爬取B站up主视频弹幕并生成词云图（2020年5月）

1. 打开一个视频

2. 查看抓包信息找到弹幕接口

3. 请求弹幕接口即可得到包含弹幕的xml

4. 弹幕接口url带有oid参数，需要先获取oid，在网页源代码中搜索oid的值发现cid就是oid，可以通过正则来提取cid

5. 代码实现

# coding=utf-8

import re
import requests
import jieba
import pandas
import matplotlib.pyplot as plt

from lxml import etree
from wordcloud import WordCloud

url = input('请输入B站短视频链接：')
# url = 'https://www.bilibili.com/video/BV11p411o73u'
headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/81.0.4044.92 Safari/537.36'}

# 获取cid和视频标题
html_text = requests.get(url=url, headers=headers).text
cid = re.search('cid=(\d+)&aid=\d+', html_text).group(1)
html = etree.HTML(html_text)
title = html.xpath("//span[@class='tit']/text()")[0]

# 获取弹幕列表
response = requests.get('https://api.bilibili.com/x/v1/dm/list.so?oid={}'.format(cid), headers=headers)
xml = etree.fromstring(response.content)
danmu_list = xml.xpath("/i/d/text()")

# 把弹幕存到csv
dm_df = pandas.DataFrame(danmu_list)
dm_df.to_csv('弹幕-{}.csv'.format(title), encoding='utf_8_sig', header=None, index=None)   # 'utf_8_sig'解决保存到csv乱码问题
with open('弹幕-{}.csv'.format(title), 'r', encoding='utf-8') as f:
    txt = f.read()

# 结巴分词
txt = jieba.lcut(txt)
txt = ' '.join(txt)

# 读取词云图背景图片
mask_ima = plt.imread('heart.jpg')
# 生成词云图
wordcloud = WordCloud(font_path='msyh.ttc',  # 字体微软雅黑
                      min_font_size=10,  # 最小字体大小
                      max_words=1000,  # 最大词数
                      max_font_size=150,  # 最大字体大小
                      stopwords={' '},  # 屏蔽词语
                      # width=1000,  # 词云图宽度
                      # height=1000,  # 词云图高度
                      mask = mask_ima,  # 自定义背景图
                      background_color='black' # 背景颜色
).generate(txt)
# 保存词云图
wordcloud.to_file('B站弹幕词云图.png')

保存到csv的弹幕数据

词云图效果