def parseLines(self, lines):
"""Parse list of lines in RAWXYZ format.
Return Structure object or raise StructureFormatError.
"""
linefields = [l.split() for l in lines]
# prepare output structure
stru = Structure()
# find first valid record
start = 0
for field in linefields:
if len(field) == 0 or field[0] == "#":
start += 1
else:
break
# find the last valid record
stop = len(lines)
while stop > start and len(linefields[stop-1]) == 0:
stop -= 1
# get out for empty structure
if start >= stop:
return stru
# here we have at least one valid record line
# figure out xyz layout from the first line for plain and raw formats
floatfields = [ isfloat(f) for f in linefields[start] ]
nfields = len(linefields[start])
if nfields not in (3, 4):
emsg = ("%d: invalid RAWXYZ format, expected 3 or 4 columns" %
(start + 1))
raise StructureFormatError(emsg)
if floatfields[:3] == [True, True, True]:
el_idx, x_idx = (None, 0)
elif floatfields[:4] == [False, True, True, True]:
el_idx, x_idx = (0, 1)
else:
emsg = "%d: invalid RAWXYZ format" % (start + 1)
raise StructureFormatError(emsg)
# now try to read all record lines
try:
p_nl = start
for fields in linefields[start:] :
p_nl += 1
if fields == []:
continue
elif len(fields) != nfields:
emsg = ('%d: all lines must have ' +
'the same number of columns') % p_nl
raise StructureFormatError, emsg
element = el_idx is not None and fields[el_idx] or ""
xyz = [ float(f) for f in fields[x_idx:x_idx+3] ]
if len(xyz) == 2:
xyz.append(0.0)
stru.addNewAtom(element, xyz=xyz)
except ValueError:
emsg = "%d: invalid number" % p_nl
exc_type, exc_value, exc_traceback = sys.exc_info()
raise StructureFormatError, emsg, exc_traceback
return stru
def parseLines(self, lines):
"""Parse list of lines in PDB format.
Return Structure object or raise StructureFormatError.
"""
xcfg_Number_of_particles = None
xcfg_A = None
xcfg_H0 = numpy.zeros((3, 3), dtype=float)
xcfg_H0_set = numpy.zeros((3, 3), dtype=bool)
xcfg_NO_VELOCITY = False
xcfg_entry_count = None
p_nl = 0
p_auxiliary_re = re.compile(r"^auxiliary\[(\d+)\] =")
p_auxiliary = {}
stru = Structure()
# ignore trailing blank lines
stop = len(lines)
for line in reversed(lines):
if line.strip():
break
stop -= 1
# iterator over the valid data lines
ilines = iter(lines[:stop])
try:
# read XCFG header
for line in ilines:
p_nl += 1
stripped_line = line.strip()
# blank lines and lines starting with # are ignored
if stripped_line == "" or line[0] == '#':
continue
elif xcfg_Number_of_particles is None:
if line.find("Number of particles =") != 0:
emsg = ("%d: first line must " +
"contain 'Number of particles ='") % p_nl
raise StructureFormatError(emsg)
xcfg_Number_of_particles = int(line[21:].split(None, 1)[0])
p_natoms = xcfg_Number_of_particles
elif line.find("A =") == 0:
xcfg_A = float(line[3:].split(None, 1)[0])
elif line.find("H0(") == 0:
i, j = (int(line[3]) - 1, int(line[5]) - 1)
xcfg_H0[i, j] = float(line[10:].split(None, 1)[0])
xcfg_H0_set[i, j] = True
elif line.find(".NO_VELOCITY.") == 0:
xcfg_NO_VELOCITY = True
elif line.find("entry_count =") == 0:
xcfg_entry_count = int(line[13:].split(None, 1)[0])
elif p_auxiliary_re.match(line):
m = p_auxiliary_re.match(line)
idx = int(m.group(1))
p_auxiliary[idx] = line[m.end():].split(None, 1)[0]
else:
break
# check header for consistency
if numpy.any(xcfg_H0_set == False):
emsg = "H0 tensor is not properly defined"
raise StructureFormatError(emsg)
p_auxnum = len(p_auxiliary) and max(p_auxiliary.keys()) + 1
for i in range(p_auxnum):
if not i in p_auxiliary:
p_auxiliary[i] = "aux%d" % i
sorted_aux_keys = sorted(p_auxiliary.keys())
if p_auxnum != 0:
stru.xcfg = {
'auxiliaries': [p_auxiliary[k] for k in sorted_aux_keys]
}
ecnt = len(p_auxiliary) + (3 if xcfg_NO_VELOCITY else 6)
if ecnt != xcfg_entry_count:
emsg = ("%d: auxiliary fields are "
"not consistent with entry_count") % p_nl
raise StructureFormatError(emsg)
# define proper lattice
stru.lattice.setLatBase(xcfg_H0)
# here we are inside the data block
p_element = None
for line in ilines:
p_nl += 1
words = line.split()
# ignore atom mass
if len(words) == 1 and isfloat(words[0]):
continue
# parse element allowing empty symbol
elif len(words) <= 1:
w = line.strip()
p_element = w[:1].upper() + w[1:].lower()
elif len(words) == xcfg_entry_count and p_element is not None:
fields = [float(w) for w in words]
xyz = [xcfg_A * xi for xi in fields[:3]]
stru.addNewAtom(p_element, xyz=xyz)
a = stru[-1]
_assign_auxiliaries(a,
fields,
auxiliaries=p_auxiliary,
no_velocity=xcfg_NO_VELOCITY)
else:
emsg = "%d: invalid record" % p_nl
raise StructureFormatError(emsg)
if len(stru) != p_natoms:
emsg = "expected %d atoms, read %d" % (p_natoms, len(stru))
raise StructureFormatError(emsg)
except (ValueError, IndexError):
emsg = "%d: file is not in XCFG format" % p_nl
exc_type, exc_value, exc_traceback = sys.exc_info()
raise StructureFormatError, emsg, exc_traceback
return stru
def parseLines(self, lines):
"""Parse list of lines in PDB format.
Return Structure object or raise StructureFormatError.
"""
xcfg_Number_of_particles = None
xcfg_A = None
xcfg_H0 = numpy.zeros((3,3), dtype=float)
xcfg_H0_set = numpy.zeros((3,3), dtype=bool)
xcfg_NO_VELOCITY = False
xcfg_entry_count = None
xcfg_auxiliary = []
p_nl = 0
p_auxiliary_re = re.compile(r"^auxiliary\[(\d+)\] =")
p_auxiliary = {}
try:
stru = Structure()
# ignore trailing blank lines
stop = len(lines)
while stop>0 and lines[stop-1].strip() == "":
stop -= 1
ilines = iter(lines[:stop])
# read XCFG header
for line in ilines:
p_nl += 1
stripped_line = line.strip()
# blank lines and lines starting with # are ignored
if stripped_line == "" or line[0] == '#':
continue
elif xcfg_Number_of_particles is None:
if line.find("Number of particles =") != 0:
emsg = ("%d: first line must " +
"contain 'Number of particles ='") % p_nl
raise StructureFormatError(emsg)
xcfg_Number_of_particles = int(line[21:].split(None, 1)[0])
p_natoms = xcfg_Number_of_particles
elif line.find("A =") == 0:
xcfg_A = float(line[3:].split(None, 1)[0])
elif line.find("H0(") == 0:
i, j = ( int(line[3])-1 , int(line[5])-1 )
xcfg_H0[i,j] = float(line[10:].split(None, 1)[0])
xcfg_H0_set[i,j] = True
elif line.find(".NO_VELOCITY.") == 0:
xcfg_NO_VELOCITY = True
elif line.find("entry_count =") == 0:
xcfg_entry_count = int(line[13:].split(None, 1)[0])
elif p_auxiliary_re.match(line):
m = p_auxiliary_re.match(line)
idx = int(m.group(1))
p_auxiliary[idx] = line[m.end():].split(None, 1)[0]
else:
break
# check header for consistency
if numpy.any(xcfg_H0_set == False):
emsg = "H0 tensor is not properly defined"
raise StructureFormatError(emsg)
p_auxnum = len(p_auxiliary) and max(p_auxiliary.keys())+1
for i in range(p_auxnum):
if not i in p_auxiliary:
p_auxiliary[i] = "aux%d" % i
sorted_aux_keys = p_auxiliary.keys()
sorted_aux_keys.sort()
if p_auxnum != 0:
stru.xcfg = {
'auxiliaries' : [ p_auxiliary[k]
for k in sorted_aux_keys ]
}
if 6-3*xcfg_NO_VELOCITY+len(p_auxiliary) != xcfg_entry_count:
emsg = ("%d: auxiliary fields " +
"not consistent with entry_count") % p_nl
raise StructureFormatError(emsg)
# define proper lattice
stru.lattice.setLatBase(xcfg_H0)
# build p_assign_atom function to assign entries to proper fields
p_exprs = [ "a.xyz[0]=fields[0]",
"a.xyz[1]=fields[1]",
"a.xyz[2]=fields[2]" ]
if not xcfg_NO_VELOCITY:
p_exprs += [ "a.v=numpy.zeros(3, dtype=float)",
"a.v[0]=fields[3]",
"a.v[1]=fields[4]",
"a.v[2]=fields[5]" ]
for idx in sorted_aux_keys:
prop = p_auxiliary[idx]
col = idx + 6 - 3*xcfg_NO_VELOCITY
if prop == "Uiso":
p_exprs.append("a.U[0,0]=a.U[1,1]=a.U[2,2]=" +
"fields[%d]" % col)
elif re.match(r"^U\d\d$", prop) \
and 1<=int(prop[1])<=3 and 1<=int(prop[2])<=3 :
i, j = int(prop[1])-1, int(prop[2])-1
if i==j:
p_exprs.append("a.U[%i,%i]=fields[%d]" % (i, j, col) )
else:
p_exprs.append("a.U[%i,%i]=a.U[%i,%i]=fields[%d]" % \
(i, j, j, i, col) )
else:
p_exprs.append( "a.__dict__[%r]=fields[%d]" % \
(prop, col) )
p_assign_expr = "pass; " + "; ".join(p_exprs[3:])
exec "def p_assign_atom(a, fields) : %s" % p_assign_expr
# here we are inside data
p_element = None
p_nl -= 1
for line in lines[p_nl:stop]:
p_nl += 1
words = line.split()
# ignore atom mass
if len(words) == 1 and isfloat(words[0]):
continue
# parse element allowing empty symbol
elif len(words) <= 1:
w = line.strip()
p_element = w[:1].upper() + w[1:].lower()
elif len(words) == xcfg_entry_count and p_element is not None:
fields = [ float(w) for w in words ]
stru.addNewAtom(p_element, fields[:3])
a = stru.getLastAtom()
a.xyz *= xcfg_A
p_assign_atom(a, fields)
else:
emsg = "%d: invalid record" % p_nl
raise StructureFormatError(emsg)
if len(stru) != p_natoms:
emsg = "expected %d atoms, read %d" % (p_natoms, len(stru))
raise StructureFormatError(emsg)
except (ValueError, IndexError):
emsg = "%d: file is not in XCFG format" % p_nl
exc_type, exc_value, exc_traceback = sys.exc_info()
raise StructureFormatError, emsg, exc_traceback
return stru
