read_files.py

import cloudant
import json
import urllib3.request as p
import requests
import struct


def interpret_file(url):
    """
    File structure is:
       bytes 0..3: length of json header N (excluding header word)
       bytes 4..4+N: json header (ASCII data)
       bytes 4+N+1..EOF: binary data of channels

    The binary data format depends on what's in the json header:
      header["channel_list"] ---> ordered list of channels
      header["byte_depth"]    ---> size of binary word
      header["bit_shift"]    ---> amount to shift right

	Every channel is listed one after another for each time point (fully
    interlaced)

    """
    bits = requests.get(file_url, headers={'Range': 'bytes=0-3'}).content
    header_length = struct.unpack("<L", bits)[0] 
    bits = requests.get(file_url, headers={'Range': 'bytes=4-{}'.format(4+header_length)}).content
    hdr = json.loads(bits)
    try:
        bit_depth = hdr["bit_depth"]
    except:
        bit_depth = hdr["byte_depth"]        	
    bit_shift = hdr["bit_shift"]
    print(json.dumps(hdr, indent=4))

def build_url(doc):
   cl = doc["file_servers"]["cluster-local"] 
   return cl + "?" + p.urlencode(dict(
     id = doc["_id"],
     filename=doc["external_docs"].values()[0]["ondiskname"],
     db="nedm/measurements"
   )) 

# Authentication
acct = cloudant.Account(uri="http://10.155.59.15:5984")
res = acct.login("nedm_user", """pw""") 
assert res.status_code == 200

# Grab the correct database
db = acct["nedm%2Fmeasurements"]

# Reads all data from a certain time
document = db.document("c85a276eac9ceb54174368c97915f25c")
res = document.get().json()
#print(json.dumps(res, indent=4))
file_url = build_url(res)

interpret_file(file_url)