python3 beautiful爬取安居客的二手房信息

今天有一位同学找我爬取安居客的信息，安居客网站有反爬虫机制，所以不能简单用requests来做，这里用selenium来模拟获取url网页。

环境

mac, python3.7 beautifulsoup

代码

#加载packages
from selenium import webdriver
from lxml import etree
import urllib.request

#设置headers信息
url="https://shanghai.anjuke.com/fangjia/"
req=urllib.request.Request(url)
req.add_header("User-Agent",
                   "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.81 Safari/537.36")
req.add_header("Accept", "*/*")
req.add_header("Accept-Language", "zh-CN,zh;q=0.8")

#声明使用Chrome
driver = webdriver.Chrome()

#获取2010-2018年全国房价的链接
link1=[]
link1_all=[]
for i in range(2010,2019):
    link1='https://www.anjuke.com/fangjia/quanguo'+str(i)+'/'
    link1_all.append(link1)

import queue
q = queue.Queue()
for link in link1_all:
    q.put(link)

%%time
from tqdm import tqdm
from bs4 import BeautifulSoup
file=open('output.csv','w+')
name=['年份','月份','房价','涨跌幅','市','链接']

info_list=[]
prices=[]
zones=[]
tendencies=[]
links=[]
years=[]
file.write(','.join(name))
file.write('\n')
while not q.empty():
    current_url=q.get()
    print("retrive:"+current_url)
    try:
        driver.get(current_url)
    except:
        continue
        pass
    page=etree.HTML(driver.page_source)
    html_text=etree.tostring(page,encoding=str, pretty_print=True)
    bsObj=BeautifulSoup(html_text)
    info_up=bsObj.find_all('li',class_='clearfix up');
    info_nochange=bsObj.find_all('li',class_='clearfix nochange')
    info_down=bsObj.find_all('li',class_='clearfix down');
    info=[]
    info.extend(info_up)
    info.extend(info_down)
    info.extend(info_nochange)
    city_arr=bsObj.find_all('h2');
    city=city_arr[0].string.split('年')[1].split('房')[0]
#     for item in city_arr:
#         a_label=item.select('a')
#         if(not a_label):
#             city=item.string
# #             print(city)
#             break
      
    for line in info:
        item_list=[]
        tendency=line.find('em').string
        year_zone=line.find('b').string
        year,month_zone=year_zone.split('年')
        line_arr=month_zone.split('月');
        url=line.a.attrs['href']
        if('javascript' not in url):
            q.put(url)
        if(len(line_arr)<2):
            continue
        else:
            month=line_arr[0]
            zone=line_arr[1]
        price=line.find('span').string
        years.append(year)
        zones.append(zone)
        prices.append(price)
        tendencies.append(tendency)
        links.append(current_url)
        item_list.append(year)
        item_list.append(month)
        item_list.append(price)
        item_list.append(tendency)
        item_list.append(city)
        item_list.append(current_url)
        file.write(','.join(item_list))
        file.write('\n')
        
file.close()

参考文献

[1].python用BeautifulSoup用抓取a标签内所有数据.https://blog.csdn.net/suibianshen2012/article/details/62040460

[2].从零开始学网络爬虫之BeautifulSoap. https://blog.csdn.net/lxmanutd/article/details/53513103