hikvision-snapshot-downloader.py

#!/usr/bin/python
import xml.etree.ElementTree as ET 
import requests 
import re 
import sys 
from datetime import date 
from dateutil.parser import parse
from dateutil.rrule import rrule, DAILY

# suppresses SSL warning
from requests.packages.urllib3.exceptions import InsecureRequestWarning
requests.packages.urllib3.disable_warnings(InsecureRequestWarning)

config =        {
  "username": "<USERNAME>",
  "password": "<PASSWORD>",
  "protocol": "https",
  "ip": "<IP>",
  "url": "/ISAPI/ContentMgmt/search",
  "resultkeyword": ".//{http://www.hikvision.com/ver20/XMLSchema}playbackURI",
  "xml": """
     <CMSearchDescription>
     <searchID>C837920F-0F00-0001-A7BF-10BE33B0E3C0</searchID>
     <trackIDList>
      <trackID>103</trackID>
     </trackIDList>
     <timeSpanList>
      <timeSpan>
         <startTime>DATET00:00:01Z</startTime>
         <endTime>DATET23:59:59Z</endTime>
      </timeSpan>
     </timeSpanList>
     <contentTypeList>
      <contentType>metadata</contentType>
     </contentTypeList>
     <maxResults>1000</maxResults>
     <searchResultPostion>0</searchResultPostion>
     <metadataList>
      <metadataDescriptor>//recordType.meta.std-cgi.com/CMR</metadataDescriptor>
     </metadataList>
     </CMSearchDescription>
""", 
  "datepattern":".*?starttime=(\d\d\d\d\d\d\d\d)T(.*)Z&endtime.*"
}
ip_no_auth = config.get("protocol") + "://" + config.get("ip") 
full_url = ip_no_auth  + config.get("url")

s = requests.Session()
s.auth=(config.get("username"), config.get("password"))


def extractDateFromPlaybackURI(playbackuri):
  date_search = re.search(config.get("datepattern"),playbackuri,re.IGNORECASE)
  if date_search:
    return date_search.group(1) + "-" + date_search.group(2)
  else:
    return "NOTFOUND"

def getImage(playbackuri, session):
  r = session.get(playbackuri, allow_redirects=False)
  head = r.headers
  if not ('jpeg' in head.get('content-type').lower()):
    print "Not an image file!"
    return
  filename = extractDateFromPlaybackURI(playbackuri)
  print "  Getting image:      " + playbackuri + "    ----->    " + filename + ".jpg"
  open(filename+".jpg", 'wb').write(r.content)

def getXML(requestxml):
  r = s.post( full_url, data=requestxml, verify=False)

  if r.status_code == 200:
    print "--> Request OK"
    loopoverXML(ET.fromstring(r.content))
  else:
    print "--> Request NOK"

def loopoverXML(responsexml):
  snapshots = []
  for item in responsexml.findall(config.get("resultkeyword")):
    snapshots.append(item.text)

  print "Downloading..."
  for snap in snapshots:
    print "Item " + str(snapshots.index(snap)+1) + "/" + str(len(snapshots))
    getImage(snap,s)

def getDates(startdate):
  print "Input start date is "+ startdate.strftime('%Y%m%d')
  # check if we need multiple runs or just one for today
  if startdate == date.today():
    print "  Collecting images only for today"
    startdatestring = startdate.strftime('%Y%m%d')
    requestxml = config.get("xml")
  # replace date in template XML with actual date
    requestxml = requestxml.replace("DATE",startdatestring)
    getXML(requestxml)
  else:
    startdatestring = startdate.strftime('%Y%m%d')
    print "  Collecting images from "+startdatestring + " to " + date.today().strftime('%Y%m%d')

    # loop over every date between start date and today
    for dt in rrule(DAILY, dtstart=startdate, until=date.today()):
      requestxml = config.get("xml")
      # replace date
      curdatestring = dt.strftime('%Y%m%d')
      requestxml = requestxml.replace("DATE",curdatestring)
      getXML(requestxml)



if __name__ == "__main__":
  if len(sys.argv) > 1:
    # If argument given - and this exists - we will retrieve all images from that day until today per 24 hours (limit of hikvision)
    arg_start_date = parse(sys.argv[1])
    getDates(arg_start_date)
  else:
    # Default to last day
    getDates(date.today())