Python2 Python3 爬取赶集网租房信息,带源码分析

*之前偶然看了某个腾讯公开课的视频,写的爬取赶集网的租房信息,这几天突然想起来,于是自己分析了一下赶集网的信息,然后自己写了一遍,写完又用用Python3重写了一遍.之中也遇见了少许的坑.记一下.算是一个总结.*

python2 爬取赶集网租房信息与网站分析

1. 分析目标网站url
2. 寻找目标标签
3. 获取,并写入csv文件

#-*- coding: utf-8 -*-
from bs4 import BeautifulSoup
from urlparse import urljoin
import requests
import csv

URL = 'http://jn.ganji.com/fang1/o{page}p{price}/'
# 首先最基本的是 jn,ganji.com/fang1 其中jn为济南,也就是我的城市,默认登录后为此
# 而fang1 位租房信息 fang5 为二手房信息,zhaopin 为招聘模块等,我们这次只查找fang1
# 不过这个链接还可以更复杂 
#比如http://jn.ganji.com/fang1/tianqiao/h1o1p1/ 或者
#  http://jn.ganji.com/fang1/tianqiao/b1000e1577/
#    其中h:房型,o页面,p价格区间,其中h,p后的数字与网站相应菜单的排列顺序相对应 
# 而s与e则为对应的自己输入的价格区间
#                          h: house o:page p:price
# jn jinan  fang1 zufang tiaoqiao:tianqiaoqu b:begin 1000  e:end start 1755

ADDR = 'http://bj.ganji.com/'
start_page =1
end_page = 5
price =1

# 注意wb格式打开写入可能会导致csv文件每次写入前面多一个空格
# 对此你可以参考这篇文章:http://blog.csdn.net/pfm685757/article/details/47806469
with open('info.csv','wb') as f :
    csv_writer = csv.writer(f,delimiter=',')

    print 'starting'
    while start_page<end_page:
        start_page+=1
        # 通过分析标签可知我们要获取的标签信息必须要通过多个class确认才能保证唯一性
        # 之后是获取信息的具体设置
        print 'get{0}'.format(URL.format(page = start_page,price=price))
        response = requests.get(URL.format(page = start_page,price=price))
        html=BeautifulSoup(response.text,'html.parser')
        house_list = html.select('.f-list > .f-list-item > .f-list-item-wrap')
        #check house_list
        if not house_list:
            print 'No house_list'
            break
        for house in house_list:
            house_title = house.select('.title > a')[0].string.encode('utf-8')
            house_addr = house.select('.address > .area > a')[-1].string.encode('utf-8')
            house_price = house.select('.info > .price > .num')[0].string.encode('utf-8')
            house_url = urljoin(ADDR,house.select('.title > a ')[0]['href'])
            # 写入csv文件
            csv_writer.writerow([house_title,house_addr,house_price,house_url])
    print 'ending'

Python3 爬取赶集网i租房信息

要注意的点

1. urlparse.urljoin 改为urllib.urlparse.urljoin

# python2
from urlparse import urljoin
# Python3
from urllib.parse  import urljoin

1. Python3中csv对bytes和str两种类型进行了严格区分,open的写入格式应该进行改变wb->w
2. 设置utf8编码格式

with open('info.csv','w',encoding='utf8') as f :
    csv_writer = csv.writer(f,delimiter=',')

完整代码如下

#-*- coding: utf-8 -*-
from bs4 import BeautifulSoup
from urllib.parse  import urljoin
import requests
import csv

URL = 'http://jn.ganji.com/fang1/o{page}p{price}/'
#                                h: house o:page p:price
#  http://jn.ganji.com/fang1/tianqiao/b1000e1577/
# jn jinan  fang1 zufang tiaoqiao:tianqiaoqu b:begin 1000  e:end start 1755
# fang5 为二手房 zhipin 为 招聘 赶集网的url划分的都很简单,时间充足完全可以获取非常多的信息

ADDR = 'http://bj.ganji.com/'
start_page =1
end_page = 5

price =1

'''
URL = 'http://jn.ganji.com/fang1/h{huxing}o{page}b{beginPrice}e{endPrice}/'
# 选择户型为h1-h5
# 输入价位为 begin or end
price='b1000e2000'

# 户型为

'''
# 默认为utf8打开,否则会以默认编码GBK写入
with open('info.csv','w',encoding='utf8') as f :
    csv_writer = csv.writer(f,delimiter=',')

    print('starting')
    while start_page<end_page:
        start_page+=1
        print('get{0}'.format(URL.format(page = start_page,price=price)))
        response = requests.get(URL.format(page = start_page,price=price))
        html=BeautifulSoup(response.text,'html.parser')
        house_list = html.select('.f-list > .f-list-item > .f-list-item-wrap')
        #check house_list
        if not house_list:
            print('No house_list')
            break
        for house in house_list:
            house_title = house.select('.title > a')[0].string
            house_addr = house.select('.address > .area > a')[-1].string
            house_price = house.select('.info > .price > .num')[0].string
            house_url = urljoin(ADDR, house.select('.title > a ')[0]['href'])
            csv_writer.writerow([house_title,house_addr,house_price,house_url])
    print('ending')

最后的csv文件展示一下: