【Python爬虫系列】Python 爬取上海链家二手房数据

先上代码，后续再详细解释

Python 爬虫代码：

# -*- coding:utf-8 -*-
############################################################################
# 程序：上海链家网爬虫
# 功能：抓取上海链家二手房在售、成交数据 ，大约各5万记录；小区2万多个
# 创建时间：2016/11/10
# 更新历史：2016/11/26
#          2016.11.27:增加地铁找房；更新区域参数；拆分模块 ，以便于单独调用
#          2016.12.06 加入多线程处理
#          2016.12.28 增加按照面积、户型、总价等明细条件，以便扩大爬数范围
# 使用库：requests、BeautifulSoup4、MySQLdb
# 作者：yuzhucu
#############################################################################
import requests
from bs4 import BeautifulSoup
import time
import MySQLdb
import urllib
import urllib2
import json
import cookielib
import re
import zlib
from threading import Thread
from Queue import Queue
from time import sleep
#登录，不登录不能爬取三个月之内的数据
#import LianJiaLogIn
# 获取当前时间

def getCurrentTime():
    return time.strftime('[%Y-%m-%d %H:%M:%S]', time.localtime(time.time()))

def getFangCondition():
    result = []
    for a in range(1,9):#面积
        for l in  range(1,7):#户型
            for p in  range(1,9):#总价
                cond = {}
                cond['url']='a'+str(a)+'l'+str(l)+'p'+str(p)
                cond['a']='a'+str(a)
                cond['l']='l'+str(l)
                cond['p']='p'+str(p)
                #print cond['url']
                result.append(cond)
    return result

def getFangTransCondition():
    result = []
    for a in range(1,9):#面积
        for l in  range(1,7):#户型
            cond = {}
            cond['url']='a'+str(a)+'l'+str(l)
            cond['a']='a'+str(a)
            cond['l']='l'+str(l)
            #print  cond['url']
            result.append(cond)
    return result


def getURL(url, tries_num=50, sleep_time=0, time_out=10):
    headers = {'content-type': 'application/json',
               'User-Agent': 'Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:22.0) Gecko/20100101 Firefox/22.0'}
    proxies ={ "http": "10.11.12.13:8080", "https": "http://10.11.12.13:8080" }#替换为可用代理IP
    sleep_time_p = sleep_time
    time_out_p = time_out
    tries_num_p = tries_num
    try:
        res = requests.Session()
        if isproxy==1:
            res = requests.get(url, headers=headers, timeout=time_out,proxies=proxies)
        else:
            res = requests.get(url, headers=headers, timeout=time_out)
        res.raise_for_status()  # 如果响应状态码不是 200，就主动抛出异常
    except requests.RequestException as e:
        sleep_time_p = sleep_time_p + 10
        time_out_p = time_out_p + 10
        tries_num_p = tries_num_p -1
        # 设置重试次数，最大timeout 时间和 最长休眠时间
        #print tries_num_p
        if tries_num_p >0 :
            time.sleep(sleep_time_p)
            print getCurrentTime(), url, 'URL Connection Error: 第', max_retry- tries_num_p, u'次 Retry Connection', e
            res = getURL(url, tries_num_p, sleep_time_p, time_out_p)
            if res.status_code == 200:
               print getCurrentTime(), url, 'URL Connection Success: 共尝试',  max_retry- tries_num_p, u'次', ',sleep_time:', sleep_time_p, ',time_out:', time_out_p
            else:
               print getCurrentTime(), url, 'URL Connection Error: 共尝试',  max_retry- tries_num_p, u'次', ',sleep_time:', sleep_time_p, ',time_out:', time_out_p
               pass

    return res

def getXiaoquList(fang_url):
    result = {}
    base_url = 'http://sh.lianjia.com'
    # res=requests.get(fang_url)
    res = getURL(fang_url)
    res.encoding = 'utf-8'
    soup = BeautifulSoup(res.text, 'html.parser')
    for fang in soup.select('.info-panel'):
        if (len(fang) > 0):
            try:
                result['xiaoqu_key'] = fang.select('h2')[0].a['key'].strip().lstrip().strip(" ")
                result['xiaoqu_name'] = fang.select('h2')[0].text.strip()
                result['xiaoqu_url'] = base_url + fang.select('h2')[0].a['href'].strip()
                result['quyu'] = fang.select('.con')[0].contents[1].text.strip()
                result['bankuai'] = fang.select('.con')[0].contents[3].text.strip()
                result['price'] = fang.select('.price')[0].span.text.strip() + fang.select('.price')[0].contents[2].strip()
                result['age'] = ''
                result['subway'] = ''
                result['onsale_num'] = ''
                result['fang_url'] = ''
                if len(fang.select('.con')[0].contents) >= 5:
                    result['age'] = fang.select('.con')[0].contents[-1].string.strip()
                if len(fang.select('.fang-subway-ex')) > 0:
                    result['subway'] = fang.select('.fang-subway-ex')[0].text.strip()
                if len(fang.select('.square')) > 0:
                    result['onsale_num'] = fang.select('.square')[0].a.text.strip()
                if len(fang.select('.square')) > 0:
                    result['fang_url'] = base_url + fang.select('.square')[0].a['href'].strip()
                    getLianjiaList(result['fang_url'])
                result['updated_date']=time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
                mySQL.insertData('lianjia_fang_xiaoqu', result)
                print getCurrentTime(), u'小区：', result['xiaoqu_key'], result['xiaoqu_name'], result['age'], result[ 'quyu'],result['bankuai'], \
                                                     result['subway'], result['xiaoqu_url'], result['price'], result['onsale_num'], result['fang_url']
                getLianjiaList(result['fang_url'])
            except Exception, e:
                print  getCurrentTime(), u"Exception:%d: %s" % (e.args[0], e.args[1])
    return result

def getLianjiaList(fang_url):
    result = {}
    base_url = 'http://sh.lianjia.com'
    # res=requests.get(fang_url)
    res = getURL(fang_url)
    res.encoding = 'utf-8'
    soup = BeautifulSoup(res.text, 'html.parser')
    for fang in soup.select('.info-panel'):
        if (len(fang) > 0):
            result['fang_key'] = fang.select('h2')[0].a['key'].strip()
            result['fang_desc'] = fang.select('h2')[0].text.strip()
            result['fang_url'] = base_url + fang.select('h2')[0].a['href'].strip()
            result['price'] = fang.select('.price')[0].text.strip()
            result['price_pre'] = fang.select('.price-pre')[0].text.strip()
            result['xiaoqu