danmu/102201214 许莎莎.py

from bs4 import BeautifulSoup
import  re
import pandas as pd
import jieba
import requests
import imageio
import wordcloud
from openpyxl import load_workbook
from collections import Counter

#获取User-Agent和cookie
headers = {
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/128.0.0.0 Safari/537.36 Edg/128.0.0.0",
        "cookie": "buvid3=0C047DB7-FB67-6565-B853-68B19196AEE053166infoc; buvid4=D2E32722-EB31-8B5B-8BC7-420F049CDE3657801-022071821-mG8+jYWtWHQ35A9yqIgZIA%3D%3D; buvid_fp=60e37bdf4fe67cde89d283db25adff46; _uuid=FCEA6C48-BB82-123A-61106-3F5410106BB410B03170infoc; b_nut=100; header_theme_version=CLOSE; enable_web_push=DISABLE; bsource=search_bing; CURRENT_FNVAL=4048; SESSDATA=aa6a6590%2C1742210524%2C7b1c4%2A92CjCxud8rqp6tuF7AYkzmJF0YS7_L4_80iMI3NuY5q-M7BEW3cf0_bVyhIcnZMJapP7YSVnJiQ2NVcTJZZ1ZIMFduRURJXzZXOWtaTTl2WnBFSHkwckM0UzdwY2xHMG9MNVl4c1pUSHlFaFJ4RnQ5WjY3ZHRtcm5qcDhNSVo3eXZORDczc0VlYlF3IIEC; bili_jct=5232d057d308c18c1419d19271a3b85e; DedeUserID=1576579979; DedeUserID__ckMd5=da7d6054e70acbba; home_feed_column=5; browser_resolution=1528-748; bp_t_offset_1576579979=978508367389523968; bili_ticket=eyJhbGciOiJIUzI1NiIsImtpZCI6InMwMyIsInR5cCI6IkpXVCJ9.eyJleHAiOjE3MjY5MjQzMzUsImlhdCI6MTcyNjY2NTA3NSwicGx0IjotMX0.9m3fjjjWd1wCsWNsTPwS9afVCknRz7dWtL6JV0CTQgI; bili_ticket_expires=1726924275; b_lsid=E4323108A_19205503541; sid=4q83ttnl; rpdid=|(u))kkY|mmJ0J'u~kYYYmmml"
}
#检查文本是否包含“ai”和“人工智能”字样
#此函数用于检查
def contains_ai_or_artificial_intelligence(text):
    ai_pattern = re.compile( r'(\bai\b|人工智能|([\u4e00-\u9fff]|\s|^)ai([\u4e00-\u9fff]|\s|$))', re.IGNORECASE)
    return re.search(ai_pattern, text)

#获取html文本
#该函数发送HTTP GET请求到指定URL，并返回网页的源码
def get_html(url):
    response = requests.get(url,headers=headers)
    response.encoding = 'utf-8'
    html=response.text
    return html

# 查找正确的api链接
#解析HTML数据，提取标签
def seek_api_urls(html_data):
        soup = BeautifulSoup(html_data, 'html.parser')
        urls = set()
        a_tags=soup.find_all('a', href=True)
        for a_link in a_tags:
            # 获取href的值
            link = a_link['href']
            urls.add(link)
        # 筛选正确的链接
        pattern = re.compile(r'https://api\.bilibili\.com/x/v1/dm/list\.so\?')
        api_urls = [url_find for url_find in urls if pattern.match(url_find)]
        #返回链接值
        return api_urls

# 获取弹幕接口链接函数
def get_api_urls(url):
    response = requests.get(url, headers=headers)
    if response.status_code == 200:
        html_data=response.text
        api_urls=seek_api_urls(html_data)
        return api_urls
    else:
        return [] 
 
# 获取视频接口函数
def get_urls(page):
    # 获得搜索页面url
    url = f"https://search.bilibili.com/video?keyword=%E5%B7%B4%E9%BB%8E%E5%A5%A5%E8%BF%90%E4%BC%9A&from_source=webtop_search&spm_id_from=333.1007&search_source=2&page={page}"
    html_data=get_html(url)
    soup = BeautifulSoup(html_data, 'html.parser')
    # 创建列表储存筛选完的内容
    urls = set()
    a_tags=soup.find_all('a', href=True)
    for a_link in a_tags:
        link = a_link['href']
        # 补全链接
        full_link=f'https:{link}'
        urls.add(full_link)
    # 筛选正确的链接
    pattern = re.compile(r'https://www\.bilibili\.com/video')
    #7x42=294，前七页全部读取
    if page != 8:
        vedieo_urls_f = [url_find for url_find in urls if pattern.match(url_find)]
        return vedieo_urls_f
    #第8页只读6个
    else: vedieo_urls_f = []
    num = 0
    for url_find in urls:
            if pattern.match(url_find):
                num = num + 1
                vedieo_urls_f.append(url_find)
                if num == 6:
                    return vedieo_urls_f

#获取接口链接
def vedio_transform_port(url):
    html_data = get_html(url)
    soup = BeautifulSoup(html_data,"html.parser")
    page_num = []   #储存总共的分p数
    span_tag = None #用做判断有无分p的flag

    div_tags = soup.findAll("div",attrs={"class":"head-left"}) #找到class=head-left的div
    for tag in div_tags:
        span_tag=(tag.findAll("span",attrs={"class":"cur-page"}))   #再从中找到class=cur-page的span

    if span_tag == None:    #值为None则为单个视频
        port_url = url.replace("bilibili.com", "ibilibili.com")
        port_urls.add(port_url)
    else:  
        for page in span_tag:
            pages = jieba.lcut(page.get_text()) #取得span的内容“（x/y)"，用jieba拆分成'(','x','/','y',')',其中y即为分p总数
            page_num = pages[3] #取得y的值
            # 替换每个分p视频的链接
            for page in range(1,int(page_num)+1):
                port_url = f"{url}?p={page}"
                port_urls.add(port_url.replace("bilibili.com", "ibilibili.com"))

for page in range(1,8):
    vedio_urls=get_urls(page)
    port_urls=set() 
    for vedio_url in vedio_urls:
        port_url = vedio_transform_port(vedio_url)


    for url in port_urls:
        api_urls=get_api_urls(url)
        if api_urls:  
            api_url = api_urls[0]
            html_data = get_html(api_url)
            soup = BeautifulSoup(html_data, 'html.parser')
            content_list =re.findall('<d p=".*?">(.*?)</d>',html_data)
            content='\n'.join(content_list)
            with open('弹幕.txt',mode='a',encoding='utf-8') as f:
                f.write(content)

ai_list = []    
most_common_barrages = []   

with open('弹幕.txt', 'r', encoding='utf-8') as file:
    content_txt = file.readlines()  
for barrage in content_txt:
    if contains_ai_or_artificial_intelligence(barrage): 
        ai_list.append(barrage.strip())  
# 使用Counter统计每个弹幕的出现次数
counter = Counter(ai_list)
# 获取出现次数最多的前8个弹幕
most_common_barrages = counter.most_common(8)
#转变类型才可以写入excel
ai_list1 = counter.most_common()
# 输出结果
for barrage, count in most_common_barrages:
    print(f'弹幕: {barrage} 出现次数: {count}')

df = pd.DataFrame(ai_list1, columns=['弹幕', '出现次数'])
# 写入Excel文件
excel_path = '弹幕统计.xlsx'
df.to_excel(excel_path, index=False, engine='openpyxl')

wb = load_workbook(excel_path)
ws = wb.active

ws.column_dimensions['A'].width = 60

wb.save(excel_path)

ai_str = '\n'.join(ai_list) #分割成字符型