Python爬虫代码--链家杭州二手房

# -*- coding: utf-8 -*-
"""
Created on Sat Oct 14 19:21:38 2017

@author: night
"""
from bs4 import BeautifulSoup
import requests
import json
import re
import time
from collections import Counter
import pandas as pd


starturl_list = ['https://hz.lianjia.com/ershoufang/jianggan/',
                 'https://hz.lianjia.com/ershoufang/xihu/',
                 'https://hz.lianjia.com/ershoufang/xiacheng/',
                 'https://hz.lianjia.com/ershoufang/gongshu/',
                 'https://hz.lianjia.com/ershoufang/shangcheng/',
                 'https://hz.lianjia.com/ershoufang/binjiang/',
                 'https://hz.lianjia.com/ershoufang/yuhang/',
                 'https://hz.lianjia.com/ershoufang/xiaoshan/',
                 'https://hz.lianjia.com/ershoufang/xiasha/']
#获取最大页面数量
#request = requests.get('https://hz.lianjia.com/ershoufang/xiasha/')
#soup = BeautifulSoup(request.text,'html.parser')
#num = json.loads(soup.find('div',{'class':"page-box house-lst-page-box"}).get('page-data'))['totalPage']

#get_pageurls(starturl_list[4])


#获取二手房每一页的url

def get_pageurls(url):
    request = requests.get(url)
    soup = BeautifulSoup(request.text,'html.parser')
    totalnum = json.loads(soup.find('div',{'class':"page-box house-lst-page-box"}).get('page-data'))['totalPage']+1
#    pageurls_list = []
    pageurls_list.append(url)
    for num in range(2,totalnum):
        newurl = url + 'pg{}/'.format(num)
        pageurls_list.append(newurl)
#    return pageurls_list

#get_pageurls(starturl_list[0])


#获取每一页的二手房url
def get_eachurls(pageurl):  
    request = requests.get(pageurl)
    soup = BeautifulSoup(request.text,'html.parser')
    for i in soup.find_all('li',{'class':'clear'}):
#        if i.a.get('href') not in eachurl_list:
        eachurl_list.append(i.a.get('href'))







#获取二手房信息
allinfo = []
def houseinformation(houseurl):
    global allinfo
    request = requests.get(houseurl)
    soup = BeautifulSoup(request.text,'html.parser')
    info = {}
    list = soup.find_all('script')
#    try:
    house_info_pre = list[26].text.encode('utf-8')
    pattern_position = '''resblockPosition.*\''''
    pos = re.search(re.compile(pattern_position),house_info_pre).group(0).split("'")[1]
    longi = pos.split(',')[0]
    lati = pos.split(',')[1]
    try:
        info[u'经度'] = longi
    except:
        info[u'经度'] = None
    try:
        info[u'纬度'] = lati
    except:
        info[u'纬度'] = None
#获取标题、单价、总价
    try:
        info[u'标题'] = unicode(soup.find('div',{'class':'title'}).contents[1].get('title'))
    except:
        info[u'标题'] = None       
    try:
        info[u'副标题'] = unicode(soup.find('div',{'class':'title'}).contents[3].string)
    except:
        info[u'副标题'] = None        
    try:
        info[u'总价'] = soup.find('div',{'class':'price'}).find('span',{'class':'total'}).string + soup.find('div',{'class':'price'}).find('span',{'class':'unit'}).string
    except:
        info[u'总价'] = None      
    try:
        info[u'单价'] = soup.find('span',{'class':'unitPriceValue'}).get_text()
    except:
        info[u'单价'] = None
#获取基本属性
    base = soup.find('div',{'class':'introContent'}).contents[1].ul.find_all('li')
    try:
        info[u'房屋类型'] = unicode(base[0].contents[1].string)
    except:
        info[u'房屋类型'] = None
    try:
        info[u'所在楼层'] = unicode(base[1].contents[1].string)
    except:
        info[u'所在楼层'] = None
    try:
        info[u'建筑面积'] = unicode(base[2].contents[1].string)
    except:
        info[u'建筑面积'] = None
    try:
        info[u'户型结构'] = unicode(base[3].contents[1].string)
    except:
        info[u'户型结构'] = None
    try:
        info[u'套内面积'] = unicode(base[4].contents[1].string)
    except:
        info[u'套内面积'] = None
    try:
        info[u'建筑类型'] = unicode(base[5].contents[1].string)
    except:
        info[u'建筑类型'] = None
    try:
        info[u'房屋朝向'] = unicode(base[6].contents[1].string)
    except:
        info[u'房屋朝向'] = None
    try:
        info[u'建筑结构'] = unicode(base[7].contents[1].string)
    except:
        info[u'建筑结构'] = None
    try:
        info[u'配备电梯'] = unicode(base[8].contents[1].string)
    except:
        info[u'配备电梯'] = None
#获取交易属性
    trans = soup.find('div',{'class':'introContent'}).contents[3].ul.find_all('li')
    try:
        info[u'挂牌时间'] = unicode(trans[0].contents[1].string)
    except:
        info[u'挂牌时间'] = None
    try:
        info[u'交易属性'] = unicode(trans[1].contents[1].string)
    except:
        info[u'交易属性'] = None
    try:
        info[u'上次交易'] = unicode(trans[2].contents[1].string)
    except:
        info[u'上次交易'] = None
    try:
        info[u'房屋用途'] = unicode(trans[3].contents[1].string)
    except:
        info[u'房屋用途'] = None
    try:
        info[u'房屋年限'] = unicode(trans[4].contents[1].string)
    except:
        info[u'房屋年限'] = None
    try:
        info[u'产权所属'] = unicode(trans[5].contents[1].string)
    except:
        info[u'产权所属'] = None
    try:
        info[u'抵押信息'] = unicode(trans[6].contents[1].string)
    except:
        info[u'抵押信息'] = None
    try:
        info[u'房本备件'] = unicode(trans[7].contents[1].string)
    except:
        info[u'房本备件'] = None
    try:
        info[u'房源编码'] = unicode(trans[8].contents[1].string)
    except:
        info[u'房源编码'] = None
#获取小区信息
    try:
        info[u'小区名称'] = re.search("resblockName:'(.*?)'",request.text).group(1)
    except:
        info[u'小区名称'] = None
    try:
        info[u'网址'] = houseurl      
    except:
        info[u'网址'] = None
    allinfo.append(info)

    

#########################################################################
#get_pageurls(starturl_list[0])
pageurls_list = []
eachurl_list = []

get_pageurls(starturl_list[7])

n = 1
for i in pageurls_list:
    get_eachurls(i)
    print '储存第{}页网址'.format(n)
    n+=1

for i in range(len(eachurl_list)):    
    houseinformation(eachurl_list[i])
    print u'抓取第{}条信息，房源名称为:{}'.format(i+1,allinfo[i][u'标题'])
#    time.sleep(0.5)

df = pd.DataFrame(allinfo)
df.to_csv(r"C:\test\data_lianjia_shangcheng.csv",encoding='gb18030')