scrap from blocket.se

blocket.py

from requests import get
from lxml import html
from operator import itemgetter

citycodes = {'skane' : '23', 'lund': '23_10', 'eslov' : '23_2'}
result = []

def list(city, keyword, limit):
    url = 'http://www.blocket.se/' + city + '?q=' + keyword + \
               '&cg=0&w=0&st=s&ca=' + citycodes[city] + '&is=1&l=0&md=th'
    print "Looking for", keyword, "in", city, url
    text = get(url)
    doc = html.fromstring(text.content)
    articles = doc.xpath('//*/article/div')
    l = 0
    for v in articles:
        if l >= limit:
            break
        l += 1
        if not v.get("itemprop"):
            continue
        match = {}
        match['keyword'] = keyword
        match['date'] = v[0][1].get("datetime")
        match['text'] = v[1][0].text
        price = v[2].text
        if price:
            price = v[2].text.replace(":-", "").replace(" ", "")
        else:
            price = "?"
        match['price'] = price
        match['city'] = v[0][0][1].text
        match['link'] = v[1][0].get("href")
        result.append(match)

favorites = ["nintendo", "wii," "3ds", "gamecube", "sega", "dreamcast", "xbox", "x-box", "ps3", "ps4", "playstation", "lego", "playmobil", "psvita", "ps vita", "psp"]
query = ' ELLER '.join(favorites)
print query
list("skane", query, len(favorites) * 5)
#for f in favorites:
#    list("skane", f, 5)

sorted_result = sorted(result, key=itemgetter('date'), reverse=True)

for m in sorted_result:
    print '\033[96m' + m['date'],
    #print '%-16s' % ('\033[95m' + m['keyword']),
    print '\033[94m' + m['text'],
    print '\033[93m' + m['price'],
    print '\033[92m' + m['city'],
    print '\033[91m' + m['link'],
    print