def process_genotyping_manifest_file_from_bucket(bucket_name,
genotyping_folder_name):
bucket_stat_list = cloudstorage_api.listbucket('/' + bucket_name)
if not bucket_stat_list:
logging.info('No files in cloud bucket %r.' % bucket_name)
return None
bucket_stat_list = [
s for s in bucket_stat_list
if s.filename.lower().endswith('.csv') and '%s' %
genotyping_folder_name in s.filename
]
if not bucket_stat_list:
logging.info('No CSVs in cloud bucket %r folder %r (all files: %s).' %
(bucket_name, genotyping_folder_name, bucket_stat_list))
return None
bucket_stat_list.sort(key=lambda s: s.st_ctime)
path = bucket_stat_list[-1].filename
timestamp = datetime.datetime.utcfromtimestamp(
bucket_stat_list[-1].st_ctime)
csv_file = cloudstorage_api.open(path)
logging.info('Opening latest genotyping manifest CSV in %r: %r.',
bucket_name + '/' + genotyping_folder_name, path)
now = clock.CLOCK.now()
if now - timestamp > _MAX_INPUT_AGE:
logging.info(
'Input %r (timestamp %s UTC) is > 24h old (relative to %s UTC), not processing.'
% (path, timestamp, now))
return None
update_sample_info_from_genotyping_manifest_file(csv_file)
def process_genomic_manifest_result_file_from_bucket():
bucket_name = config.getSetting(config.BIOBANK_SAMPLES_BUCKET_NAME)
result_folder_name = config.getSetting(GENOMIC_BIOBANK_MANIFEST_RESULT_FOLDER_NAME)
bucket_stat_list = cloudstorage_api.listbucket('/' + bucket_name)
if not bucket_stat_list:
logging.info('No files in cloud bucket %r.' % bucket_name)
return None
bucket_stat_list = [s for s in bucket_stat_list if s.filename.lower().endswith('.csv')
and '%s' % result_folder_name in s.filename]
if not bucket_stat_list:
logging.info(
'No CSVs in cloud bucket %r folder %r (all files: %s).' % (bucket_name, result_folder_name,
bucket_stat_list))
return None
bucket_stat_list.sort(key=lambda s: s.st_ctime)
path = bucket_stat_list[-1].filename
csv_file = cloudstorage_api.open(path)
filename = path.replace('/' + bucket_name + '/' + result_folder_name + '/', '')
logging.info('Opening latest genomic manifest result CSV in %r: %r.', bucket_name + '/' + result_folder_name,
path)
timestamp = timestamp_from_filename(filename)
now = clock.CLOCK.now()
if now - timestamp > _MAX_INPUT_AGE:
logging.info('Input %r (timestamp %s UTC) is > 24h old (relative to %s UTC), not processing.'
% (filename, timestamp, now))
print('Input %r (timestamp %s UTC) is > 24h old (relative to %s UTC), not processing.'
% (filename, timestamp, now))
return None
genomic_set_id = _get_genomic_set_id_from_filename(filename)
update_package_id_from_manifest_result_file(genomic_set_id, csv_file)
def _find_latest_genomic_set_csv(self, cloud_bucket_name, keyword=None):
bucket_stat_list = cloudstorage_api.listbucket('/' + cloud_bucket_name)
if not bucket_stat_list:
raise RuntimeError('No files in cloud bucket %r.' %
cloud_bucket_name)
bucket_stat_list = [
s for s in bucket_stat_list if s.filename.lower().endswith('.csv')
]
if not bucket_stat_list:
raise RuntimeError('No CSVs in cloud bucket %r (all files: %s).' %
(cloud_bucket_name, bucket_stat_list))
if keyword:
buckt_stat_keyword_list = []
for item in bucket_stat_list:
if keyword in item.filename:
buckt_stat_keyword_list.append(item)
if buckt_stat_keyword_list:
buckt_stat_keyword_list.sort(key=lambda s: s.st_ctime)
return buckt_stat_keyword_list[-1].filename
else:
raise RuntimeError(
'No CSVs in cloud bucket %r with keyword %s (all files: %s).'
% (cloud_bucket_name, keyword, bucket_stat_list))
bucket_stat_list.sort(key=lambda s: s.st_ctime)
return bucket_stat_list[-1].filename
def update_ehr_status():
"""
Entrypoint, executed as a cron job
"""
now = clock.CLOCK.now()
cutoff_date = (now - datetime.timedelta(days=1)).date()
bucket_name = _get_curation_bucket_name()
try:
organization_info_list = _get_organization_info_list(
cloudstorage_api.listbucket('/' + bucket_name), cutoff_date)
except config.MissingConfigException as e:
LOG.info(str(e))
return
for org_info in organization_info_list:
deferred.defer(_do_update_for_organization, *org_info)
def _find_latest_genomic_set_csv(cloud_bucket_name):
"""Returns the full path (including bucket name) of the most recently created CSV in the bucket.
Raises:
RuntimeError: if no CSVs are found in the cloud storage bucket.
"""
bucket_stat_list = cloudstorage_api.listbucket('/' + cloud_bucket_name)
if not bucket_stat_list:
raise FileNotFoundError('No files in cloud bucket %r.' % cloud_bucket_name)
# GCS does not really have the concept of directories (it's just a filename convention), so all
# directory listings are recursive and we must filter out subdirectory contents.
bucket_stat_list = [s for s in bucket_stat_list if s.filename.lower().endswith('.csv')
and '%s' % _RESULT_FILE_SUFFIX not in s.filename]
if not bucket_stat_list:
raise FileNotFoundError(
'No CSVs in cloud bucket %r (all files: %s).' % (cloud_bucket_name, bucket_stat_list))
bucket_stat_list.sort(key=lambda s: s.st_ctime)
return bucket_stat_list[-1].filename
