1.用于爬取58上的租房信息,限成都,其他地方的,可以把网址改改;
2.这个爬虫有一点问题,就是没用多线程,因为我用了之后总是会报: 'module' object has no attribute '_strptime'这个奇怪的错误,挣扎了许久,放弃;
如有大神看到这篇帖子,希望可以指点一二,不胜感激,谢谢。
3.我本来打算做成EXE文件的,但是在中文处理方面总是乱码,需要进一步研究;
以下为代码:
#!/usr/bin/python
# -*- encoding:utf-8 -*-
import requests
from bs4 import BeautifulSoup
from multiprocessing.dummy import Pool as ThreadPool
import re
import datetime
import sys
# from datetime import datetime
reload(sys)
sys.setdefaultencoding('utf-8')
#得到soup,因后文通用,直接放这儿就行了
def urlBS(url):
response=requests.get(url)
response.encoding = 'utf-8'
soup = BeautifulSoup(response.text,"lxml")
return soup
#通过交互的方式让搜索人输入想要的房屋条件,不输的话有默认值
def get_source_url():
base_url='http://cd.58.com/zufang/' #首先,锁定为整租:/zufang/,然后限定为个人房源:/0/,0为个人,1为经纪人
# real_url='http://cd.58.com/zufang/?isreal=true'
try:
source_key=input('请按序号输入你想要的房屋来源,1为不限,2为个人房源,3为经纪人(默认为2个人房源):\n')
except:
source_key=2
source_from={1:'',2:'0/',3:'1/'} # 4:'?isreal=true',4为诚信房源专区'
try:
price_min=str(input('请输入你期望的价格下限(不输默认为500):\n'))
except:
price_min='500'
try:
price_max=str(input('请输入你期望的价格上限(不输默认为1000):\n'))
except:
price_max='1000'
price='minprice='+price_min+'_'+price_max
try:
room_key=input('请输入你想要的房子间数:0为不限,1为1室,2为2室,3为3室,4为4室,5为4室以上(不输默认为1室):\n')
except:
room_key=1
room_num={0:'',1:'j1/',2:'j2/',3:'j3/',4:'j4/',5:'j5/'}
key_words=raw_input('请输入你想要搜索的其他关键词,如小区名称,地铁位置等(不输默认为空):\n')
source_url=base_url+source_from[source_key]+room_num[room_key]+'?'+price+'&key='+key_words
# print new_url
return source_url
# new_url='http://cd.58.com/zufang/0/j1/?minprice=600_800&PGTID=0d300008-0006-6cd9-6ba7-a7672ec996c3&ClickID=3'
#找到下一页的地址,因为58的网站很坑,它并没有显示共多少页,所以只能通过爬取他的下一页对应的href来得到下一页的链接
#但是,更坑的是,他的页面进去后第一次的下一页按了后和当前页是一样的,所以我就在确定有下一页的情况下,直接用当前页+1得到下一页的url
def get_new_list(source_url):
new_url=source_url
new_url_list=[new_url]
while True:
soup=urlBS(new_url)
cp=re.compile(r'/pn(.)/')
finder=soup.find('a',{'class':'next'})
if finder:
next_url=finder['href']
now_page=cp.findall(source_url)
next_page='http://cd.58.com'+next_url
if now_page:
now_page=now_page[0]
newpage=str(int(now_page)+1)
new_page=cp.sub(newpage,next_page)
else:
now_page='1'
newpage='2'
new_page='http://cd.58.com'+next_url
new_url=new_page
else:
new_page=''
break
# else:
# print 'dont have next page'
# print new_url
if len(new_url_list)==1:
new_url_list.append(new_url)
elif new_page==new_url_list[-1]:
break
else:
new_url_list.append(new_url)
# print new_url_list
return new_url_list
#得到房屋信息页的链接
def get_house_url(new_url):
soup = urlBS(new_url)
href_list=soup.select('div[class="img_list"] a')
house_url_list=[]
for each in href_list:
href=each['href']
#print href
house_url_list.append(href)
return house_url_list
#爬取房屋信息,同时不要那些骗子的信息,以及一个月前更新的信息
def house_info(house_url):
# house_url='http://cd.58.com/zufang/26364594760504x.shtml?version=A&psid=162154127192148068945806804&entinfo=26364594760504_0'
# print house_url
soup=urlBS(house_url)
try:
tel=soup.find('span',{'class':'tel-num tel-num-geren pl30 f30'}).text #个人房源
except:
tel=soup.find('span',{'class':'tel-num pl30 f30'}).text #中介
match_tel=re.search(r'^1\d{5}.*',tel) #排除所有电话号码以0开始的人,即留固定电话的人,因为我们认为,固定房源的人是不会留固定电话的
situation=soup.find('div',{'class':'description-content'}).text.strip()
# print situation
match_si=re.search(u'(我是房东|男士勿扰|男生勿扰|限女生|微信|男士|男性|男生|女性|女的|姐妹|"+")',situation)
#更新时间
update_time=soup.find('span',{'class':'pl10'}).text
update_date = datetime.datetime.strptime(update_time.replace('更新时间:',''), "%Y-%m-%d").date()
thirtyDayAgo=datetime.date.today() + datetime.timedelta(days=-30)
day_line=(update_date - thirtyDayAgo).days
if not match_tel: #认为隐藏了电话号码的,电话号码以0开始的,都是骗子,不要他
# print '电话号码有问题,骗子'
pass
elif match_si: #认为含有某些字的全部为骗子,把这些排除掉
# print '内容有问题,骗子'
pass
elif day_line<0: #取近30天更新的数据,时间太长了的数据没啥意义
# print '已经是一个月之前的消息了'
pass
else:
print house_url
print situation
print tel
#标题
title=soup.find('h1',{'class':'main-title font-heiti'}).text
print title
#价格
p=re.compile(r'\n|\t|\r| ')
rent_price=soup.find('i',{'class':'ncolor'}).text
price=p.sub('',rent_price)
print price
#房屋大小
house=soup.find_all('li',{'class':'house-primary-content-li clearfix'})
house_content=p.sub('',house[0].text)
print house_content
#小区
try:
house_Community=p.sub('',house[1].text)
except:
house_Community=''
print house_Community
#位置
try:
house_place=p.sub('',house[2].text)
except:
house_place=''
print house_place
#设施
try:
facility=soup.find('li',{'class':'house-primary-content-li clearfix person-config'})
facility=p.sub('',facility.text)
except:
facility=''
print facility
#联系人
contact=soup.find('li',{'class':'house-primary-content-li clearfix person-contact'}).text
contact=p.sub('',contact)
print contact
print update_time+'\n\n\n'
# a=[house_url,price,house_content,house_Community,house_place,title,situation,facility]
f.write('----------------------------------------------------------------------------------\n')
f.write(house_url+'\n'+price+'\n'+house_content+'\n'+house_Community+'\n'+house_place+'\n'+title+'\n'+situation+'\n'+facility+'\n\n')
if __name__=='__main__':
source_url=get_source_url()
print source_url
# source_url='http://cd.58.com/zufang/0/?minprice=500_1500&key=四河'
get_new_list=get_new_list(source_url)
# print get_new_list
f=open("house_rent.txt", "w") #先清空,然后再打开,再写入,写入时的方式是a(追加)
# f.truncate()
# f.close()
#
# f=open("house_rent.text", "a")
print '正在下载,请稍候。。。\n\n'
# pool = ThreadPool(4)
for new_url in get_new_list:
new_url=new_url.encode('utf-8').decode('utf-8')
# print new_url
house_url_list=get_house_url(new_url)
# print house_url_list
for each in house_url_list: #本来打算使用多线程,但是总是会报: 'module' object has no attribute '_strptime'这个奇怪的错误,挣扎了许久,放弃
house_info(each)
# results = pool.map(house_info, house_url_list)
# pool.close()
# pool.join()
f.close()