def open_input_file(self, input_path):
encoding = detect_encoding(input_path)
if input_path.endswith(".gz"):
f = gzip.open(input_path, mode="rt", encoding=encoding)
else:
f = open(input_path, encoding=encoding)
return f
def _loop_definition(self):
encoding = detect_encoding(self.path)
#sys.stderr.write('loop definition [' + self.path + ']. encoding=' + str(encoding) + '\n')
f = open(self.path, encoding=encoding)
for l in f:
l = l.rstrip().lstrip()
if l.startswith('#'):
yield l
else:
break
f.close()
def first_input_file(self):
if self.pipeinput == False:
input_path = self.input_paths[0]
encoding = detect_encoding(input_path)
if input_path.endswith(".gz"):
f = gzip.open(input_path, mode="rt", encoding=encoding)
else:
f = open(input_path, encoding=encoding)
else:
f = sys.stdin
return f
def __init__(self, path, seekpos=None, chunksize=None):
super().__init__(path)
self.seekpos = seekpos
self.chunksize = chunksize
self.encoding = detect_encoding(self.path)
self.annotator_name = ''
self.annotator_displayname = ''
self.annotator_version = ''
self.no_aggregate_cols = []
self.index_columns = []
self.report_substitution = None
self._setup_definition()
def _loop_data(self):
encoding = detect_encoding(self.path)
#sys.stderr.write('loop data [' + self.path + ']. encoding=' + str(encoding) + '\n')
f = open(self.path, 'rb')
lnum = 0
for l in f:
l = l.decode(encoding)
lnum += 1
l = l.rstrip('\r\n')
if l.startswith('#'):
continue
else:
yield lnum, l
f.close()
def setup(self):
""" Do necesarry pre-run tasks """
if self.ready_to_convert: return
# Open file handle to input path
for input_path in self.input_paths:
encoding = detect_encoding(input_path)
if input_path.endswith('.gz'):
f = gzip.open(input_path, mode='rt', encoding=encoding)
else:
f = open(input_path, encoding=encoding)
self.input_files.append(f)
# Read in the available converters
self._initialize_converters()
# Select the converter that matches the input format
self._select_primary_converter()
# Open the output files
self._open_output_files()
self.ready_to_convert = True
def setup (self):
""" Do necesarry pre-run tasks """
if self.ready_to_convert: return
# Open file handle to input path
for input_path in self.input_paths:
encoding = detect_encoding(input_path)
self.input_files.append(open(input_path, encoding=encoding))
# Read in the available converters
self._initialize_converters()
# Select the converter that matches the input format
self._select_primary_converter()
# A correct .crv file is not processed.
#todo handle this for multiple inputs. have to convert them so they can be merged
# if self.input_format == 'crv' and \
# self.input_paths[0].split('.')[-1] == 'crv':
# self.logger.info('Input file is already a crv file. Exiting converter.')
# exit(0)
# Open the output files
self._open_output_files()
self.ready_to_convert = True
def write_preface(self, level):
self.level = level
if self.wf is not None:
self.wf.close()
if level != 'variant':
return
self.wf = open(self.filename, 'w', encoding='utf-8', newline='')
lines = [
'#fileformat=VCFv4.2',
'#OpenCRAVATFileDate=' +
datetime.datetime.now().strftime('%Y%m%d'),
]
self.write_preface_lines(lines)
self.vcflines = {}
self.input_path_dict = {}
if self.input_format == 'vcf':
if self.args.inputfiles is not None:
if type(self.args.inputfiles) is str:
self.args.inputfiles = [self.args.inputfiles]
for i in range(len(self.args.inputfiles)):
self.input_path_dict[self.args.inputfiles[i]] = i
written_headers = []
self.samples = []
num_inputfiles = len(self.args.inputfiles)
for inputfile in self.args.inputfiles:
inputfile_prefix = os.path.basename(inputfile).split(
'.')[0]
input_path_no = self.input_path_dict[inputfile]
encoding = detect_encoding(inputfile)
if inputfile.endswith('.gz'):
import gzip
f = gzip.open(inputfile, 'rt', encoding=encoding)
else:
f = open(inputfile)
lineno = 0
self.vcflines[input_path_no] = {}
for line in f:
lineno += 1
if line.startswith('##fileformat='):
continue
if line.startswith('##'):
if not line in written_headers:
self.wf.write(line)
written_headers.append(line)
elif line.startswith('#CHROM'):
toks = line[:-1].split('\t')
if len(toks) >= 10:
if num_inputfiles == 1:
self.samples.extend([v for v in toks[9:]])
else:
self.samples.extend([
inputfile_prefix + '_' + v
for v in toks[9:]
])
elif line.startswith('#') == False:
self.vcflines[input_path_no][lineno] = line.rstrip(
'\n').rstrip('\r')
f.close()
else:
self.cursor2.execute(
'select distinct(base__sample_id) from sample')
self.samples = []
rows = self.cursor2.fetchall()
if rows is None or len(rows) == 0:
self.samples.append('NOSAMPLEID')
else:
for row in rows:
v = row[0]
if v is None:
v = 'NOSAMPLEID'
self.samples.append(v)