本次爬取自如網房源信息所用到的知識點:

1. requests get請求

2. lxml解析html

3. Xpath

4. MongoDB存儲

正文

1.分析目標站點

1. url: http://hz.ziroom.com/z/nl/z3.html?p=2 的p參數控制分頁

2. get請求

2.獲取單頁源碼

 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.解析單頁源碼

1. 解析html文檔, 目的: 測試XPath表達式

將獲取的源碼保存到當前文件夾下的"result.html"中, 然後通過XPath對其進行相應內容的提取, 當然你也可以使用某些在線工具.

 1 from lxml import etree
 2 #解析html文檔
 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方法將列表中的內容以" "字符連接
 9     nearby = result.xpath("./div/div/p[2]/span/text()")[0]
10     print(title)
11     print(location)
12     print(area)
13     print(nearby)

2. 解析源代碼

 1 from lxml import etree
 2 def parse_one_page(sourcehtml):
 3     ‘‘‘解析單頁源碼‘‘‘
 4     contentTree = etree.HTML(sourcehtml)   #解析源代碼
 5     results = contentTree.xpath(‘//ul[@id="houseList"]/li‘) #利用XPath提取相應內容
 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方法將列表中的內容以" "字符連接
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.獲取多個頁面

 1 def parse_one_page(sourcehtml):
 2     ‘‘‘解析單頁源碼‘‘‘
 3     contentTree = etree.HTML(sourcehtml)   #解析源代碼
 4     results = contentTree.xpath(‘//ul[@id="houseList"]/li‘) #利用XPath提取相應內容
 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方法將列表中的內容以" "字符連接
 9         #nearby = result.xpath("./div/div/p[2]/span/text()")[0].strip()這裏需要加判斷, 改寫為下句
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     """得到總頁數"""
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("共頁"))
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. 存儲到MongoDB中

需確保MongoDB已啟動服務, 否則必然會存儲失敗

 1 def save_to_mongodb(result):
 2     """存儲到MongoDB中"""
 3     # 創建數據庫連接對象, 即連接到本地
 4     client = pymongo.MongoClient(host="localhost")
 5     # 指定數據庫,這裏指定ziroom
 6     db = client.iroomz
 7     # 指定表的名稱, 這裏指定roominfo
 8     db_table = db.roominfo
 9     try:
10         #存儲到數據庫
11         if db_table.insert(result):
12             print("---存儲到數據庫成功---",result)
13     except Exception:
14         print("---存儲到數據庫失敗---",result)

6.完整代碼

 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     ‘‘‘獲取單頁源碼‘‘‘
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     ‘‘‘解析單頁源碼‘‘‘
25     contentTree = etree.HTML(sourcehtml)   #解析源代碼
26     results = contentTree.xpath(‘//ul[@id="houseList"]/li‘) #利用XPath提取相應內容
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方法將列表中的內容以" "字符連接
31         #nearby = result.xpath("./div/div/p[2]/span/text()")[0].strip()這裏需要加判斷, 改寫為下句
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     """得到總頁數"""
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("共頁"))
47     return pages
48 def save_to_mongodb(result):
49     """存儲到MongoDB中"""
50     # 創建數據庫連接對象, 即連接到本地
51     client = pymongo.MongoClient(host="localhost")
52     # 指定數據庫,這裏指定ziroom
53     db = client.iroomz
54     # 指定表的名稱, 這裏指定roominfo
55     db_table = db.roominfo
56     try:
57         #存儲到數據庫
58         if db_table.insert(result):
59             print("---存儲到數據庫成功---",result)
60     except Exception:
61         print("---存儲到數據庫失敗---",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)

7.最終結果

總結

1. 在第三步中XPath使用註意事項

title = result.xpath("./div/h3/a/text()")
此處的點‘.‘不能忘記, 它表示當前節點, 如果不加‘.‘, ‘/‘就表示從根節點開始選取

2. 在第四步獲取多個頁面時出現索引超出範圍錯誤

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

IndexError: list index out of range

造成這種錯誤原因有兩種:

1) [index] index超出list範圍

2) [index] index索引內容為空

因為這裏的nearby的index是0, 排除第一種情況, 那麽這裏就是空行了, 加句if判斷就可以解決

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

以上主要是對爬蟲過程學習的總結, 若有不對的地方, 還請指正, 謝謝!

因還在學習中, 價格部分會用到圖片識別, 有待完善.

Python爬蟲項目--爬取自如網房源信息