Downloading files from MyTardis using its RESTful API

mytardis_download.py

#!/usr/bin/env python
"""
The following Python script demonstrates how to download files within a
private MyTardis dataset using API key authentication.

WARNING: This script does not check MD5 sums or verify file sizes etc. and
it only retrieves the first page of datafile records
(see datafiles['meta']['limit'] and datafiles['meta']['next']).

Replace "https://mytardis.example.com" with a real MyTardis URL.

Replace "username:api_key" below with valid credentials for your
MyTardis server, otherwise you will get an HTTP 401 (Unauthorized) error.

Replace 1234 below with a dataset ID you want to download.  When viewing
a dataset in your web browser, you can see its ID in your browser's address
bar, e.g. http://mytardis.example.com/dataset/1234
"""
import requests
import os
mytardis_url = "https://mytardis.example.com"
dataset_id = 1234
url = "%s/api/v1/dataset/%d/files?format=json" \
    % (mytardis_url, dataset_id)
headers = {'Accept': 'application/json',
           'Authorization': 'ApiKey username:api_key'}
response = requests.get(url, headers=headers)
if response.status_code != 200:
  raise Exception("HTTP %d" % response.status_code, response.text)
datafiles = response.json()
# datafiles['meta']['next'] will tell us if there are
# additional pages of datafiles to be retreived:
print str(datafiles['meta'])
dataset_dir = "dataset-%d" % dataset_id
if not os.path.exists(dataset_dir):
  os.makedirs(dataset_dir)

for datafile in datafiles['objects']:
  if datafile['directory'] != "":
    if not os.path.exists(os.path.join(dataset_dir, datafile['directory'])):
      os.makedirs(os.path.join(dataset_dir, datafile['directory']))
                               
  print "Downloading %s..." % os.path.join(dataset_dir, datafile['directory'],
                                           datafile['filename'])
  url = "%s/%sdownload/" % (mytardis_url, datafile['resource_uri'])
  response = requests.get(url, headers=headers)
  f = open(os.path.join(dataset_dir, datafile['directory'],
                        datafile['filename']), 'wb')
  for chunk in response.iter_content(chunk_size=512 * 1024): 
    if chunk: # filter out keep-alive new chunks
      f.write(chunk)
  f.close()