Downloads all photos from Miss Photo NSU-2010 web site (which is *really* unusable)

miss.py

# -*- coding: utf-8 -*-

import urllib, sgmllib
import re
import os, shutil

class IDParser(sgmllib.SGMLParser):
    '''Returns list of contestant IDs from main page'''

    def __init__(self):
        sgmllib.SGMLParser.__init__(self)
        self.ids = []

    def parse(self, s):
        self.feed(s)
        self.close()
        return self.ids

    def reset(self):
        sgmllib.SGMLParser.reset(self)
        self.ids = []

    def start_a(self, attrs):
        for k,v in attrs:
            if k == "class" and v == "list":
                self.__parse_href(attrs)

    def __parse_href(self, attrs):
        for k,v in attrs:
            if k == "href":
                id = re.findall(r"\?n=(\d+)\&sort=date", v)[0]
                self.ids += [ int(id) ]

class NameParser(sgmllib.SGMLParser):
    '''Returns name and latin alias from contestant's page'''

    def __init__(self):
        sgmllib.SGMLParser.__init__(self)
        self.reset()

    def reset(self):
        sgmllib.SGMLParser.reset(self)
        self.name = ""
        self.latin = ""
        self.in_name = 0

    def parse(self, s):
        self.feed(s)
        self.close()
        return self.name, self.latin

    def start_h1(self, attrs):
        for k,v in attrs:
            if k == "class" and v == "name":
                self.in_name = 1

    def end_h1(self):
        self.in_name = 0

    def handle_data(self, data):
        if self.in_name:
            self.name = data

    def start_img(self, attrs):
        for k,v in attrs:
            if k == "src":
                self.latin = re.findall(r"foto/(.*?)-\d.jpg", v)[0]

class Miss:
    baseurl = 'http://missphoto.nsu.ru/2010/'

    def __init__(self, imgdir = 'img'):
        self.imgdir = imgdir

    def download(self):
        f = urllib.urlopen(self.baseurl + 'list.php');
        main = f.read()
        f.close()

        id_parser = IDParser()
        ids = id_parser.parse(main)

        maxid = max(ids)

        if os.path.exists(self.imgdir):
            shutil.rmtree(self.imgdir)

        for id in sorted(ids):
            print "%i/%i\t" % (id, maxid),
            self.__parse_id(id)

    def __parse_id(self, id):
        f = urllib.urlopen(self.baseurl + 'view.php?n=%s' % id );
        page = f.read()
        f.close()
        name_parser = NameParser()
        name, latin = name_parser.parse(page)

        uname = name.decode('cp1251')
        print "%-*s \t %s" % (15, latin, uname),

        os.makedirs(self.imgdir + "/" + "%s" % uname)
        for i in range(1, 4):
            file = '%s-%s.jpg' % (latin, i)
            img = urllib.urlopen(self.baseurl + 'foto/' + file)
            img_file = open(self.imgdir + "/" + uname + "/" + file, 'w+')
            img_file.write(img.read())
            img.close()
            img_file.close()
        print

if __name__ == "__main__":
    Miss().download()