parsehub 2.1.10__tar.gz → 2.2.0__tar.gz

{parsehub-2.1.10/src/parsehub.egg-info → parsehub-2.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: parsehub
-Version: 2.1.10
+Version: 2.2.0
 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
@@ -22,7 +22,7 @@ Requires-Dist: pydub>=0.25.1
 Requires-Dist: python-dotenv>=1.0.1
 Requires-Dist: tenacity>=8.5.0
 Requires-Dist: urlextract>=1.9.0
-Requires-Dist: yt-dlp[default]
+Requires-Dist: yt-dlp[default]>=2026.8.18.122307.dev0
 Requires-Dist: lxml>=5.3.0
 Requires-Dist: pydantic>=1.10.19
 Requires-Dist: markdownify>=1.1.0
@@ -93,6 +93,7 @@ Dynamic: license-file
 | **小黑盒**      |  ✅  |  ✅  |                     |
 | **Snapchat**    |  ✅  |      |                     |
 | **知乎**        |  ✅  |  ✅  | 🐶 问答, 专栏, 圈子 |
+| **豆瓣**        |  ✅  |  ✅  | 👥 小组话题         |
 
 ## 📦 安装
 
@@ -215,6 +216,7 @@ print(result)
 - `快手`
 - `小红书`
 - `知乎`
+- `豆瓣`
 
 ```python
 from parsehub import ParseHub
@@ -313,6 +315,16 @@ except ParseError as exc:
 
 ---
 
+### 启用日志
+
+```python
+from loguru import logger
+
+logger.enable('parsehub')
+```
+
+---
+
 ## 🤝 参与贡献
 
 欢迎提交 Pull Request 或 Issue!

{parsehub-2.1.10 → parsehub-2.2.0}/README.md

@@ -51,6 +51,7 @@
 | **小黑盒**      |  ✅  |  ✅  |                     |
 | **Snapchat**    |  ✅  |      |                     |
 | **知乎**        |  ✅  |  ✅  | 🐶 问答, 专栏, 圈子 |
+| **豆瓣**        |  ✅  |  ✅  | 👥 小组话题         |
 
 ## 📦 安装
 
@@ -173,6 +174,7 @@ print(result)
 - `快手`
 - `小红书`
 - `知乎`
+- `豆瓣`
 
 ```python
 from parsehub import ParseHub
@@ -271,6 +273,16 @@ except ParseError as exc:
 
 ---
 
+### 启用日志
+
+```python
+from loguru import logger
+
+logger.enable('parsehub')
+```
+
+---
+
 ## 🤝 参与贡献
 
 欢迎提交 Pull Request 或 Issue!

{parsehub-2.1.10 → parsehub-2.2.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "parsehub"
-version = "2.1.10"
+version = "2.2.0"
 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"
@@ -25,7 +25,7 @@ dependencies = [
     "python-dotenv>=1.0.1",
     "tenacity>=8.5.0",
     "urlextract>=1.9.0",
-    "yt-dlp[default]",
+    "yt-dlp[default]>=2026.8.18.122307.dev0",
     "lxml>=5.3.0",
     "pydantic>=1.10.19",
     "markdownify>=1.1.0",

{parsehub-2.1.10 → parsehub-2.2.0}/src/parsehub/parsers/parser/__init__.py

@@ -1,6 +1,7 @@
 # ruff: noqa: F403
 from .bilibili import *
 from .coolapk import *
+from .douban import *
 from .douyin import *
 from .facebook import *
 from .instagram import *

parsehub-2.2.0/src/parsehub/parsers/parser/douban.py

@@ -0,0 +1,86 @@
+from pathlib import Path
+
+from ...provider_api.douban import IMAGE_REFERER, Douban, DoubanError, DoubanPhoto, DoubanVideo
+from ...types import (
+    AniRef,
+    AnyMediaRef,
+    DownloadResult,
+    ImageRef,
+    ParseError,
+    Platform,
+    ProgressCallback,
+    RichTextParseResult,
+    VideoRef,
+)
+from ...utils.helpers import UA
+from ..base.base import BaseParser
+
+
+class DoubanParser(BaseParser):
+    __platform__ = Platform.DOUBAN
+    __supported_type__ = ["视频", "图文"]
+    __match__ = r"^(http(s)?://)?(((www|m)\.)?douban\.com/((group/)?topic/\d+|doubanapp/dispatch)|douc\.cc/.+)"
+    __redirect_keywords__ = ["douc.cc", "doubanapp/dispatch"]
+
+    async def _do_parse(self, raw_url: str) -> "DoubanRichTextParseResult":
+        try:
+            topic = await Douban(proxy=self.proxy, cookie=self.cookie.get_value()).parse(raw_url)
+        except DoubanError as e:
+            raise ParseError(f"豆瓣解析失败: {e.msg}") from e
+        except Exception as e:
+            raise ParseError("豆瓣解析失败: 未知错误") from e
+
+        # 图片和视频在正文里有位置关系, 交给 RichText 由 markdown 保留顺序
+        media: list[AnyMediaRef] = [
+            *([self.to_video_ref(topic.video)] if topic.video else []),
+            *(self.to_media_ref(p) for p in topic.photos),
+        ]
+        return DoubanRichTextParseResult(title=topic.title, media=media, markdown_content=topic.markdown_content)
+
+    @staticmethod
+    def to_video_ref(video: DoubanVideo) -> VideoRef:
+        return VideoRef(
+            url=video.url,
+            thumb_url=video.thumb_url,
+            width=video.width,
+            height=video.height,
+            duration=video.duration,
+        )
+
+    @staticmethod
+    def to_media_ref(photo: DoubanPhoto) -> ImageRef | AniRef:
+        ref_type = AniRef if photo.is_animated else ImageRef
+        return ref_type(
+            url=photo.url,
+            ext=photo.ext,
+            thumb_url=photo.thumb_url,
+            width=photo.width,
+            height=photo.height,
+        )
+
+
+class DoubanRichTextParseResult(RichTextParseResult):
+    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,
+    ) -> "DownloadResult":
+        headers = {"User-Agent": UA, "Referer": IMAGE_REFERER}
+        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,
+        )
+
+
+__all__ = ["DoubanParser", "DoubanRichTextParseResult"]

parsehub-2.2.0/src/parsehub/provider_api/douban.py

@@ -0,0 +1,159 @@
+import re
+from dataclasses import dataclass, field
+from typing import Any, cast
+
+import httpx
+from bs4 import BeautifulSoup
+from markdown import markdown
+from markdownify import MarkdownConverter
+
+TOPIC_API = "https://m.douban.com/rexxar/api/v2/group/topic/{}"
+IMAGE_REFERER = "https://www.douban.com/"
+"""豆瓣图床有防盗链, 下载图片时必须带上 Referer"""
+
+TOPIC_URL_RE = r"douban\.com/(?:group/)?topic/(\d+)"
+"""/group/topic/<id> 与 /topic/<id> 共用同一套 ID. 锚定域名避免误匹配 /gallery/topic/<id> (另一套 ID)"""
+
+# rexxar 是移动版网页的内部接口, 无 UA 时返回 418
+MOBILE_UA = (
+    "Mozilla/5.0 (iPhone; CPU iPhone OS 17_0 like Mac OS X) AppleWebKit/605.1.15 "
+    "(KHTML, like Gecko) Version/17.0 Mobile/15E148 Safari/604.1"
+)
+
+
+class Douban:
+    def __init__(self, proxy: str | None = None, cookie: dict[str, str] | None = None):
+        self.proxy = proxy
+        self.cookie = cookie
+
+    async def parse(self, url: str) -> "DoubanTopic":
+        return DoubanTopic.parse(await self.fetch_topic_data(url))
+
+    @staticmethod
+    def get_topic_id(url: str) -> str:
+        if match := re.search(TOPIC_URL_RE, url):
+            return match.group(1)
+        raise DoubanError("暂不支持该豆瓣链接, 目前仅支持小组话题")
+
+    async def fetch_topic_data(self, url: str) -> dict[str, Any]:
+        headers = {"User-Agent": MOBILE_UA, "Referer": "https://m.douban.com/"}
+        async with httpx.AsyncClient(proxy=self.proxy, cookies=self.cookie, timeout=30) as cli:
+            result = await cli.get(TOPIC_API.format(self.get_topic_id(url)), headers=headers)
+
+        if result.status_code != 200:
+            fallback = f"获取话题内容失败: HTTP {result.status_code}"
+            try:
+                error = result.json()
+            except Exception as e:
+                raise DoubanError(fallback) from e
+            # 缺少 localized_message 时 msg 只是英文错误码 (如 need_permission), 需补中文上下文
+            if localized := error.get("localized_message"):
+                raise DoubanError(localized)
+            if msg := error.get("msg"):
+                raise DoubanError(f"获取话题内容失败: {msg}")
+            raise DoubanError(fallback)
+
+        return cast(dict[str, Any], result.json())
+
+
+@dataclass
+class DoubanVideo:
+    url: str
+    thumb_url: str | None = None
+    width: int = 0
+    height: int = 0
+    duration: int = 0
+
+
+@dataclass
+class DoubanPhoto:
+    url: str
+    ext: str = "jpg"
+    thumb_url: str | None = None
+    width: int = 0
+    height: int = 0
+    is_animated: bool = False
+
+
+@dataclass
+class DoubanTopic:
+    title: str
+    markdown_content: str
+    text_content: str
+    video: DoubanVideo | None = None
+    photos: list[DoubanPhoto] = field(default_factory=list)
+
+    @classmethod
+    def parse(cls, data: dict) -> "DoubanTopic":
+        content = data.get("content") or ""
+        markdown_content = MarkdownConverter(heading_style="ATX").convert(content).strip() if content else ""
+        text_content = "".join(BeautifulSoup(markdown(markdown_content), "lxml").find_all(string=True)).strip()
+
+        video_info = data.get("video_info") or {}
+        return cls(
+            title=data.get("title") or "",
+            markdown_content=markdown_content,
+            text_content=text_content,
+            video=parse_video(video_info) if video_info else None,
+            photos=[p for photo in data.get("photos") or [] if (p := parse_photo(photo))],
+        )
+
+
+def parse_photo(photo: dict) -> DoubanPhoto | None:
+    image = photo.get("image") or {}
+    large = image.get("large") or {}
+    normal = image.get("normal") or {}
+
+    # 动图的 large 是体积极大的原始 GIF, 豆瓣同时提供了等效的 mp4
+    if image.get("is_animated") and (video := image.get("video")) and video.get("url"):
+        return DoubanPhoto(
+            url=video["url"],
+            ext="mp4",
+            thumb_url=normal.get("url"),
+            width=video.get("width") or 0,
+            height=video.get("height") or 0,
+            is_animated=True,
+        )
+
+    source = large or normal
+    if not (url := source.get("url")):
+        return None
+    thumb_url = normal.get("url")
+    return DoubanPhoto(
+        url=url,
+        thumb_url=thumb_url if thumb_url != url else None,
+        width=source.get("width") or 0,
+        height=source.get("height") or 0,
+    )
+
+
+def parse_video(video_info: dict) -> DoubanVideo | None:
+    if not (url := video_info.get("video_url")):
+        return None
+    return DoubanVideo(
+        url=url,
+        thumb_url=video_info.get("cover_url"),
+        width=video_info.get("video_width") or 0,
+        height=video_info.get("video_height") or 0,
+        duration=parse_duration(video_info.get("duration")),
+    )
+
+
+def parse_duration(duration: Any) -> int:
+    """把 ``HH:MM:SS`` / ``MM:SS`` 形式的时长转换为秒"""
+    if isinstance(duration, int):
+        return duration
+    if not isinstance(duration, str):
+        return 0
+    seconds = 0
+    for part in duration.split(":"):
+        if not part.isdigit():
+            return 0
+        seconds = seconds * 60 + int(part)
+    return seconds
+
+
+class DoubanError(Exception):
+    def __init__(self, msg: str):
+        self.msg = msg
+        super().__init__(msg)

{parsehub-2.1.10 → parsehub-2.2.0}/src/parsehub/types/platform.py

@@ -6,6 +6,7 @@ class Platform(Enum):
 
     BILIBILI = ("bilibili", "Bilibili")
     COOLAPK = ("coolapk", "酷安")
+    DOUBAN = ("douban", "豆瓣")
     DOUYIN = ("douyin", "抖音")
     TIKTOK = ("tiktok", "TikTok")
     FACEBOOK = ("facebook", "Facebook")

{parsehub-2.1.10 → parsehub-2.2.0/src/parsehub.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: parsehub
-Version: 2.1.10
+Version: 2.2.0
 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
@@ -22,7 +22,7 @@ Requires-Dist: pydub>=0.25.1
 Requires-Dist: python-dotenv>=1.0.1
 Requires-Dist: tenacity>=8.5.0
 Requires-Dist: urlextract>=1.9.0
-Requires-Dist: yt-dlp[default]
+Requires-Dist: yt-dlp[default]>=2026.8.18.122307.dev0
 Requires-Dist: lxml>=5.3.0
 Requires-Dist: pydantic>=1.10.19
 Requires-Dist: markdownify>=1.1.0
@@ -93,6 +93,7 @@ Dynamic: license-file
 | **小黑盒**      |  ✅  |  ✅  |                     |
 | **Snapchat**    |  ✅  |      |                     |
 | **知乎**        |  ✅  |  ✅  | 🐶 问答, 专栏, 圈子 |
+| **豆瓣**        |  ✅  |  ✅  | 👥 小组话题         |
 
 ## 📦 安装
 
@@ -215,6 +216,7 @@ print(result)
 - `快手`
 - `小红书`
 - `知乎`
+- `豆瓣`
 
 ```python
 from parsehub import ParseHub
@@ -313,6 +315,16 @@ except ParseError as exc:
 
 ---
 
+### 启用日志
+
+```python
+from loguru import logger
+
+logger.enable('parsehub')
+```
+
+---
+
 ## 🤝 参与贡献
 
 欢迎提交 Pull Request 或 Issue!

{parsehub-2.1.10 → parsehub-2.2.0}/src/parsehub.egg-info/SOURCES.txt

@@ -21,6 +21,7 @@ src/parsehub/parsers/base/ytdlp.py
 src/parsehub/parsers/parser/__init__.py
 src/parsehub/parsers/parser/bilibili.py
 src/parsehub/parsers/parser/coolapk.py
+src/parsehub/parsers/parser/douban.py
 src/parsehub/parsers/parser/douyin.py
 src/parsehub/parsers/parser/facebook.py
 src/parsehub/parsers/parser/instagram.py
@@ -41,6 +42,7 @@ src/parsehub/parsers/parser/zuiyou.py
 src/parsehub/provider_api/__init__.py
 src/parsehub/provider_api/bilibili.py
 src/parsehub/provider_api/coolapk.py
+src/parsehub/provider_api/douban.py
 src/parsehub/provider_api/douyin.py
 src/parsehub/provider_api/instagram.py
 src/parsehub/provider_api/kuaishou.py

{parsehub-2.1.10 → parsehub-2.2.0}/src/parsehub.egg-info/requires.txt

@@ -5,7 +5,7 @@ pydub>=0.25.1
 python-dotenv>=1.0.1
 tenacity>=8.5.0
 urlextract>=1.9.0
-yt-dlp[default]
+yt-dlp[default]>=2026.8.18.122307.dev0
 lxml>=5.3.0
 pydantic>=1.10.19
 markdownify>=1.1.0

{parsehub-2.1.10 → parsehub-2.2.0}/test/test_core_offline.py

@@ -1,13 +1,25 @@
 import unittest
-from unittest.mock import patch
+from pathlib import Path
+from unittest.mock import AsyncMock, patch
 from urllib.parse import parse_qs, urlparse
 
+import httpx
+
 from parsehub import ParseHub
 from parsehub.errors import ParseError, UnknownPlatform
 from parsehub.parsers.base import BaseParser
+from parsehub.parsers.parser.douban import DoubanParser, DoubanRichTextParseResult
 from parsehub.parsers.parser.douyin import parse_video_info
+from parsehub.provider_api.douban import (
+    IMAGE_REFERER,
+    Douban,
+    DoubanError,
+    DoubanPhoto,
+    DoubanTopic,
+    DoubanVideo,
+)
 from parsehub.provider_api.douyin import DouyinMobileCrawler, DouyinMobileDevice
-from parsehub.types import ImageParseResult, ImageRef, Platform, VideoParseResult, VideoRef
+from parsehub.types import AniRef, ImageParseResult, ImageRef, ParseResult, Platform, VideoParseResult, VideoRef
 from parsehub.utils.helpers import SecretCookie, match_url, run_sync
 
 
@@ -296,6 +308,237 @@ class TestDouyinStorySupport(unittest.TestCase):
         self.assertEqual(info["duration"], 9682)
 
 
+class TestDoubanTopicParsing(unittest.TestCase):
+    @staticmethod
+    def _photo(image: dict, width: int = 500, height: int = 400) -> dict:
+        return {"id": "1", "image": image, "size": {"width": width, "height": height}}
+
+    def test_error_messages_prefer_chinese_and_contextualise_raw_codes(self):
+        cases = [
+            # 豆瓣给了中文提示时直接透出
+            ({"localized_message": "这篇内容不存在了", "msg": "topic not found"}, "这篇内容不存在了"),
+            # 只有英文错误码时补上中文上下文, 避免用户只看到 need_permission
+            ({"msg": "need_permission"}, "获取话题内容失败: need_permission"),
+            ({}, "获取话题内容失败: HTTP 403"),
+        ]
+        for payload, expected in cases:
+            with self.subTest(payload=payload):
+                response = httpx.Response(403, json=payload)
+                with patch.object(httpx.AsyncClient, "get", new=AsyncMock(return_value=response)):
+                    with self.assertRaises(DoubanError) as ctx:
+                        run_sync(Douban().fetch_topic_data("https://www.douban.com/group/topic/1/"))
+                self.assertEqual(ctx.exception.msg, expected)
+
+    def test_get_topic_id_accepts_both_topic_url_forms(self):
+        # /group/topic/<id> 与 /topic/<id> 共用同一套 ID 与接口
+        self.assertEqual(Douban.get_topic_id("https://www.douban.com/group/topic/495373106/"), "495373106")
+        self.assertEqual(Douban.get_topic_id("https://www.douban.com/topic/492821052/"), "492821052")
+
+    def test_get_topic_id_rejects_non_topic_url(self):
+        for url in (
+            "https://movie.douban.com/subject/1292052/",
+            # 话题广场是另一套 ID, 不能被 /topic/ 规则误匹配
+            "https://www.douban.com/gallery/topic/125573/",
+            # 广播是另一种内容类型, 接口与数据结构均不同
+            "https://m.douban.com/people/182691094/status/9372433345/",
+        ):
+            with self.subTest(url=url), self.assertRaises(DoubanError):
+                Douban.get_topic_id(url)
+
+    def test_photo_topic_prefers_large_and_keeps_inline_markdown(self):
+        large = "https://img3.doubanio.com/view/group_topic/l/public/p1.jpg"
+        small = "https://img3.doubanio.com/view/group_topic/m/public/p1.jpg"
+        topic = DoubanTopic.parse(
+            {
+                "title": "标题",
+                "content": f"<div id='content'><p>正文</p><img src=\"{large}\"/></div>",
+                "photos": [
+                    self._photo(
+                        {
+                            "is_animated": False,
+                            "large": {"url": large, "width": 500, "height": 482},
+                            "normal": {"url": small, "width": 200, "height": 193},
+                        }
+                    )
+                ],
+            }
+        )
+
+        self.assertEqual(topic.text_content, "正文")
+        self.assertIn(large, topic.markdown_content)
+        # 取 large 作正片, normal 退为缩略图
+        self.assertEqual(topic.photos, [DoubanPhoto(url=large, ext="jpg", thumb_url=small, width=500, height=482)])
+
+    def test_animated_photo_uses_mp4_variant(self):
+        topic = DoubanTopic.parse(
+            {
+                "title": "动图",
+                "content": "",
+                "photos": [
+                    self._photo(
+                        {
+                            "is_animated": True,
+                            # large 是体积极大的原始 GIF, 应优先取 video
+                            "large": {"url": "https://img3.doubanio.com/view/group_topic/raw/public/p2.jpg"},
+                            "normal": {"url": "https://img3.doubanio.com/view/group_topic/l/public/p2.jpg"},
+                            "video": {
+                                "url": "https://img3.doubanio.com/view/group_topic/l/public/p2.mp4",
+                                "width": 500,
+                                "height": 291,
+                            },
+                        }
+                    )
+                ],
+            }
+        )
+
+        self.assertEqual(
+            topic.photos,
+            [
+                DoubanPhoto(
+                    url="https://img3.doubanio.com/view/group_topic/l/public/p2.mp4",
+                    ext="mp4",
+                    thumb_url="https://img3.doubanio.com/view/group_topic/l/public/p2.jpg",
+                    width=500,
+                    height=291,
+                    is_animated=True,
+                )
+            ],
+        )
+
+    def test_video_topic_converts_duration_to_seconds(self):
+        topic = DoubanTopic.parse(
+            {
+                "title": "视频",
+                "content": "<div id='content'><p>说明</p></div>",
+                "photos": [],
+                "video_info": {
+                    "video_url": "https://sv1.doubanio.com/example.mp4",
+                    "cover_url": "https://sv1.doubanio.com/example_cover.png",
+                    "duration": "01:05",
+                    "video_width": 720,
+                    "video_height": 1280,
+                },
+            }
+        )
+
+        self.assertEqual(
+            topic.video,
+            DoubanVideo(
+                url="https://sv1.doubanio.com/example.mp4",
+                thumb_url="https://sv1.doubanio.com/example_cover.png",
+                width=720,
+                height=1280,
+                duration=65,
+            ),
+        )
+
+    def test_video_info_without_url_is_ignored(self):
+        """video_info 缺 video_url 时不该当成视频话题"""
+        topic = DoubanTopic.parse({"title": "t", "content": "", "photos": [], "video_info": {"duration": "00:12"}})
+
+        self.assertEqual(topic.photos, [])
+
+
+class TestDoubanParserResultTypes(unittest.IsolatedAsyncioTestCase):
+    @staticmethod
+    def _topic(**kwargs) -> DoubanTopic:
+        defaults: dict = {
+            "title": "标题",
+            "markdown_content": "",
+            "text_content": "",
+            "video": None,
+            "photos": [],
+        }
+        return DoubanTopic(**{**defaults, **kwargs})
+
+    @staticmethod
+    async def _parse(topic: DoubanTopic):
+        with patch.object(Douban, "parse", new=AsyncMock(return_value=topic)):
+            return await DoubanParser()._do_parse("https://www.douban.com/group/topic/1/")
+
+    async def test_topic_always_returns_richtext(self):
+        """图 / 视频 / 纯文字 任意组合都走 RichText, 不再区分纯图"""
+        photo = DoubanPhoto(url="https://img3.doubanio.com/p1.jpg", width=500, height=482)
+        video = DoubanVideo(url="https://sv1.doubanio.com/a.mp4", duration=39)
+        img_md = "正文\n\n![](https://img3.doubanio.com/p1.jpg)"
+        cases = {
+            # markdown 必须原样带过去: 图片内嵌的位置信息只存在于 markdown 里
+            "图 + 正文": (self._topic(markdown_content=img_md, text_content="正文", photos=[photo]), 1, img_md),
+            "纯图无正文": (self._topic(markdown_content="![](x)", photos=[photo]), 1, "![](x)"),
+            "纯文字": (self._topic(markdown_content="只有文字", text_content="只有文字"), 0, "只有文字"),
+            "纯视频": (self._topic(video=video), 1, ""),
+        }
+        for label, (topic, media_count, expected_md) in cases.items():
+            with self.subTest(case=label):
+                result = await self._parse(topic)
+                self.assertIsInstance(result, DoubanRichTextParseResult)
+                self.assertEqual(len(result.media or []), media_count)
+                self.assertEqual(result.markdown_content, expected_md)
+
+    async def test_video_with_photos_keeps_video_first(self):
+        result = await self._parse(
+            self._topic(
+                text_content="说明",
+                markdown_content="说明",
+                video=DoubanVideo(url="https://sv1.doubanio.com/a.mp4", duration=39),
+                photos=[DoubanPhoto(url="https://img3.doubanio.com/p1.jpg", width=500, height=482)],
+            )
+        )
+
+        self.assertIsInstance(result, DoubanRichTextParseResult)
+        self.assertEqual(
+            result.media,
+            [
+                VideoRef(url="https://sv1.doubanio.com/a.mp4", duration=39),
+                ImageRef(url="https://img3.doubanio.com/p1.jpg", width=500, height=482),
+            ],
+        )
+
+    async def test_animated_photo_becomes_ani_ref(self):
+        result = await self._parse(
+            self._topic(
+                photos=[
+                    DoubanPhoto(
+                        url="https://img3.doubanio.com/p2.mp4",
+                        ext="mp4",
+                        thumb_url="https://img3.doubanio.com/p2.jpg",
+                        width=500,
+                        height=291,
+                        is_animated=True,
+                    )
+                ]
+            )
+        )
+
+        self.assertEqual(
+            result.media,
+            [
+                AniRef(
+                    url="https://img3.doubanio.com/p2.mp4",
+                    ext="mp4",
+                    thumb_url="https://img3.doubanio.com/p2.jpg",
+                    width=500,
+                    height=291,
+                )
+            ],
+        )
+
+    async def test_provider_error_is_wrapped_in_parse_error(self):
+        with patch.object(Douban, "parse", new=AsyncMock(side_effect=DoubanError("这篇内容不存在了"))):
+            with self.assertRaisesRegex(ParseError, "豆瓣解析失败: 这篇内容不存在了"):
+                await DoubanParser()._do_parse("https://www.douban.com/group/topic/1/")
+
+    async def test_download_injects_douban_referer(self):
+        result = DoubanRichTextParseResult(title="标题", media=[ImageRef(url="https://img3.doubanio.com/p1.jpg")])
+
+        with patch.object(ParseResult, "_do_download", new=AsyncMock()) as mocked:
+            await result._do_download(output_dir=Path("/tmp/does-not-matter"))
+
+        # 豆瓣图床无 Referer 时返回 418
+        self.assertEqual(mocked.await_args.kwargs["headers"]["Referer"], IMAGE_REFERER)
+
+
 class TestPlatformUrlMatching(unittest.TestCase):
     def test_supported_platform_url_formats(self):
         parsehub = ParseHub()
@@ -314,6 +557,14 @@ class TestPlatformUrlMatching(unittest.TestCase):
                 "https://www.coolapk.com/feed/70163953",
                 "https://www.coolapk.com/picture/123456",
             ],
+            Platform.DOUBAN: [
+                "https://www.douban.com/group/topic/495373106/",
+                "https://m.douban.com/group/topic/495373106/",
+                "douban.com/group/topic/495373106/?_spm_id=MTU0MzM0ODY2&_i=5304694LhuE3jh",
+                "https://www.douban.com/topic/492821052/?_spm_id=MTQ3NTcyNQ&dt_dapp=1",
+                "https://www.douban.com/doubanapp/dispatch?uri=/group/topic/495373106/",
+                "https://douc.cc/2Yx4Ol",
+            ],
             Platform.DOUYIN: [
                 "https://www.douyin.com/video/7615533976798727464",
                 "https://www.douyin.com/note/7615533976798727464",
@@ -419,6 +670,11 @@ class TestPlatformUrlMatching(unittest.TestCase):
     def test_known_unsupported_url_formats_are_not_matched(self):
         parsehub = ParseHub()
         urls = [
+            "https://movie.douban.com/subject/1292052/",
+            "https://book.douban.com/subject/1084336/",
+            "https://www.douban.com/people/154334866/",
+            "https://www.douban.com/group/657759/",
+            "https://www.douban.com/gallery/topic/125573/",
             "https://www.douyin.com/share/user/MS4wLjABAAAA",
             "https://www.douyin.com/qishui/share/video/123456",
             "https://www.tiktok.com/share/user/123456",