#!/usr/bin/env python

import arvados
import hashlib
import re

arvados.job_setup.one_task_per_input_file(if_sequence=0, and_end_task=True)

this_job = arvados.current_job()
this_task = arvados.current_task()
this_task_input = this_task['parameters']['input']

if 'algorithm' in this_job['script_parameters']:
    alg = this_job['script_parameters']
else:
    alg = 'md5'
digestor = hashlib.new(alg)

input_stream = arvados.DataReader(this_task_input)
while True:
    buf = input_stream.read(2**20)
    if len(buf) == 0:
        break
    digestor.update(buf)

hexdigest = digestor.hexdigest()
file_name = re.sub(r'^[^/]+/', '', this_task_input)

input_stream.close()

this_task.set_output("%s %s\n" % (hexdigest, file_name))