Selenium模拟浏览器获取爬取QQ音乐歌词、评论等。

不念不忘少年蓝@ 2022-11-15 14:17 243阅读 0赞

### 目录 ###

*  说明
 *  Code

# 说明 #

请使用爬虫Selenium模拟浏览器获取爬取QQ音乐中你喜欢的某位歌手（可以是任意歌手）最受欢迎的前5首歌曲的歌词、流派、歌曲发行时间、评论条数、评论时间、评论点赞次数、评论内容具体(每一首歌的评论>=500条)。

如下图所示：  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80NTczNTI0Mg_size_16_color_FFFFFF_t_70]

**歌词、流派、歌曲发行时间、评论条数保存在: music\_info.csv文件中。  
评论时间、评论点赞次数、评论内容具体保存在: comments\_info.csv文件中。  
统计每首歌的每个评论点赞次数保存在: series.csv文件中。**

请按照如下形式组织代码:  
![在这里插入图片描述][20210405170612628.png]

# Code #

**只需改main.py其他不用改。**  
红色框中，歌曲链接。  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80NTczNTI0Mg_size_16_color_FFFFFF_t_70 1]

from crawler.yqq.get_info import get_music_info
    
    """
    main.py
    """
    
    # 存放文件的目录
    path = r'D:\Study\Python'
    # 歌曲的链接
    url_list = ['https://y.qq.com/n/yqq/song/000T5eoR4YpW4t.html', 'https://y.qq.com/n/yqq/song/001afu2a1qjiik.html']
    # 每个歌曲的评论页数
    num_comment_page = [2, 2]
    # 爬取睡眠时间, 建议长一点, 防止页面未加载
    time_sleep = 5
    
    get_music_info(path, url_list, num_comment_page, time_sleep)

from selenium import webdriver
    import time
    from crawler.yqq.commentcrawler import get_commnets_praise
    from crawler.yqq.info2csv import tocsv
    from crawler.yqq.normalinfocrawler import normal_info
    
    """
    get_info.py
    """
    
    
    def get_music_info(path, url_list, num_comment_page, time_sleep=1):
        assert len(url_list) == len(num_comment_page), 'len(url_list) != len(num_comment_page)'
        music_dict = {
        '歌词': [], '流派': [], '歌曲发行时间': [], '评论条数': []}
        commnets_praise_dict = {
        '评论': [], '点赞数': [], '评论时间': []}
    
        driver = webdriver.Chrome()
    
        for i in range(len(url_list)):
            driver.get(url_list[i])
            time.sleep(time_sleep)
    
            m_d = normal_info(driver)
            for key in music_dict:
                music_dict[key].append(m_d[key])
    
            c_p_d_page = {
        '评论': [], '点赞数': [], '评论时间': []}
            for j in range(num_comment_page[i]):
                c_p_d = get_commnets_praise(driver)
                for key in c_p_d_page:
                    c_p_d_page[key].extend(c_p_d[key])
                botton = driver.find_element_by_class_name('next')
                botton.click()
                time.sleep(time_sleep)
    
            for key in commnets_praise_dict:
                commnets_praise_dict[key].append(c_p_d_page[key])
    
        driver.close()
        tocsv(path, music_dict, commnets_praise_dict)
        # print(music_dict)
        # print(commnets_praise_dict)

import bs4
    
    """
    commentcrawler.py
    """
    
    
    def get_commnets_praise(driver):
        pageSource = driver.page_source
        soup = bs4.BeautifulSoup(pageSource, 'html.parser')
        items = soup.find('ul', class_='js_all_list')
        info_dict = {
        '评论': [], '点赞数': [], '评论时间': []}
        for item in items.find_all('li'):
            info_dict['评论'].append(item.find('p', class_='js_hot_text').text)
            info_dict['点赞数'].append(item.find('span', class_='js_praise_num').text)
            info_dict['评论时间'].append(item.find('span', class_='c_tx_thin').text)
    
        return info_dict

import bs4
    import re
    
    """
    normalinfocrawler.py
    """
    
    
    def normal_info(driver):
        pageSource = driver.page_source
        soup = bs4.BeautifulSoup(pageSource, 'html.parser')
    
        data = soup.find('div', class_='main')
    
        info_dict = {
        '歌词': [], '流派': [], '歌曲发行时间': [], '评论条数': []}
    
        lyric = ''
        for p in data.find('div', class_='lyric__cont_box').find_all('p'):
            if p:
                lyric += p.text
    
        info_dict['歌词'].append(lyric)
        info_dict['流派'].append(data.find('li', class_='js_genre').text.split('：')[-1])
        info_dict['歌曲发行时间'].append(
            data.find('li', class_='js_public_time').text.split('：')[-1])
        info_dict['评论条数'].append(re.findall(r'\d+', data.find('a', class_='js_into_comment').text)[0])
    
        return info_dict

import pandas as pd
    
    """
    info2csv.py
    """
    
    
    def tocsv(path, music_dict, commnets_praise_dict):
        music_info = pd.DataFrame(columns=('歌词', '流派', '歌曲发行时间', '评论条数'))
        comments_info = pd.DataFrame(columns=('评论', '点赞数', '评论时间'))
        for key in music_dict:
            music_info[key] = (music_dict[key])
    
        for key in commnets_praise_dict:
            values = []
            for item in commnets_praise_dict[key]:
                values.extend(item)
            comments_info[key] = values
    
        music_info.to_csv(path + r'\music_info.csv', encoding='utf_8_sig')
        comments_info.to_csv(path + r'\comments_info.csv', encoding='utf_8_sig')
    
        ser = pd.Series()
        for i in range(len(commnets_praise_dict['点赞数'])):
            ser['song{}'.format(i)] = sum(list(map(int, commnets_praise_dict['点赞数'][i])))
    
        ser.to_csv(path + r'\series.csv', encoding='utf_8_sig')

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80NTczNTI0Mg_size_16_color_FFFFFF_t_70]: /images/20221022/b245b23cffef4deb81a5ce12df9a8a41.png
[20210405170612628.png]: /images/20221022/7927d8ebf60f4ef4985f1c89f7c8b41a.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80NTczNTI0Mg_size_16_color_FFFFFF_t_70 1]: /images/20221022/91c73dd26c7e4199a8da33b14e91ced0.png