python爬取7w+『赘婿』弹幕,发现弹幕比剧还精彩!
1前言
在上一篇文章【以『赘婿』为实战案例,手把手教会你用python爬取『爱奇艺』视频弹幕】,教会了大家如何爬取爱奇艺弹幕!
本文将在上文的基础上继续完善,爬取更多的弹幕数据进行可视化分析!
同样还是以『赘婿』为例,目前已经更新到28集,下面将爬取这28集的全部弹幕数据,约7w+条数据!!!
2采集数据
1.寻找url
上文介绍到,每一集都需要通过查看数据把获取弹幕的url
因此,我们需要去获取这28级的弹幕url!!!
tv_name_list =[
'',
'/54/00/7973227714515400',
'/57/00/4779805474835700',
'/37/00/1016845483273700',
'/77/00/8679935826337700',
'/46/00/7197533339804600',
'/48/00/8042213977424800',
'/98/00/2262609044749800',
'/94/00/1699488619659400',
'/47/00/1805374511564700',
'/46/00/1933721047194600',
'/08/00/7232026471250800',
'/59/00/8982352350925900',
'/43/00/4702797553454300',
'/38/00/2151107991923800',
'/93/00/8357465155589300',
'/29/00/2071693573022900',
'/71/00/4646645944127100',
'/39/00/1182091647913900',
'/31/00/7711721648193100',
'/58/00/2099769377685800',
'/83/00/3042314248738300',
'/21/00/2889100571832100',
'/98/00/3374410909698000',
'/37/00/4335405595243700',
'/32/00/5215381530163200',
'/11/00/2379725258541100',
'/48/00/4872856713204800',
'/08/00/1488519001760800',
]
以上就是28集的弹幕url参数!!!
2.请求数据
def get_data():
for k in range(1,len(tv_name_list)):#29个 1-28
url_id = tv_name_list[k]
for x in range(1,11):
# x是从1到11,11怎么来的,这一集总共46分钟,爱奇艺每5分钟会加载新的弹幕,46除以5向上取整
try:
url = 'https://cmts.iqiyi.com/bullet'+str(url_id)+'_300_' + str(x) + '.z'
xml = download_xml(url)
# 把编码好的文件分别写入个xml文件中(类似于txt文件),方便后边取数据
with open('./lyc/zx'+str(k) +'-'+ str(x) + '.xml', 'a+', encoding='utf-8') as f:
f.write(xml)
except:
pass
这样就可以将含有的弹幕信息的xml文件下载到本地!
3.合并数据到excel
import openpyxl
outwb = openpyxl.Workbook() # 打开一个将写的文件
outws = outwb.create_sheet(index=0) # 在将写的文件创建sheet
"""
import xlwt
# # 创建一个workbook 设置编码
workbook = xlwt.Workbook(encoding = 'utf-8')
# # 创建一个worksheet
worksheet = workbook.add_sheet('sheet1')
#
# # 写入excel
# # 参数对应 行, 列, 值
# worksheet.write(0,0, label='index')
# worksheet.write(0,1, label='tvname')
# worksheet.write(0,2, label='uid')
# worksheet.write(0,3, label='content')
# worksheet.write(0,4, label='likeCount')
"""
outws.cell(row = 1 , column = 1 , value = "index")
outws.cell(row = 1 , column = 2 , value = "tvname")
outws.cell(row = 1 , column = 3 , value = "uid")
outws.cell(row = 1 , column = 4 , value = "content")
outws.cell(row = 1 , column = 5 , value = "likeCount")
避坑:
之前我们使用xlwt来保存数据到excel,但是最多写到65535行,这次我们采用openpyxl来写入到excel!!!
def xml_parse(file_name,tv__name):
global count
DOMTree = xml.dom.minidom.parse(file_name)
collection = DOMTree.documentElement
# 在集合中获取所有entry数据
entrys = collection.getElementsByTagName("entry")
for entry in entrys:
uid = entry.getElementsByTagName('uid')[0]
content = entry.getElementsByTagName('content')[0]
likeCount = entry.getElementsByTagName('likeCount')[0]
#print(uid.childNodes[0].data)
#print(content.childNodes[0].data)
#print(likeCount.childNodes[0].data)
# 写入excel
# 参数对应 行, 列, 值
outws.cell(row=count, column=1, value=str(count))
outws.cell(row=count, column=2, value=str("第"+str(tv__name)+"集"))
outws.cell(row=count, column=3, value=str(uid.childNodes[0].data))
outws.cell(row=count, column=4, value=str(content.childNodes[0].data))
outws.cell(row=count, column=5, value=str(likeCount.childNodes[0].data))
count=count+1
这样就可以将xml里的数据保存到excel
def combine_data():
for k in range(1,29):
for x in range(1,11):
try:
xml_parse("./lyc/zx"+str(k) +"-"+ str(x) + ".xml",k)
print(str(k) + "-" + str(x))
except:
pass
# 保存
#workbook.save('弹幕数据集-李运辰.xls')
outwb.save("弹幕数据集-李运辰.xls") # 保存结果
这样7w+条弹幕数据可以完全写入到excel中,命名为 弹幕数据集-李运辰.xls
3数据可视化
1.浏览数据
# 导包
import pandas as pd
#读入数据
df_all = pd.read_csv("弹幕数据集-李运辰.csv",encoding="gbk")
df = df_all.copy()
# 重置索引
df = df.reset_index(drop=True)
print(df.head())
说明:1.index序号、2.tvname集数、3.uid用户id、4.content评论、5.likeCount评论点赞数
2.累计发送弹幕数的用户
#累计发送弹幕数的用户
def an1():
danmu_counts = df.groupby('uid')['content'].count().sort_values(ascending=False).reset_index()
danmu_counts.columns = ['用户id', '累计发送弹幕数']
name = danmu_counts['用户id']
name = (name[0:10]).tolist()
dict_values = danmu_counts['累计发送弹幕数']
dict_values = (dict_values[0:10]).tolist()
# 链式调用
c = (
Bar(
init_opts=opts.InitOpts( # 初始配置项
theme=ThemeType.MACARONS,
animation_opts=opts.AnimationOpts(
animation_delay=1000, animation_easing="cubicOut" # 初始动画延迟和缓动效果
))
)
.add_xaxis(xaxis_data=name) # x轴
.add_yaxis(series_name="累计发送弹幕数的用户", yaxis_data=dict_values) # y轴
.set_global_opts(
title_opts=opts.TitleOpts(title='', subtitle='', # 标题配置和调整位置
title_textstyle_opts=opts.TextStyleOpts(
font_family='SimHei', font_size=25, font_weight='bold', color='red',
), pos_left="90%", pos_top="10",
),
xaxis_opts=opts.AxisOpts(name='用户id', axislabel_opts=opts.LabelOpts(rotate=45)),
# 设置x名称和Label rotate解决标签名字过长使用
yaxis_opts=opts.AxisOpts(name='累计发送弹幕数'),
)
.render("累计发送弹幕数的用户.html")
)
3.查看某个用户评论情况
#查看某个用户评论情况
def an2():
df_top1 = df[df['uid'] == 2127950839].sort_values(by="likeCount", ascending=False).reset_index()
print(df_top1.head(20))
4.用户(2127950839)每一集的评论数
#查看用户(2127950839)每一集的评论数
def an3():
df_top1 = df[df['uid'] == 2127950839].sort_values(by="likeCount", ascending=False).reset_index()
data_top1 = df_top1.groupby('tvname')['content'].count()
print(data_top1)
name = data_top1.index.tolist()
dict_values = data_top1.values.tolist()
# 链式调用
c = (
Bar(
init_opts=opts.InitOpts( # 初始配置项
theme=ThemeType.MACARONS,
animation_opts=opts.AnimationOpts(
animation_delay=1000, animation_easing="cubicOut" # 初始动画延迟和缓动效果
))
)
.add_xaxis(xaxis_data=name) # x轴
.add_yaxis(series_name="查看用户(2127950839)每一集的评论数", yaxis_data=dict_values) # y轴
.set_global_opts(
title_opts=opts.TitleOpts(title='', subtitle='', # 标题配置和调整位置
title_textstyle_opts=opts.TextStyleOpts(
font_family='SimHei', font_size=25, font_weight='bold', color='red',
), pos_left="90%", pos_top="10",
),
xaxis_opts=opts.AxisOpts(name='集数', axislabel_opts=opts.LabelOpts(rotate=45)),
# 设置x名称和Label rotate解决标签名字过长使用
yaxis_opts=opts.AxisOpts(name='评论数'),
)
.render("查看用户(2127950839)每一集的评论数.html")
)
5.剧集评论点赞数最多的评论内容
#剧集评论点赞数最多的评论内容
def an4():
df_like = df[df.groupby(['tvname'])['likeCount'].rank(method="first", ascending=False) == 1].reset_index()[['tvname', 'content', 'likeCount']]
df_like.columns = ['集', '弹幕内容', '点赞数']
print(df_like)
6.评论内容词云
#评论内容词云
def an5():
contents = (df_all['content']).tolist()
text = "".join(contents)
with open("stopword.txt", "r", encoding='UTF-8') as f:
stopword = f.readlines()
for i in stopword:
print(i)
i = str(i).replace("\r\n", "").replace("\r", "").replace("\n", "")
text = text.replace(i, "")
word_list = jieba.cut(text)
result = " ".join(word_list) # 分词用 隔开
# 制作中文云词
icon_name = 'fas fa-play'
gen_stylecloud(text=result, icon_name=icon_name, font_path='simsun.ttc',
output_name="评论内容词云.png") # 必须加中文字体,否则格式错误
4总结
1.爬取了7w+『赘婿』弹幕,保存到excel(数据分享给大家)!
2.通过pandas读取excel并进行相关统计计算!
3.以可视化方式当分析好的数据进行可视化展示!
如果大家对本文代码源码感兴趣,扫码关注『Python爬虫数据分析挖掘』后台回复:赘婿可视化 ,获取完整代码和数据集!
------------- 推荐文章 -------------
1.以『赘婿』为实战案例,手把手教会你用python爬取『爱奇艺』视频弹幕
耐得住寂寞,才能登得顶
Gitee码云:https://gitee.com/lyc96/projects