import os
import re
from collections import Counter
import pandas as pd


#提取txt文件里面的弹幕，并存于列表中
def read_danmu_files(directory):
    danmu_data = []
    for filename in os.listdir(directory):
        if filename.endswith('paris_olympics_danmak.txt'):
            with open(os.path.join(directory, filename), 'r', encoding='utf-8') as file:
                danmu_data.extend(file.readlines())
    return danmu_data

def count_danmu():
    danmu_data = read_danmu_files('./')  # 当前目录下
    #用正则表达式提取含有ai的弹幕
    ai_danmaku = [danmaku for danmaku in danmu_data if
                  re.search(r'(?<![a-zA-Z])((?:AI|人工智能))(?![a-zA-Z])', danmaku)]
    unique_ai_danmaku = list(set(ai_danmaku))
    #获取与ai弹幕相关的数量
    danmaku_counts = [(danmaku, ai_danmaku.count(danmaku)) for danmaku in unique_ai_danmaku]
    #对与ai相关弹幕按照数量数量进行排序
    sorted_danmaku_counts = sorted(danmaku_counts, key=lambda x: x[1], reverse=True)
    #得出排名前八的弹幕
    top_8_danmaku = sorted_danmaku_counts[:8]
    #将数据导入excel表
    data = {'弹幕内容': [danmaku[0] for danmaku in top_8_danmaku]}
    df = pd.DataFrame(data)
    df.to_excel('2025巴黎奥运AI弹幕统计.xlsx', index=False)