#!/usr/bin/env python
#coding: utf-8
import urllib2,urllib
import re
import sqlite3
import os,sys
import time

db = sqlite3.connect('/var/www/simplecd.old/verycd.sqlite3.db')
db.text_factory = str

path = os.path.dirname(os.path.realpath(sys.argv[0]))

#proxies = {'http':'http://vpn.unswbbs.com:3128'}
proxies = {'http':'http://localhost:3128'}
#proxies = {'http':'http://obmem.com:30000'}
proxy_support = urllib2.ProxyHandler(proxies)
opener = urllib2.build_opener(proxy_support, urllib2.HTTPHandler)
urllib2.install_opener(opener)

def clear_idcache(id):
	#clear idcache:
	try:
		l1 = str(long(id)%10)
		l2 = str(long(id)/10%100)
		l3 = str(long(id)/1000)
		cachefile = path + '/idcache/%s/%s/%s.html'%(l1,l2,l3)
		if os.path.exists(cachefile):
			os.remove(cachefile)
	except:
		pass

def doubanapi(id,conn=db):
	apikey = '047ca5407fe8b8a40d75efc17fba29ec'
	title = '幸运星'
	cat = 'movie'
	url = 'http://api.douban.com/%s/subjects?q=%s&start-index=1&max-results=1&alt=json&apikey=%s'%(cat,title,apikey)
	print url
	import json
	dbj = urllib2.urlopen(url).read()
	dbj = json.loads(dbj)
	for x in dbj['entry'][0].keys():
		print x,dbj['entry'][0][x]
	print dbj['entry'][0]['title']['$t']

def douban(id,conn=db):
	print 'updating',id
	id = long(id)

	c = conn.cursor()
	c.execute('select title,category1,brief,content from verycd where verycdid=?',(id,))
	(title,cat,brief,content) = c.fetchone()
	c.close()

	dbed = re.compile('Douban\d+').search(brief)
	imdbid = re.compile(r'http://www\.imdb\.com/title/(tt\d+)').search(content)
	if imdbid:
		imdbid = imdbid.group(1)
	print imdbid
	if (cat not in ['电影','音乐','剧集','动漫','综艺']) or dbed:
		print '...not movie or already updated'
		clear_idcache(id)
		return

	if title.find("《") == -1:
		name = title
	else:
		name = title[ title.find("《")+3: title.find("》") ]

	if cat == '音乐':
		durl = 'http://www.douban.com/subject_search?search_text='+name+'&cat=1003'
	elif imdbid:
		durl = 'http://www.douban.com/subject_search?search_text='+imdbid+'&cat=1002'
	else:
		durl = 'http://www.douban.com/subject_search?search_text='+name+'&cat=1002'
	
	durl = durl.replace(' ','+')
	print '...fetching douban page',name
	print durl
	try:
		ds = urllib2.urlopen(durl).read()
	except:
		print '...fetch error'
		return

	dbscore = re.compile(r'href=".*?/subject/(\d+)/".*?"rating_nums">([0-9.]+)<.*?(\d+)',re.DOTALL).search(ds)
	if not dbscore:
		print '...not found'
		return
	ins = brief.find('votes)')
	dbb = 'Douban%s:%s(%s votes)'%(dbscore.group(1),dbscore.group(2),dbscore.group(3))
	if ins == -1:
		brief = dbb + brief
	else:
		brief = brief[:ins+6]+dbb+brief[ins+6:]

	c = conn.cursor()
	try:
		c.execute('update verycd set brief=? where verycdid=?',(brief,id))
		c.close()
		conn.commit()
	except:
		c.close()
		conn.commit()
		print '...db update failure'
		return

	print '...updated',dbb
	clear_idcache(id)
	return True


def updateall():
	c = db.cursor()
	c.execute('select verycdid from verycd where category1="电影" and verycdid>6000 order by updtime desc')
	ids = c.fetchall()
	for id in ids:
		update = douban(id[0])
		if update:
			time.sleep(3)
		

def update(num=20):
	c = db.cursor()
	c.execute('select verycdid from verycd order by updtime desc limit ?',(num,))
	ids = c.fetchall()
	for id in ids:
		douban(id[0])

if __name__ == '__main__':
	update(20)
	#douban(2802148)
	#updateall()