = sqrt((1 - cosab) / 2)
ty = sqrt((1 - cosab) / 6)
tz = sqrt((1 + 2 * cosab) / 3)
tmol.setVec([[tx, -ty, tz], [0, 2 * ty, tz], [-tx, -ty, tz]])
elif ibrav == '-5':
# trigonal, alternative
checkCellVal(cosab, 'cosab')
tx = sqrt((1 - cosab) / 2)
ty = sqrt((1 - cosab) / 6)
tz = sqrt((1 + 2 * cosab) / 3)
u = (tz - 2 * sqrt(2) * ty) / sqrt(3)
v = (tz + sqrt(2) * ty) / sqrt(3)
tmol.setVec([[u, v, v], [v, u, v], [v, v, u]])
elif ibrav == '6':
# simple tetragonal
checkCellVal(c, 'c')
tmol.setVec([[1, 0, 0], [0, 1, 0], [0, 0, c]])
elif ibrav == '7':
# body centered tetragonal
checkCellVal(c, 'c')
tmol.setVec([[0.5, -0.5, c * 0.5],
[0.5, 0.5, c * 0.5],
[-0.5, -0.5, c * 0.5]])
elif ibrav == '8':
# simple orthorhombic
checkCellVal(c, 'c')
checkCellVal(b, 'b')
tmol.setVec([[1, 0, 0], [0, b, 0], [0, 0, c]])
elif ibrav == '9':
# basis centered orthorhombic
checkCellVal(c, 'c')
checkCellVal(b, 'b')
tmol.setVec([[0.5, b * 0.5, 0], [-0.5, b * 0.5, 0], [0, 0, c]])
elif ibrav == '10':
# face centered orthorhombic
checkCellVal(c, 'c')
checkCellVal(b, 'b')
tmol.setVec([[0.5, 0, c * 0.5],
[0.5, b * 0.5, 0],
[0, b * 0.5, c * 0.5]])
elif ibrav == '11':
# body centered orthorhombic
checkCellVal(c, 'c')
checkCellVal(b, 'b')
tmol.setVec([[0.5, b * 0.5, c * 0.5],
[-0.5, b * 0.5, c * 0.5],
[-0.5, -b * 0.5, c * 0.5]])
elif ibrav == '12':
# simple monoclinic
checkCellVal(c, 'c')
checkCellVal(b, 'b')
checkCellVal(cosab, 'cosab')
tmol.setVec([[1, 0, 0],
[b * cosab, b * sqrt(1 - cosab**2), 0],
[0, 0, c]])
elif ibrav == '-12':
# simple monoclinic, alternate definition
checkCellVal(c, 'c')
checkCellVal(b, 'b')
checkCellVal(cosac, 'cosac')
tmol.setVec([[1, 0, 0], [0, b, 0],
[c * cosac, 0, c * sqrt(1 - cosac**2)]])
elif ibrav == '13':
# base centered monoclinic
checkCellVal(c, 'c')
checkCellVal(b, 'b')
checkCellVal(cosab, 'cosab')
tmol.setVec([[0.5, 0, -c * 0.5],
[b * cosab, b * sqrt(1 - cosab**2), 0],
[0.5, 0, c * 0.5]])
elif ibrav == '14':
# triclinic
checkCellVal(c, 'c')
checkCellVal(b, 'b')
checkCellVal(cosab, 'cosab')
checkCellVal(cosac, 'cosac')
checkCellVal(cosbc, 'cosbc')
singam = sqrt(1 - cosab**2)
tmol.setVec([[1, 0, 0], [b * cosab, b * singam, 0],
[c * cosac, c * (cosbc - cosac * cosab) / singam,
c * sqrt(1 + 2 * cosbc * cosac * cosab - cosbc * cosbc -
cosac * cosac - cosab * cosab) / singam]])
# scale atoms after creating cell:
tmol.setFmt(fmt, scale=True)
# delete nat and ntype before returning
del tparam['&system']['nat']
del tparam['&system']['ntyp']
return tmol, tparam
