Python爬虫实战之爬取微博热搜

前言

在开始之前，我们需要了解一些基本的爬虫知识。Python爬虫是一种自动化获取网页内容的技术，可以模拟浏览器行为，获取网页源代码，并从中提取所需的信息。在爬取微博热搜榜单时，我们需要发送HTTP请求获取网页源代码，然后使用正则表达式或者解析库对源代码进行解析和提取。

爬取目标（效果展示）

在使用Python进行爬虫的过程中，经常需要获取微博热搜榜单的数据。微博热搜榜单是一个非常有价值的信息源，可以了解当前社会热点事件和用户关注度。本文将介绍使用Python爬取微博热搜榜单的方法。

效果展示：在这里插入图片描述

爬取的内容是：标题、榜单、热度值、新闻类型、时间戳、url地址等

准备工作

我用的是python3.8，VScode编辑器，所需的库有：requests、etree、time

开头导入所需用到的导入的库：

python
复制代码
import requests # python基础爬虫库
from lxml import etree # 可以将网页转换为Elements对象
import time # 防止爬取过快可以睡眠一秒

建表：

CREATE TABLE "WB_HotList" (
	"id" INT IDENTITY(1,1) PRIMARY key,
	"batch" NVARCHAR(MAX),
	"daydate" SMALLDATETIME,
	"star_word" NVARCHAR(MAX),
	"title" NVARCHAR(MAX),
	"category" NVARCHAR(MAX),
	"num" NVARCHAR(MAX),
	"subject_querys" NVARCHAR(MAX),
	"flag" NVARCHAR(MAX),
	"icon_desc" NVARCHAR(MAX),
	"raw_hot" NVARCHAR(MAX),
	"mid" NVARCHAR(MAX),
	"emoticon" NVARCHAR(MAX),
	"icon_desc_color" NVARCHAR(MAX),
	"realpos" NVARCHAR(MAX),
	"onboard_time" SMALLDATETIME,
	"topic_flag" NVARCHAR(MAX),
	"ad_info" NVARCHAR(MAX),
	"fun_word" NVARCHAR(MAX),
	"note" NVARCHAR(MAX),
	"rank" NVARCHAR(MAX),
	"url" NVARCHAR(MAX)	
)

为防止，字段给的不够，直接给个MAX！在这里插入图片描述

代码分析

先讲讲我的整体思路在逐步分析：

第一步：发送请求，获取网页信息
第二步：解析数据，提取我们所需要的数据
第三步：添加入库批次号
第四步：把数据存入数据库

第一步

发送请求，获取网页信息

提供了数据的接口，所以我们直接访问接口就行，如下图（json格式）：

# 接口地址：https://weibo.com/ajax/statuses/hot_band

在这里插入图片描述

def __init__(self) :
	self.url = "https://weibo.com/ajax/statuses/hot_band"
	self.headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/95.0.4638.69 Safari/537.36"}
# 发送请求，获取相应 
def parse_url(self):
	response = requests.get(self.url,headers=self.headers)
	time.sleep(2) # 休息两秒
	
	return response.content.decode()

第二步

解析数据，提取我们所需要的数据

接口中的数据格式化如下（只需提取我们所需要的）：在这里插入图片描述

for i in range(50):
	ban_list = json_data['data']['band_list'][i]
	batch = f'第{a}批'
	try:
	    star_word = ban_list['star_word']
	except Exception as e:
	    print(e)
	try:
	    title = ban_list['word']
	except Exception as e:
	    print(e)
	try:
	    category = ban_list['category']
	except Exception as e:
	    print(e)
	try:
	    num = ban_list['num']
	except Exception as e:
	    print(e)
	try:
	    subject_querys = ban_list['subject_querys']
	except Exception as e:
	    print(e)
	try:
	    flag = ban_list['flag']
	except Exception as e:
	    print(e)
	try:
	    icon_desc = ban_list['icon_desc']
	except Exception as e:
	    print(e)  
	try:
	    raw_hot = ban_list['raw_hot']
	except Exception as e:
	    print(e)      
	try:
	    mid = ban_list['mid']
	except Exception as e:
	    print(e) 
	try:
	    emoticon = ban_list['emoticon']
	except Exception as e:
	    print(e)
	try:
	    icon_desc_color = ban_list['icon_desc_color']
	except Exception as e:
	    print(e)
	try:
	    realpos = ban_list['realpos']
	except Exception as e:
	    print(e)
	try:
	    onboard_time = ban_list['onboard_time']
	    onboard_time = datetime.datetime.fromtimestamp(onboard_time)
	except Exception as e:
	    print(e)
	try:
	    topic_flag = ban_list['topic_flag']
	except Exception as e:
	    print(e)
	try:
	    ad_info = ban_list['ad_info']
	except Exception as e:
	    print(e)
	try:
	    fun_word = ban_list['fun_word']
	except Exception as e:
	    print(e)   
	try:
	    note = ban_list['note']
	except Exception as e:
	    print(e)      
	try:
	    rank = ban_list['rank'] + 1
	except Exception as e:
	    print(e) 
	try:
	    url = json_data['data']['band_list'][i]['mblog']['text']
	    url = re.findall('href="(.*?)"',url)[0]

第三步

数据库的batch用于判断，每次插入的批次（50个一批），如果爬虫断了，写个方法还能接着上次的批次

如图：在这里插入图片描述

# 把数据库batch列存入列表并返回（用于判断批次号）
def batch(self):
	conn=pymssql.connect('.', 'sa', 'yuan427', 'test')
	cursor=conn.cursor()
	
	cursor.execute("select batch from WB_HotList") #向数据库发送SQL命令
	rows=cursor.fetchall()
	batchlist=[]
	for list in rows:
	    batchlist.append(list[0]) 
	
	return batchlist

第四步

把数据存入数据库

# 连接数据库服务,创建游标对象
db = pymssql.connect('.', 'sa', 'yuan427', 'test') #服务器名,账户,密码,数据库名
if db:
    print("连接成功!")    
cursor= db.cursor()

try:
	# 插入sql语句
	sql = "insert into test4(batch,daydate,star_word,title,category,num,subject_querys,flag,icon_desc,raw_hot,mid,emoticon,icon_desc_color,realpos,onboard_time, \
	        topic_flag,ad_info,fun_word,note,rank,url) values (%s,getdate(),%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s)"
	
	# 执行插入操作
	cursor.execute(sql,(batch,star_word,title,category,num,subject_querys,flag,icon_desc,raw_hot,mid,emoticon,icon_desc_color,realpos,onboard_time,topic_flag,ad_info, \
	            fun_word,note,rank,url))
	db.commit()
	
	print('成功载入......' )
	
	except Exception as e:
	db.rollback()
	print(str(e))
    
# 关闭游标，断开数据库
cursor.close()
db.close()

完整代码

import requests,pymssql,time,json,re,datetime
from threading import Timer

class Spider:
    def __init__(self) :
        self.url = "https://weibo.com/ajax/statuses/hot_band"
        self.headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/95.0.4638.69 Safari/537.36"}

    # 发送请求，获取相应 
    def parse_url(self):
        response = requests.get(self.url,headers=self.headers)
        time.sleep(2)
        
        return response.content.decode()

    # 解析数据，入库
    def parse_data(self,data,a):
        json_data = json.loads(data)

        # 连接数据库服务,创建游标对象
        db = pymssql.connect('.', 'sa', 'yuan427', 'test') #服务器名,账户,密码,数据库名   
        cursor= db.cursor()
     
        for i in range(50):
            ban_list = json_data['data']['band_list'][i]
            
            
            batch = f'第{a}批'
            
            try:
                star_word = ban_list['star_word']
            except Exception as e:
                print(e)
            
            
            try:
                title = ban_list['word']
            except Exception as e:
                print(e)

            try:
                category = ban_list['category']
            except Exception as e:
                print(e)
            
            try:
                num = ban_list['num']
            except Exception as e:
                print(e)
        
            try:
                subject_querys = ban_list['subject_querys']
            except Exception as e:
                print(e)

            try:
                flag = ban_list['flag']
            except Exception as e:
                print(e)

            try:
                icon_desc = ban_list['icon_desc']
            except Exception as e:
                print(e)  

            try:
                raw_hot = ban_list['raw_hot']
            except Exception as e:
                print(e)      
            
            try:
                mid = ban_list['mid']
            except Exception as e:
                print(e) 
            
            try:
                emoticon = ban_list['emoticon']
            except Exception as e:
                print(e)
            
            try:
                icon_desc_color = ban_list['icon_desc_color']
            except Exception as e:
                print(e)
            
            try:
                realpos = ban_list['realpos']
            except Exception as e:
                print(e)
            
            try:
                onboard_time = ban_list['onboard_time']
                onboard_time = datetime.datetime.fromtimestamp(onboard_time)
            except Exception as e:
                print(e)
            
            try:
                topic_flag = ban_list['topic_flag']
            except Exception as e:
                print(e)
            
            try:
                ad_info = ban_list['ad_info']
            except Exception as e:
                print(e)
            
            try:
                fun_word = ban_list['fun_word']
            except Exception as e:
                print(e)   
            
            try:
                note = ban_list['note']
            except Exception as e:
                print(e)      
        
            try:
                rank = ban_list['rank'] + 1
            except Exception as e:
                print(e) 
            
            try:
                url = json_data['data']['band_list'][i]['mblog']['text']
                url = re.findall('href="(.*?)"',url)[0]
            except Exception as e:
                print(e)
           
            try:
                # 插入sql语句
                sql = "insert into test4(batch,daydate,star_word,title,category,num,subject_querys,flag,icon_desc,raw_hot,mid,emoticon,icon_desc_color,realpos,onboard_time, \
                        topic_flag,ad_info,fun_word,note,rank,url) values (%s,getdate(),%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s)"

                # 执行插入操作
                cursor.execute(sql,(batch,star_word,title,category,num,subject_querys,flag,icon_desc,raw_hot,mid,emoticon,icon_desc_color,realpos,onboard_time,topic_flag,ad_info, \
                            fun_word,note,rank,url))
                db.commit()

                print('成功载入......' )
            
            except Exception as e:
                db.rollback()
                print(str(e))
            
        # 关闭游标，断开数据库
        cursor.close()
        db.close()
         
    # 把数据库batch列存入列表并返回（用于判断批次号）
    def batch(self):
        conn=pymssql.connect('.', 'sa', 'yuan427', 'test')

        cursor=conn.cursor()

        cursor.execute("select batch from WB_HotList") #向数据库发送SQL命令

        rows=cursor.fetchall()
        batchlist=[]
        for list in rows:
            batchlist.append(list[0]) 

        return batchlist    
             
    # 实现主要逻辑 
    def run(self, a):
        
        # 根据数据库批次号给定a的值
        batchlist = self.batch()
        if len(batchlist) != 0:
            batch = batchlist[len(batchlist) -1]
            a = re.findall('第(.*?)批',batch)
            a = int(a[0]) + 1

        data = self.parse_url()

        self.parse_data(data,a)
        a +=1
        # 定时调用
        t = Timer(1800, self.run, (a, )) # 1800表示1800秒，半小时调用一次
        t.start()

    
if __name__ == "__main__": 
    spider = Spider()
    spider.run(1)