def record_to_csv(self,
tarfile_name,
filepath,
fullpath,
archive_path,
filesize,
result,
ext=None):
flowcell_id = SFHelper.get_flowcell_id(tarfile_name)
normalized_filepath = fullpath.split("uploads/")[-1]
if filepath.endswith('html'):
head, sep, tail = fullpath.partition('all/')
path = head.split(flowcell_id + '/')[-1]
else:
path = SFUtils.get_meta_path(fullpath, False)
includes_csv = open(self.includes_csv_path, "a")
includes_csv.write(tarfile_name + ", " + normalized_filepath + ", " +
archive_path + ", " + flowcell_id + ", " +
SFHelper.get_pi_name(path) + ", " +
SFHelper.get_project_id(path) + ", " +
SFHelper.get_project_name(path, tarfile_name, ext) +
", " + SFHelper.get_sample_name(path) + ", " +
SFHelper.get_run_name(tarfile_name) + ", " +
SFHelper.get_sequencing_platform(tarfile_name) +
"," + str(filesize) + "," + result + "\n")
includes_csv.close()
def main(args):
if len(sys.argv) < 6:
print(
"\n Usage: python app.py tarfile_list tarfile_dir extract_path audit_dir dryrun include_file_size"
)
return
# The file containing the tarlist
tarfile_list = args[1]
# The directory containing the tarfiles
tarfile_dir = args[2]
# path containing the extracted file
extract_path = args[3]
# sub-directory to hold the log and audit files
audit_dir = args[4]
#If this is a dryrun or not
dryrun = args[5].lower() == 'true'
#If filesize has to be recorded or not - applicable only for dry run
if dryrun:
if len(sys.argv) > 6:
record_file_size = args[6]
else:
record_file_size = False
else:
record_file_size = True
bytes_stored = 0
files_registered = 0
#if (args[6] is not None):
# bytes_stored = args[6]
# if (args[7] is not None):
# files_registered = args[7]
sf_audit = SFAudit(audit_dir, extract_path, bytes_stored, files_registered)
sf_audit.prep_for_audit()
for line_filepath in open(tarfile_list).readlines():
tarfile_name = line_filepath.rstrip()
tarfile_path = tarfile_dir + '/' + tarfile_name.rstrip()
# This is a valid tarball, so process
logging.info("Processing file: " + tarfile_path)
if (tarfile_name.endswith("supplement.tar")
or 'singlecell' in tarfile_name or '10x' in tarfile_name):
# Register PI collection
register_collection(tarfile_path, "PI_Lab", tarfile_name, False,
sf_audit, dryrun)
# Register Flowcell collection with Project type parent
register_collection(tarfile_path, "Flowcell", tarfile_name, True,
sf_audit, dryrun)
# create Object metadata with Flowcell type parent and register object
register_object(tarfile_path, "Flowcell", tarfile_name, False,
tarfile_path, sf_audit, dryrun)
logging.info('Done processing file: ' + tarfile_path)
continue
#Get or create list file (if not present)
tarfile_contents = SFUtils.get_tarball_contents(
tarfile_name, tarfile_dir, sf_audit)
if tarfile_contents is None:
continue
# Extract all files and store in extract_path directory
if record_file_size:
if not (SFUtils.extract_files_from_tar(tarfile_path,
extract_path)):
# Something wrong with this file path, go to
# next one and check logs later
continue
#loop through each line in the contents file of this tarball
#We need to do an upload for each fastq.gz or BAM file
for line in tarfile_contents.readlines():
logging.info('processing line in tarfile: ' + line)
if (line.rstrip().endswith("/")):
#This is a directory, nothing to do
continue
#Get full path of the extracted file
filepath = SFUtils.get_filepath_to_archive(line.rstrip(),
extract_path)
logging.info('filepath to archive: ' + filepath)
#if SFUtils.path_contains_exclude_str(tarfile_name, line.rstrip()):
exclusion_list = ['10X', 'demux', 'demultiplex']
if any(ext in line.rstrip() for ext in exclusion_list):
sf_audit.record_exclusion(
tarfile_name, line.rstrip(), filepath,
'Path contains substring from exclusion list')
continue
if filepath.endswith('fastq') or filepath.endswith(
'fastq.gz') or filepath.endswith('fastq.gz.md5'):
# Extract the info for PI metadata
path = SFUtils.get_meta_path(filepath)
ext = SFUtils.get_unaligned_ext(filepath)
# Register PI collection
register_collection(path, "PI_Lab", tarfile_name, False,
sf_audit, dryrun)
#Register Flowcell collection with Project type parent
register_collection(path, "Flowcell", tarfile_name, True,
sf_audit, dryrun, ext)
#create Object metadata with Sample type parent and register object
register_object(path, "Sample", tarfile_name, True, filepath,
sf_audit, dryrun, ext)
elif line.rstrip().endswith(
'laneBarcode.html'
) and '/all/' in line and not 'Control_Sample' in line:
#Remove the string after the first '/all' because metadata path if present will be before that
head, sep, tail = line.partition('all/')
#Remove everything upto the Flowcell_id, because metadata path if present will be after that
flowcell_id = SFHelper.get_flowcell_id(tarfile_name)
if flowcell_id in head:
path = head.split(flowcell_id + '/')[-1]
ext = SFUtils.get_unaligned_ext(filepath)
#Ensure that metadata path does not have the Sample sub-directory and that it is valid
if path.count('/') == 1 and '_' in path:
#Register the html in flowcell collection
path = path + 'laneBarcode.html'
logging.info('metadata base: ' + path)
# Register PI collection
register_collection(path, "PI_Lab", tarfile_name,
False, sf_audit, dryrun)
# Register Flowcell collection with Project type parent
register_collection(path, "Flowcell", tarfile_name,
True, sf_audit, dryrun, ext)
# create Object metadata with Flowcell type parent and register object
register_object(path, "Flowcell", tarfile_name, False,
filepath, sf_audit, dryrun, ext)
else:
# ignore this html
sf_audit.record_exclusion(
tarfile_name, line.rstrip(), filepath,
'html path not valid - may have other sub-directory'
)
continue
else:
#ignore this html
sf_audit.record_exclusion(
tarfile_name, line.rstrip(), filepath,
'html path not valid - could not extract flowcell_id')
continue
else:
#For now, we ignore files that are not fastq.gz or html
sf_audit.record_exclusion(tarfile_name, line.rstrip(),
filepath,
'Not fastq.gz or valid html file')
logging.info('Done processing file: ' + tarfile_path)
# delete the extracted tar file
if record_file_size:
os.system("rm -rf " + extract_path + "*")
sf_audit.audit_summary()
def main(args):
if len(sys.argv) < 5:
print(
"\n Usage: python app.py tarfile_list tarfile_dir extract_path audit_dir dryrun initial_bytes initial_file_count"
)
return
# The file containing the tarlist
tarfile_list = args[1]
# The directory containing the tarfiles
tarfile_dir = args[2]
# path containing the extracted file
extract_path = args[3]
# sub-directory to hold the log and audit files
audit_dir = args[4]
#If this is a dryrun or not
dryrun = args[5].lower() == 'true'
sf_audit = SFAudit(audit_dir, extract_path, 0, 0)
sf_audit.prep_for_audit()
for line_filepath in open(tarfile_list).readlines():
tarfile_name = line_filepath.rstrip()
tarfile_path = tarfile_dir + '/' + tarfile_name.rstrip()
# This is a valid tarball, so process
logging.info("Processing file: " + tarfile_path)
if (tarfile_name.endswith("supplement.tar")
or 'singlecell' in tarfile_name or '10x' in tarfile_name):
# Register Flowcell collection
register_collection(tarfile_path, "Flowcell", tarfile_name, False,
tarfile_path, sf_audit, dryrun)
logging.info('Done processing file: ' + tarfile_path)
continue
tarfile_contents = SFUtils.get_tarball_contents(
tarfile_name, tarfile_dir, sf_audit)
if tarfile_contents is None:
continue
#loop through each line in the contents file of this tarball
#We need to do an upload for each fatq.gz or BAM file
for line in tarfile_contents.readlines():
if (line.rstrip().endswith("/")):
#This is a directory, nothing to do
continue
#if SFUtils.path_contains_exclude_str(tarfile_name, line.rstrip()):
exclusion_list = ['10X', 'Phix', 'PhiX', 'demux', 'demultiplex']
if any(ext in line.rstrip() for ext in exclusion_list):
sf_audit.record_exclusion(
tarfile_name, line.rstrip(),
'Path contains substring from exclusion list')
continue
filepath = SFUtils.get_filepath_to_archive(line.rstrip(),
extract_path)
if filepath.endswith('fastq.gz') or filepath.endswith(
'fastq.gz.md5'):
# Extract the info for PI metadata
path = SFUtils.get_meta_path(filepath)
#Register Flowcell collection with Project type parent
register_collection(path, "Flowcell", tarfile_name, False,
filepath, sf_audit, dryrun)
else:
#For now, we ignore files that are not fastq.gz or html
sf_audit.record_exclusion(tarfile_name, line.rstrip(),
'Not fastq.gz or valid html file')
logging.info('Done processing file: ' + tarfile_path)