def run(self):
zip_files = glob.glob(os.path.join(self.args.src,
ALEXA_ZIP_FILE_FORMAT))
zip_files.sort()
for zip_f in zip_files:
basename = os.path.basename(zip_f)
date = midas.parse_tstamp(basename, ALEXA_TS_FORMAT)
tstamp = midas.serialize_tstamp(date)
dst_fname = 'top_1m_{0}'.format(tstamp)
dst_f = os.path.join(self.args.dst, dst_fname)
if os.path.isfile(dst_f) or os.stat(zip_f).st_size == 0:
self.out('Skipping {0}'.format(basename))
else:
with open(dst_f, 'w') as fp:
for site, rank in iter_alexa_zip_file(zip_f):
fp.write('{0}\t{1}\t{2}\n'.format(site, tstamp, rank))
self.out('Processed {0}'.format(basename))
def ids_to_samples(self):
if self.args.samples:
directory = self.args.samples
else:
directory = self.config['samples']
if os.path.isfile(directory):
files = [directory, ]
else:
files = []
make_abs = functools.partial(os.path.join, directory)
for path in imap(make_abs, os.listdir(directory)):
if os.path.isfile(path):
files.append(path)
samples = dict()
for f in files:
for site, tstamp, code in csv_file_reader(f, delimiter='\t'):
tstamp = parse_tstamp(tstamp)
site_id = self.sites_to_ids[site]
samples[site_id] = (site, tstamp, code)
return samples