parsehub 2.0.15__tar.gz → 2.0.16__tar.gz

{parsehub-2.0.15/src/parsehub.egg-info → parsehub-2.0.16}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: parsehub
-Version: 2.0.15
+Version: 2.0.16
 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
@@ -34,6 +34,7 @@ Requires-Dist: pillow>=12.1.0
 Requires-Dist: python-slugify[unidecode]>=8.0.4
 Requires-Dist: opencv-python-headless>=4.13.0.92
 Requires-Dist: cryptography>=46.0.6
+Requires-Dist: gmssl>=3.2.2
 Dynamic: license-file
 
 <div align="center">
@@ -105,7 +106,8 @@ print(result)
 | **Facebook**    | ✅  |    |       |
 | **Threads**     | ✅  | ✅  |       |
 | **Bilibili**    | ✅  |    | 📝 动态 |
-| **抖音 / TikTok** | ✅  | ✅  |       |
+| **抖音**          | ✅  | ✅  |       |
+| **TikTok**      | ✅  | ✅  |       |
 | **微博**          | ✅  | ✅  |       |
 | **小红书**         | ✅  | ✅  |       |
 | **贴吧**          | ✅  | ✅  |       |
@@ -164,7 +166,13 @@ ParseHub(cookie={"key1": "value1", "key2": "value2"})
 
 目前支持 Cookie 登录的平台:
 
-`Twitter` · `Instagram` · `Kuaishou` · `Bilibili` · `YouTube`
+- `Twitter`
+- `Instagram`
+- `Kuaishou`
+- `Bilibili`
+- `YouTube`
+- `抖音`
+- `TikTok`
 
 ### 全局配置
 

{parsehub-2.0.15 → parsehub-2.0.16}/README.md

@@ -67,7 +67,8 @@ print(result)
 | **Facebook**    | ✅  |    |       |
 | **Threads**     | ✅  | ✅  |       |
 | **Bilibili**    | ✅  |    | 📝 动态 |
-| **抖音 / TikTok** | ✅  | ✅  |       |
+| **抖音**          | ✅  | ✅  |       |
+| **TikTok**      | ✅  | ✅  |       |
 | **微博**          | ✅  | ✅  |       |
 | **小红书**         | ✅  | ✅  |       |
 | **贴吧**          | ✅  | ✅  |       |
@@ -126,7 +127,13 @@ ParseHub(cookie={"key1": "value1", "key2": "value2"})
 
 目前支持 Cookie 登录的平台:
 
-`Twitter` · `Instagram` · `Kuaishou` · `Bilibili` · `YouTube`
+- `Twitter`
+- `Instagram`
+- `Kuaishou`
+- `Bilibili`
+- `YouTube`
+- `抖音`
+- `TikTok`
 
 ### 全局配置
 

{parsehub-2.0.15 → parsehub-2.0.16}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "parsehub"
-version = "2.0.15"
+version = "2.0.16"
 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"
@@ -37,6 +37,7 @@ dependencies = [
     "python-slugify[unidecode]>=8.0.4",
     "opencv-python-headless>=4.13.0.92",
     "cryptography>=46.0.6",
+    "gmssl>=3.2.2",
 ]
 
 [dependency-groups]

{parsehub-2.0.15 → parsehub-2.0.16}/src/parsehub/config/config.py

@@ -1,7 +1,7 @@
 import sys
 from pathlib import Path
 
-from pydantic import BaseModel, ConfigDict, HttpUrl
+from pydantic import BaseModel, ConfigDict
 
 
 class _GlobalConfig(BaseModel):
@@ -11,8 +11,6 @@ class _GlobalConfig(BaseModel):
         "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
         "AppleWebKit/537.36 (KHTML, like Gecko) Chrome/144.0.0.0 Safari/537.36"
     )
-    douyin_api: HttpUrl = "https://douyin.wtf/"
-    """抖音解析API, 建议自行部署: https://github.com/Evil0ctal/Douyin_TikTok_Download_API"""
     default_save_dir: Path = Path(sys.argv[0]).parent / "downloads"
     """默认下载目录"""
 

parsehub-2.0.16/src/parsehub/parsers/parser/douyin.py

@@ -0,0 +1,251 @@
+from dataclasses import dataclass
+from enum import Enum
+from pathlib import Path
+from typing import Self, Union
+
+from ... import ProgressCallback
+from ...provider_api.douyin import DouyinWebCrawler
+from ...types import (
+    DownloadResult,
+    ImageParseResult,
+    ImageRef,
+    LivePhotoRef,
+    MultimediaParseResult,
+    ParseError,
+    Platform,
+    VideoParseResult,
+    VideoRef,
+)
+from ..base.base import BaseParser
+
+
+class DouyinParser(BaseParser):
+    __platform__ = Platform.DOUYIN
+    __supported_type__ = ["视频", "图文"]
+    __match__ = r"^(http(s)?://)?.+douyin.com/(?!share/user|qishui).+"
+    __redirect_keywords__ = ["v.douyin", "iesdouyin"]
+    __reserved_parameters__ = ["modal_id"]
+
+    async def _do_parse(self, raw_url: str) -> Union["VideoParseResult", "ImageParseResult", "MultimediaParseResult"]:
+        result = await self._fetch_api_result(raw_url)
+
+        match result.type:
+            case DouyinMediaType.VIDEO:
+                return self._build_video_result(result)
+            case DouyinMediaType.IMAGE:
+                return self._build_image_result(result)
+
+    async def _fetch_api_result(self, url: str) -> "DouyinApiResult":
+        """获取并解析抖音 API 结果"""
+        if not self.cookie:
+            raise ParseError("抖音 Cookie 未配置")
+
+        crawler = DouyinWebCrawler(proxy=self.proxy, cookie=self.cookie)
+        response = await crawler.parse(url)
+        return DouyinApiResult.parse(response)
+
+    @staticmethod
+    def _build_video_result(result: "DouyinApiResult") -> VideoParseResult:
+        """构建视频解析结果"""
+        return DouyinVideoParseResult(
+            title=result.desc,
+            video=result.video,
+        )
+
+    @staticmethod
+    def _build_image_result(result: "DouyinApiResult") -> ImageParseResult:
+        """构建图片解析结果"""
+        return ImageParseResult(
+            title=result.desc,
+            photo=result.image_list,
+        )
+
+
+class DouyinVideoParseResult(VideoParseResult):
+    async def _do_download(
+        self,
+        *,
+        output_dir: str | Path,
+        callback: ProgressCallback | None = None,
+        callback_args: tuple = (),
+        callback_kwargs: dict | None = None,
+        proxy: str | None = None,
+        headers: dict | None = None,
+    ) -> "DownloadResult":
+        headers = {
+            "Referer": "https://www.douyin.com/",
+        }
+        return await super()._do_download(
+            output_dir=output_dir,
+            callback=callback,
+            callback_args=callback_args,
+            callback_kwargs=callback_kwargs,
+            proxy=proxy,
+            headers=headers,
+        )
+
+
+def remove_video_watermark(url: str) -> str:
+    """移除抖音视频水印标识 (playwm -> play)"""
+    return url.replace("playwm", "play")
+
+
+def parse_video_info(video_data: dict) -> dict:
+    bit_rates = video_data.get("bit_rate")
+    if not bit_rates:
+        raise ParseError("抖音解析失败: 未获取到视频下载地址")
+
+    # 按分辨率降序排列，选择最高质量
+    bit_rates.sort(
+        key=lambda x: x.get("play_addr", {}).get("width", 0) * x.get("play_addr", {}).get("height", 0),
+        reverse=True,
+    )
+    best_quality = bit_rates[0]
+
+    play_addr = best_quality.get("play_addr", {})
+    video_url_list = play_addr.get("url_list", [])
+    if not video_url_list:
+        raise ParseError("抖音解析失败: 视频下载地址为空")
+
+    video_url = remove_video_watermark(video_url_list[0])
+
+    cover = video_data.get("cover", {})
+    thumb_url_list = cover.get("url_list", [])
+    thumb_url = thumb_url_list[-1] if thumb_url_list else None
+
+    return {
+        "video_url": video_url,
+        "thumb_url": thumb_url,
+        "duration": best_quality.get("duration", 0),
+        "width": play_addr.get("width", 0),
+        "height": play_addr.get("height", 0),
+    }
+
+
+class DouyinMediaType(Enum):
+    """抖音媒体类型"""
+
+    VIDEO = "video"
+    IMAGE = "image"  # 实况图片 + 图片
+
+
+@dataclass
+class DouyinApiResult:
+    """抖音 API 解析结果"""
+
+    type: DouyinMediaType
+    video: VideoRef = None
+    desc: str = ""
+    image_list: list[ImageRef | LivePhotoRef] = None
+
+    @classmethod
+    def parse(cls, json_dict: dict) -> Self:
+        data = json_dict.get("aweme_detail")
+        if not data:
+            raise ParseError("抖音解析失败: 未获取到作品详情")
+
+        desc = data.get("desc", "")
+
+        if images := data.get("images"):
+            return cls._parse_images(images, desc)
+        elif image_post_info := data.get("image_post_info"):
+            return cls._parse_image_post_info(image_post_info, desc)
+        else:
+            return cls._parse_video(data, desc)
+
+    @classmethod
+    def _parse_images(cls, images: list[dict], desc: str) -> Self:
+        """解析旧版图片格式 (images 字段)
+
+        支持普通图片和实况照片 (LivePhoto)
+        """
+        has_live_photos = any(img.get("video") for img in images)
+
+        if has_live_photos:
+            image_list = []
+            for image in images:
+                if video := image.get("video"):
+                    video_info = parse_video_info(video)
+                    image_list.append(
+                        LivePhotoRef(
+                            url=video_info["thumb_url"],
+                            video_url=video_info["video_url"],
+                            width=int(video_info["width"]),
+                            height=int(video_info["height"]),
+                            duration=int(video_info["duration"]) or 3,
+                        )
+                    )
+                else:
+                    url_list = image.get("url_list", [])
+                    if url_list:
+                        image_list.append(
+                            ImageRef(
+                                url=url_list[-1],
+                                height=image.get("height", 0),
+                                width=image.get("width", 0),
+                            )
+                        )
+        else:
+            image_list = [
+                ImageRef(
+                    url=img["url_list"][-1],
+                    height=img.get("height", 0),
+                    width=img.get("width", 0),
+                )
+                for img in images
+                if img.get("url_list")
+            ]
+
+        return cls(
+            type=DouyinMediaType.IMAGE,
+            desc=desc,
+            image_list=image_list,
+        )
+
+    @classmethod
+    def _parse_image_post_info(cls, image_post_info: dict, desc: str) -> Self:
+        """解析新版图片格式 (image_post_info 字段)"""
+        images = image_post_info.get("images", [])
+        image_list = []
+
+        for image in images:
+            display_image = image.get("display_image", {})
+            url_list = display_image.get("url_list", [])
+            if url_list:
+                image_list.append(
+                    ImageRef(
+                        url=url_list[-1],
+                        height=display_image.get("height", 0),
+                        width=display_image.get("width", 0),
+                    )
+                )
+
+        return cls(
+            type=DouyinMediaType.IMAGE,
+            image_list=image_list,
+            desc=desc,
+        )
+
+    @classmethod
+    def _parse_video(cls, data: dict, desc: str) -> Self:
+        """解析视频"""
+        video_data = data.get("video")
+        if not video_data:
+            raise ParseError("抖音解析失败: 未获取到视频数据")
+
+        video_info = parse_video_info(video_data)
+
+        return cls(
+            type=DouyinMediaType.VIDEO,
+            video=VideoRef(
+                url=video_info["video_url"],
+                thumb_url=video_info["thumb_url"],
+                width=video_info["width"],
+                height=video_info["height"],
+                duration=video_info["duration"],
+            ),
+            desc=desc,
+        )
+
+
+__all__ = ["DouyinParser"]

parsehub-2.0.16/src/parsehub/parsers/parser/tiktok.py

@@ -0,0 +1,228 @@
+from dataclasses import dataclass
+from enum import Enum
+from pathlib import Path
+from typing import Self, Union
+
+from ... import ProgressCallback
+from ...config import GlobalConfig
+from ...provider_api.tiktok import TikTokWebCrawler
+from ...types import (
+    DownloadResult,
+    ImageParseResult,
+    ImageRef,
+    LivePhotoRef,
+    MultimediaParseResult,
+    ParseError,
+    Platform,
+    VideoParseResult,
+    VideoRef,
+)
+from ..base.base import BaseParser
+
+
+class TikTokParser(BaseParser):
+    __platform__ = Platform.TIKTOK
+    __supported_type__ = ["视频", "图文"]
+    __match__ = r"^(http(s)?://)?.+tiktok.com/(?!share/user|qishui).+"
+    __redirect_keywords__ = ["vt.tiktok"]
+
+    async def _do_parse(self, raw_url: str) -> Union["VideoParseResult", "ImageParseResult", "MultimediaParseResult"]:
+        result = await self._fetch_api_result(raw_url)
+
+        match result.type:
+            case TikTokMediaType.VIDEO:
+                return self._build_video_result(result)
+            case TikTokMediaType.IMAGE:
+                return self._build_image_result(result)
+
+    async def _fetch_api_result(self, url: str) -> "TikTokApiResult":
+        crawler = TikTokWebCrawler(proxy=self.proxy, cookie=self.cookie)
+        try:
+            response = await crawler.parse(url)
+            return TikTokApiResult.parse(response)
+        except ParseError:
+            raise
+        except Exception as e:
+            raise ParseError(f"TikTok 解析失败: {e}") from e
+
+    @staticmethod
+    def _build_video_result(result: "TikTokApiResult") -> VideoParseResult:
+        return TikTokVideoParseResult(
+            title=result.desc,
+            video=result.video,
+        )
+
+    @staticmethod
+    def _build_image_result(result: "TikTokApiResult") -> ImageParseResult:
+        return ImageParseResult(
+            title=result.desc,
+            photo=result.image_list,
+        )
+
+
+class TikTokVideoParseResult(VideoParseResult):
+    async def _do_download(
+        self,
+        *,
+        output_dir: str | Path,
+        callback: ProgressCallback | None = None,
+        callback_args: tuple = (),
+        callback_kwargs: dict | None = None,
+        proxy: str | None = None,
+        headers: dict | None = None,
+    ) -> "DownloadResult":
+        headers = {
+            "User-Agent": GlobalConfig.ua,
+            "Referer": "https://www.tiktok.com/",
+        }
+        return await super()._do_download(
+            output_dir=output_dir,
+            callback=callback,
+            callback_args=callback_args,
+            callback_kwargs=callback_kwargs,
+            proxy=proxy,
+            headers=headers,
+        )
+
+
+def first_url(data: dict | None) -> str | None:
+    url_list = (data or {}).get("url_list") or (data or {}).get("UrlList") or []
+    return next((url for url in url_list if url), None)
+
+
+def as_int(value) -> int:
+    try:
+        return int(value or 0)
+    except (TypeError, ValueError):
+        return 0
+
+
+def pick_cover(video_data: dict) -> str | None:
+    for key in ("origin_cover", "cover", "dynamic_cover", "originCover", "dynamicCover"):
+        cover_url = first_url(video_data.get(key))
+        if cover_url:
+            return cover_url
+    cover = video_data.get("cover")
+    return cover if isinstance(cover, str) else None
+
+
+def parse_video_info(video_data: dict) -> dict:
+    bit_rates = video_data.get("bit_rate") or video_data.get("bitrateInfo") or []
+    candidates = []
+
+    for bit_rate in bit_rates:
+        play_addr = bit_rate.get("play_addr") or bit_rate.get("PlayAddr") or {}
+        video_url = first_url(play_addr)
+        if not video_url:
+            continue
+
+        width = as_int(play_addr.get("width") or play_addr.get("Width") or video_data.get("width"))
+        height = as_int(play_addr.get("height") or play_addr.get("Height") or video_data.get("height"))
+        bitrate = as_int(bit_rate.get("bit_rate") or bit_rate.get("Bitrate") or bit_rate.get("bitrate"))
+        data_size = as_int(play_addr.get("data_size") or play_addr.get("DataSize") or bit_rate.get("data_size"))
+        duration = as_int(play_addr.get("duration") or play_addr.get("Duration") or video_data.get("duration"))
+
+        candidates.append(
+            {
+                "video_url": video_url,
+                "thumb_url": pick_cover(video_data),
+                "duration": duration,
+                "width": width,
+                "height": height,
+                "quality": (width * height, bitrate, data_size),
+            }
+        )
+
+    if not candidates:
+        play_addr = video_data.get("play_addr") or video_data.get("playAddr") or {}
+        video_url = first_url(play_addr)
+        if video_url:
+            width = as_int(play_addr.get("width") or video_data.get("width"))
+            height = as_int(play_addr.get("height") or video_data.get("height"))
+            candidates.append(
+                {
+                    "video_url": video_url,
+                    "thumb_url": pick_cover(video_data),
+                    "duration": as_int(play_addr.get("duration") or video_data.get("duration")),
+                    "width": width,
+                    "height": height,
+                    "quality": (width * height, 0, 0),
+                }
+            )
+
+    if not candidates:
+        raise ParseError("TikTok 解析失败: 未获取到无水印视频下载地址")
+
+    return max(candidates, key=lambda x: x["quality"])
+
+
+class TikTokMediaType(Enum):
+    VIDEO = "video"
+    IMAGE = "image"
+
+
+@dataclass
+class TikTokApiResult:
+    type: TikTokMediaType
+    video: VideoRef = None
+    desc: str = ""
+    image_list: list[ImageRef | LivePhotoRef] = None
+
+    @classmethod
+    def parse(cls, json_dict: dict) -> Self:
+        if not json_dict:
+            raise ParseError("TikTok 解析失败: 未获取到作品详情")
+
+        desc = json_dict.get("desc", "")
+        image_post_info: dict = json_dict.get("image_post_info", {}) or json_dict.get("imagePost", {})
+        if image_post_info:
+            return cls._parse_image_post(image_post_info, desc)
+        return cls._parse_video(json_dict, desc)
+
+    @classmethod
+    def _parse_image_post(cls, image_post_info: dict, desc: str) -> Self:
+        image_list = []
+
+        for image in image_post_info.get("images", []):
+            display_image = image.get("display_image") or image.get("displayImage") or image.get("image") or {}
+            url = first_url(display_image)
+            if url:
+                image_list.append(
+                    ImageRef(
+                        url=url,
+                        height=as_int(display_image.get("height") or display_image.get("Height")),
+                        width=as_int(display_image.get("width") or display_image.get("Width")),
+                    )
+                )
+
+        if not image_list:
+            raise ParseError("TikTok 解析失败: 未获取到无水印图文下载地址")
+
+        return cls(
+            type=TikTokMediaType.IMAGE,
+            desc=desc,
+            image_list=image_list,
+        )
+
+    @classmethod
+    def _parse_video(cls, data: dict, desc: str) -> Self:
+        video_data = data.get("video", {})
+        if not video_data:
+            raise ParseError("TikTok 解析失败: 未获取到视频数据")
+
+        video_info = parse_video_info(video_data)
+
+        return cls(
+            type=TikTokMediaType.VIDEO,
+            video=VideoRef(
+                url=video_info["video_url"],
+                thumb_url=video_info["thumb_url"],
+                width=video_info["width"],
+                height=video_info["height"],
+                duration=video_info["duration"],
+            ),
+            desc=desc,
+        )
+
+
+__all__ = ["TikTokParser"]