用Python爬取《后浪》弹幕，绘制生成专属词云，看看“后浪们”都在评论些什么pythonqq43735106的博客-

请查收，最近B站献给新一代的青年宣言片。
国家一级演员何冰走上舞台，以青年宣言《后浪》为词，认可、赞美与寄语年轻一代。在UP主们的青春混剪中，属于年轻人的光芒正在闪耀。“你们有幸 遇见这样的时代 但时代更有幸 遇见这样的你们”
用Python爬取《后浪》弹幕，看看“后浪”都在评论些什么？

一、找到评论链接
进入B站《后浪》播放页面，按F12键后，刷新页面。

评论链接在红色标记下对应的包，蓝线上方Request URL即为评论链接https://api.bilibili.com/x/v1/dm/list.so?oid=188273397。注意，Hide data URLs勾选为All。

二、敲代码

1、利用requests库请求获取网页

def getHtmlText(url): try:         response = requests.get(url)         response.raise_for_status()         response.encoding = response.apparent_encoding         data = response.content.decode('utf-8')#utf-8转码，不然会出现乱码 return data     except: return '' 

2、找到所有评论并将其存入二维列表

所有评论都在标签下的文本中，而所有标签都在标签下。利用bs4库中的BeautifulSoup函数将1中得到的页面进行解析。找到标签下的文本存入列表。（这里也可以用正则表达式）

def fillList(html,list):     soup = BeautifulSoup(html,'html.parser')     itotal = soup.find('i')     dtotals = itotal.find_all('d') for dtotal in dtotals:         danmu = dtotal.text         List.append([danmu]) 

3、将二维列表中的所有数据下载到文本

def down_danmu(list): for list in List: with open('后浪弹幕.txt', 'a', encoding='utf-8-sig') as f:             s = str(list).replace('[','').replace(']','') + 'n' #去除[],这两行按数据不同，可以选择，每行末尾加换行符             s = s.replace("'", '').replace(',', '')#去除单引号，逗号，每行末尾追加换行符             f.write(s) print("文件写入成功！") 

4、生成专属词云图

def showPic():     img = np.array(image.open('bili.jpg'))#numpy库与PIL库引入自定义图片 with open('后浪弹幕.txt','r',encoding='utf-8') as f:         text = f.read()         w = wordcloud.WordCloud(font_path="msyh.ttc", mask=img, scale=15,stopwords=' ', width=1000, height=700, background_color='white')         w.generate(text)         w.to_file("Houlang.png") 

完整代码

import requests from bs4 import BeautifulSoup import wordcloud import PIL.Image as image import numpy as np  def getHtmlText(url): try:         response = requests.get(url)         response.raise_for_status()         response.encoding = response.apparent_encoding         html = response.content.decode('utf-8') return html     except: return '' def fillList(html,list):     soup = BeautifulSoup(html,'html.parser')     itotal = soup.find('i')     dtotals = itotal.find_all('d') for dtotal in dtotals:         danmu = dtotal.text         List.append([danmu]) def down_danmu(list): for list in List: with open('./后浪弹幕.txt', 'a', encoding='utf-8-sig') as f:             s = str(list).replace('[','').replace(']','') + 'n'             s = s.replace("'", '').replace(',', '')             f.write(s) print("文件写入成功！") def showPic():     img = np.array(image.open('bili.jpg')) with open('后浪弹幕.txt','r',encoding='utf-8') as f:         text = f.read()         w = wordcloud.WordCloud(font_path="msyh.ttc", mask=img, scale=15,stopwords=' ', width=1000, height=700, background_color='white')         w.generate(text)         w.to_file("Houlang.png")  List = [] url = 'https://api.bilibili.com/x/v1/dm/list.so?oid=186803402' html = getHtmlText(url) fillList(html,List) down_danmu(List) showPic() 

效果图