Python爬虫项目--爬取自如网房源信息

本次爬取自如网房源信息所用到的知识点:

1. requests get请求

2. lxml解析html

3. xpath

4. mongodb存储

正文

1.分析目标站点

1. url: http://hz.ziroom.com/z/nl/z3.html?p=2 的p参数控制分页

2. get请求

2.获取单页源码

 1 # -*- coding: utf-8 -*-
 2 import requests
 3 import time
 4 from requests.exceptions import requestexception
 5 def get_one_page(page):
 6     try:
 7         url = "http://hz.ziroom.com/z/nl/z2.html?p=" + str(page)
 8         headers = {
 9             'referer':'http://hz.ziroom.com/',
10             'upgrade-insecure-requests':'1',
11             'user-agent':'mozilla/5.0(windowsnt6.3;win64;x64)applewebkit/537.36(khtml,likegecko)chrome/68.0.3440.106safari/537.36'
12         }
13         res = requests.get(url,headers=headers)
14         if res.status_code == 200:
15             print(res.text)
16     except requestexception:
17         return none
18 def main():
19     page = 1
20     get_one_page(page)
21 if __name__ == '__main__':
22     main()
23     time.sleep(1)

3.解析单页源码

1. 解析html文档, 目的: 测试xpath表达式

将获取的源码保存到当前文件夹下的"result.html"中, 然后通过xpath对其进行相应内容的提取, 当然你也可以使用某些在线工具.

 1 from lxml import etree
 2 #解析html文档
 3 html = etree.parse("./resul.html",etree.htmlparser())
 4 results = html.xpath('//ul[@id="houselist"]/li')
 5 for result in results[1:]:
 6     title = result.xpath("./div/h3/a/text()")[0][5:] if len(result.xpath("./div/h3/a/text()")[0]) >5 else ""
 7     location = result.xpath("./div/h4/a/text()")[0].replace("[","").replace("]",'')
 8     area = " ".join(result.xpath("./div/div/p[1]/span/text()")).replace(" ","",1) #使用join方法将列表中的内容以" "字符连接
 9     nearby = result.xpath("./div/div/p[2]/span/text()")[0]
10     print(title)
11     print(location)
12     print(area)
13     print(nearby)

2. 解析源代码

 1 from lxml import etree
 2 def parse_one_page(sourcehtml):
 3     '''解析单页源码'''
 4     contenttree = etree.html(sourcehtml)   #解析源代码
 5     results = contenttree.xpath('//ul[@id="houselist"]/li') #利用xpath提取相应内容
 6     for result in results[1:]:
 7         title = result.xpath("./div/h3/a/text()")[0][5:] if len(result.xpath("./div/h3/a/text()")[0]) > 5 else ""
 8         location = result.xpath("./div/h4/a/text()")[0].replace("[", "").replace("]", '')
 9         area = " ".join(result.xpath("./div/div/p[1]/span/text()")).replace(" ", "", 1)  # 使用join方法将列表中的内容以" "字符连接
10         nearby = result.xpath("./div/div/p[2]/span/text()")[0]
11         yield {
12         "title": title,
13         "location": location,
14         "area": area,
15         "nearby": nearby
16         }
17 def main():
18     page = 1
19     html = get_one_page(page)
20     print(type(html))
21     parse_one_page(html)
22     for item in parse_one_page(html):
23       print(item)
24 
25 if __name__ == '__main__':
26     main()
27     time.sleep(1)

4.获取多个页面

 1 def parse_one_page(sourcehtml):
 2     '''解析单页源码'''
 3     contenttree = etree.html(sourcehtml)   #解析源代码
 4     results = contenttree.xpath('//ul[@id="houselist"]/li') #利用xpath提取相应内容
 5     for result in results[1:]:
 6         title = result.xpath("./div/h3/a/text()")[0][5:] if len(result.xpath("./div/h3/a/text()")[0]) > 5 else ""
 7         location = result.xpath("./div/h4/a/text()")[0].replace("[", "").replace("]", '')
 8         area = " ".join(result.xpath("./div/div/p[1]/span/text()")).replace(" ", "", 1)  # 使用join方法将列表中的内容以" "字符连接
 9         #nearby = result.xpath("./div/div/p[2]/span/text()")[0].strip()这里需要加判断, 改写为下句
10         nearby = result.xpath("./div/div/p[2]/span/text()")[0].strip() if len(result.xpath("./div/div/p[2]/span/text()"))>0 else ""
11         yield {
12         "title": title,
13         "location": location,
14         "area": area,
15         "nearby": nearby
16         }
17         print(nearby)
18     #yield {"pages":pages}
19 def get_pages():
20     """得到总页数"""
21     page = 1
22     html = get_one_page(page)
23     contenttree = etree.html(html)
24     pages = int(contenttree.xpath('//div[@class="pages"]/span[2]/text()')[0].strip("共页"))
25     return pages
26 def main():
27     pages = get_pages()
28     print(pages)
29     for page in range(1,pages+1):
30         html = get_one_page(page)
31         for item in parse_one_page(html):
32             print(item)
33 
34 if __name__ == '__main__':
35     main()
36     time.sleep(1)

5. 存储到mongodb中

需确保mongodb已启动服务, 否则必然会存储失败

 1 def save_to_mongodb(result):
 2     """存储到mongodb中"""
 3     # 创建数据库连接对象, 即连接到本地
 4     client = pymongo.mongoclient(host="localhost")
 5     # 指定数据库,这里指定ziroom
 6     db = client.iroomz
 7     # 指定表的名称, 这里指定roominfo
 8     db_table = db.roominfo
 9     try:
10         #存储到数据库
11         if db_table.insert(result):
12             print("---存储到数据库成功---",result)
13     except exception:
14         print("---存储到数据库失败---",result)

6.完整代码

 1 # -*- coding: utf-8 -*-
 2 
 3 import requests
 4 import time
 5 import pymongo
 6 from lxml import etree
 7 from requests.exceptions import requestexception
 8 def get_one_page(page):
 9     '''获取单页源码'''
10     try:
11         url = "http://hz.ziroom.com/z/nl/z2.html?p=" + str(page)
12         headers = {
13             'referer':'http://hz.ziroom.com/',
14             'upgrade-insecure-requests':'1',
15             'user-agent':'mozilla/5.0(windowsnt6.3;win64;x64)applewebkit/537.36(khtml,likegecko)chrome/68.0.3440.106safari/537.36'
16         }
17         res = requests.get(url,headers=headers)
18         if res.status_code == 200:
19             return res.text
20         return none
21     except requestexception:
22         return none
23 def parse_one_page(sourcehtml):
24     '''解析单页源码'''
25     contenttree = etree.html(sourcehtml)   #解析源代码
26     results = contenttree.xpath('//ul[@id="houselist"]/li') #利用xpath提取相应内容
27     for result in results[1:]:
28         title = result.xpath("./div/h3/a/text()")[0][5:] if len(result.xpath("./div/h3/a/text()")[0]) > 5 else ""
29         location = result.xpath("./div/h4/a/text()")[0].replace("[", "").replace("]", '')
30         area = " ".join(result.xpath("./div/div/p[1]/span/text()")).replace(" ", "", 1)  # 使用join方法将列表中的内容以" "字符连接
31         #nearby = result.xpath("./div/div/p[2]/span/text()")[0].strip()这里需要加判断, 改写为下句
32         nearby = result.xpath("./div/div/p[2]/span/text()")[0].strip() if len(result.xpath("./div/div/p[2]/span/text()"))>0 else ""
33         data = {
34         "title": title,
35         "location": location,
36         "area": area,
37         "nearby": nearby
38         }
39         save_to_mongodb(data)
40     #yield {"pages":pages}
41 def get_pages():
42     """得到总页数"""
43     page = 1
44     html = get_one_page(page)
45     contenttree = etree.html(html)
46     pages = int(contenttree.xpath('//div[@class="pages"]/span[2]/text()')[0].strip("共页"))
47     return pages
48 def save_to_mongodb(result):
49     """存储到mongodb中"""
50     # 创建数据库连接对象, 即连接到本地
51     client = pymongo.mongoclient(host="localhost")
52     # 指定数据库,这里指定ziroom
53     db = client.iroomz
54     # 指定表的名称, 这里指定roominfo
55     db_table = db.roominfo
56     try:
57         #存储到数据库
58         if db_table.insert(result):
59             print("---存储到数据库成功---",result)
60     except exception:
61         print("---存储到数据库失败---",result)
62 
63 def main():
64     pages = get_pages()
65     print(pages)
66     for page in range(1,pages+1):
67         html = get_one_page(page)
68         parse_one_page(html)
69 
70 if __name__ == '__main__':
71     main()
72     time.sleep(1)

点击查看

7.最终结果

总结

1. 在第三步中xpath使用注意事项

title = result.xpath("./div/h3/a/text()")
此处的点'.'不能忘记, 它表示当前节点, 如果不加'.', '/'就表示从根节点开始选取

 2. 在第四步获取多个页面时出现索引超出范围错误

nearby = result.xpath("./div/div/p[2]/span/text()")[0].strip()

indexerror: list index out of range

   造成这种错误原因有两种:

   1) [index] index超出list范围

   2) [index] index索引内容为空

   因为这里的nearby的index是0, 排除第一种情况, 那么这里就是空行了, 加句if判断就可以解决

nearby = result.xpath("./div/div/p[2]/span/text()")[0].strip()
#改写以后:
nearby = result.xpath("./div/div/p[2]/span/text()")[0].strip() if len(result.xpath("./div/div/p[2]/span/text()"))>0 else ""

 以上主要是对爬虫过程学习的总结, 若有不对的地方, 还请指正, 谢谢!

因还在学习中, 价格部分会用到图片识别, 有待完善.