def dictFromColumns(fN, keyValCols, keyValTypes, assumeUnique = True):
'''Cols and Types are 2-tuple with key first, val second
assumeUnique will raise Error True and non-unique value arises'''
keyCasteFxn = cgLuckyCharmsFlat.getCasteFunction(keyValTypes[0])
valCasteFxn = cgLuckyCharmsFlat.getCasteFunction(keyValTypes[1])
keyCol = keyValCols[0]
valCol = keyValCols[1]
key_val = {}
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
key, val = keyCasteFxn(ls[keyCol]), valCasteFxn(ls[valCol])
if assumeUnique:
if key in key_val:
raise NameError("Mapping is not 1 to 1")
else:
key_val[key] = val
else:
key_val.setdefault(key, []).append(val)
f.close()
return key_val
def dictFromColumns(fN, keyValCols, keyValTypes, assumeUnique = True):
'''Cols and Types are 2-tuple with key first, val second
assumeUnique will raise Error True and non-unique value arises'''
keyCasteFxn = cgLuckyCharmsFlat.getCasteFunction(keyValTypes[0])
valCasteFxn = cgLuckyCharmsFlat.getCasteFunction(keyValTypes[1])
keyCol = keyValCols[0]
valCol = keyValCols[1]
key_val = {}
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
key, val = keyCasteFxn(ls[keyCol]), valCasteFxn(ls[valCol])
if assumeUnique:
if key in key_val:
raise NameError("Mapping is not 1 to 1")
else:
key_val[key] = val
else:
key_val.setdefault(key, []).append(val)
f.close()
return key_val
def loadTranscriptionInfo(self, attNames): '''loads caste fxns, column positions, default values for each selected attribute''' for attName in attNames: dataField = getattr(self._dataClass, attName) self._attName_casteFromFxn[attName] = cgLuckyCharms.getCasteFunction(dataField.dataType, True) self._attName_casteToFxn[attName] = cgLuckyCharms.getCasteFunction(dataField.dataType, False) self._attName_columnPosition[attName] = dataField.dataSlot self._attName_defaultValue[attName] = dataField.dataDefault
def loadTranscriptionInfo(self):
'''loads caste fxns, column positions, default values for each ALL attributes in format file'''
for attName in self._attName__formatInfo:
dataSlot, dataType, dataDefault = self._attName__formatInfo[attName]
self._attName_casteFromFxn[attName] = cgLuckyCharms.getCasteFunction(dataType, True)
self._attName_casteToFxn[attName] = cgLuckyCharms.getCasteFunction(dataType, False)
self._attName_columnPosition[attName] = dataSlot
self._attName_defaultValue[attName] = dataDefault
def loadTranscriptionInfo(self, attNames):
'''loads caste fxns, column positions, default values for each selected attribute'''
for attName in attNames:
dataField = getattr(self._dataClass, attName)
self._attName_casteFromFxn[
attName] = cgLuckyCharms.getCasteFunction(
dataField.dataType, True)
self._attName_casteToFxn[attName] = cgLuckyCharms.getCasteFunction(
dataField.dataType, False)
self._attName_columnPosition[attName] = dataField.dataSlot
self._attName_defaultValue[attName] = dataField.dataDefault
def listFromColumns(fN, columns, valTypes, mergeType = 'lol', naToZero = False):
'''multiple columns should either go into multiple lists (list of lists)
or merge into the same list
mergeType = lol or merge (list of list or merge)'''
#checks
if len(columns) != len(valTypes):
raise NameError("Must provide Types for ALL columns")
colNum_casteFxn = dict( (columns[i], cgLuckyCharmsFlat.getCasteFunction(valTypes[i])) for i in range(len(columns)) )
lol = [list() for i in columns]
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
for i, colNum in enumerate(columns):
if ls[colNum] == "NA":
ls[colNum] = "0"
lol[i].append(colNum_casteFxn[colNum](ls[colNum]))
f.close()
if len(lol) == 1:
return lol[0]
elif mergeType == 'lol':
return lol
elif mergeType == 'merge':
mergedList = []
[mergedList.extend(x) for x in lol]
return mergedList
else:
raise NameError("WTH?!")
def listFromColumns(fN, columns, valTypes, mergeType = 'lol'):
'''multiple columns should either go into multiple lists (list of lists)
or merge into the same list
mergeType = lol or merge (list of list or merge)'''
#checks
if len(columns) != len(valTypes):
raise NameError("Must provide Types for ALL columns")
colNum_casteFxn = dict( (i, cgLuckyCharmsFlat.getCasteFunction(valTypes[i])) for i in range(len(columns)) )
lol = [list() for i in range(len(columns))]
f = open(fN, 'r')
for line in f:
ls = line.strip().split('\t')
for i in columns:
lol[i].append(ls[i])
f.close()
if len(lol) == 1:
return lol[0]
elif mergeType == 'lol':
return lol
elif mergeType == 'merge':
mergedList = []
[mergedList.extend(x) for x in lol]
return mergedList
else:
raise NameError("WTH?!")
def loadFormatInfo(self):
'''from column format file, get positions and such
0 is always the id so start from 1'''
#handle quickFormat
if type(self._dataFormatFN) == type([]):
for formatLine in self._dataFormatFN:
colNum, attName, theType, defValue = formatLine.strip().split(' ')
colNum = int(colNum)
#check for empty lists:
if 'List' in theType and defValue == '.':
defValue = list()
else:
defValue = cgLuckyCharms.getCasteFunction(theType)(defValue)
self._attName__formatInfo[attName] = (colNum, theType, defValue)
#handle file
else:
f = open(self._dataFormatFN, 'r')
for i, line in enumerate(f):
#blank line means skipped data
if line.strip() == '': continue
#get formatting info
attName, theType, defValue = line.strip().split('\t')
#check for empty lists:
if 'List' in theType and defValue == '.':
defValue = list()
else:
defValue = cgLuckyCharms.getCasteFunction(theType)(defValue)
self._attName__formatInfo[attName] = (i + 1, theType, defValue)
f.close()
