HeXi/hexi/plugins/nonebot_plugin_video_analysis/handlers/douyin.py

"""抖音视频/图文解析编排层"""

import re
import urllib.parse
from pathlib import Path
from typing import Optional, Union

import httpx
from nonebot import logger

from ..services.fetchers.douyin_api import fetch_douyin_content
from ..services.fetchers.douyin_ssr import MOBILE_UA, fetch_douyin_note_ssr
from ..models import DouyinFetchError
from ..utils import get_data_dir, parse_netscape_cookies
from .sender import PendingMedia, _as_paths

SHORT_LINK_PATTERN = re.compile(r"(v\.douyin\.com/[A-Za-z0-9_\-]+)")


async def _resolve_short_link(url: str) -> Optional[str]:
    """短链重定向：httpx 取 Location（最多 3 跳），失败返回 None"""
    try:
        async with httpx.AsyncClient(
            headers={"User-Agent": MOBILE_UA},
            follow_redirects=False,
            timeout=10,
        ) as client:
            current = url
            for _ in range(3):
                resp = await client.get(current)
                if resp.status_code >= 400:
                    return None
                location = resp.headers.get("Location")
                if not location:
                    final = str(resp.url)
                    # 200 但仍是短链本身（如 JS 挑战壳页）→ 静默失败，显式记录；
                    # 图文短链若此处失败将退化到 playwright 旧链路
                    if "v.douyin.com" in final:
                        logger.warning(f"短链返回挑战页（未跳转）: {final}")
                    return final
                # Location 可能是相对路径，需拼上当前 URL
                current = urllib.parse.urljoin(current, location)
            return current
    except Exception:
        logger.warning(f"短链重定向失败: {url}")
        return None


async def parse_douyin(
    url: str,
) -> tuple[Optional[str], Optional[Union[Path, list[Path]]], bool]:
    """解析抖音链接（短链 / 全链接），返回 (title, file_paths, is_image_post)

    分派规则（不能依赖 URL 路径：短链重定向后图文也统一变成
    iesdouyin.com/share/video/{id} 形态）：
       - note 形态链接 → SSR 静态解析，失败不回退 playwright
         （其图文链路存在 aweme/post 取到作者其他作品的缺陷）
       - 其余形态 → 先试 SSR 按内容判定：有 images 即图文 → SSR 秒级下载；
         真视频 / SSR 失败 → playwright 拦截（保留全部清晰度能力）
    """
    # 优先匹配短链接 v.douyin.com/xxx
    short = SHORT_LINK_PATTERN.search(url)
    if short:
        target_url = "https://" + short.group(1)
        # 短链先重定向拿到全链接，确定图文/视频类型后分派
        resolved = await _resolve_short_link(target_url)
        if resolved:
            logger.info(f"短链重定向: {target_url} -> {resolved}")
            target_url = resolved
    else:
        # 全链接: www.douyin.com/note/xxx 或 www.douyin.com/video/xxx
        m = re.search(
            r"(https?://(?:www\.)?douyin\.com/(?:note|video)/\d+)", url
        )
        if not m:
            return None, None, False
        target_url = m.group(1)

    try:
        is_img_post = False
        cookies_path = get_data_dir() / "cookies.txt"
        cookies = parse_netscape_cookies(cookies_path)
        # 先试 SSR（秒级、免浏览器），失败一律回退 playwright 链路。
        # 2026-08-13 起抖音对 iesdouyin SSR 端点整体降级（登录态也拿不到
        # videoInfoRes，只返回 33KB 壳页），note/video 形态统一回退，
        # 图文作品的"取到作者其他作品"缺陷由 douyin_api 按 aweme_id 精确匹配修复。
        try:
            title, file_path = await fetch_douyin_note_ssr(target_url, cookies)
        except DouyinFetchError:
            logger.info("SSR 解析失败，回退 playwright 链路")
            title, file_path = None, None
        if file_path is None:
            title, file_path = await fetch_douyin_content(
                target_url, cookies, 10, False
            )
        if isinstance(file_path, list):
            is_img_post = True
        return title, file_path, is_img_post
    except Exception:
        logger.exception(f"获取直链失败 {target_url}")
    return None, None, False


async def process_douyin_res(
    title: str,
    file_paths: Union[Path, list[Path]],
    is_private: bool,
    image_post: bool,
    plan: str | None = None,
) -> tuple[Optional[PendingMedia], Optional[str]]:
    """下载已完成 → 打包为待发送媒体（不上传、不发送、不清理）

    多级发送（temp 本地 → S3 链接 → 回退本地）由 send_pending_media 统一处理。
    """
    if not file_paths:
        return None, None
    return (
        PendingMedia(
            files=_as_paths(file_paths),
            image_post=image_post,
            is_private=is_private,
            plan=plan,
            title=title,
        ),
        None,
    )