def readColumns(filename, columns=[1, 2]):
# Read file and clear out comments
with open(filename, 'r') as file:
cleanStr = file.read()
comments = pp.ZeroOrMore(pp.pythonStyleComment).setParseAction(
pp.replaceWith(''))
try:
cleanStr = comments.transformString(cleanStr)
except pp.ParseException as pe:
print(('Parsing Error using pyparsing: invalid input:', pe))
sys.exit()
# Define grammar for ncols of data based on number of entries in first row
floating = pp.Word(pp.nums + ".+-E").setParseAction(lambda t: float(t[0]))
EOL = pp.LineEnd().suppress()
row1entry = floating.copy().setWhitespaceChars(" \t")
row1 = pp.Group(pp.ZeroOrMore(EOL) + pp.OneOrMore(row1entry) + EOL)
row = pp.Forward()
def defineTotalCols(toks):
ncols = len(toks[0])
row << pp.Group(floating * ncols)
return None
row1.addParseAction(defineTotalCols)
text = row1 + pp.ZeroOrMore(row)
try:
data = text.parseString(cleanStr).asList()
except pp.ParseException as pe:
print(('Parsing Error using pyparsing: invalid input:', pe))
sys.exit()
cols = list(map(list, list(zip(*data))))
return [cols[i - 1] for i in columns]
def readDym(dymFilename):
integer = pp.Word(pp.nums).setParseAction(lambda t: int(t[0]))
floating = pp.Word(pp.nums + ".+-E").setParseAction(lambda t: float(t[0]))
vector = pp.Group(floating * 3)
matrix = pp.Group(vector * 3)
dymtype = integer
ncluster = integer.copy() # Uses unique ParseAction for Forward
atNums = pp.Forward() # Placeholders for grabbing ncluster entries
atMasses = pp.Forward()
atCoords = pp.Forward()
def countedParseAction(toks):
n = toks[0]
atNums << pp.Group(integer * n)
atMasses << pp.Group(floating * n)
atCoords << pp.Group(vector * n)
return None
ncluster.addParseAction(countedParseAction)
dm = pp.Group(pp.OneOrMore(pp.Group(integer * 2 + matrix)))
specifics = pp.SkipTo(pp.stringEnd)
text = dymtype + ncluster + atNums + atMasses + atCoords + dm + specifics
with open(dymFilename, 'r') as f:
try:
data = text.parseString(f.read()).asList()
except pp.ParseException as pe:
print(('Parsing Error using pyparsing: invalid input:', pe))
sys.exit()
# Construct dym object
dym = {'dymType': data[0]}
dym['nAt'] = data[1]
dym['atNums'] = data[2]
dym['atMasses'] = data[3]
dym['atCoords'] = data[4]
dym['dm'] = [[None for j in range(dym['nAt'])] for i in range(dym['nAt'])]
for i, j in ((i, j) for i in range(dym['nAt']) for j in range(dym['nAt'])):
index = i * dym['nAt'] + j
if data[5][index][0] == i + 1 and data[5][index][1] == j + 1:
dym['dm'][i][j] = data[5][index][2]
# Parse dymtype-specific items
if dym['dymType'] == 2:
nUnique = integer
nCell = integer
centralZAtom = pp.Group(integer + integer + vector)
zAtomList = pp.Group(integer + pp.countedArray(centralZAtom))
typeSpecific = nUnique + nCell + pp.Group(pp.OneOrMore(zAtomList))
# Parse
try:
data = typeSpecific.parseString(data[6]).asList()
except pp.ParseException as pe:
print(('Parsing Error using pyparsing: invalid input:', pe))
sys.exit()
# Store type-specific information
if dym['dymType'] == 2:
dym['nTypeAt'] = data[0]
dym['nCellAt'] = data[1]
dym['centerAt'] = {}
for zAtomList in data[2]:
z = zAtomList[0]
dym['centerAt'][z] = []
for centralZAtom in zAtomList[1]:
atomInfo = {}
atomInfo['clustIndex'] = centralZAtom[0]
atomInfo['cellIndex'] = centralZAtom[1]
atomInfo['coord'] = centralZAtom[2]
dym['centerAt'][z].append(atomInfo)
return dym