您的位置：首页 > Web前端 > JavaScript

selenium+PhantomJS爬取豆瓣读书

2019-08-26 17:18 786 查看

本文实例为大家分享了selenium+PhantomJS爬取豆瓣读书的具体代码，供大家参考，具体内容如下

获取关于Python的全部书籍信息；
通过代码测试 request携带‘User-Agent'及 ‘data'数据信息的方式均无法获取到相关信息，获取数据时，部分数据为空，导致获取过程中报错，无法获取全部数据，初步判定豆瓣读书的反爬机制较为严格；通过selenium 模拟浏览器请求的方法测试后发现，可利用 selenium 方法请求获取数据；

#导入需要的模块
from selenium import webdriver
import time
from lxml import etree
import pymysql
import re

#创建一个函数
def my_browers(url, page):

# 获取浏览器对象
browers = webdriver.PhantomJS(executable_path=r'd:\Desktop\pythonjs\phantomjs-2.1.1-windows\bin\phantomjs.exe')

# 用浏览器发起请求
browers.get(url)

#休息两秒，频率低一点，爬的时间久一点，安全就多一点
time.sleep(2)

# 获取页面信息
html = browers.page_source

# 调用页面解析函数
parse_html(html)

# 解析页面信息
def parse_html(html):

# 生成一个xpath对象
html = etree.HTML(html)

# 获取所有的书籍信息列表
books = html.xpath('//div[contains(@class,"sc-bZQynM")]')

# 遍历每一本书籍 然后拿到我们想要的数据
for book in books:
# 创建一个存书字典存数据用
book_dict = {}

# 获取封面信息
pic = book.xpath('//img/@src')
if pic:
book_dict['pic'] = pic[0]
else:
book_dict['pic'] = ''
# print(pic)

# 获取书名
book_name = book.xpath('//div[@class="title"]/a/text()')
# print(book_name)
if book_name:
book_name = book_name[0]
# 删除书名中最后出现的引号，
#由于存数据库的时候书名最后面的引号会导致数据库报错，删除可以使代码更健壮
if '"' in book_name:
pattern = re.compile(r'"')
book_name = pattern.sub('', book_name)
if "'" in book_name:
pattern = re.compile(r"'")
book_name = pattern.sub('', book_name)
# 删除书名中最后出现的\，存数据的时候书名最后的\会把sql语句最后的引号转义，
#删除可以使代码更健壮
if '\\' in book_name:
book_name = book_name[:-1]
book_dict['book_name'] = book_name
else:
book_dict['book_name'] = ''

# 获取书籍详情连接
book_url = book.xpath('//div[@class="title"]/a/@href')
if book_url:
book_dict['book_url'] = book_url[0]
else:
book_dict['book_url'] = ''

# 获取评分信息
score_book = book.xpath('//span[@class="rating_nums"]/text()')
if score_book:
book_dict['score_book'] = score_book[0]
else:
book_dict['score_book'] = ''

# 获取出版社信息
book_detail = book.xpath('//div[@class="meta abstract"]/text()')
if book_detail:
# 删除书详情中最后出现的引号；
book_detail = book_detail[0]
if "'" in book_detail:
pattern = re.compile(r"'")
book_detail = pattern.sub('', book_detail)

book_dict['book_detail'] = book_detail
else:
book_dict['book_detail'] = ''
print(book_dict)

# 调用数据库函数
insert_mysql(book_dict)

# 插入数据库
def insert_mysql(book_dict):
# 连接数据库
conn = pymysql.connect('localhost', 'root', 'root', 'test', charset='utf8')

# 创建操作数据库的对象
cursor = conn.cursor()

pic = book_dict['pic']
book_name = book_dict['book_name']
book_url = book_dict['book_url']
score = book_dict['score_book']
book_detail = book_dict['book_detail']

sql = f"insert into python_book (pic,book_name,book_url,score,book_detail) " \
f"VALUE ('{pic}','{book_name}','{book_url}','{score}','{book_detail}')"

# 执行并提交
cursor.execute(sql)
conn.commit()

if __name__ == '__main__':
for i in range(0, 199):
print('=================下载第{}页========================'.format(i + 1))
page = i * 15
base_url = 'https://book.douban.com/subject_search?search_text=python&cat=1001&start={}'.format(page)
my_browers(base_url, page)

以上就是本文的全部内容，希望对大家的学习有所帮助

您可能感兴趣的文章:

内容来自用户分享和网络整理，不保证内容的准确性，如有侵权内容，可联系管理员处理

标签： selenium PhantomJS 爬取豆瓣

相关文章推荐

新的分享

章节导航