# -*- coding: UTF-8 -*-
import mysql.connector as db
import time
import json
import re
import sys

def get_userName(html):
	ptn1 = '<div class="userName"><strong><a href="/(?P<account>[^"]+)" title="[^"]+">(?P<nickName>.+?)</a><a href="[^"]+" title="(?P<certification>[^"]+)" target="_blank" class="[^"]+"></a><a href="[^"]+" class="[^"]+" user="[^"]+"><em>(?P<level>\d+)</em></a></strong></div>'
	group = re.search(ptn1, html, re.I | re.S)
	if group is not None:
		return group.groupdict()
	ptn2 = '<div class="userName"><strong><a href="/(?P<account>[^"]+)" title="[^"]+">(?P<nickName>.+?)</a><a href="[^"]+" class="[^"]+" user="[^"]+"><em>(?P<level>\d+)</em></a></strong></div>'
	group = re.search(ptn2, html, re.I | re.S)
	if group is None:
		return None
	match = group.groupdict()
	match['certification'] = ''
	return match
def get_userNums(html):
	ptn1 = '<div class="userNums"><span><span class="cNote" >(?P<address>.+?)</span><a href="[^"]+">听众<strong id="[^"]+">(?P<follower>\d+)</strong>人</a></span><span><a href="[^"]+">收听<strong>(?P<following>\d+)</strong>人</a></span></div>'
	group = re.search(ptn1, html, re.I | re.S)
	if group is not None:
		return group.groupdict()
	ptn2 = '<div class="userNums"><span><span class="cNote" style="margin-right:0px" ></span><a href="[^"]+">听众<strong id="[^"]+">(?P<follower>\d+)</strong>人</a></span><span><a href="[^"]+">收听<strong>(?P<following>\d+)</strong>人</a></span></div>'
	group = re.search(ptn2, html, re.I | re.S)
	if group is None:
		return None
	match = group.groupdict()
	match['address'] = ''
	return match
def get_pubInfo(html):
	ptn = '<div class="pubInfo"><p class="pubTime">(?P<weiboTime>.+?)(?P<weiboFrom>\<.+?)</p><p><a href="[^"]+">(?P<lastWeibo>.*?)</a></p></div>'
	group = re.search(ptn, html, re.I | re.S)
	if group is None:
		return None
	return group.groupdict()

def get_source(html):
	ptn = '<div class="cNote source">通过<a boss="[^"]+" href="[^"]+">(?P<weiboClient>.+?)</a>收听</div>'
	group = re.search(ptn, html, re.I | re.S)
	if group is None:
		return None
	return group.groupdict()['weiboClient']

def parse_info(infos):
	datas = []
	for info in infos:
		data = {'account' : '','nick_name' : '','address' : '','follower' : 0,'following' : 0,'last_weibo' : '','level' : 1,'weibo_from' : '','weibo_client' : '','weibo_time' : '','certification' : '',}
		userName = get_userName(info)
		userNums = get_userNums(info)
		pubInfo = get_pubInfo(info)
		if userName is None:
			continue
		data['account'] = userName['account']
		data['nick_name'] = userName['nickName']
		data['level'] = int(userName['level'])
		data['certification'] = userName['certification']
		if userNums is not None:
			data['address'] = userNums['address']
			data['follower'] = int(userNums['follower'])
			data['following'] = int(userNums['following'])
		if pubInfo is not None:
			data['last_weibo'] = pubInfo['lastWeibo']
			data['weibo_from'] = re.sub('<[^>]+>', '', pubInfo['weiboFrom'])
			data['weibo_time'] = pubInfo['weiboTime']
		data['weibo_client'] = get_source(info)
		datas.append(data)
	return datas

if __name__ == '__main__':
	connect = db.connect(user='root',db='collection',password='',host="127.0.0.1")
	cursor = connect.cursor()
	cursor.execute("SET SQL_MODE = 'TRADITIONAL'")
	
	sqlInsertFollower = "INSERT IGNORE INTO `follower`(`account`, `nick_name`, `address`, `follower`, `following`, `last_weibo`, `level`, `weibo_from`, `weibo_client`, `weibo_time`, `certification`, `following_account`) VALUES(%s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s); ";
	fJson = open('dir.log', 'r')
	while True:
		jsonFile = fJson.readline().strip()
		if jsonFile == '':
			break
		fp = open("follower\\%s" % jsonFile, 'r')
		following_account = re.sub('^follower_(?P<account>.+?)\.json$', '\g<account>', jsonFile)
		print("开始保存 %s 的听众！" % following_account)
		fp.readline()
		while True:
			line = fp.readline()
			if line == '':
				break
			if line.strip() == '':
				continue
			rs = json.loads(line)
			group = re.search('<ul class="LC">(?P<LC>.+?)</ul>', rs['info'], re.S | re.I)
			if group is None:
				break
			infos = re.sub('>\s+<', '><', group.groupdict()['LC'].replace('\n', '').replace('\r', '')).strip().replace('</li><li','</li>\r\n<li').split('\r\n')
			for data in  parse_info(infos):
				cursor.execute(sqlInsertFollower, (data['account'], data['nick_name'], data['address'], data['follower'], data['following'], data['last_weibo'], data['level'], data['weibo_from'], data['weibo_client'], data['weibo_time'], data['certification'], following_account))
			connect.commit()
		fp.close()
	fJson.close()
	connect.commit()
	cursor.close()
	connect.close()