gistfile1.py

#coding=utf-8

from bs4 import BeautifulSoup
import urllib2

request_header = dict()
request_header['User-Agent'] = "kanjianbox spider"
url = "http://www.kanjian.com/"

request = urllib2.Request(url, None, request_header)

req = urllib2.urlopen(request)
doc = req.read()

soup = BeautifulSoup(doc, "lxml")

recommend_list = soup.select("div#recommendArtist .bd ul li.item")

result = []

for item in recommend_list:
    user = dict()

    avatar_tag = item.select("a.userAvatar")[0]
    page_url = avatar_tag["href"]
    uid = page_url.rstrip("/").split("/")[-1]
    user["uid"] = int(uid)

    avatar_img_tag = avatar_tag.select("img")
    user["avatarUrl"] = avatar_img_tag[0]["src"]
    user["name"] = avatar_img_tag[0]["alt"]

    style_a_tags = item.select("p.style a")
    styles = [x.text for x in style_a_tags]
    user["styles"] = "/".join(styles)

    result.append(user)

print result