import datetime

def create_assist_date(datestart = None,dateend = None):
	# 创建日期辅助表

	if datestart is None:
		datestart = '2016-01-01'
	if dateend is None:
		dateend = datetime.datetime.now().strftime('%Y-%m-%d')

	# 转为日期格式
	datestart=datetime.datetime.strptime(datestart,'%Y-%m-%d')
	dateend=datetime.datetime.strptime(dateend,'%Y-%m-%d')
	date_list = []
	date_list.append(datestart.strftime('%Y-%m-%d'))
	while datestart<dateend:
		# 日期叠加一天
	    datestart+=datetime.timedelta(days=+1)
	    # 日期转字符串存入列表
	    date_list.append(datestart.strftime('%Y-%m-%d'))
	return date_list

import requests

# 这里填入自己的User-agent和cookie信息
headers = {
            'User-Agent': "",       
             'Cookie': ""
}
def getDanmuKu(url):
    danmu = requests.get(url, headers=headers)
    # 解码弹幕内容为utf-8
    danmu = danmu.content.decode('utf-8')
    # 用BeatifulSoup提取弹幕信息
    danmu = BeautifulSoup(danmu)
    danmus=danmu.findAll('d')

    # 将弹幕信息和内容分别用两个列表保存
    danmu_info = []
    danmu_text = []
    for iter in danmus:
        all_info.append(iter['p'])
        all_text.append(iter.text)
        
    return all_info, all_text

保存弹幕到csv文件中
在当前文件目录下创建一个空文件夹csv，通过csv.writer 的writerow()方法写入弹幕数据

def saveInCsv(all_info, all_text, date):
    # 将弹幕文件保存在csv文件中
    f=open('csv3/danmuinfo_' + str(date)+ '.csv','w', encoding='utf-8')
    csv_writer = csv.writer(f)
    csv_writer.writerow(["时间", "弹幕模式", "字号大小", "颜色", "Unix格式时间戳", "弹幕种类", "发送者ID", "rowID"])
    for single_info in all_info:
        single_info = str(single_info).split(',')
        csv_writer.writerow(single_info)
    f.close()
    f = open('csv3/danmutext_' + str(date) + '.csv', 'w', encoding='utf-8')
    csv_writer=csv.writer(f)
    for single_text in all_text:
        csv_writer.writerow(single_text)
    f.close()def saveInCsv(all_info, all_text, date):
    # 将弹幕文件保存在csv文件中
    f=open('csv3/danmuinfo_' + str(date)+ '.csv','w', encoding='utf-8')
    csv_writer = csv.writer(f)
    csv_writer.writerow(["时间", "弹幕模式", "字号大小", "颜色", "Unix格式时间戳", "弹幕种类", "发送者ID", "rowID"])
    for single_info in all_info:
        single_info = str(single_info).split(',')
        csv_writer.writerow(single_info)
    f.close()
    f = open('csv3/danmutext_' + str(date) + '.csv', 'w', encoding='utf-8')
    csv_writer=csv.writer(f)
    for single_text in all_text:
        csv_writer.writerow(single_text)
    f.close()

import jieba
import re,string
from zhon.hanzi import punctuation
import os
import csv


danmuCount = dict()
danmuNum = 0
punc = '~`!#$%^&*()_+-=|';":/.,?><~·！@#￥%……&*（）——+-=“：’；、。？》《{} oh1O○〇●哈'
with open('danmuku3.csv', 'a', encoding='utf-8') as savefile:
    writer = csv.writer(savefile)
    writer.writerow(['name','type','value','date'])
    for date in dateLists:
        with open('csv3/danmutext_'+ date + '.csv', 'r', encoding='utf-8') as csvfile:
            print('---分析日期', date, '弹幕...n')
            reader = csv.reader(csvfile)
            for line in reader:
                danmuNum = danmuNum + 1
                line = "".join(line)
                line = re.sub(r"[%s]+" % punc, "", line)
                # words_list = jieba.lcut(line)
                # for word in words_list:
                    # data[line] = data[line] + 1
                # line = line.lower()
                if len(line) >= 2 and len(line) <= 15:
                    if danmuCount.get(line):
                        danmuCount[line] = danmuCount[line] + 1
                    else:
                        danmuCount[line] = 1

        sortList = sorted(danmuCount.items(), key=lambda item:item[1], reverse=True)

        if len(sortList)>10:
            pltLists = sortList[:10]

            for plttuple in pltLists:
                saveLine = []
                saveLine.append(plttuple[0])
                saveLine.append('Chinese')
                saveLine.append(plttuple[1])
                saveLine.append(date)
                writer.writerow(saveLine)