Undefined-bot 3.6.5__tar.gz → 3.6.6__tar.gz

{undefined_bot-3.6.5 → undefined_bot-3.6.6}/CHANGELOG.md

@@ -1,3 +1,14 @@
+## v3.6.6 抖音视频提取与媒体信息模式
+
+本版本把抖音视频纳入自动提取、工具调用和文件分析链路：可识别短链、长链与 aweme_id，下载并发送视频合并转发，也可只返回附件 UID 或元信息。同时补齐 B 站与 arXiv 工具的信息模式，并加固抖音真实视频地址探测。
+
+- 新增 Douyin 视频链路。新增 `douyin` 模块与 `douyin_video` 工具，支持抖音短链、`douyin.com/video/<id>` 和裸 aweme_id 解析，可按 `send` / `uid` / `info` 输出发送合并转发、注册附件 UID 或返回视频标题、作者、时长、简介等信息。
+- 接入抖音自动处理管线。普通消息命中抖音链接或 aweme_id 后，可按 `[douyin]` 配置自动下载并发送两节点合并转发；支持独立代理开关、时长与文件大小限制、清晰度探测顺序、群聊/私聊白名单和单条消息处理上限。
+- 加固抖音下载与真实视频探测。下载链路从 SSR share 页提取 `video.play_addr` token，再按清晰度探测 `aweme/v1/play/`；探测优先读取 `Content-Range` 总长度并按文件大小去重，ttid 获取失败时可回退直接请求 share 页，受限视频会返回更明确的过滤原因。
+- 扩展媒体工具信息输出。`bilibili_video` 与 `arxiv_paper` 新增 `output_mode=info`，可只获取视频或论文元信息而不发送/下载附件；`file_analysis_agent` 提示同步覆盖抖音、B 站和 arXiv 的 `uid` / `info` 使用方式。
+
+---
+
 ## v3.6.5 Firecrawl 搜索、代理作用域与热更新连接修复
 
 本版本聚焦联网能力与出站配置治理：`web_agent` 新增 Firecrawl 搜索提供商和搜索优先级控制，代理配置从全局开关收敛为模型与功能段独立启用，并修复配置热更新时旧模型连接可能被提前关闭的问题。

{undefined_bot-3.6.5 → undefined_bot-3.6.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: Undefined-bot
-Version: 3.6.5
+Version: 3.6.6
 Summary: QQ bot platform with cognitive memory architecture and multi-agent Skills, via OneBot V11.
 Project-URL: Repository, https://github.com/69gg/Undefined
 Project-URL: Issues, https://github.com/69gg/Undefined/issues

{undefined_bot-3.6.5 → undefined_bot-3.6.6}/config.toml.example

@@ -1233,6 +1233,34 @@ auto_extract_group_ids = []
 # en: Private chat allowlist for auto-extraction (empty = follow global access.allowed_private_ids).
 auto_extract_private_ids = []
 
+# zh: 抖音视频自动提取配置。
+# en: Douyin video auto-extraction settings.
+[douyin]
+# zh: 是否让抖音短链解析、share 页、ttid 注册和视频流请求使用 [proxy] 代理地址。默认关闭。
+# en: Whether Douyin short-link, share page, ttid registration, and video stream requests use proxy addresses from [proxy]. Disabled by default.
+use_proxy = false
+# zh: 是否启用自动提取（检测到抖音短链、长链或 aweme_id 时自动下载并发送）。
+# en: Enable auto-extraction (auto-download and send when Douyin short links, long links, or aweme IDs are detected).
+auto_extract_enabled = false
+# zh: 最大视频时长（秒），超过则不下载，仅发送信息卡片。0=不限。
+# en: Max video duration (seconds); exceeding this sends info card only. 0=unlimited.
+max_duration = 600
+# zh: 最大文件大小（MB），超过则仅发送信息卡片。0=不限。
+# en: Max file size (MB); exceeding this sends info card only. 0=unlimited.
+max_file_size = 100
+# zh: 清晰度探测顺序。游客 share 页无 bit_rate 时会按该列表探测 play 端点，并按文件大小去重。
+# en: Quality probing order. Visitor share pages lack bit_rate, so play endpoints are probed in this order and deduplicated by file size.
+prefer_ratios = ["1080p", "720p", "540p", "360p"]
+# zh: 自动提取功能的群聊白名单（空=跟随全局 access.allowed_group_ids）。
+# en: Group allowlist for auto-extraction (empty = follow global access.allowed_group_ids).
+auto_extract_group_ids = []
+# zh: 自动提取功能的私聊白名单（空=跟随全局 access.allowed_private_ids）。
+# en: Private chat allowlist for auto-extraction (empty = follow global access.allowed_private_ids).
+auto_extract_private_ids = []
+# zh: 单条消息最多自动处理几个抖音视频。<=0 回退 3，>10 截断到 10。
+# en: Max number of Douyin videos to auto-process from one message. <=0 falls back to 3, >10 is capped to 10.
+auto_extract_max_items = 3
+
 # zh: arXiv 论文自动提取配置。
 # en: arXiv paper auto-extraction settings.
 [arxiv]

{undefined_bot-3.6.5 → undefined_bot-3.6.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "Undefined-bot"
-version = "3.6.5"
+version = "3.6.6"
 description = "QQ bot platform with cognitive memory architecture and multi-agent Skills, via OneBot V11."
 readme = "README.md"
 authors = [

{undefined_bot-3.6.5 → undefined_bot-3.6.6}/src/Undefined/__init__.py

@@ -24,7 +24,7 @@ if TYPE_CHECKING:
     from .skills.registry import BaseRegistry as BaseRegistry
     from .skills.tools import ToolRegistry as ToolRegistry
 
-__version__: str = "3.6.5"
+__version__: str = "3.6.6"
 
 # symbol -> (module_path, attribute_name)；首次访问时才 importlib 加载
 _LAZY_IMPORTS: dict[str, tuple[str, str]] = {

{undefined_bot-3.6.5 → undefined_bot-3.6.6}/src/Undefined/arxiv/sender.py

@@ -120,6 +120,25 @@ def _build_info_message(
     return "\n".join(lines)
 
 
+def format_paper_info(
+    info: PaperInfo,
+    *,
+    author_preview_limit: int,
+    summary_preview_chars: int,
+) -> str:
+    """Format arXiv metadata for tool results."""
+    lines = [
+        _build_info_message(
+            info,
+            author_preview_limit=author_preview_limit,
+            summary_preview_chars=summary_preview_chars,
+        )
+    ]
+    if info.pdf_url:
+        lines.append(f"PDF: {info.pdf_url}")
+    return "\n".join(lines)
+
+
 async def _send_text_message(
     sender: "MessageSender",
     target_type: Literal["group", "private"],

{undefined_bot-3.6.5 → undefined_bot-3.6.6}/src/Undefined/bilibili/sender.py

@@ -14,7 +14,7 @@ from Undefined.bilibili.downloader import (
     download_video,
     get_video_info,
 )
-from Undefined.bilibili.models import DanmakuItem
+from Undefined.bilibili.models import DanmakuItem, VideoStats
 from Undefined.bilibili.parser import normalize_to_bvid
 from Undefined.utils.io import get_file_size
 
@@ -53,6 +53,19 @@ def _format_progress(progress_ms: int) -> str:
     return _format_duration(seconds)
 
 
+def _format_stats_line(stats: VideoStats) -> str:
+    return (
+        "数据: "
+        f"播放 {_format_count(stats.view)} | "
+        f"点赞 {_format_count(stats.like)} | "
+        f"投币 {_format_count(stats.coin)} | "
+        f"收藏 {_format_count(stats.favorite)} | "
+        f"弹幕 {_format_count(stats.danmaku)} | "
+        f"评论 {_format_count(stats.reply)} | "
+        f"分享 {_format_count(stats.share)}"
+    )
+
+
 def _chunked(items: list[DanmakuItem], size: int) -> Iterable[list[DanmakuItem]]:
     size = max(1, size)
     for index in range(0, len(items), size):
@@ -84,16 +97,7 @@ def _build_info_segments(
             f"「{info.title}」",
             f"UP主: {info.up_name or '未知'}",
             f"时长: {_format_duration(info.duration)}",
-            (
-                "数据: "
-                f"播放 {_format_count(stats.view)} | "
-                f"点赞 {_format_count(stats.like)} | "
-                f"投币 {_format_count(stats.coin)} | "
-                f"收藏 {_format_count(stats.favorite)} | "
-                f"弹幕 {_format_count(stats.danmaku)} | "
-                f"评论 {_format_count(stats.reply)} | "
-                f"分享 {_format_count(stats.share)}"
-            ),
+            _format_stats_line(stats),
         ]
     )
     desc = info.desc.strip()
@@ -121,16 +125,7 @@ def _build_video_history_message(
         f"[Bilibili] 「{info.title}」",
         f"UP主: {info.up_name}",
         f"时长: {_format_duration(info.duration)}",
-        (
-            "数据: "
-            f"播放 {_format_count(stats.view)} | "
-            f"点赞 {_format_count(stats.like)} | "
-            f"投币 {_format_count(stats.coin)} | "
-            f"收藏 {_format_count(stats.favorite)} | "
-            f"弹幕 {_format_count(stats.danmaku)} | "
-            f"评论 {_format_count(stats.reply)} | "
-            f"分享 {_format_count(stats.share)}"
-        ),
+        _format_stats_line(stats),
     ]
     if quality_name and file_size_mb is not None:
         lines.append(f"清晰度: {quality_name} | 大小: {file_size_mb:.1f}MB")
@@ -143,6 +138,26 @@ def _build_video_history_message(
     return "\n".join(lines)
 
 
+def format_bilibili_video_info(info: "VideoInfo") -> str:
+    """Format Bilibili metadata for tool results."""
+    stats = info.stats
+    lines = [
+        f"「{info.title}」",
+        f"BV: {info.bvid}",
+        f"AV: av{info.aid}",
+        f"UP主: {info.up_name or '未知'}",
+        f"时长: {_format_duration(info.duration)}",
+        _format_stats_line(stats),
+    ]
+    desc = info.desc.strip()
+    if desc:
+        lines.extend(["---", desc])
+    if info.cover_url:
+        lines.append(f"封面: {info.cover_url}")
+    lines.append(info.url)
+    return "\n".join(lines)
+
+
 def _build_uid_message(
     info: "VideoInfo",
     *,

{undefined_bot-3.6.5 → undefined_bot-3.6.6}/src/Undefined/config/config_class.py

@@ -187,6 +187,15 @@ class Config:
     bilibili_danmaku_max_count: int
     bilibili_auto_extract_group_ids: list[int]
     bilibili_auto_extract_private_ids: list[int]
+    # Douyin 视频提取
+    douyin_use_proxy: bool
+    douyin_auto_extract_enabled: bool
+    douyin_max_duration: int
+    douyin_max_file_size: int
+    douyin_prefer_ratios: list[str]
+    douyin_auto_extract_group_ids: list[int]
+    douyin_auto_extract_private_ids: list[int]
+    douyin_auto_extract_max_items: int
     # arXiv 论文提取
     arxiv_use_proxy: bool
     arxiv_auto_extract_enabled: bool
@@ -250,6 +259,16 @@ class Config:
         init=False,
         repr=False,
     )
+    _douyin_group_ids_set: set[int] = dataclass_field(
+        default_factory=set,
+        init=False,
+        repr=False,
+    )
+    _douyin_private_ids_set: set[int] = dataclass_field(
+        default_factory=set,
+        init=False,
+        repr=False,
+    )
     _arxiv_group_ids_set: set[int] = dataclass_field(
         default_factory=set,
         init=False,
@@ -272,6 +291,9 @@ class Config:
     )
 
     def __post_init__(self) -> None:
+        self._refresh_runtime_sets()
+
+    def _refresh_runtime_sets(self) -> None:
         # 访问控制属于高频热路径，启动后缓存为 set 降低重复构建开销。
         normalized_mode = str(self.access_mode).strip().lower()
         if normalized_mode not in {"off", "blacklist", "allowlist", "legacy"}:
@@ -287,6 +309,12 @@ class Config:
         self._bilibili_private_ids_set = {
             int(item) for item in self.bilibili_auto_extract_private_ids
         }
+        self._douyin_group_ids_set = {
+            int(item) for item in self.douyin_auto_extract_group_ids
+        }
+        self._douyin_private_ids_set = {
+            int(item) for item in self.douyin_auto_extract_private_ids
+        }
         self._arxiv_group_ids_set = {
             int(item) for item in self.arxiv_auto_extract_group_ids
         }
@@ -457,6 +485,18 @@ class Config:
         # 功能白名单为空时跟随全局 access 控制
         return self.is_private_allowed(user_id)
 
+    def is_douyin_auto_extract_allowed_group(self, group_id: int) -> bool:
+        """群聊是否允许 Douyin 自动提取。"""
+        if self._douyin_group_ids_set:
+            return int(group_id) in self._douyin_group_ids_set
+        return self.is_group_allowed(group_id)
+
+    def is_douyin_auto_extract_allowed_private(self, user_id: int) -> bool:
+        """私聊是否允许 Douyin 自动提取。"""
+        if self._douyin_private_ids_set:
+            return int(user_id) in self._douyin_private_ids_set
+        return self.is_private_allowed(user_id)
+
     def is_arxiv_auto_extract_allowed_group(self, group_id: int) -> bool:
         """群聊是否允许 arXiv 自动提取。"""
         if self._arxiv_group_ids_set:
@@ -535,6 +575,8 @@ class Config:
             if old_value != new_value:
                 setattr(self, name, new_value)
                 changes[name] = (old_value, new_value)
+        if changes:
+            self._refresh_runtime_sets()
         return changes
 
     def reload(self, strict: bool = False) -> dict[str, tuple[Any, Any]]:

{undefined_bot-3.6.5 → undefined_bot-3.6.6}/src/Undefined/config/load_sections/integrations.py

@@ -14,6 +14,7 @@ from ..coercers import (
     _coerce_int,
     _coerce_int_list,
     _coerce_str,
+    _coerce_str_list,
     _get_value,
 )
 
@@ -77,6 +78,44 @@ def load_integrations(
         _get_value(data, ("bilibili", "auto_extract_private_ids"), None)
     )
 
+    # Douyin 配置
+    douyin_use_proxy = _coerce_bool(
+        _get_value(data, ("douyin", "use_proxy"), "DOUYIN_USE_PROXY"), False
+    )
+    douyin_auto_extract_enabled = _coerce_bool(
+        _get_value(data, ("douyin", "auto_extract_enabled"), None), False
+    )
+    douyin_max_duration = _coerce_int(
+        _get_value(data, ("douyin", "max_duration"), None), 600
+    )
+    if douyin_max_duration < 0:
+        douyin_max_duration = 600
+    douyin_max_file_size = _coerce_int(
+        _get_value(data, ("douyin", "max_file_size"), None), 100
+    )
+    if douyin_max_file_size < 0:
+        douyin_max_file_size = 100
+    douyin_prefer_ratios = _coerce_str_list(
+        _get_value(data, ("douyin", "prefer_ratios"), None)
+    )
+    allowed_ratios = ("1080p", "720p", "540p", "360p")
+    douyin_prefer_ratios = [
+        ratio for ratio in douyin_prefer_ratios if ratio in allowed_ratios
+    ] or list(allowed_ratios)
+    douyin_auto_extract_group_ids = _coerce_int_list(
+        _get_value(data, ("douyin", "auto_extract_group_ids"), None)
+    )
+    douyin_auto_extract_private_ids = _coerce_int_list(
+        _get_value(data, ("douyin", "auto_extract_private_ids"), None)
+    )
+    douyin_auto_extract_max_items = _coerce_int(
+        _get_value(data, ("douyin", "auto_extract_max_items"), None), 3
+    )
+    if douyin_auto_extract_max_items <= 0:
+        douyin_auto_extract_max_items = 3
+    if douyin_auto_extract_max_items > 10:
+        douyin_auto_extract_max_items = 10
+
     # arXiv 配置
     arxiv_use_proxy = _coerce_bool(
         _get_value(data, ("arxiv", "use_proxy"), "ARXIV_USE_PROXY"), False
@@ -272,6 +311,14 @@ def load_integrations(
         "bilibili_danmaku_max_count": bilibili_danmaku_max_count,
         "bilibili_auto_extract_group_ids": bilibili_auto_extract_group_ids,
         "bilibili_auto_extract_private_ids": bilibili_auto_extract_private_ids,
+        "douyin_use_proxy": douyin_use_proxy,
+        "douyin_auto_extract_enabled": douyin_auto_extract_enabled,
+        "douyin_max_duration": douyin_max_duration,
+        "douyin_max_file_size": douyin_max_file_size,
+        "douyin_prefer_ratios": douyin_prefer_ratios,
+        "douyin_auto_extract_group_ids": douyin_auto_extract_group_ids,
+        "douyin_auto_extract_private_ids": douyin_auto_extract_private_ids,
+        "douyin_auto_extract_max_items": douyin_auto_extract_max_items,
         "arxiv_use_proxy": arxiv_use_proxy,
         "arxiv_auto_extract_enabled": arxiv_auto_extract_enabled,
         "arxiv_max_file_size": arxiv_max_file_size,

undefined_bot-3.6.6/src/Undefined/douyin/__init__.py

@@ -0,0 +1,15 @@
+"""Douyin video extraction module."""
+
+from Undefined.douyin.parser import (
+    canonical_share_url,
+    extract_douyin_ids,
+    extract_from_json_message,
+    normalize_aweme_id,
+)
+
+__all__ = [
+    "canonical_share_url",
+    "extract_douyin_ids",
+    "extract_from_json_message",
+    "normalize_aweme_id",
+]