• spider csdn blog part II


    继续上次的笔记, 继续完善csdn博文的提取.

    发现了非常好的模块. html2docx

    结果展示:

    运行之后, 直接生成docx文档. 截个图如下:
    结果已经基本满意了!!!
    在编写过程中的一些感想.

    1. 获取网站响应:
      决定放弃requests, 采用 selenium.webdriver.
      后者就是模拟浏览器操作. 可以应对许多需要登录的, 防止爬取的网站
      超时控制(等待网站响应), 操作网页等功能也非常强大.

    2. 定位页面元素:
      在定位页面元素方面: 有太多的方法可以选择. 最后决定就用一种. webdriver方法.
      放弃etree, BeautifulSoup, 还有直接用re模块的提取.

    3. 好好学习并掌握webdriver一种办法, 就可以了. 它的功能已经非常强大了,
      也支持用xpath来锁定页面元素.

    4. webdriver支持 网页面里注入javascript脚本来完成任务. (网络开发里的前端技术)

    5. 为了与html2docx衔接, 这里利用了
      selenium.webdriver.WebElemnt.get_attribute('outHTML')方法获取元素的html
      BeautifulSoup对象的prettify()方法, 来生成合法的完整的页面元素的html源码.

    代码:

    
    
    
    
    import os; type(os)
    import time; type(time)
    import re
    anys = '.*?' # 任意长的字符串, 贪婪型的
    
    import random; type(random)
    
    #import requests
    #from lxml import etree
    from selenium import webdriver
    chrome_options = webdriver.ChromeOptions()
    chrome_options.binary_location = r'C:UsersAdministratorAppDataRoaming360se6Application360se.exe'
    chrome_options.add_argument(r'--lang=zh-CN') # 这里添加一些启动的参数
    
    import logging
    logging.basicConfig(level=logging.INFO, 
        format= '%(asctime)s - %(name)s - %(levelname)s : %(message)s', 
        #format='%(asctime)s %(filename)s [line:%(lineno)d] %(levelname)s %(message)s', 
        )
    logger = logging.getLogger(__name__)
    
    #logger.info("Start print log")
    #logger.debug("Do something")
    #logger.warning("Something maybe fail.")
    #logger.info("Finish")
    
    from bs4 import BeautifulSoup
    from html2docx import html2docx
    
    
    from selenium.webdriver.support.ui import WebDriverWait
    from selenium.webdriver.common.by import By
    from selenium.webdriver.support import expected_conditions as EC
    type(By)
    type(EC)
    
    def itm(): return int(time.time())
    
    def insert_title(title:'html w/o Body tag', 
                     article:'html with Body tag'):
        '''在article这个完整的合法的html网页里, 在它的头部插入标题部分(title).
        标题部分: 包括文章标题行+作者行+发布日期
        '''
        merged = re.sub('(<html>
    s*<body>)',
                '<html>
     <body>
    ' + title, 
                article, 
                flags=re.S,)
        return merged
    
    
    def get_it_wait_untill(browser, element_func='title', sleep_time=80, arg=''):
        '''
        selenium内核的锁定页面元素, 然后取之. 比如:
            获取网页标题
            获取整个网页的源文件
            获取指定页面元素: 
                by_id
                by_xpath
        Example:
            >>> get_it_wait_untill(browser, 'title')
            >>> get_it_wait_untill(browser, 'page_source')
            >>> get_it_wait_untill(browser, element_func='find_element_by_id', 
                            arg='content_views',
                            )
            >>> get_it_wait_untill(browser, element_func='find_element_by_xpath', 
                            arg='//section[@class="content article-content"]',
                            )
        '''
        prop = str(type(getattr(browser, element_func)))
        #python的很多内置的函数, 是使用C语言写出来的,要看C语言的源代码
        
        if prop == "<class 'str'>":
            element = WebDriverWait(browser, sleep_time).until(
                    lambda x: getattr(x, element_func)
                    )
        #elif callable(getattr(browser, element_func)):
        elif prop == "<class 'method'>":
            element = WebDriverWait(browser, sleep_time).until(
                    lambda x: getattr(x, element_func)(arg)
                    )
            
        return element
    
    
    
    
    def get_csdn_blog(
            url='https://blog.csdn.net/Lili_0820/article/details/70155949' 
            ,
            sleep_time=40
            ,
            ):
        '''
        爬取csdn blog文章
        参数:
        url: str,
        sleep_time: int, wait time in seconds
        Example:
            >>> get_csdn_blog()
        '''
        logger.info(f'当前网页的url: {url}')
        browser = webdriver.Chrome(options=chrome_options)
        browser.implicitly_wait(200)
        #timeout_wait = WebDriverWait(browser, 2*5) # 10sec
        browser.get(url)
        timeout_wait = WebDriverWait(browser, sleep_time) # 10sec;
        type(timeout_wait)
        
        '''
        我们需要确保: 网页信息已经全部加载, 否则可能提取不到有用信息.
        
        Sets a sticky timeout to implicitly wait for an element to be found,
        or a command to complete. 
        This method only needs to be called one time per session.
        当浏览器(webdriver实例)在定位元素的时候, 
        我们可以设置一个隐式的超时等待时间, 
        如果超过这个设定的时间还不能锁定元素, 那么就报错或者继续执行.
        本方法在整个对话期内, 只需调用一次.
        
        '''
        title = WebDriverWait(browser, sleep_time).until(lambda x: x.title)
        logger.info(f'提取网页标题: {title}')
        
        html= WebDriverWait(browser, sleep_time).until(lambda x: x.page_source)
        #html = browser.page_source 
        #需要花点时间
        #time.sleep(sleep_time) # 太粗暴简单了
        
        title      = browser.find_element_by_xpath('//h1[@class="title-article"]').text
        pub_date   = browser.find_element_by_xpath('//div[@class="article-bar-top"]').text
        author_url = browser.find_element_by_xpath('//div[@class="article-bar-top"] /a[1]').get_attribute('href')
        pub_date   = re.findall('
    (.*?)阅读数.*?收藏', pub_date,re.S)[0]    
        author, pub_date = re.findall('(.*?) (发布.*?) ', pub_date, re.S)[0]    
        
        insertion = f'''
        <h1> {title} </h1>
        <p> {author} ({author_url}) </p>
        <p> {pub_date} </p>
        '''
    
        content_we = browser.find_element_by_id('content_views') # selenium.webelement
        text = content_we.text; type(text)
        logger.info('网页源码的长度和博文的长度分别是: {1} {0}'.
              format(len(text), len(html))
        )
        content_html = content_we.get_attribute('outerHTML')
        content_html = BeautifulSoup(content_html, 'lxml').prettify()
        content_html = insert_title(insertion, content_html)
    
        # 规范化: 输出文件名
    #    if '|' in title: title2=title.replace('|', '')
    #    title2 = title2.replace('QuantStart','')
    #    title2 = title2.replace(' ','_')
        outf=f'{title}_{itm()}.docx'
        buffer = html2docx(content_html, title=title)
        
        with open(outf, "wb") as fh: fh.write(buffer.getvalue())
        if os.path.exists(outf):  print( f'{outf} created!!!')
            
    
    
        # re方法
        '''
        pattern = 'id="content_views" class="markdown_views.*?>' + 
            '(.*?)' + 
            '<link href="https://csdnimg.cn/release/' + 
            'phoenix/mdeditor/markdown_views'
        a = re.findall(pattern, html, re.S)
        a = a[0]
        a = re.findall(f'{anys}(<p>{anys})</div>{anys}', a, re.S)[0]
        '''
        
        #  etree方法
        '''
        tree = etree.HTML(html)    
        cv_etree = tree.xpath('//div[@id="content_views"]')[0]
        text = cv_etree.xpath('*/text()')
        cv_html = etree.tostring(cv_etree, encoding='unicode')
        '''
        browser.close()
        browser.quit()
        #return a
    
    
    if __name__=='__main__':
        pass
    #    url='https://blog.csdn.net/Lili_0820/article/details/70155949' 
    #    get_csdn_blog(url, sleep_time=80)
    
    
    
  • 相关阅读:
    [Leetcode] Rotate Image
    [Leetcode] Permutation Sequence
    [Leetcode] Palindrome Partitioning
    [Leetcode] Letter Combinations of a Phone Number
    Java里的多线程
    css学习2----css动态菜单
    css学习1----css超链接效果
    javascript判断身份证是否合法
    RMI(Remote Method Invocation,远程方法调用)
    struts.properties配置详解
  • 原文地址:https://www.cnblogs.com/duan-qs/p/12068896.html
Copyright © 2020-2023  润新知