parsehub 2.3.16__tar.gz → 2.4.0__tar.gz

{parsehub-2.3.16/src/parsehub.egg-info → parsehub-2.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: parsehub
-Version: 2.3.16
+Version: 2.4.0
 Summary: 轻量、异步、开箱即用的社交媒体聚合解析库
 Author-email: 梓澪 <zilingmio@gmail.com>
 License: MIT
@@ -82,7 +82,7 @@ Dynamic: license-file
 | **YouTube**     |  ✅  |      | 🎵 音乐                   |
 | **Facebook**    |  ✅  |      |                           |
 | **Threads**     |  ✅  |  ✅  |                           |
-| **Bilibili**    |  ✅  |      | 📝 动态                   |
+| **Bilibili**    |  ✅  |      | 📝 动态, 🎵 音乐          |
 | **抖音**        |  ✅  |  ✅  | ☀️日常                    |
 | **TikTok**      |  ✅  |  ✅  |                           |
 | **微博**        |  ✅  |  ✅  |                           |

{parsehub-2.3.16 → parsehub-2.4.0}/README.md

@@ -37,7 +37,7 @@
 | **YouTube**     |  ✅  |      | 🎵 音乐                   |
 | **Facebook**    |  ✅  |      |                           |
 | **Threads**     |  ✅  |  ✅  |                           |
-| **Bilibili**    |  ✅  |      | 📝 动态                   |
+| **Bilibili**    |  ✅  |      | 📝 动态, 🎵 音乐          |
 | **抖音**        |  ✅  |  ✅  | ☀️日常                    |
 | **TikTok**      |  ✅  |  ✅  |                           |
 | **微博**        |  ✅  |  ✅  |                           |

{parsehub-2.3.16 → parsehub-2.4.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "parsehub"
-version = "2.3.16"
+version = "2.4.0"
 description = "轻量、异步、开箱即用的社交媒体聚合解析库"
 readme = "README.md"
 requires-python = ">=3.12.0"

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/parsers/base/ytdlp.py

@@ -411,7 +411,7 @@ class YtParser(BaseParser, register=False):
         title = dl["title"]
         duration = dl.get("duration", 0)
         thumbnail = dl.get("thumbnail")
-        description = dl["description"]
+        description = dl.get("description", "")
         width = dl.get("width", 0)
         height = dl.get("height", 0)
         return YtVideoInfo(

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/parsers/parser/bilibili.py

@@ -12,6 +12,8 @@ from ...types import (
     ImageParseResult,
     ImageRef,
     LivePhotoRef,
+    MusicParseResult,
+    MusicRef,
     ParseError,
     Platform,
     ProgressCallback,
@@ -25,17 +27,21 @@ from ..base.ytdlp import YtParser, YtVideoParseResult
 
 class BiliParse(BaseParser):
     __platform__ = Platform.BILIBILI
-    __supported_type__ = ["视频", "动态"]
-    __match__ = r"^(http(s)?://)?((((w){3}.|(m).|(t).)?bilibili\.com)/(video|opus|\b\d{18,19}\b)|b23.tv|bili2233.cn).*"
+    __supported_type__ = ["视频", "动态", "音乐"]
+    __match__ = r"^(http(s)?://)?((((w){3}.|(m).|(t).)?bilibili\.com)/(video|opus|audio/au\d+|\b\d{18,19}\b)|b23.tv|bili2233.cn).*"
     __reserved_parameters__ = ["p"]
     __redirect_keywords__ = ["b23.tv", "bili2233.cn"]
 
-    async def _do_parse(self, raw_url: str) -> YtVideoParseResult | BiliVideoParseResult | ImageParseResult:
+    async def _do_parse(
+        self, raw_url: str
+    ) -> YtVideoParseResult | BiliVideoParseResult | BiliMusicParseResult | ImageParseResult:
         parsed = urlparse(raw_url)
         if parsed.hostname == "mall.bilibili.com":
             raise ParseError("不支持解析商品")
         if parsed.path.startswith("/bangumi/"):
             raise ParseError("不支持解析番剧和影视")
+        if BiliAPI.get_music_id(raw_url):
+            return await self.bili_music_parse(raw_url)
         if await self.is_dynamic(raw_url):
             dynamic = await self.get_dynamic_info(raw_url)
             content = self.hashtag_handler(dynamic.content or "")
@@ -142,6 +148,36 @@ class BiliParse(BaseParser):
             ),
         )
 
+    async def bili_music_parse(self, url: str) -> BiliMusicParseResult:
+        async with BiliAPI(proxy=self.proxy) as bili:
+            info_response = await bili.get_music_info(url)
+            if not (info := info_response.get("data")):
+                if info_response.get("code") == 4511001:
+                    raise ContentUnavailableError("音乐未找到或已下架")
+                raise ParseError(f"获取音乐信息失败: {info_response.get('message') or '未知错误'}")
+
+            play_response = await bili.get_music_url(str(info["id"]))
+            cdns: list[str] = (play_response.get("data") or {}).get("cdns") or []
+            if not cdns:
+                raise ParseError("获取音乐链接失败")
+
+        return BiliMusicParseResult(
+            title=info.get("title") or "",
+            music=MusicRef(
+                url=cdns[0],
+                ext=self._music_ext(cdns[0]),
+                thumb_url=info.get("cover") or None,
+                backup_urls=cdns[1:],
+                duration=int(info.get("duration") or 0),
+            ),
+            content=info.get("intro") or "",
+        )
+
+    @staticmethod
+    def _music_ext(url: str) -> str:
+        suffix = Path(urlparse(url).path).suffix
+        return suffix.removeprefix(".") or "m4a"
+
     async def ytp_parse(self, url: str) -> YtVideoParseResult:
         return await BiliYtParse(proxy=self.proxy, cookie=self.cookie)._do_parse(url)
 
@@ -205,7 +241,34 @@ class BiliVideoParseResult(VideoParseResult):
         )
 
 
+class BiliMusicParseResult(MusicParseResult):
+    async def _do_download(
+        self,
+        *,
+        output_dir: Path,
+        callback: ProgressCallback | None = None,
+        callback_args: tuple = (),
+        callback_kwargs: dict | None = None,
+        proxy: str | None = None,
+        headers: dict | None = None,
+        connections: int = 4,
+        concurrency: int = 4,
+    ) -> DownloadResult:
+        headers = {"referer": "https://www.bilibili.com", "User-Agent": UA}
+        return await super()._do_download(
+            output_dir=output_dir,
+            callback=callback,
+            callback_args=callback_args,
+            callback_kwargs=callback_kwargs,
+            proxy=proxy,
+            headers=headers,
+            connections=connections,
+            concurrency=concurrency,
+        )
+
+
 __all__ = [
     "BiliParse",
     "BiliVideoParseResult",
+    "BiliMusicParseResult",
 ]

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/parsers/parser/instagram.py

@@ -71,11 +71,12 @@ class InstagramParser(BaseParser):
                         title=post.title,
                         content=post.caption,
                         video=VideoRef(
-                            url=info.url,
+                            # yt-dlp 的 info.url 是帖子页，provider 的 video_url 才是媒体直链
+                            url=item.video_url or info.url,
                             thumb_url=item.display_url or info.thumbnail,
                             width=info.width,
                             height=info.height,
-                            duration=int(info.duration or item.duration),
+                            duration=int(item.duration or info.duration),
                         ),
                     )
                 if not item.video_url:

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/parsers/parser/zuiyou.py

@@ -15,7 +15,7 @@ class ZuiYouParser(BaseParser):
         return MultimediaParseResult(
             content=zy.content,
             media=[
-                VideoRef(url=i.url, thumb_url=i.thumb_url)
+                VideoRef(url=i.url, thumb_url=i.thumb_url, duration=i.duration)
                 if i.type == MediaType.VIDEO
                 else ImageRef(url=i.url, thumb_url=i.thumb_url)
                 for i in zy.media

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/provider_api/bilibili.py

@@ -125,6 +125,27 @@ class BiliAPI:
         )
         return cast(dict[str, Any], response.json())
 
+    async def get_music_info(self, url: str) -> dict:
+        """获取音乐详细信息"""
+        sid = self.get_music_id(url)
+        if not sid:
+            raise ValueError(f"Invalid url: {url}")
+        response = await self._get_client().get(
+            "https://www.bilibili.com/audio/music-service-c/web/song/info",
+            params={"sid": sid},
+        )
+        response.raise_for_status()
+        return cast(dict[str, Any], response.json())
+
+    async def get_music_url(self, sid: str) -> dict:
+        """获取音乐播放链接"""
+        response = await self._get_client().get(
+            "https://www.bilibili.com/audio/music-service-c/web/url",
+            params={"sid": sid},
+        )
+        response.raise_for_status()
+        return cast(dict[str, Any], response.json())
+
     async def get_buvid(self):
         """获取 buvid"""
         response = await self._get_client().get(
@@ -188,6 +209,11 @@ class BiliAPI:
             tmp = tmp * BASE + idx
         return f"av{(tmp & MASK_CODE) ^ XOR_CODE}"
 
+    @staticmethod
+    def get_music_id(url: str) -> str | None:
+        m = re.search(r"/audio/au(\d+)", url)
+        return m.group(1) if m else None
+
     def get_bvid(self, url: str):
         m_bv = re.search(r"BV[0-9A-Za-z]{10,}", url)
         if m_bv:

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/provider_api/instagram.py

@@ -116,11 +116,54 @@ class InstagramAPI:
                 cookies=self.cookie or None,
             ) as session:
                 if self.cookie:
-                    return await self._get_cookie_post(session, shortcode)
-                return await self._get_logged_out_post(session, shortcode)
+                    post = await self._get_cookie_post(session, shortcode)
+                else:
+                    post = await self._get_logged_out_post(session, shortcode)
+                return await self._fill_missing_video_fields(session, post, shortcode)
         except requests.RequestsError as exc:
             raise InstagramAPIError(f"请求 Instagram 失败: {exc}") from exc
 
+    async def _fill_missing_video_fields(
+        self,
+        session: requests.AsyncSession,
+        post: InstagramPost,
+        shortcode: str,
+    ) -> InstagramPost:
+        """用 embed 补齐 v1 / GraphQL 缺失的视频直链和时长。"""
+        if not any(
+            (post.type is InstagramMediaType.VIDEO or item.is_video)
+            and (not item.video_url or item.duration <= 0)
+            for item in post.media
+        ):
+            return post
+
+        try:
+            node = await self._fetch_embed(session, shortcode)
+        except requests.RequestsError:
+            return post
+        if node is None:
+            return post
+
+        try:
+            embed = InstagramPost.from_graph(node)
+        except InstagramAPIError:
+            return post
+
+        for index, item in enumerate(post.media):
+            if post.type is not InstagramMediaType.VIDEO and not item.is_video:
+                continue
+            source = embed.media[index] if index < len(embed.media) else None
+            if source is None and len(post.media) == 1 and len(embed.media) == 1:
+                source = embed.media[0]
+            if source is None:
+                continue
+            if not item.video_url and source.video_url:
+                item.video_url = source.video_url
+                item.is_video = True
+            if item.duration <= 0 and source.duration > 0:
+                item.duration = source.duration
+        return post
+
     async def _get_logged_out_post(self, session: requests.AsyncSession, shortcode: str) -> InstagramPost:
         detail = "未返回帖子数据"
         media_id = str(self.shortcode_to_media_id(shortcode))

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/provider_api/zuiyou.py

@@ -15,6 +15,7 @@ class Media:
     url: str
     thumb_url: str
     type: MediaType
+    duration: int = 0
 
 
 @dataclass
@@ -34,8 +35,15 @@ class ZuiYouPost:
             id_ = img["id"]
             img_url = list(img["urls"].values())[-1]["urls"][-1]
             if img.get("video"):
-                video_url = videos[str(id_)]["url"]
-                media.append(Media(url=video_url, thumb_url=img_url, type=MediaType.VIDEO))
+                video = videos[str(id_)]
+                media.append(
+                    Media(
+                        url=video["url"],
+                        thumb_url=img_url,
+                        type=MediaType.VIDEO,
+                        duration=int(video.get("dur") or 0),
+                    )
+                )
                 continue
             media.append(Media(url=img_url, thumb_url=img_url, type=MediaType.PHOTO))
 

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/types/__init__.py

@@ -8,8 +8,8 @@ from ..errors import (
     ParseError,
 )
 from .callback import ProgressCallback, ProgressUnit
-from .media_file import AniFile, AnyMediaFile, ImageFile, LivePhotoFile, MediaFile, VideoFile
-from .media_ref import AniRef, AnyMediaRef, ImageRef, LivePhotoRef, MediaRef, VideoRef
+from .media_file import AniFile, AnyMediaFile, ImageFile, LivePhotoFile, MediaFile, MusicFile, VideoFile
+from .media_ref import AniRef, AnyMediaRef, ImageRef, LivePhotoRef, MediaRef, MusicRef, VideoRef
 from .platform import Platform
 from .post import PostType
 from .result import (
@@ -17,6 +17,7 @@ from .result import (
     DownloadResult,
     ImageParseResult,
     MultimediaParseResult,
+    MusicParseResult,
     ParseResult,
     RichTextParseResult,
     VideoParseResult,
@@ -33,12 +34,14 @@ __all__ = [
     "MediaRef",
     "AniRef",
     "VideoRef",
+    "MusicRef",
     "ImageRef",
     "LivePhotoRef",
     "DownloadResult",
     "ParseResult",
     "ImageParseResult",
     "VideoParseResult",
+    "MusicParseResult",
     "MultimediaParseResult",
     "RichTextParseResult",
     "AnyMediaRef",
@@ -49,6 +52,7 @@ __all__ = [
     "ImageFile",
     "AniFile",
     "LivePhotoFile",
+    "MusicFile",
     "AnyMediaFile",
     "ProgressCallback",
     "ProgressUnit",

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/types/media_file.py

@@ -78,6 +78,20 @@ class AniFile(MediaFile):
             self.duration = info.duration
 
 
+@dataclass(kw_only=True)
+class MusicFile(MediaFile):
+    """本地音乐
+    Attributes:
+        path: 路径
+        duration: 音乐时长，单位: 秒
+    """
+
+    duration: int = 0
+
+    def __post_init__(self) -> None:
+        """音乐不做宽高探测；时长由解析结果传入。"""
+
+
 @dataclass(kw_only=True)
 class LivePhotoFile(MediaFile):
     """本地 Live Photo
@@ -100,6 +114,6 @@ class LivePhotoFile(MediaFile):
             self.duration = info.duration
 
 
-AnyMediaFile = MediaFile | VideoFile | ImageFile | AniFile | LivePhotoFile
+AnyMediaFile = MediaFile | VideoFile | ImageFile | AniFile | LivePhotoFile | MusicFile
 
-__all__ = ["MediaFile", "VideoFile", "ImageFile", "AniFile", "LivePhotoFile", "AnyMediaFile"]
+__all__ = ["MediaFile", "VideoFile", "ImageFile", "AniFile", "LivePhotoFile", "MusicFile", "AnyMediaFile"]

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/types/media_ref.py

@@ -37,6 +37,22 @@ class VideoRef(MediaRef):
     duration: int = 0
 
 
+@dataclass(kw_only=True)
+class MusicRef(MediaRef):
+    """音乐
+    Attributes:
+        url: URL
+        ext: 默认扩展名
+        thumb_url: 缩略图 URL
+        width: 宽度
+        height: 高度
+        duration: 音乐时长，单位: 秒
+    """
+
+    ext: str = "m4a"
+    duration: int = 0
+
+
 @dataclass(kw_only=True)
 class ImageRef(MediaRef):
     """图片
@@ -87,6 +103,6 @@ class LivePhotoRef(MediaRef):
     duration: int = 3
 
 
-AnyMediaRef = ImageRef | VideoRef | AniRef | LivePhotoRef
+AnyMediaRef = ImageRef | VideoRef | AniRef | LivePhotoRef | MusicRef
 
-__all__ = ["MediaRef", "VideoRef", "ImageRef", "AniRef", "LivePhotoRef", "AnyMediaRef"]
+__all__ = ["MediaRef", "VideoRef", "MusicRef", "ImageRef", "AniRef", "LivePhotoRef", "AnyMediaRef"]

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/types/post.py

@@ -9,3 +9,4 @@ class PostType(StrEnum):
     VIDEO = "video"
     MULTIMEDIA = "multimedia"
     RICHTEXT = "richtext"
+    MUSIC = "music"

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub/types/result.py

@@ -20,8 +20,8 @@ from ..utils.downloader import download
 from ..utils.fallback import FallbackStage, run_fallback
 from ..utils.helpers import run_sync
 from .callback import ProgressCallback
-from .media_file import AniFile, AnyMediaFile, ImageFile, LivePhotoFile, VideoFile
-from .media_ref import AniRef, AnyMediaRef, ImageRef, LivePhotoRef, VideoRef
+from .media_file import AniFile, AnyMediaFile, ImageFile, LivePhotoFile, MusicFile, VideoFile
+from .media_ref import AniRef, AnyMediaRef, ImageRef, LivePhotoRef, MusicRef, VideoRef
 from .platform import Platform
 from .post import PostType
 
@@ -155,6 +155,8 @@ class ParseResult(ABC):  # noqa: B024
                     mf = ImageFile(path=f, width=media.width, height=media.height)
                 case VideoRef():
                     mf = VideoFile(path=f, width=media.width, height=media.height, duration=media.duration)
+                case MusicRef():
+                    mf = MusicFile(path=f, duration=media.duration)
                 case AniRef():
                     mf = AniFile(path=f, width=media.width, height=media.height, duration=media.duration)
                 case LivePhotoRef():
@@ -361,6 +363,21 @@ class VideoParseResult(ParseResult):
         )
 
 
+class MusicParseResult(ParseResult):
+    """单个音乐"""
+
+    type = PostType.MUSIC
+
+    def __init__(
+        self,
+        title: str = "",
+        music: str | MusicRef | None = None,
+        content: str = "",
+    ):
+        music = MusicRef(url=music) if isinstance(music, str) else music
+        super().__init__(title=title, media=music, content=content)
+
+
 class ImageParseResult(ParseResult):
     """单图 / 多图 / 图集 / 实况照片"""
 
@@ -464,4 +481,4 @@ class DownloadResult:
         return f"{self.__class__.__name__}(media={media_count}, output_dir={self.output_dir})"
 
 
-AnyParseResult = VideoParseResult | ImageParseResult | MultimediaParseResult | RichTextParseResult
+AnyParseResult = VideoParseResult | ImageParseResult | MultimediaParseResult | RichTextParseResult | MusicParseResult

{parsehub-2.3.16 → parsehub-2.4.0/src/parsehub.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: parsehub
-Version: 2.3.16
+Version: 2.4.0
 Summary: 轻量、异步、开箱即用的社交媒体聚合解析库
 Author-email: 梓澪 <zilingmio@gmail.com>
 License: MIT
@@ -82,7 +82,7 @@ Dynamic: license-file
 | **YouTube**     |  ✅  |      | 🎵 音乐                   |
 | **Facebook**    |  ✅  |      |                           |
 | **Threads**     |  ✅  |  ✅  |                           |
-| **Bilibili**    |  ✅  |      | 📝 动态                   |
+| **Bilibili**    |  ✅  |      | 📝 动态, 🎵 音乐          |
 | **抖音**        |  ✅  |  ✅  | ☀️日常                    |
 | **TikTok**      |  ✅  |  ✅  |                           |
 | **微博**        |  ✅  |  ✅  |                           |

{parsehub-2.3.16 → parsehub-2.4.0}/src/parsehub.egg-info/SOURCES.txt

@@ -76,8 +76,11 @@ src/parsehub/utils/downloader.py
 src/parsehub/utils/fallback.py
 src/parsehub/utils/helpers.py
 src/parsehub/utils/media_info.py
+test/test_bilibili_music_offline.py
 test/test_bilibili_offline.py
 test/test_cli.py
 test/test_cli_config.py
 test/test_core_offline.py
-test/test_downloader.py
+test/test_downloader.py
+test/test_instagram_duration_offline.py
+test/test_zuiyou_duration_offline.py

parsehub-2.4.0/test/test_bilibili_music_offline.py

@@ -0,0 +1,127 @@
+import unittest
+from pathlib import Path
+from unittest.mock import AsyncMock, patch
+
+from parsehub.errors import ContentUnavailableError, ParseError
+from parsehub.parsers.parser.bilibili import BiliMusicParseResult, BiliParse
+from parsehub.provider_api.bilibili import BiliAPI
+from parsehub.types import MusicFile, MusicParseResult, MusicRef, ParseResult
+
+MUSIC_INFO = {
+    "code": 0,
+    "data": {
+        "id": 2854883,
+        "title": "残 酷 月 光",
+        "cover": "http://i0.hdslb.com/bfs/music/bc477834dc9b74abdf9869989fc7a0937b25e020.jpg",
+        "intro": "突然想唱 就录了一下",
+        "duration": 200,
+    },
+}
+
+MUSIC_PLAY = {
+    "code": 0,
+    "data": {"cdns": ["https://cdn.example/music-192k.m4a"], "size": 150},
+}
+
+
+class FakeBiliMusicAPI:
+    """响应取自线上实测: song/info 正常响应、4511001 与 -400 错误响应、单条 cdn"""
+
+    info_response: dict = MUSIC_INFO
+    play_response: dict = MUSIC_PLAY
+    last_sid: str | None = None
+
+    get_music_id = staticmethod(BiliAPI.get_music_id)  # _do_parse 经 BiliAPI 路由, 替换后需保留
+
+    def __init__(self, proxy=None):
+        self.proxy = proxy
+
+    async def __aenter__(self):
+        return self
+
+    async def __aexit__(self, *args):
+        return None
+
+    async def get_music_info(self, url):
+        return FakeBiliMusicAPI.info_response
+
+    async def get_music_url(self, sid):
+        FakeBiliMusicAPI.last_sid = sid
+        return FakeBiliMusicAPI.play_response
+
+
+class TestBiliMusicParse(unittest.IsolatedAsyncioTestCase):
+    async def test_mobile_music_url_parses_to_music_result(self):
+        with patch("parsehub.parsers.parser.bilibili.BiliAPI", FakeBiliMusicAPI):
+            result = await BiliParse().parse("https://m.bilibili.com/audio/au2854883?share_source=copy")
+
+        self.assertIsInstance(result, BiliMusicParseResult)
+        self.assertIsInstance(result, MusicParseResult)
+        self.assertEqual(result.to_dict()["type"], "music")
+        self.assertEqual(result.title, "残 酷 月 光")
+        self.assertEqual(result.content, "突然想唱 就录了一下")
+        self.assertEqual(
+            result.media,
+            MusicRef(
+                url="https://cdn.example/music-192k.m4a",
+                thumb_url="http://i0.hdslb.com/bfs/music/bc477834dc9b74abdf9869989fc7a0937b25e020.jpg",
+                duration=200,
+            ),
+        )
+        self.assertEqual(result.raw_url, "https://m.bilibili.com/audio/au2854883")
+        self.assertEqual(FakeBiliMusicAPI.last_sid, "2854883")
+
+    async def test_missing_music_is_content_unavailable(self):
+        with (
+            patch("parsehub.parsers.parser.bilibili.BiliAPI", FakeBiliMusicAPI),
+            patch.object(
+                FakeBiliMusicAPI,
+                "info_response",
+                {"code": 4511001, "data": None, "message": "音乐未找到或已下架"},
+            ),
+        ):
+            with self.assertRaises(ContentUnavailableError):
+                await BiliParse()._do_parse("https://www.bilibili.com/audio/au2854883")
+
+    async def test_api_error_keeps_server_message(self):
+        with (
+            patch("parsehub.parsers.parser.bilibili.BiliAPI", FakeBiliMusicAPI),
+            patch.object(FakeBiliMusicAPI, "info_response", {"code": -400, "message": "请求错误"}),
+        ):
+            with self.assertRaisesRegex(ParseError, "获取音乐信息失败: 请求错误"):
+                await BiliParse()._do_parse("https://www.bilibili.com/audio/au2854883")
+
+    async def test_missing_playurl_is_parse_error(self):
+        with (
+            patch("parsehub.parsers.parser.bilibili.BiliAPI", FakeBiliMusicAPI),
+            patch.object(
+                FakeBiliMusicAPI, "play_response", {"code": 4511001, "data": None, "message": "音乐未找到或已下架"}
+            ),
+        ):
+            with self.assertRaisesRegex(ParseError, "获取音乐链接失败"):
+                await BiliParse()._do_parse("https://www.bilibili.com/audio/au2854883")
+
+
+class TestMusicResultDownload(unittest.IsolatedAsyncioTestCase):
+    async def test_music_result_downloads_with_m4a_and_keeps_duration(self):
+        result = MusicParseResult(title="clip", music=MusicRef(url="https://cdn.example/a", duration=12))
+
+        with patch("parsehub.types.result.download", new=AsyncMock(return_value="/tmp/clip.m4a")) as mocked:
+            out = await result._do_download(output_dir=Path("/tmp/does-not-matter"))
+
+        self.assertIsInstance(out.media, MusicFile)
+        self.assertEqual(out.media.duration, 12)
+        self.assertTrue(Path(mocked.await_args.args[1]).name.endswith("clip.m4a"))
+
+    async def test_download_injects_bilibili_referer(self):
+        result = BiliMusicParseResult(title="clip", music=MusicRef(url="https://cdn.example/a"))
+
+        with patch.object(ParseResult, "_do_download", new=AsyncMock()) as mocked:
+            await result._do_download(output_dir=Path("/tmp/x"))
+
+        # 音乐 CDN 无 Referer 返回 403
+        self.assertEqual(mocked.await_args.kwargs["headers"]["referer"], "https://www.bilibili.com")
+
+
+if __name__ == "__main__":
+    unittest.main()

{parsehub-2.3.16 → parsehub-2.4.0}/test/test_core_offline.py

@@ -131,6 +131,7 @@ class TestParserRegistry(unittest.TestCase):
         self.assertEqual(len(by_id), len(platforms))
         self.assertEqual(by_id["tieba"]["name"], Platform.TIEBA.display_name)
         self.assertIn("图文", by_id["tieba"]["supported_types"])
+        self.assertIn("音乐", by_id["bilibili"]["supported_types"])
         self.assertEqual(parsehub.get_platform("https://tieba.baidu.com/p/9939510114"), Platform.TIEBA)
         self.assertIsNone(parsehub.get_platform("https://example.invalid/not-supported"))
 
@@ -554,6 +555,8 @@ class TestPlatformUrlMatching(unittest.TestCase):
                 "https://www.bilibili.com/video/av123456",
                 "https://www.bilibili.com/opus/1234567890123456789",
                 "https://t.bilibili.com/1234567890123456789",
+                "https://www.bilibili.com/audio/au2854883",
+                "https://m.bilibili.com/audio/au2854883",
                 "https://b23.tv/abc123",
                 "https://bili2233.cn/abc123",
             ],
@@ -687,6 +690,8 @@ class TestPlatformUrlMatching(unittest.TestCase):
             "https://www.youtube.com/live/1h_uc3K4Cpg",
             "https://www.youtube.com/post/Ugkxexample",
             "https://www.youtube.com/@example",
+            "https://www.bilibili.com/audio/home",
+            "https://www.bilibili.com/audio/am10624",
         ]
 
         for url in urls:

parsehub-2.4.0/test/test_instagram_duration_offline.py

@@ -0,0 +1,134 @@
+import unittest
+from unittest.mock import AsyncMock, patch
+
+from parsehub.parsers.base.ytdlp import YtParser, YtVideoInfo
+from parsehub.parsers.parser.instagram import InstagramParser
+from parsehub.provider_api.instagram import InstagramAPI, InstagramMedia, InstagramMediaType, InstagramPost
+
+
+class TestInstagramParserDuration(unittest.IsolatedAsyncioTestCase):
+    async def test_video_prefers_provider_url_and_duration(self):
+        post = InstagramPost(
+            shortcode="abc",
+            type=InstagramMediaType.VIDEO,
+            media=[
+                InstagramMedia(
+                    is_video=True,
+                    display_url="https://cdn.example/thumb.jpg",
+                    video_url="https://cdn.example/v.mp4",
+                    duration=18.018,
+                    width=1920,
+                    height=1080,
+                )
+            ],
+        )
+        info = YtVideoInfo(
+            title="t",
+            description="",
+            thumbnail=None,
+            url="https://www.instagram.com/p/abc/",
+            info_json={},
+            duration=0,
+            width=1920,
+            height=1080,
+        )
+        parser = InstagramParser()
+
+        with (
+            patch.object(InstagramParser, "_parse", AsyncMock(return_value=post)),
+            patch(
+                "parsehub.parsers.parser.instagram.InstagramYtParse._parse",
+                AsyncMock(return_value=info),
+            ),
+        ):
+            result = await parser._do_parse("https://www.instagram.com/reel/abc/")
+
+        self.assertEqual(result.media.url, "https://cdn.example/v.mp4")
+        self.assertEqual(result.media.duration, 18)
+
+
+class TestInstagramAPIDuration(unittest.IsolatedAsyncioTestCase):
+    @staticmethod
+    def _embed_node() -> dict:
+        return {
+            "__typename": "GraphVideo",
+            "shortcode": "abc",
+            "display_url": "https://cdn.example/thumb.jpg",
+            "video_url": "https://cdn.example/v.mp4",
+            "video_duration": 30.933,
+            "dimensions": {"width": 1080, "height": 1920},
+        }
+
+    async def test_fills_duration_from_embed(self):
+        api = InstagramAPI()
+        post = InstagramPost(
+            shortcode="abc",
+            type=InstagramMediaType.VIDEO,
+            media=[
+                InstagramMedia(
+                    is_video=True,
+                    display_url="https://cdn.example/thumb.jpg",
+                    video_url="https://cdn.example/v.mp4",
+                    duration=0,
+                )
+            ],
+        )
+
+        with patch.object(InstagramAPI, "_fetch_embed", AsyncMock(return_value=self._embed_node())):
+            result = await api._fill_missing_video_fields(object(), post, "abc")
+
+        self.assertEqual(result.media[0].duration, 30.933)
+
+    async def test_fills_missing_video_url_from_embed(self):
+        api = InstagramAPI()
+        post = InstagramPost(
+            shortcode="abc",
+            type=InstagramMediaType.VIDEO,
+            media=[
+                InstagramMedia(
+                    is_video=False,
+                    display_url="https://cdn.example/thumb.jpg",
+                    duration=0,
+                )
+            ],
+        )
+
+        with patch.object(InstagramAPI, "_fetch_embed", AsyncMock(return_value=self._embed_node())):
+            result = await api._fill_missing_video_fields(object(), post, "abc")
+
+        self.assertEqual(result.media[0].video_url, "https://cdn.example/v.mp4")
+        self.assertTrue(result.media[0].is_video)
+        self.assertEqual(result.media[0].duration, 30.933)
+
+    async def test_skips_embed_when_complete(self):
+        api = InstagramAPI()
+        post = InstagramPost(
+            shortcode="abc",
+            type=InstagramMediaType.VIDEO,
+            media=[
+                InstagramMedia(
+                    is_video=True,
+                    display_url="https://cdn.example/thumb.jpg",
+                    video_url="https://cdn.example/v.mp4",
+                    duration=18,
+                )
+            ],
+        )
+
+        with patch.object(InstagramAPI, "_fetch_embed", AsyncMock()) as fetch:
+            await api._fill_missing_video_fields(object(), post, "abc")
+
+        fetch.assert_not_awaited()
+
+
+class _FakeYtParser(YtParser, register=False):
+    async def _extract_info(self, url: str) -> dict:
+        return {"title": "t", "duration": 10, "webpage_url": url}
+
+
+class TestYtParserDescription(unittest.IsolatedAsyncioTestCase):
+    async def test_missing_description_does_not_raise(self):
+        info = await _FakeYtParser()._parse("https://example.com/v")
+
+        self.assertEqual(info.description, "")
+        self.assertEqual(info.duration, 10)

parsehub-2.4.0/test/test_zuiyou_duration_offline.py

@@ -0,0 +1,69 @@
+import unittest
+from unittest.mock import AsyncMock, patch
+
+from parsehub.parsers.parser.zuiyou import ZuiYouParser
+from parsehub.provider_api.zuiyou import Media, MediaType, ZuiYou, ZuiYouPost
+
+
+class TestZuiyouDuration(unittest.TestCase):
+    def test_parse_reads_video_dur(self):
+        payload = {
+            "data": {
+                "post": {
+                    "content": "x",
+                    "imgs": [
+                        {
+                            "id": 1,
+                            "urls": {"a": {"urls": ["https://cdn.example/thumb.jpg"]}},
+                            "video": True,
+                        }
+                    ],
+                    "videos": {"1": {"dur": 5, "url": "https://cdn.example/v.mp4"}},
+                }
+            }
+        }
+
+        post = ZuiYouPost.parse(payload)
+
+        self.assertEqual(post.media[0].type, MediaType.VIDEO)
+        self.assertEqual(post.media[0].duration, 5)
+        self.assertEqual(post.media[0].url, "https://cdn.example/v.mp4")
+
+    def test_photo_duration_defaults_to_zero(self):
+        payload = {
+            "data": {
+                "post": {
+                    "imgs": [
+                        {
+                            "id": 1,
+                            "urls": {"a": {"urls": ["https://cdn.example/p.jpg"]}},
+                        }
+                    ]
+                }
+            }
+        }
+
+        post = ZuiYouPost.parse(payload)
+
+        self.assertEqual(post.media[0].duration, 0)
+
+
+class TestZuiyouParserDuration(unittest.IsolatedAsyncioTestCase):
+    async def test_parser_sets_video_duration(self):
+        zy = ZuiYouPost(
+            content="x",
+            media=[
+                Media(
+                    url="https://cdn.example/v.mp4",
+                    thumb_url="https://cdn.example/thumb.jpg",
+                    type=MediaType.VIDEO,
+                    duration=19,
+                )
+            ],
+            raw={},
+        )
+
+        with patch.object(ZuiYou, "parse", AsyncMock(return_value=zy)):
+            result = await ZuiYouParser()._do_parse("https://s.izuiyou.com/s/abc")
+
+        self.assertEqual(result.media[0].duration, 19)