scrape.py

#!/usr/bin/python3

import urllib.request
import lxml.html
import untangle
import sys

def rchop(the_string, sub_str):
  if (the_string.endswith(sub_str)):
    return the_string[:-len(sub_str)]
  return the_string

def GetImages(assetID):
    param_str = '&Width=800&Height=600'
    url = 'http://www.brydenfinancing.com/~i'+(str(assetID))+'-110'
    req = urllib.request.Request(url, headers={'User-Agent': 'Mozilla/5.0'})
    con = urllib.request.urlopen(req)
    dom = lxml.html.fromstring(con.read());

    images = []

    for link in dom.xpath('//img/@src'):
        if (param_str in link):
            img_src = rchop(link, param_str)
            images.append(img_src)

    return images



doc = untangle.parse('bryden.xml')

# use python's handy 'for item in list' syntax to traverse the list of Assets
for asset in doc.root.Dealership.Assets.Asset:
    # save the VIN cdata value to a variable and print it
    vin = asset.VIN.cdata
    assetid = asset.AssetID.cdata
    stockno = asset.StockNo.cdata
    if 'ROB' in stockno:   # Show Rob's Stock Numbers Only
        print ('StockNo: '+stockno+' | VIN: '+vin+' | ID : '+assetid,end="")
        # some Assets have no pictures which causes an IndexError exception when you try to access the object
        # we surround the "dangerous" code with a try: block
        try:
            # essentially the same procedure as above for Assets... get python to give us each picture object
            # for this asset in turn
            for picture in asset.Pictures.Picture:
                print ('\t'+picture['Url']+'\n')
        # if the above code throws IndexError, we catch it and use that opportunity to inform that there are no pictures
        except IndexError:
            image_list = GetImages(assetid)
            print ("| Found",len(image_list),"pictures")
            sys.stdout.flush()
            print ("\t",image_list[0])