nonebot-plugin-parser 2.4.3__tar.gz → 2.5.0__tar.gz

{nonebot_plugin_parser-2.4.3 → nonebot_plugin_parser-2.5.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: nonebot-plugin-parser
-Version: 2.4.3
+Version: 2.5.0
 Summary: NoneBot2 链接分享解析 Alconna 版, 现支持B站|抖音|快手|微博|小红书|YouTube|TikTok|Twitter|AcFun|NGA
 Keywords: acfun,bilibili,douyin,kuaishou,nga,nonebot,nonebot2,tiktok,twitter,video,weibo,xiaohongshu,youtube
 Author: fllesser
@@ -36,8 +36,8 @@ Requires-Dist: nonebot-plugin-uninfo>=0.10.1,<1.0.0
 Requires-Dist: nonebot-plugin-htmlkit>=0.1.0rc4 ; extra == 'all'
 Requires-Dist: nonebot-plugin-htmlrender>=0.6.7 ; extra == 'all'
 Requires-Dist: yt-dlp[default]>=2026.2.21 ; extra == 'all'
-Requires-Dist: emosvg>=0.1.6 ; extra == 'all'
-Requires-Dist: emosvg>=0.1.6 ; extra == 'emosvg'
+Requires-Dist: emosvg>=0.1.7 ; extra == 'all'
+Requires-Dist: emosvg>=0.1.7 ; extra == 'emosvg'
 Requires-Dist: nonebot-plugin-htmlkit>=0.1.0rc4 ; extra == 'htmlkit'
 Requires-Dist: nonebot-plugin-htmlrender>=0.6.7 ; extra == 'htmlrender'
 Requires-Dist: yt-dlp[default]>=2025.2.21 ; extra == 'ytdlp'
@@ -300,8 +300,8 @@ parser_render_type="common"
 parser_append_url=False
 
 # [可选] 自定义渲染字体
-# 配置字体文件名，并将字体文件放置于 localstore 生成的插件 data 目录下
-# 例如: ./data/nonebot_plugin_parser/
+# 配置字体文件名，并将字体文件放置于 localstore 生成的插件 config 目录下
+# 例如: ./config/nonebot_plugin_parser/
 parser_custom_font="LXGWZhenKaiGB-Regular.ttf"
 
 # [可选] 是否需要转发媒体内容(超过 4 项时始终使用合并转发)
@@ -452,16 +452,6 @@ images = self.create_image_contents([
 ])
 ```
 
-> 构建图文内容(适用于类似 Bilibili 动态图文混排)
-
-```python
-graphics = self.create_graphics_content(
-    image_url="https://example.com/image.jpg",
-    text="图片前的文字说明",  # 可选
-    alt="图片描述"            # 可选，居中显示
-)
-```
-
 > 创建动图内容（GIF)，平台一般只提供视频（后续插件会做自动转为 gif 的处理)
 
 ```python

{nonebot_plugin_parser-2.4.3 → nonebot_plugin_parser-2.5.0}/README.md

@@ -246,8 +246,8 @@ parser_render_type="common"
 parser_append_url=False
 
 # [可选] 自定义渲染字体
-# 配置字体文件名，并将字体文件放置于 localstore 生成的插件 data 目录下
-# 例如: ./data/nonebot_plugin_parser/
+# 配置字体文件名，并将字体文件放置于 localstore 生成的插件 config 目录下
+# 例如: ./config/nonebot_plugin_parser/
 parser_custom_font="LXGWZhenKaiGB-Regular.ttf"
 
 # [可选] 是否需要转发媒体内容(超过 4 项时始终使用合并转发)
@@ -398,16 +398,6 @@ images = self.create_image_contents([
 ])
 ```
 
-> 构建图文内容(适用于类似 Bilibili 动态图文混排)
-
-```python
-graphics = self.create_graphics_content(
-    image_url="https://example.com/image.jpg",
-    text="图片前的文字说明",  # 可选
-    alt="图片描述"            # 可选，居中显示
-)
-```
-
 > 创建动图内容（GIF)，平台一般只提供视频（后续插件会做自动转为 gif 的处理)
 
 ```python

{nonebot_plugin_parser-2.4.3 → nonebot_plugin_parser-2.5.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "nonebot-plugin-parser"
-version = "2.4.3"
+version = "2.5.0"
 description = "NoneBot2 链接分享解析 Alconna 版, 现支持B站|抖音|快手|微博|小红书|YouTube|TikTok|Twitter|AcFun|NGA"
 readme = "README.md"
 requires-python = ">=3.10"
@@ -53,7 +53,6 @@ dependencies = [
   "nonebot-plugin-apscheduler>=0.5.0,<1.0.0",
   "nonebot-plugin-localstore>=0.7.4,<1.0.0",
   "nonebot-plugin-uninfo>=0.10.1,<1.0.0",
-
 ]
 
 [project.urls]
@@ -65,12 +64,12 @@ Repository = "https://github.com/fllesser/nonebot-plugin-parser"
 htmlkit = ["nonebot-plugin-htmlkit>=0.1.0rc4"]
 htmlrender = ["nonebot-plugin-htmlrender>=0.6.7"]
 ytdlp = ["yt-dlp[default]>=2025.2.21"]
-emosvg = ["emosvg>=0.1.6"]
+emosvg = ["emosvg>=0.1.7"]
 all = [
   "nonebot-plugin-htmlkit>=0.1.0rc4",
   "nonebot-plugin-htmlrender>=0.6.7",
   "yt-dlp[default]>=2026.2.21",
-  "emosvg>=0.1.6",
+  "emosvg>=0.1.7",
 ]
 
 [dependency-groups]
@@ -82,7 +81,7 @@ dev = [
   { include-group = "test" },
 ]
 extras = [
-  "emosvg>=0.1.6",
+  "emosvg>=0.1.7",
   "nonebot-plugin-htmlkit>=0.1.0rc4",
   "nonebot-plugin-htmlrender>=0.6.7",
   "yt-dlp[default]>=2026.2.21",
@@ -90,7 +89,7 @@ extras = [
 ]
 test = [
   "nonebug>=0.4.3,<1.0.0",
-  "poethepoet>=0.42.0",
+  "poethepoet>=0.42.1",
   "pytest-asyncio>=1.3.0,<1.4.0",
   "pytest-cov>=7.0.0",
   "pytest-xdist>=3.8.0,<4.0.0",
@@ -114,12 +113,12 @@ conflicts = [
 ]
 
 [tool.uv.sources]
-nonebug = { git = "https://github.com/nonebot/nonebug" }
+nonebug = { git = "https://github.com/nonebot/nonebug", rev = "master" }
 
 [tool.bumpversion]
 tag = true
 commit = true
-current_version = "2.4.3"
+current_version = "2.5.0"
 message = "release: bump vesion from {current_version} to {new_version}"
 
 [[tool.bumpversion.files]]

{nonebot_plugin_parser-2.4.3 → nonebot_plugin_parser-2.5.0}/src/nonebot_plugin_parser/config.py

@@ -1,6 +1,6 @@
 from pathlib import Path
 
-from nonebot import require, get_driver, get_plugin_config
+from nonebot import logger, require, get_driver, get_plugin_config
 from apilmoji import ELK_SH_CDN, EmojiStyle
 from pydantic import BaseModel
 from bilibili_api.video import VideoCodecs, VideoQuality
@@ -143,7 +143,22 @@ class Config(BaseModel):
     @property
     def custom_font(self) -> Path | None:
         """自定义字体"""
-        return (self.data_dir / self.parser_custom_font) if self.parser_custom_font else None
+        if self.parser_custom_font:
+            font_path = self.config_dir / self.parser_custom_font
+            if font_path.exists():
+                return font_path
+
+            # 尝试从旧路径迁移字体文件
+            old_path = self.data_dir / self.parser_custom_font
+            if old_path.exists():
+                try:
+                    old_path.rename(font_path)
+                    logger.info(f"字体文件 {old_path} 成功迁移到 {font_path}")
+                except OSError:
+                    logger.error(f"字体文件迁移失败, 请手动将其移动到 {font_path}")
+                    return old_path
+
+                return font_path
 
     @property
     def need_forward_contents(self) -> bool:

{nonebot_plugin_parser-2.4.3 → nonebot_plugin_parser-2.5.0}/src/nonebot_plugin_parser/helper.py

@@ -101,8 +101,7 @@ class UniHelper:
         """文件 Seg"""
         if not display_name:
             display_name = file.name
-        if not display_name:
-            raise ValueError("文件名不能为空")
+
         if pconfig.use_base64:
             return File(raw=file.read_bytes(), name=display_name)
         else:

{nonebot_plugin_parser-2.4.3 → nonebot_plugin_parser-2.5.0}/src/nonebot_plugin_parser/parsers/__init__.py

@@ -23,7 +23,6 @@ from .data import (
     ImageContent,
     VideoContent,
     DynamicContent,
-    GraphicsContent,
 )
 
 __all__ = [
@@ -31,7 +30,6 @@ __all__ = [
     "Author",
     "BaseParser",
     "DynamicContent",
-    "GraphicsContent",
     "ImageContent",
     "ParseResult",
     "Platform",

{nonebot_plugin_parser-2.4.3 → nonebot_plugin_parser-2.5.0}/src/nonebot_plugin_parser/parsers/base.py

@@ -4,9 +4,9 @@ from typing import TYPE_CHECKING, Any, TypeVar, ClassVar, cast
 from asyncio import Task
 from pathlib import Path
 from collections.abc import Callable, Coroutine
-from typing_extensions import Unpack
+from typing_extensions import Unpack, final
 
-from .data import Platform, ParseResult, ParseResultKwargs
+from .data import Platform, ParseResult, ImageContent, ParseResultKwargs
 from ..config import pconfig as pconfig
 from ..download import DOWNLOADER
 from ..constants import IOS_HEADER, COMMON_HEADER, ANDROID_HEADER, COMMON_TIMEOUT
@@ -83,9 +83,11 @@ class BaseParser:
         """获取所有已注册的 Parser 类"""
         return cls._registry
 
+    @final
     async def parse(self, keyword: str, searched: Match[str]) -> ParseResult:
         return await self._handlers[keyword](self, searched)
 
+    @final
     async def parse_with_redirect(
         self,
         url: str,
@@ -191,8 +193,6 @@ class BaseParser:
         image_urls: list[str],
     ):
         """创建图片内容列表"""
-        from .data import ImageContent
-
         contents: list[ImageContent] = []
         for url in image_urls:
             task = DOWNLOADER.download_img(url, ext_headers=self.headers)
@@ -202,14 +202,13 @@ class BaseParser:
     def create_image_content(
         self,
         url_or_task: str | Task[Path],
+        alt: str | None = None,
     ):
         """创建图片内容"""
-        from .data import ImageContent
-
         if isinstance(url_or_task, str):
             url_or_task = DOWNLOADER.download_img(url_or_task, ext_headers=self.headers)
 
-        return ImageContent(url_or_task)
+        return ImageContent(url_or_task, alt=alt)
 
     def create_dynamic_contents(
         self,
@@ -237,17 +236,9 @@ class BaseParser:
 
         return AudioContent(url_or_task, duration)
 
-    def create_graphics_content(
-        self,
-        image_url: str,
-        text: str | None = None,
-        alt: str | None = None,
-    ):
-        """创建图文内容 图片不能为空 文字可空 渲染时文字在前 图片在后"""
-        from .data import GraphicsContent
-
-        image_task = DOWNLOADER.download_img(image_url, ext_headers=self.headers)
-        return GraphicsContent(image_task, text, alt)
+    def create_empty_graphics(self) -> list[str | ImageContent]:
+        """创建空的图片内容列表"""
+        return []
 
     @property
     def downloader(self):

{nonebot_plugin_parser-2.4.3 → nonebot_plugin_parser-2.5.0}/src/nonebot_plugin_parser/parsers/bilibili/__init__.py

@@ -22,6 +22,7 @@ from ..base import (
 )
 from ..data import Platform, ImageContent, MediaContent
 from ..cookie import ck2dict
+from .dynamic import DynamicInfo
 
 # 选择客户端
 select_client("curl_cffi")
@@ -160,20 +161,31 @@ class BilibiliParser(BaseParser):
         """解析动态或图文"""
         from bilibili_api.dynamic import Dynamic
 
-        from .dynamic import DynamicData
+        from .dynamic import DynamicWrapper
 
         dynamic = Dynamic(dynamic_id, await self.credential)
         if await dynamic.is_article():
             return await self._parse_bilibli_api_opus(dynamic.turn_to_opus())
 
-        dynamic_info = convert(await dynamic.get_info(), DynamicData).item
-        author = self.create_author(dynamic_info.name, dynamic_info.avatar)
+        dynamic_info = convert(await dynamic.get_info(), DynamicWrapper).item
+        return await self._parse_dynamic_info(dynamic_info)
+
+    async def _parse_dynamic_info(self, dynamic_info: DynamicInfo):
+        if dynamic_info.is_video():
+            if (major := dynamic_info.modules.major) and (archive := major.archive):
+                result = await self.parse_video(bvid=archive.bvid)
+                result.text = dynamic_info.text
+                result.extra["content_type"] = "动态"
+                return result
 
         # 下载图片
+        author = self.create_author(dynamic_info.name, dynamic_info.avatar)
         contents: list[MediaContent] = []
-        for image_url in dynamic_info.image_urls:
-            img_task = self.downloader.download_img(image_url, ext_headers=self.headers)
-            contents.append(ImageContent(img_task))
+        contents.extend(self.create_image_contents(dynamic_info.image_urls))
+
+        repost = None
+        if dynamic_info.type == "DYNAMIC_TYPE_FORWARD" and dynamic_info.orig is not None:
+            repost = await self._parse_dynamic_info(dynamic_info.orig)
 
         return self.result(
             title=dynamic_info.title,
@@ -181,6 +193,8 @@ class BilibiliParser(BaseParser):
             timestamp=dynamic_info.timestamp,
             author=author,
             contents=contents,
+            repost=repost,
+            extra={"content_type": "动态"},
         )
 
     async def parse_opus_by_id(self, opus_id: int):
@@ -191,7 +205,7 @@ class BilibiliParser(BaseParser):
     async def _parse_bilibli_api_opus(self, bili_opus: Opus):
         """解析图文动态(Opus)"""
 
-        from .opus import OpusItem, TextNode, ImageNode
+        from .opus import OpusItem
 
         opus_info = await bili_opus.get_info()
         if not isinstance(opus_info, dict):
@@ -201,23 +215,19 @@ class BilibiliParser(BaseParser):
         logger.debug(f"opus_data: {opus_data}")
         author = self.create_author(*opus_data.name_avatar)
 
-        # 按顺序处理图文内容（参考 parse_read 的逻辑）
-        contents: list[MediaContent] = []
-        current_text = ""
-
+        # 按顺序处理图文内容
+        graphics = self.create_empty_graphics()
         for node in opus_data.extract_nodes():
-            if isinstance(node, ImageNode):
-                contents.append(self.create_graphics_content(node.url, current_text.strip(), node.alt))
-                current_text = ""
-            elif isinstance(node, TextNode):
-                current_text += node.text
+            if isinstance(node, str):
+                graphics.append(node)
+            else:
+                graphics.append(self.create_image_content(node.url, alt=node.alt))
 
         return self.result(
             title=opus_data.title,
             author=author,
             timestamp=opus_data.timestamp,
-            contents=contents,
-            text=current_text.strip(),
+            graphics=graphics,
         )
 
     async def parse_live(self, room_id: int):
@@ -266,11 +276,18 @@ class BilibiliParser(BaseParser):
 
         favdata = convert(fav_dict, FavData)
 
+        author = self.create_author(favdata.info.upper.name, favdata.info.upper.face)
+
+        graphics: list[str | ImageContent] = []
+        for fav in favdata.medias:
+            graphics.append(self.create_image_content(fav.cover, alt=fav.desc))
+            graphics.append(fav.desc)
+
         return self.result(
             title=favdata.title,
             timestamp=favdata.timestamp,
-            author=self.create_author(favdata.info.upper.name, favdata.info.upper.face),
-            contents=[self.create_graphics_content(fav.cover, fav.desc) for fav in favdata.medias],
+            author=author,
+            graphics=graphics,
         )
 
     async def _get_video(self, *, bvid: str | None = None, avid: int | None = None) -> Video:

{nonebot_plugin_parser-2.4.3 → nonebot_plugin_parser-2.5.0}/src/nonebot_plugin_parser/parsers/bilibili/dynamic.py

@@ -26,11 +26,26 @@ class VideoArchive(Struct):
     title: str
     desc: str
     cover: str
-    # duration_text: str
+    duration_text: str = ""
     # jump_url: str
     # stat: dict[str, str]
     # badge: dict[str, Any] | None = None
 
+    @property
+    def duration_seconds(self) -> float:
+        """将 duration_text（如 '3:42'）解析为秒数"""
+        if not self.duration_text:
+            return 0.0
+        parts = self.duration_text.split(":")
+        try:
+            if len(parts) == 2:
+                return int(parts[0]) * 60 + int(parts[1])
+            elif len(parts) == 3:
+                return int(parts[0]) * 3600 + int(parts[1]) * 60 + int(parts[2])
+        except ValueError:
+            pass
+        return 0.0
+
 
 class OpusImage(Struct):
     """图文动态图片信息"""
@@ -73,6 +88,8 @@ class DynamicMajor(Struct):
         """获取标题"""
         if self.type == "MAJOR_TYPE_ARCHIVE" and self.archive:
             return self.archive.title
+        if self.type == "MAJOR_TYPE_OPUS" and self.opus:
+            return self.opus.title
         return None
 
     @property
@@ -102,6 +119,13 @@ class DynamicMajor(Struct):
             return self.archive.cover
         return None
 
+    @property
+    def duration(self) -> float:
+        """获取视频时长（秒）"""
+        if self.type == "MAJOR_TYPE_ARCHIVE" and self.archive:
+            return self.archive.duration_seconds
+        return 0.0
+
 
 class DynamicModule(Struct):
     """动态模块"""
@@ -110,6 +134,8 @@ class DynamicModule(Struct):
     module_dynamic: dict[str, Any] | None = None
     module_stat: dict[str, Any] | None = None
 
+    _cached_major: DynamicMajor | None = None
+
     @property
     def author_name(self) -> str:
         """获取作者名称"""
@@ -126,7 +152,7 @@ class DynamicModule(Struct):
         return self.module_author.pub_ts
 
     @property
-    def major_info(self) -> dict[str, Any] | None:
+    def _major_info(self) -> dict[str, Any] | None:
         """获取主要内容信息"""
         if self.module_dynamic:
             if major := self.module_dynamic.get("major"):
@@ -135,6 +161,24 @@ class DynamicModule(Struct):
             return self.module_dynamic
         return None
 
+    @property
+    def major(self) -> DynamicMajor | None:
+        """获取缓存的 DynamicMajor 实例"""
+        if self._cached_major is None:
+            major_info = self._major_info
+            if major_info:
+                self._cached_major = convert(major_info, DynamicMajor)
+        return self._cached_major
+
+    @property
+    def desc_text(self) -> str | None:
+        """获取动态自身的文字描述（非 major 内容的文字）"""
+        if self.module_dynamic:
+            desc = self.module_dynamic.get("desc")
+            if desc and isinstance(desc, dict):
+                return desc.get("text")
+        return None
+
 
 class DynamicInfo(Struct):
     """动态信息"""
@@ -144,6 +188,7 @@ class DynamicInfo(Struct):
     visible: bool
     modules: DynamicModule
     basic: dict[str, Any] | None = None
+    orig: "DynamicInfo | None" = None
 
     @property
     def name(self) -> str:
@@ -163,41 +208,31 @@ class DynamicInfo(Struct):
     @property
     def title(self) -> str | None:
         """获取标题"""
-        major_info = self.modules.major_info
-        if major_info:
-            major = convert(major_info, DynamicMajor)
+        if major := self.modules.major:
             return major.title
-        return None
 
     @property
     def text(self) -> str | None:
-        """获取文本内容"""
-        major_info = self.modules.major_info
-        if major_info:
-            major = convert(major_info, DynamicMajor)
+        """获取文本内容（优先取动态自身文字，回退到 major 的文字）"""
+        # 优先取动态自身描述（如发视频时附带的文字）
+        if desc_text := self.modules.desc_text:
+            return desc_text
+        # 回退到 major 的文字（图文摘要、视频简介等）
+        if major := self.modules.major:
             return major.text
-        return None
 
     @property
     def image_urls(self) -> list[str]:
         """获取图片URL列表"""
-        major_info = self.modules.major_info
-        if major_info:
-            major = convert(major_info, DynamicMajor)
+        if major := self.modules.major:
             return major.image_urls
         return []
 
-    @property
-    def cover_url(self) -> str | None:
-        """获取封面URL"""
-        major_info = self.modules.major_info
-        if major_info:
-            major = convert(major_info, DynamicMajor)
-            return major.cover_url
-        return None
-
+    def is_video(self) -> bool:
+        """判断是否为视频动态"""
+        major = self.modules.major
+        return major is not None and major.archive is not None
 
-class DynamicData(Struct):
-    """动态项目"""
 
+class DynamicWrapper(Struct):
     item: DynamicInfo

{nonebot_plugin_parser-2.4.3 → nonebot_plugin_parser-2.5.0}/src/nonebot_plugin_parser/parsers/bilibili/opus.py

@@ -1,25 +1,9 @@
 from typing import Any
-from collections.abc import Generator
+from dataclasses import dataclass
 
 from msgspec import Struct
 
 
-class TextNode(Struct, tag="TextNode"):
-    """图文动态文本节点"""
-
-    text: str
-    """文本内容"""
-
-
-class ImageNode(Struct, tag="ImageNode"):
-    """图文动态图片节点"""
-
-    url: str
-    """图片链接"""
-    alt: str | None = None
-    """图片描述"""
-
-
 class Author(Struct):
     """图文动态作者信息"""
 
@@ -102,6 +86,16 @@ class Info(Struct):
     basic: Basic | None = None
 
 
+@dataclass(slots=True)
+class ImageNode:
+    """图文动态图片节点"""
+
+    url: str
+    """图片链接"""
+    alt: str | None = None
+    """图片描述"""
+
+
 class OpusItem(Struct):
     """图文动态项目"""
 
@@ -124,30 +118,45 @@ class OpusItem(Struct):
                 return module.module_author.pub_ts
         return None
 
-    def extract_nodes(self) -> Generator[TextNode | ImageNode, None, None]:
-        """生成图文节点（保持顺序）"""
+    def extract_nodes(self):
+        """提取图文节点（保持顺序）"""
         for module in self.item.modules:
             if module.module_type == "MODULE_TYPE_CONTENT" and module.module_content:
-                for paragraph in module.module_content.paragraphs:
+                iterator = iter(module.module_content.paragraphs)
+                for paragraph in iterator:
                     # 处理文本段落
                     if paragraph.text and paragraph.text.nodes:
-                        text_content = self._extract_text_from_nodes(paragraph.text.nodes)
-                        text_content = text_content.strip()
-                        if text_content:
-                            yield TextNode(text="\n\n" + text_content)
-
+                        cur_text = "".join(
+                            text for text, _ in self._extract_texts_from_nodes(paragraph.text.nodes)
+                        ).strip()
+                        if cur_text:
+                            yield cur_text
                     # 处理图片段落
                     if paragraph.pic and paragraph.pic.pics:
                         for pic in paragraph.pic.pics:
-                            yield ImageNode(url=pic.url)
-
-    def _extract_text_from_nodes(self, nodes: list[dict[str, Any]]) -> str:
+                            image_node = ImageNode(url=pic.url)
+                            next_text = ""
+                            if (next_par := next(iterator, None)) and next_par.text and next_par.text.nodes:
+                                for text, color in self._extract_texts_from_nodes(next_par.text.nodes):
+                                    if color == "#999999":
+                                        image_node.alt = text
+                                    else:
+                                        next_text += text
+                            yield image_node
+                            next_text = next_text.strip()
+                            if next_text:
+                                yield next_text
+
+    def _extract_texts_from_nodes(self, nodes: list[dict[str, Any]]) -> list[tuple[str, str | None]]:
         """从节点列表中提取文本内容"""
-        text_content = ""
+        texts: list[tuple[str, str | None]] = []
         for node in nodes:
-            if node.get("type") in [
+            if node.get("type") in (
                 "TEXT_NODE_TYPE_WORD",
                 "TEXT_NODE_TYPE_RICH",
-            ] and node.get("word"):
-                text_content += node["word"].get("words", "")
-        return text_content
+            ) and node.get("word"):
+                text = node["word"]["words"]
+                color = node["word"]["color"]
+                texts.append((text, color))
+
+        return texts