Python 爬取51cto博客数据存入MySQL
实验环境
1.安装Python 3.7
2.安装requests, bs4，pymysql 模块
实验步骤
1.安装环境及模块
可参考博客https://blog.51cto.com/13760351/2500048
2.编写代码
# 51cto 博客页面数据插入mysql数据库
# 导入模块
import re
import bs4
import pymysql
import requests
# 连接数据库账号密码
db = pymysql.connect(host=‘172.171.13.229‘,
user=‘root‘, passwd=‘abc123‘,
db=‘test‘, port=3306,
charset=‘utf8‘)
# 获取游标
cursor = db.cursor()
def open_url(url):
# 连接模拟网页访问
headers = {
‘user-agent‘: ‘Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) ‘
‘Chrome/57.0.2987.98 Safari/537.36‘}
res = requests.get(url, headers=headers)
return res
# 爬取网页内容
def find_text(res):
soup = bs4.BeautifulSoup(res.text, ‘html.parser‘)
# 博客名
titles = []
targets = soup.find_all("a", class_="tit")
for each in targets:
each = each.text.strip()
if "置顶" in each:
each = each.split(‘ ‘)[0]
titles.append(each)
# 阅读量
reads = []
read1 = soup.find_all("p", class_="read fl on")
read2 = soup.find_all("p", class_="read fl")
for each in read1:
reads.append(each.text)
for each in read2:
reads.append(each.text)
# 评论数
comment = []
targets = soup.find_all("p", class_=‘comment fl‘)
for each in targets:
comment.append(each.text)
# 收藏
collects = []
targets = soup.find_all("p", class_=‘collect fl‘)
for each in targets:
collects.append(each.text)
# 发布时间
dates=[]
targets = soup.find_all("a", class_=‘time fl‘)
for each in targets:
each = each.text.split(‘：‘)[1]
dates.append(each)
# 插入sql 语句
sql = """insert into blog (blog_title,read_number,comment_number, collect, dates)
values( ‘%s‘, ‘%s‘, ‘%s‘, ‘%s‘, ‘%s‘);"""
# 替换页面 \xa0
for titles, reads, comment, collects, dates in zip(titles, reads, comment, collects, dates):
reads = re.sub(‘\s‘, ‘‘, reads)
comment = re.sub(‘\s‘, ‘‘, comment)
collects = re.sub(‘\s‘, ‘‘, collects)
cursor.execute(sql % (titles, reads, comment, collects，dates))
db.commit()
pass
# 统计总页数
def find_depth(res):
soup = bs4.BeautifulSoup(res.text, ‘html.parser‘)
depth = soup.find(‘li‘, class_=‘next‘).previous_sibling.previous_sibling.text
return int(depth)
# 主函数
def main():
host = "https://blog.51cto.com/13760351"
res = open_url(host) # 打开首页链接
depth = find_depth(res) # 获取总页数
# 爬取其他页面信息
for i in range(1, depth + 1):
url = host + ‘/p‘ + str(i) # 完整链接
res = open_url(url) # 打开其他链接
find_text(res) # 爬取数据
# 关闭游标
cursor.close()
# 关闭数据库连接
db.close()
if __name__ == ‘__main__‘:
main()3..MySQL创建对应的表
CREATE TABLE `blog` ( `row_id` int(11) NOT NULL AUTO_INCREMENT COMMENT ‘主键‘, `blog_title` varchar(52) DEFAULT NULL COMMENT ‘博客标题‘, `read_number` varchar(26) DEFAULT NULL COMMENT ‘阅读数量‘, `comment_number` varchar(16) DEFAULT NULL COMMENT ‘评论数量‘, `collect` varchar(16) DEFAULT NULL COMMENT ‘收藏数量‘, `dates` varchar(16) DEFAULT NULL COMMENT ‘发布日期‘, PRIMARY KEY (`row_id`) ) ENGINE=InnoDB AUTO_INCREMENT=1 DEFAULT CHARSET=utf8;
4.运行代码，查看效果:
改进版：
改进内容：
1.数据库里面的某些字段只保留数字即可
2.默认爬取的内容都是字符串，存放数据库的某些字段，最好改为整型，方便后面数据库操作
1.代码如下：
import re
import bs4
import pymysql
import requests
# 连接数据库
db = pymysql.connect(host=‘172.171.13.229‘,
user=‘root‘, passwd=‘abc123‘,
db=‘test‘, port=3306,
charset=‘utf8‘)
# 获取游标
cursor = db.cursor()
def open_url(url):
# 连接模拟网页访问
headers = {
‘user-agent‘: ‘Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) ‘
‘Chrome/57.0.2987.98 Safari/537.36‘}
res = requests.get(url, headers=headers)
return res
# 爬取网页内容
def find_text(res):
soup = bs4.BeautifulSoup(res.text, ‘html.parser‘)
# 博客标题
titles = []
targets = soup.find_all("a", class_="tit")
for each in targets:
each = each.text.strip()
if "置顶" in each:
each = each.split(‘ ‘)[0]
titles.append(each)
# 阅读量
reads = []
read1 = soup.find_all("p", class_="read fl on")
read2 = soup.find_all("p", class_="read fl")
for each in read1:
reads.append(each.text)
for each in read2:
reads.append(each.text)
# 评论数
comment = []
targets = soup.find_all("p", class_=‘comment fl‘)
for each in targets:
comment.append(each.text)
# 收藏
collects = []
targets = soup.find_all("p", class_=‘collect fl‘)
for each in targets:
collects.append(each.text)
# 发布时间
dates=[]
targets = soup.find_all("a", class_=‘time fl‘)
for each in targets:
each = each.text.split(‘：‘)[1]
dates.append(each)
# 插入sql 语句
sql = """insert into blogs (blog_title,read_number,comment_number, collect, dates)
values( ‘%s‘, ‘%s‘, ‘%s‘, ‘%s‘, ‘%s‘);"""
# 替换页面 \xa0
for titles, reads, comment, collects, dates in zip(titles, reads, comment, collects, dates):
reads = re.sub(‘\s‘, ‘‘, reads)
reads=int(re.sub(‘\D‘, "", reads)) #匹配数字，转换为整型
comment = re.sub(‘\s‘, ‘‘, comment)
comment = int(re.sub(‘\D‘, "", comment)) #匹配数字，转换为整型
collects = re.sub(‘\s‘, ‘‘, collects)
collects = int(re.sub(‘\D‘, "", collects)) #匹配数字，转换为整型
dates = re.sub(‘\s‘, ‘‘, dates)
cursor.execute(sql % (titles, reads, comment, collects,dates))
db.commit()
pass
# 统计总页数
def find_depth(res):
soup = bs4.BeautifulSoup(res.text, ‘html.parser‘)
depth = soup.find(‘li‘, class_=‘next‘).previous_sibling.previous_sibling.text
return int(depth)
# 主函数
def main():
host = "https://blog.51cto.com/13760351"
res = open_url(host) # 打开首页链接
depth = find_depth(res) # 获取总页数
# 爬取其他页面信息
for i in range(1, depth + 1):
url = host + ‘/p‘ + str(i) # 完整链接
res = open_url(url) # 打开其他链接
find_text(res) # 爬取数据
# 关闭游标
cursor.close()
# 关闭数据库连接
db.close()
#主程序入口
if __name__ == ‘__main__‘:
main()2.创建对应表
CREATE TABLE `blogs` ( `row_id` int(11) NOT NULL AUTO_INCREMENT COMMENT ‘主键‘, `blog_title` varchar(52) DEFAULT NULL COMMENT ‘博客标题‘, `read_number` int(26) DEFAULT NULL COMMENT ‘阅读数量‘, `comment_number` int(16) DEFAULT NULL COMMENT ‘评论数量‘, `collect` int(16) DEFAULT NULL COMMENT ‘收藏数量‘, `dates` varchar(16) DEFAULT NULL COMMENT ‘发布日期‘, PRIMARY KEY (`row_id`) ) ENGINE=InnoDB AUTO_INCREMENT=1 DEFAULT CHARSET=utf8;
3.运行代码，验证
升级版
为了能让小白就可以使用这个程序，可以把这个项目打包成exe格式的文件，让其他人，使用电脑就可以运行代码，这样非常方便！
1.改进代码：
#末尾修改为：
if __name__ == ‘__main__‘:
main()
print("\n\t\t所有数据已成功存放数据库！!! \n")
time.sleep(5)2.安装打包模块pyinstaller(cmd安装）
pip install pyinstaller -i https://pypi.tuna.tsinghua.edu.cn/simple/
3.Python代码打包
1.切换到需要打包代码的路径下面
2.在cmd窗口运行 pyinstaller -F test03.py （test03为项目名称）
4.查看exe包
在打包后会出现dist目录，打好包就在这个目录里面
5.运行exe包，查看效果
检查数据库
总结：
1.这一篇博客，是在上一篇的基础上改进的，步骤是先爬取首页的信息，再爬取其他页面信息，最后在改进细节，打包exe文件
2.我们爬取网页数据大多数还是存放到数据库的，所以这种方法很实用。
3.其实在此博客的基础上还是可以改进的，重要的是掌握方法即可。