K2cr2O1
ff4a4d92a5
实现线程管理器以支持高并发场景，添加GIL-free模式提升Python 3.14下的多线程性能 新增B站API集成和本地文件服务器功能，改进镜像插件支持GIF处理 更新文档说明多线程架构和GIL-free模式的使用方法
385 lines
14 KiB
Python
385 lines
14 KiB
Python
# -*- coding: utf-8 -*-
|
||
import re
|
||
from typing import Optional, Dict, Any, List, Union
|
||
from urllib.parse import urlparse, parse_qs
|
||
|
||
from core.utils.logger import logger
|
||
from models import MessageEvent, MessageSegment
|
||
from ..base import BaseParser
|
||
from ..utils import format_duration
|
||
|
||
from bilibili_api import video, select_client, Credential
|
||
from bilibili_api.exceptions import ResponseCodeException
|
||
from core.config_loader import global_config
|
||
from core.services.local_file_server import download_to_local
|
||
|
||
# bilibili_api-python 可用性标志
|
||
BILI_API_AVAILABLE = True
|
||
|
||
# 显式指定使用 aiohttp，避免与其他库冲突
|
||
try:
|
||
select_client("aiohttp")
|
||
except Exception as e:
|
||
logger.warning(f"设置 bilibili_api 客户端失败: {e}")
|
||
|
||
|
||
class BiliParser(BaseParser):
|
||
"""
|
||
B站视频解析器（使用 bilibili-api-python 库）
|
||
"""
|
||
|
||
def __init__(self):
|
||
super().__init__()
|
||
self.name = "B站解析器"
|
||
self.url_pattern = re.compile(r"https?://(?:www\.)?(bilibili\.com/video/\w+|b23\.tv/[a-zA-Z0-9]+)")
|
||
self.nickname = "B站视频解析"
|
||
|
||
|
||
|
||
def _get_credential(self) -> Optional[Credential]:
|
||
"""获取 B 站登录凭证"""
|
||
try:
|
||
bili_config = global_config.bilibili
|
||
if bili_config.sessdata and bili_config.bili_jct and bili_config.buvid3:
|
||
return Credential(
|
||
sessdata=bili_config.sessdata,
|
||
bili_jct=bili_config.bili_jct,
|
||
buvid3=bili_config.buvid3,
|
||
dedeuserid=bili_config.dedeuserid
|
||
)
|
||
except Exception:
|
||
pass
|
||
return None
|
||
|
||
async def parse(self, url: str) -> Optional[Dict[str, Any]]:
|
||
"""
|
||
解析B站视频信息
|
||
|
||
Args:
|
||
url (str): B站视频URL
|
||
|
||
Returns:
|
||
Optional[Dict[str, Any]]: 视频信息字典，如果失败则返回None
|
||
"""
|
||
# 提取 BV 号
|
||
bvid = self.extract_bvid(url)
|
||
if not bvid:
|
||
logger.error(f"[{self.name}] 无法从 URL 提取 BV 号: {url}")
|
||
return None
|
||
|
||
try:
|
||
if BILI_API_AVAILABLE:
|
||
# 使用 bilibili-api-python 库
|
||
credential = self._get_credential()
|
||
v = video.Video(bvid=bvid, credential=credential)
|
||
info = await v.get_info()
|
||
|
||
# 处理封面 URL
|
||
cover_url = info.get('pic', '')
|
||
if cover_url:
|
||
cover_url = cover_url.split('@')[0]
|
||
if cover_url.startswith('//'):
|
||
cover_url = 'https:' + cover_url
|
||
|
||
# 处理 UP 主头像
|
||
owner = info.get('owner', {})
|
||
owner_name = owner.get('name', '未知UP主')
|
||
owner_face = owner.get('face', '')
|
||
if owner_face:
|
||
if owner_face.startswith('//'):
|
||
owner_face = 'https:' + owner_face
|
||
owner_face = owner_face.split('@')[0]
|
||
|
||
# 处理统计信息
|
||
stat = info.get('stat', {})
|
||
|
||
return {
|
||
"title": info.get('title', '未知标题'),
|
||
"bvid": bvid,
|
||
"aid": info.get('aid', 0),
|
||
"duration": info.get('duration', 0),
|
||
"cover_url": cover_url,
|
||
"play": stat.get('view', 0),
|
||
"like": stat.get('like', 0),
|
||
"coin": stat.get('coin', 0),
|
||
"favorite": stat.get('favorite', 0),
|
||
"share": stat.get('share', 0),
|
||
"danmaku": stat.get('danmaku', 0),
|
||
"owner_name": owner_name,
|
||
"owner_avatar": owner_face,
|
||
"followers": info.get('owner', {}).get('fans', 0),
|
||
"description": info.get('desc', ''),
|
||
"pubdate": info.get('pubdate', 0),
|
||
}
|
||
else:
|
||
# 备用方案：直接解析页面
|
||
return await self._parse_fallback(url, bvid)
|
||
|
||
except ResponseCodeException as e:
|
||
logger.error(f"[{self.name}] API 返回错误: {e.code} - {e.msg}")
|
||
except Exception as e:
|
||
logger.error(f"[{self.name}] 解析视频信息失败: {e}")
|
||
if BILI_API_AVAILABLE:
|
||
logger.info(f"[{self.name}] 尝试备用解析方案")
|
||
return await self._parse_fallback(url, bvid)
|
||
|
||
return None
|
||
|
||
async def _parse_fallback(self, url: str, bvid: str) -> Optional[Dict[str, Any]]:
|
||
"""
|
||
备用解析方案（不使用 bilibili-api-python）
|
||
|
||
Args:
|
||
url (str): B站视频URL
|
||
bvid (str): BV号
|
||
|
||
Returns:
|
||
Optional[Dict[str, Any]]: 视频信息字典
|
||
"""
|
||
try:
|
||
session = self.get_session()
|
||
clean_url = url.split('?')[0]
|
||
if '#/' in clean_url:
|
||
clean_url = clean_url.split('#/')[0]
|
||
|
||
async with session.get(clean_url, headers=self.HEADERS, timeout=5) as response:
|
||
response.raise_for_status()
|
||
text = await response.text()
|
||
|
||
# 提取标题
|
||
import re
|
||
title_match = re.search(r'<h1[^>]*>([^<]+)</h1>', text)
|
||
title = title_match.group(1).strip() if title_match else '未知标题'
|
||
|
||
# 提取播放量等信息
|
||
play_match = re.search(r'"view":(\d+)', text)
|
||
play = int(play_match.group(1)) if play_match else 0
|
||
|
||
like_match = re.search(r'"like":(\d+)', text)
|
||
like = int(like_match.group(1)) if like_match else 0
|
||
|
||
coin_match = re.search(r'"coin":(\d+)', text)
|
||
coin = int(coin_match.group(1)) if coin_match else 0
|
||
|
||
favorite_match = re.search(r'"favorite":(\d+)', text)
|
||
favorite = int(favorite_match.group(1)) if favorite_match else 0
|
||
|
||
share_match = re.search(r'"share":(\d+)', text)
|
||
share = int(share_match.group(1)) if share_match else 0
|
||
|
||
# 提取 UP 主信息
|
||
owner_match = re.search(r'"name":"([^"]+)"', text)
|
||
owner_name = owner_match.group(1) if owner_match else '未知UP主'
|
||
|
||
face_match = re.search(r'"face":"([^"]+)"', text)
|
||
owner_face = face_match.group(1) if face_match else ''
|
||
if owner_face:
|
||
if owner_face.startswith('//'):
|
||
owner_face = 'https:' + owner_face
|
||
owner_face = owner_face.split('@')[0]
|
||
|
||
return {
|
||
"title": title,
|
||
"bvid": bvid,
|
||
"aid": 0,
|
||
"duration": 0,
|
||
"cover_url": '',
|
||
"play": play,
|
||
"like": like,
|
||
"coin": coin,
|
||
"favorite": favorite,
|
||
"share": share,
|
||
"danmaku": 0,
|
||
"owner_name": owner_name,
|
||
"owner_avatar": owner_face,
|
||
"followers": 0,
|
||
"description": '',
|
||
"pubdate": 0,
|
||
}
|
||
|
||
except Exception as e:
|
||
logger.error(f"[{self.name}] 备用解析方案失败: {e}")
|
||
|
||
return None
|
||
|
||
def extract_bvid(self, url: str) -> Optional[str]:
|
||
"""
|
||
从 URL 中提取 BV 号
|
||
|
||
Args:
|
||
url (str): B站视频URL
|
||
|
||
Returns:
|
||
Optional[str]: BV号，如果失败则返回None
|
||
"""
|
||
# 方式1: 直接从 URL 中提取
|
||
bvid_match = re.search(r'/video/(BV\w+)', url)
|
||
if bvid_match:
|
||
return bvid_match.group(1)
|
||
|
||
# 方式2: 从短链接跳转后提取
|
||
if 'b23.tv' in url:
|
||
try:
|
||
session = self.get_session()
|
||
# 简单处理，不实际跳转，直接尝试提取
|
||
bvid_match = re.search(r'BV\w{10}', url)
|
||
if bvid_match:
|
||
return bvid_match.group(0)
|
||
except Exception:
|
||
pass
|
||
|
||
return None
|
||
|
||
async def get_real_url(self, short_url: str) -> Optional[str]:
|
||
"""
|
||
获取B站短链接的真实URL
|
||
|
||
Args:
|
||
short_url (str): B站短链接
|
||
|
||
Returns:
|
||
Optional[str]: 真实URL，如果失败则返回None
|
||
"""
|
||
try:
|
||
session = self.get_session()
|
||
async with session.head(short_url, headers=self.HEADERS, allow_redirects=False, timeout=5) as response:
|
||
if response.status == 302:
|
||
return response.headers.get('Location')
|
||
except Exception as e:
|
||
logger.error(f"[{self.name}] 获取真实URL失败: {e}")
|
||
return None
|
||
|
||
async def get_direct_video_url(self, video_url: str, bvid: str) -> Optional[str]:
|
||
"""
|
||
获取B站视频直链（通过本地文件服务器下载）
|
||
|
||
Args:
|
||
video_url (str): B站视频的完整URL
|
||
bvid (str): BV号
|
||
|
||
Returns:
|
||
Optional[str]: 本地视频 URL，如果失败则返回None
|
||
"""
|
||
if not BILI_API_AVAILABLE:
|
||
return None
|
||
|
||
try:
|
||
credential = self._get_credential()
|
||
v = video.Video(bvid=bvid, credential=credential)
|
||
# 先获取视频信息以获取 cid
|
||
info = await v.get_info()
|
||
cid = info.get('cid', 0)
|
||
|
||
if not cid:
|
||
return None
|
||
|
||
# 获取下载链接数据
|
||
download_url_data = await v.get_download_url(cid=cid)
|
||
|
||
# 使用 VideoDownloadURLDataDetecter 解析数据
|
||
detecter = video.VideoDownloadURLDataDetecter(data=download_url_data)
|
||
streams = detecter.detect_best_streams()
|
||
|
||
if streams:
|
||
# 获取视频直链
|
||
video_direct_url = streams[0].url
|
||
logger.info(f"[{self.name}] 获取到视频直链，开始下载到本地...")
|
||
|
||
# 使用本地文件服务器下载
|
||
local_url = await download_to_local(video_direct_url, timeout=120)
|
||
|
||
if local_url:
|
||
logger.success(f"[{self.name}] 视频已下载到本地: {local_url}")
|
||
return local_url
|
||
else:
|
||
logger.error(f"[{self.name}] 下载到本地失败")
|
||
return None
|
||
|
||
except Exception as e:
|
||
logger.error(f"[{self.name}] 获取视频直链失败: {e}")
|
||
|
||
return None
|
||
|
||
async def format_response(self, event: MessageEvent, data: Dict[str, Any]) -> List[Any]:
|
||
"""
|
||
格式化B站视频响应消息
|
||
|
||
Args:
|
||
event (MessageEvent): 消息事件对象
|
||
data (Dict[str, Any]): 视频信息
|
||
|
||
Returns:
|
||
List[Any]: 消息段列表
|
||
"""
|
||
# 检查视频时长
|
||
video_message: Union[str, MessageSegment]
|
||
direct_url = None
|
||
if data['duration'] > 1200: # 20分钟 = 1200秒
|
||
video_message = "视频时长超过20分钟，不进行解析。"
|
||
else:
|
||
# 构建完整的B站视频URL
|
||
video_url = f"https://www.bilibili.com/video/{data.get('bvid', '')}"
|
||
bvid = data.get('bvid', '')
|
||
direct_url = await self.get_direct_video_url(video_url, bvid)
|
||
if direct_url:
|
||
video_message = MessageSegment.video(direct_url)
|
||
else:
|
||
video_message = "视频解析失败，无法获取直链。"
|
||
|
||
text_message = (
|
||
f"BiliBili 视频解析\n"
|
||
f"--------------------\n"
|
||
f" UP主: {data['owner_name']}\n"
|
||
f" 粉丝: {self.format_count(data['followers'])}\n"
|
||
f"--------------------\n"
|
||
f" 标题: {data['title']}\n"
|
||
f" BV号: {data['bvid']}\n"
|
||
f" 时长: {format_duration(data['duration'])}\n"
|
||
f"--------------------\n"
|
||
f" 数据:\n"
|
||
f" 播放: {self.format_count(data['play'])}\n"
|
||
f" 点赞: {self.format_count(data['like'])}\n"
|
||
f" 投币: {self.format_count(data['coin'])}\n"
|
||
f" 收藏: {self.format_count(data['favorite'])}\n"
|
||
f" 转发: {self.format_count(data['share'])}\n"
|
||
f" 弹幕: {self.format_count(data.get('danmaku', 0))}\n"
|
||
)
|
||
|
||
image_message_segment = [
|
||
MessageSegment.text("B站封面："),
|
||
MessageSegment.image(data['cover_url'])
|
||
]
|
||
|
||
up_info_segment = [
|
||
MessageSegment.text("UP主头像："),
|
||
MessageSegment.image(data['owner_avatar'])
|
||
]
|
||
|
||
nodes = [
|
||
event.bot.build_forward_node(user_id=event.self_id, nickname=self.nickname, message=text_message),
|
||
event.bot.build_forward_node(user_id=event.self_id, nickname=self.nickname, message=image_message_segment),
|
||
event.bot.build_forward_node(user_id=event.self_id, nickname=self.nickname, message=up_info_segment),
|
||
event.bot.build_forward_node(user_id=event.self_id, nickname=self.nickname, message=video_message)
|
||
]
|
||
|
||
# 同时直接发送视频（如果获取到直链）
|
||
if direct_url:
|
||
try:
|
||
await event.reply(MessageSegment.video(direct_url))
|
||
except Exception as e:
|
||
logger.error(f"[{self.name}] 直接发送视频失败: {e}")
|
||
|
||
return nodes
|
||
|
||
def should_handle_url(self, url: str) -> bool:
|
||
"""
|
||
判断是否应该处理该URL
|
||
|
||
Args:
|
||
url (str): URL
|
||
|
||
Returns:
|
||
bool: 是否应该处理
|
||
"""
|
||
return bool(self.url_pattern.search(url))
|