#!/usr/bin/env python2
# -*- coding:utf-8 -*-
import httplib
from pyquery import PyQuery as pq
import re

rHTTPHost = re.compile(r"^http://([^/]+)")
rHTTPURI = re.compile(r"^http://[^/]+(/.*)$")


class HTTPConnection(object):

    def __init__(self, host):
        self.host = host
        self.conn = httplib.HTTPConnection(host)

    def reset(self, host):
        self.conn = httplib.HTTPConnection(host)

    def get_html(self, uri, method="GET"):
        self.conn.request(method, uri, headers={"Connection": "keep-alive"})
        try:
            r = self.conn.getresponse()
        except (IOError, httplib.HTTPException):
            self.reset(self.host)
            return self.get_html(uri, method)

        if r.status == 200:
            return r.read()
        elif r.status in (301, 302):
            l = r.getheader("Location")
            host = rHTTPHost.findall(l)[0]
            uri = rHTTPURI.findall(l)[0]
            self.reset(host)
            return self.get_html(uri, method)


connection = HTTPConnection("m.douban.com")
_html = connection.get_html('/movie/tag/movies?tag=2014')
douban = pq(_html)

rStar = re.compile(r'\(([0-5]).\)$')

for mlink in douban(".movie-items.list").find("a"):
    movie_uri = pq(mlink).attr("href")

    _html = connection.get_html(movie_uri)
    movie = pq(_html)

    for clink in movie("#bd > .itm > a"):
        cmt_uri = pq(clink).attr("href")
        if "comment" not in cmt_uri:
            continue

        _html = connection.get_html(cmt_uri)
        # commets is the first .list element
        comments = pq(_html)("#bd > .list ")[0]
        for cmt_wrapper in pq(comments)(".item"):
            print pq(cmt_wrapper)("span")[0].text
            mark = pq(cmt_wrapper)("span.author").text()
            if mark:
                try:
                    print rStar.findall(mark)[0]
                except:
                    print mark

# vim: ts=4 sw=4 sts=4 expandtab