def _parseHeaderLine(line):
headerLine = line[2:]
splitLine = headerLine.split(':')
if len(splitLine) != 2:
raise InvalidFormatError('Header line not understood: ' + repr(headerLine))
key, val = splitLine
key = key.lower()
val = val.strip()
if key == GENOME_HEADER:
val = urlDecodePhrase(val)
elif key not in HEADER_VAR_DICT:
if key.endswith(' '):
raise InvalidFormatError('Header variable "%s" must not end with space.' % key)
# raise InvalidFormatError('Header variable "%s" is not part of the GSuite format.' % key)
if urlDecodePhrase(key) != key:
raise InvalidFormatError('Custom header variable names in GSuite do not support URL '
'escaping. Offending header variable: "{}"'.format(key))
else:
val = val.lower()
if val not in HEADER_VAR_DICT[key].allowed:
raise InvalidFormatError('Value "%s" is not allowed for header "%s". Allowed values: %s' %
(val, key, ', '.join(HEADER_VAR_DICT[key].allowed)))
if key == FILE_TYPE_HEADER:
if val == TEXT:
val = PRIMARY
elif val == BINARY:
val = PREPROCESSED
return key, val
def _parseColumnSpecLine(line):
colNames = line[3:].lower().split('\t')
# if any(' ' in colName for colName in colNames):
# raise InvalidFormatError('Error in column specification line: %s ' % repr(line) +
# 'Please separate columns by tab, not space.')
colNames = [(col if col not in ALL_STD_COL_NAMES else col)
for col in colNames]
for colName in colNames:
if colNames.count(colName) > 1:
raise InvalidFormatError(
'Column "%s" appears multiple times in the ' % colName +
'column specification line.')
if colNames[0] == '':
raise InvalidFormatError(
'Column specification line requires at least one'
'column (the "uri" column), but none is specified.')
if colNames[0] != URI_COL:
raise InvalidFormatError('The first column must be "%s", not "%s".' %
(URI_COL, colNames[0]))
if any(colName.strip() == '' for colName in colNames):
raise InvalidFormatError('Empty column names are not allowed.')
curOptStdColIdx = -1
nonStdColsFound = []
for colName in colNames[1:]:
if colName in OPTIONAL_STD_COL_NAMES:
nextOptStdColIdx = OPTIONAL_STD_COL_NAMES.index(colName)
if nonStdColsFound:
raise InvalidFormatError(
'Non-standard columns "%s" ' % ', '.join(nonStdColsFound) +
'encountered before standard column "%s".' % colName)
elif nextOptStdColIdx <= curOptStdColIdx:
raise InvalidFormatError(
'Standard columns are not in the correct order: '
'%s.' % ', '.join('"%s"' % col
for col in OPTIONAL_STD_COL_NAMES))
curOptStdColIdx = nextOptStdColIdx
else:
if urlDecodePhrase(colName) != colName:
raise InvalidFormatError(
'Column names in GSuite do not support URL escaping. '
'Offending column name: "{}"'.format(colName))
nonStdColsFound.append(colName)
return colNames
def _parseGenomeLine(line):
genomeLine = line[4:]
splitLine = genomeLine.split('=')
if len(splitLine) != 2:
raise InvalidFormatError('Genome line not understood: ' + repr(genomeLine))
key, genome = [_.strip() for _ in splitLine]
genome = urlDecodePhrase(genome)
key = key.lower()
if key != 'genome':
raise InvalidFormatError('Key in genome line is not "genome": ' + key)
return genome
def setCustomHeader(self, headerName, headerVal):
headerName = headerName.lower()
self._customHeaders[headerName] = urlDecodePhrase(headerVal)