import requests
import hashlib
import time
import os
import csv
from hanlp_restful import HanLPClient

def md5_to_w_rid(y, a):
    input_string = y + a
    md5_hash = hashlib.md5()
    md5_hash.update(input_string.encode('utf-8'))
    w_rid = ''.join(f'{byte:02x}' for byte in md5_hash.digest())
    return w_rid

def save_to_csv(s, uname, content, emotion):
    file_exists = os.path.exists('ba数据分析.csv')
    print('...正在写入文件')
    with open('ba数据分析.csv', 'a', newline='', encoding='utf-8') as csvfile:
        fieldnames = ['情感', '楼层', '用户名', '评论内容']
        writer = csv.DictWriter(csvfile, fieldnames=fieldnames)

        if not file_exists:
            writer.writeheader()

        writer.writerow({
            '情感': emotion,
            '楼层': s,
            '用户名': uname,
            '评论内容': content
        })

def fetch_comments(oid, headers, cookies, s):
    a = 'ea1db124af3c7062474693fa704f4ff8'
    wts = int(time.time())
    session_id = "1766149779423829"
    y = f'mode=3&oid={oid}&pagination_str=%7B%22offset%22%3A%22%7B%5C%22type%5C%22%3A1%2C%5C%22direction%5C%22%3A1%2C%5C%22session_id%5C%22%3A%5C%22{session_id}%5C%22%2C%5C%22data%5C%22%3A%7B%7D%7D%22%7D&plat=1&type=1&web_location=1315875&wts={wts}'
    w_rid = md5_to_w_rid(y, a)

    url = "https://api.bilibili.com/x/v2/reply/wbi/main"
    params = {
        "oid": oid,
        "type": "1",  # 动态评论为 17，视频评论为 1
        "mode": "3",
        "plat": "1",
        "pagination_str": "{\"offset\":\"{\\\"type\\\":1,\\\"direction\\\":1,\\\"session_id\\\":\\\"" + session_id + "\\\",\\\"data\\\":{}}\"}",
        "web_location": "1315875",
        "w_rid": w_rid,
        "wts": wts
    }

    try:
        response = requests.get(url, headers=headers, cookies=cookies, params=params).json()
        comments = response.get('data', {}).get('replies', [])
        
        if not comments:
            print("没有更多评论了。")
            return s, False

        HanLP = HanLPClient('https://www.hanlp.com/api', auth='', language='zh')
        
        for comment in comments:
            uname = comment['member']['uname']
            content = comment['content']['message']
            emotion = HanLP.sentiment_analysis(content)
            
            s += 1
            print(f'第{s}楼\n用户名：{uname}\n评论内容：{content}\n\n')
            save_to_csv(s, uname, content, emotion)
        
        return s, True
    except Exception as e:
        print(f"请求失败: {e}")
        return s, False

# 主程序
if __name__ == "__main__":
    oid = 1556265189  # 视频或动态的 oid
    headers = {
        
    }
    cookies = {
    }

    s = 0  # 初始化 s 值为 0
    for i in range(20):
        s, success = fetch_comments(oid, headers, cookies, s)
        if not success:
            break
        time.sleep(1)