nonebot-plugin-parser 2.0.9__tar.gz → 2.0.11__tar.gz

{nonebot_plugin_parser-2.0.9 → nonebot_plugin_parser-2.0.11}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.3
 Name: nonebot-plugin-parser
-Version: 2.0.9
-Summary: NoneBot2 链接分享解析器自动解析, BV号/链接/小程序/卡片 | B站/抖音/快手/微博/小红书/youtube/tiktok/twitter/acfun
+Version: 2.0.11
+Summary: NoneBot2 链接分享解析 Alconna 版, 通用媒体卡片渲染(PIL 实现), 支持 B站/抖音/快手/微博/小红书/youtube/tiktok/twitter/acfun/nga
 Keywords: nonebot,nonebot2,video,bilibili,youtube,tiktok,twitter,kuaishou,acfun,weibo,xiaohongshu,nga,douyin
 Author: fllesser
 Author-email: fllesser <fllessive@gmail.com>
@@ -19,7 +19,7 @@ Requires-Dist: nonebot-plugin-localstore>=0.7.4,<1.0.0
 Requires-Dist: nonebot-plugin-apscheduler>=0.5.0,<1.0.0
 Requires-Dist: nonebot-plugin-alconna>=0.59.4
 Requires-Dist: nonebot-plugin-uninfo>=0.9.0
-Requires-Dist: nonebot-plugin-htmlkit>=0.1.0rc1 ; extra == 'htmlkit'
+Requires-Dist: nonebot-plugin-htmlkit>=0.1.0rc3 ; extra == 'htmlkit'
 Requires-Dist: jinja2>=3.1.6 ; extra == 'htmlkit'
 Requires-Python: >=3.10
 Project-URL: IssueTracker, https://github.com/fllesser/nonebot-plugin-parser/issues
@@ -52,8 +52,6 @@ Description-Content-Type: text/markdown
 
 ## 📖 介绍
 
-[nonebot-plugin-resolver](https://github.com/zhiyu1998/nonebot-plugin-resolver) 重制版
-
 | 平台    | 触发的消息形态                        | 视频 | 图集 | 音频 |
 | ------- | ------------------------------------- | ---- | ---- | ---- |
 | B站     | BV号/链接(包含短链,BV,av)/卡片/小程序 | ✅​   | ✅​   | ✅​   |
@@ -69,7 +67,7 @@ Description-Content-Type: text/markdown
 支持的链接，可参考 [测试链接](https://github.com/fllesser/nonebot-plugin-parser/blob/master/test_url.md)
 
 ## 🎨 效果图
-
+插件默认启用 PIL 实现的通用媒体卡片渲染，效果图如下
 <div align="center">
 
 <img src="https://raw.githubusercontent.com/fllesser/nonebot-plugin-parser/refs/heads/resources/resources/renderdamine/video.png" width="160" />
@@ -82,7 +80,7 @@ Description-Content-Type: text/markdown
 
 ## 💿 安装
 > [!Warning]
-> **如果你已经在使用 nonebot-plugin-resolver，请在安装此插件前卸载**
+> **如果你已经在使用 nonebot-plugin-resolver[2]，请在安装此插件前卸载**
     
 <details open>
 <summary>使用 nb-cli 安装/更新</summary>

{nonebot_plugin_parser-2.0.9 → nonebot_plugin_parser-2.0.11}/README.md

@@ -22,8 +22,6 @@
 
 ## 📖 介绍
 
-[nonebot-plugin-resolver](https://github.com/zhiyu1998/nonebot-plugin-resolver) 重制版
-
 | 平台    | 触发的消息形态                        | 视频 | 图集 | 音频 |
 | ------- | ------------------------------------- | ---- | ---- | ---- |
 | B站     | BV号/链接(包含短链,BV,av)/卡片/小程序 | ✅​   | ✅​   | ✅​   |
@@ -39,7 +37,7 @@
 支持的链接，可参考 [测试链接](https://github.com/fllesser/nonebot-plugin-parser/blob/master/test_url.md)
 
 ## 🎨 效果图
-
+插件默认启用 PIL 实现的通用媒体卡片渲染，效果图如下
 <div align="center">
 
 <img src="https://raw.githubusercontent.com/fllesser/nonebot-plugin-parser/refs/heads/resources/resources/renderdamine/video.png" width="160" />
@@ -52,7 +50,7 @@
 
 ## 💿 安装
 > [!Warning]
-> **如果你已经在使用 nonebot-plugin-resolver，请在安装此插件前卸载**
+> **如果你已经在使用 nonebot-plugin-resolver[2]，请在安装此插件前卸载**
     
 <details open>
 <summary>使用 nb-cli 安装/更新</summary>

{nonebot_plugin_parser-2.0.9 → nonebot_plugin_parser-2.0.11}/pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "nonebot-plugin-parser"
-version = "2.0.9"
-description = "NoneBot2 链接分享解析器自动解析, BV号/链接/小程序/卡片 | B站/抖音/快手/微博/小红书/youtube/tiktok/twitter/acfun"
+version = "2.0.11"
+description = "NoneBot2 链接分享解析 Alconna 版, 通用媒体卡片渲染(PIL 实现), 支持 B站/抖音/快手/微博/小红书/youtube/tiktok/twitter/acfun/nga"
 authors = [{ "name" = "fllesser", "email" = "fllessive@gmail.com" }]
 readme = "README.md"
 requires-python = ">=3.10"
@@ -41,7 +41,7 @@ dependencies = [
 ]
 
 [project.optional-dependencies]
-htmlkit = ["nonebot-plugin-htmlkit>=0.1.0rc1", "jinja2>=3.1.6"]
+htmlkit = ["nonebot-plugin-htmlkit>=0.1.0rc3", "jinja2>=3.1.6"]
 
 [project.urls]
 Repository = "https://github.com/fllesser/nonebot-plugin-parser"
@@ -54,7 +54,7 @@ dev = [
   "nonebot2[fastapi]>=2.4.3,<3.0.0",
   "nonebot-adapter-telegram>=0.1.0b20",
   "pre-commit>=4.3.0",
-  "ruff>=0.13.3,<1.0.0",
+  "ruff>=0.14.0,<1.0.0",
   "bump-my-version>=1.2.4",
 ]
 
@@ -72,7 +72,7 @@ test = [
 all_extras = ["nonebot-plugin-htmlkit>=0.1.0rc1", "jinja2>=3.1.6"]
 
 [tool.uv]
-required-version = ">=0.8.14"
+required-version = ">=0.9.2"
 default-groups = ["test", "dev", "all_extras"]
 
 [tool.nonebot]
@@ -99,7 +99,7 @@ addopts = [
 [tool.poe.tasks]
 test_others = "pytest tests/others --cov=src --cov-report=xml:coverage1.xml --junitxml=junit1.xml -n auto"
 test_parsers = "pytest tests/parsers --cov=src --cov-report=xml:coverage2.xml --junitxml=junit2.xml -n auto"
-test_render = "pytest tests/render --cov=src --cov-report=xml:coverage3.xml --junitxml=junit3.xml -n auto --log-cli-level=DEBUG"
+test_render = "pytest tests/render --cov=src --cov-report=xml:coverage3.xml --junitxml=junit3.xml"
 bump = "bump-my-version bump"
 show-bump = "bump-my-version show-bump"
 
@@ -186,7 +186,7 @@ build-backend = "uv_build"
 
 
 [tool.bumpversion]
-current_version = "2.0.9"
+current_version = "2.0.11"
 commit = true
 message = "🔖 release: bump vesion from {current_version} to {new_version}"
 tag = true

{nonebot_plugin_parser-2.0.9 → nonebot_plugin_parser-2.0.11}/src/nonebot_plugin_parser/__init__.py

@@ -11,8 +11,8 @@ from .matchers import clear_result_cache
 from .utils import safe_unlink
 
 __plugin_meta__ = PluginMetadata(
-    name="链接分享自动解析",
-    description="BV号/链接/小程序/卡片 | B站/抖音/快手/微博/小红书/youtube/tiktok/twitter/acfun",
+    name="链接分享解析 Alconna 版",
+    description="全新通用媒体卡片渲染(PIL 实现), 支持 B站/抖音/快手/微博/小红书/youtube/tiktok/twitter/acfun/nga",
     usage="发送支持平台的(BV号/链接/小程序/卡片)即可",
     type="application",
     homepage="https://github.com/fllesser/nonebot-plugin-parser",

{nonebot_plugin_parser-2.0.9 → nonebot_plugin_parser-2.0.11}/src/nonebot_plugin_parser/config.py

@@ -1,4 +1,5 @@
 from enum import Enum
+from functools import cached_property
 from pathlib import Path
 from typing import Literal
 
@@ -56,22 +57,22 @@ class Config(BaseModel):
     parser_need_forward_contents: bool = True
     """是否需要转发媒体内容"""
 
-    @property
+    @cached_property
     def nickname(self) -> str:
         """全局名称"""
         return _nickname
 
-    @property
+    @cached_property
     def cache_dir(self) -> Path:
         """插件缓存目录"""
         return _cache_dir
 
-    @property
+    @cached_property
     def config_dir(self) -> Path:
         """插件配置目录"""
         return _config_dir
 
-    @property
+    @cached_property
     def data_dir(self) -> Path:
         """插件数据目录"""
         return _data_dir
@@ -131,7 +132,7 @@ class Config(BaseModel):
         """是否在解析结果中附加原始URL"""
         return self.parser_append_url
 
-    @property
+    @cached_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

{nonebot_plugin_parser-2.0.9 → nonebot_plugin_parser-2.0.11}/src/nonebot_plugin_parser/parsers/__init__.py

@@ -1,7 +1,6 @@
 # 导出所有 Parser 类
-from .acfun import AcfunParser as AcfunParser
+from .acfun import AcfunParser as AcfunParser  # noqa: I001
 from .base import BaseParser as BaseParser
-from .bilibili import BilibiliParser as BilibiliParser
 from .data import ParseResult as ParseResult
 from .douyin import DouyinParser as DouyinParser
 from .kuaishou import KuaiShouParser as KuaiShouParser
@@ -11,6 +10,7 @@ from .twitter import TwitterParser as TwitterParser
 from .weibo import WeiBoParser as WeiBoParser
 from .xiaohongshu import XiaoHongShuParser as XiaoHongShuParser
 from .youtube import YouTubeParser as YouTubeParser
+from .bilibili import BilibiliParser as BilibiliParser
 
 __all__ = [
     "ParseResult",

{nonebot_plugin_parser-2.0.9 → nonebot_plugin_parser-2.0.11}/src/nonebot_plugin_parser/parsers/bilibili/__init__.py

@@ -460,15 +460,15 @@ class BilibiliParser(BaseParser):
         """初始化 bilibili api"""
 
         if not pconfig.bili_ck:
-            logger.warning("未配置 r_bili_ck, 无法使用哔哩哔哩 AI 总结, 可能无法解析 720p 以上画质视频")
+            logger.warning("未配置 parser_bili_ck, 无法使用哔哩哔哩 AI 总结, 可能无法解析 720p 以上画质视频")
             return None
 
         credential = Credential.from_cookies(ck2dict(pconfig.bili_ck))
         if not await credential.check_valid() and self._cookies_file.exists():
-            logger.info(f"r_bili_ck 已过期, 尝试从 {self._cookies_file} 加载")
+            logger.info(f"parser_bili_ck 已过期, 尝试从 {self._cookies_file} 加载")
             credential = Credential.from_cookies(json.loads(self._cookies_file.read_text()))
         else:
-            logger.info(f"r_bili_ck 有效, 保存到 {self._cookies_file}")
+            logger.info(f"parser_bili_ck 有效, 保存到 {self._cookies_file}")
             self._cookies_file.write_text(json.dumps(credential.get_cookies()))
 
         return credential
@@ -483,7 +483,7 @@ class BilibiliParser(BaseParser):
                 return None
 
         if not await self._credential.check_valid():
-            logger.warning("哔哩哔哩 cookies 已过期, 请重新配置 r_bili_ck")
+            logger.warning("哔哩哔哩 cookies 已过期, 请重新配置 parser_bili_ck")
             return self._credential
 
         if await self._credential.check_refresh():

nonebot_plugin_parser-2.0.11/src/nonebot_plugin_parser/parsers/xiaohongshu.py

@@ -0,0 +1,259 @@
+import json
+import re
+from typing import Any, ClassVar
+from typing_extensions import override
+from urllib.parse import urlparse
+
+import httpx
+import msgspec
+from msgspec import Struct, field
+from nonebot import logger
+
+from ..exception import ParseException
+from .base import BaseParser, Platform
+
+
+class XiaoHongShuParser(BaseParser):
+    # 平台信息
+    platform: ClassVar[Platform] = Platform(name="xiaohongshu", display_name="小红书")
+
+    # URL 正则表达式模式（keyword, pattern）
+    patterns: ClassVar[list[tuple[str, str]]] = [
+        ("xiaohongshu.com", r"https?://(?:www\.)?xiaohongshu\.com/[A-Za-z0-9._?%&+=/#@-]*"),
+        ("xhslink.com", r"https?://xhslink\.com/[A-Za-z0-9._?%&+=/#@-]*"),
+    ]
+
+    def __init__(self):
+        super().__init__()
+        explore_headers = {
+            "accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,"
+            "image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7",
+        }
+        self.headers.update(explore_headers)
+        discovery_headers = {
+            "origin": "https://www.xiaohongshu.com",
+            "x-requested-with": "XMLHttpRequest",
+            "sec-fetch-site": "same-origin",
+            "sec-fetch-mode": "cors",
+            "sec-fetch-dest": "empty",
+        }
+        self.ios_headers.update(discovery_headers)
+
+    @override
+    async def parse(self, matched: re.Match[str]):
+        """解析 URL 获取内容信息并下载资源
+
+        Args:
+            matched: 正则表达式匹配对象，由平台对应的模式匹配得到
+
+        Returns:
+            ParseResult: 解析结果
+
+        Raises:
+            ParseException: 解析失败时抛出
+        """
+        # 从匹配对象中获取原始URL
+        url = matched.group(0)
+        # 处理 xhslink 短链
+        if "xhslink" in url:
+            url = await self.get_redirect_url(url, self.ios_headers)
+            logger.debug(f"xhslink redirect url: {url}")
+
+        urlpath = urlparse(url).path
+
+        if urlpath.startswith("/explore/"):
+            xhs_id = urlpath.split("/")[-1]
+            return await self._parse_explore(url, xhs_id)
+        elif urlpath.startswith("/discovery/item/"):
+            return await self._parse_discovery(url)
+        else:
+            raise ParseException(f"不支持的小红书链接: {url}, urlpath: {urlpath}")
+
+    async def _parse_explore(self, url: str, xhs_id: str):
+        async with httpx.AsyncClient(
+            headers=self.headers,
+            timeout=self.timeout,
+        ) as client:
+            response = await client.get(url)
+            html = response.text
+            logger.info(f"url: {response.url} | status_code: {response.status_code}")
+
+        json_obj = self._extract_initial_state_json(html)
+
+        # ["note"]["noteDetailMap"][xhs_id]["note"]
+        note_data = json_obj.get("note", {}).get("noteDetailMap", {}).get(xhs_id, {}).get("note", {})
+        if not note_data:
+            raise ParseException("can't find note detail in json_obj")
+
+        class Image(Struct):
+            urlDefault: str
+
+        class User(Struct):
+            nickname: str
+            avatar: str
+
+        class NoteDetail(Struct):
+            type: str
+            title: str
+            desc: str
+            user: User
+            imageList: list[Image] = field(default_factory=list)
+            video: Video | None = None
+
+            @property
+            def nickname(self) -> str:
+                return self.user.nickname
+
+            @property
+            def avatar_url(self) -> str:
+                return self.user.avatar
+
+            @property
+            def image_urls(self) -> list[str]:
+                return [item.urlDefault for item in self.imageList]
+
+            @property
+            def video_url(self) -> str | None:
+                if self.type != "video" or not self.video:
+                    return None
+                return self.video.video_url
+
+        note_detail = msgspec.convert(note_data, type=NoteDetail)
+
+        contents = []
+        # 添加视频内容
+        if video_url := note_detail.video_url:
+            # 使用第一张图片作为封面
+            cover_url = note_detail.image_urls[0] if note_detail.image_urls else None
+            contents.append(self.create_video_content(video_url, cover_url))
+
+        # 添加图片内容
+        elif image_urls := note_detail.image_urls:
+            contents.extend(self.create_image_contents(image_urls))
+
+        # 构建作者
+        author = self.create_author(note_detail.nickname, note_detail.avatar_url)
+
+        return self.result(
+            title=note_detail.title,
+            text=note_detail.desc,
+            author=author,
+            contents=contents,
+        )
+
+    async def _parse_discovery(self, url: str):
+        async with httpx.AsyncClient(
+            headers=self.ios_headers,
+            timeout=self.timeout,
+            follow_redirects=True,
+            cookies=httpx.Cookies(),
+            trust_env=False,
+        ) as client:
+            response = await client.get(url)
+            html = response.text
+
+        json_obj = self._extract_initial_state_json(html)
+        note_data = json_obj.get("noteData")
+        if not note_data:
+            raise ParseException("can't find noteData in json_obj")
+        preload_data = note_data.get("normalNotePreloadData", {})
+        note_data = note_data.get("data", {}).get("noteData", {})
+        if not note_data:
+            raise ParseException("can't find noteData in noteData.data")
+
+        class Image(Struct):
+            url: str
+            urlSizeLarge: str | None = None
+
+        class User(Struct):
+            nickName: str
+            avatar: str
+
+        class NoteData(Struct):
+            type: str
+            title: str
+            desc: str
+            user: User
+            time: int
+            lastUpdateTime: int
+            imageList: list[Image] = []  # 有水印
+            video: Video | None = None
+
+            @property
+            def image_urls(self) -> list[str]:
+                return [item.url for item in self.imageList]
+
+            @property
+            def video_url(self) -> str | None:
+                if self.type != "video" or not self.video:
+                    return None
+                return self.video.video_url
+
+        class NormalNotePreloadData(Struct):
+            title: str
+            desc: str
+            imagesList: list[Image] = []  # 无水印, 但只有一只，用于视频封面
+
+            @property
+            def image_urls(self) -> list[str]:
+                return [item.urlSizeLarge or item.url for item in self.imagesList]
+
+        note_data = msgspec.convert(note_data, type=NoteData)
+
+        contents = []
+        if video_url := note_data.video_url:
+            if preload_data:
+                preload_data = msgspec.convert(preload_data, type=NormalNotePreloadData)
+                img_urls = preload_data.image_urls
+            else:
+                img_urls = note_data.image_urls
+            contents.append(self.create_video_content(video_url, img_urls[0]))
+        elif img_urls := note_data.image_urls:
+            contents.extend(self.create_image_contents(img_urls))
+
+        return self.result(
+            title=note_data.title,
+            author=self.create_author(note_data.user.nickName, note_data.user.avatar),
+            contents=contents,
+            text=note_data.desc,
+            timestamp=note_data.time // 1000,
+        )
+
+    def _extract_initial_state_json(self, html: str) -> dict[str, Any]:
+        pattern = r"window\.__INITIAL_STATE__=(.*?)</script>"
+        matched = re.search(pattern, html)
+        if not matched:
+            raise ParseException("小红书分享链接失效或内容已删除")
+
+        json_str = matched.group(1).replace("undefined", "null")
+        return json.loads(json_str)
+
+
+class Stream(Struct):
+    h264: list[dict[str, Any]] | None = None
+    h265: list[dict[str, Any]] | None = None
+    av1: list[dict[str, Any]] | None = None
+    h266: list[dict[str, Any]] | None = None
+
+
+class Media(Struct):
+    stream: Stream
+
+
+class Video(Struct):
+    media: Media
+
+    @property
+    def video_url(self) -> str | None:
+        stream = self.media.stream
+
+        # h264 有水印，h265 无水印
+        if stream.h265:
+            return stream.h265[0]["masterUrl"]
+        elif stream.h264:
+            return stream.h264[0]["masterUrl"]
+        elif stream.av1:
+            return stream.av1[0]["masterUrl"]
+        elif stream.h266:
+            return stream.h266[0]["masterUrl"]
+        return None

{nonebot_plugin_parser-2.0.9 → nonebot_plugin_parser-2.0.11}/src/nonebot_plugin_parser/renders/common.py

@@ -1,4 +1,5 @@
 from dataclasses import dataclass
+from functools import lru_cache
 from io import BytesIO
 from pathlib import Path
 from typing import ClassVar
@@ -19,6 +20,53 @@ class FontInfo:
     cjk_width: int
     ascii_width: int
 
+    def __hash__(self) -> int:
+        """实现哈希方法以支持 @lru_cache"""
+        return hash((self.line_height, self.cjk_width, self.ascii_width))
+
+    @lru_cache(maxsize=100)
+    def get_char_width(self, char: str) -> int:
+        """获取字符宽度，使用缓存优化"""
+        bbox = self.font.getbbox(char)
+        width = int(bbox[2] - bbox[0])
+        return width
+
+    def get_char_width_fast(self, char: str) -> int:
+        """快速获取单个字符宽度"""
+        if self._is_cjk_char(char):
+            return self.cjk_width
+        elif self._is_ascii_char(char):
+            return self.ascii_width
+        else:
+            return self.get_char_width(char)
+
+    def get_text_width(self, text: str) -> int:
+        """计算文本宽度，使用预计算的字符宽度优化性能
+
+        Args:
+            text: 要计算宽度的文本
+
+        Returns:
+            文本宽度（像素）
+        """
+        if not text:
+            return 0
+
+        total_width = 0
+        for char in text:
+            total_width += self.get_char_width_fast(char)
+        return total_width
+
+    @staticmethod
+    def _is_cjk_char(char: str) -> bool:
+        """判断是否为中日韩字符"""
+        return "\u4e00" <= char <= "\u9fff"
+
+    @staticmethod
+    def _is_ascii_char(char: str) -> bool:
+        """判断是否为ASCII字符"""
+        return ord(char) < 128
+
 
 @dataclass(eq=False, frozen=True, slots=True)
 class FontSet:
@@ -28,6 +76,7 @@ class FontSet:
     title_font: FontInfo
     text_font: FontInfo
     extra_font: FontInfo
+    indicator_font: FontInfo
 
 
 @dataclass(eq=False, frozen=True, slots=True)
@@ -169,9 +218,9 @@ class CommonRenderer(ImageRenderer):
     """转发缩放比例"""
 
     # 字体大小和行高
-    FONT_SIZES: ClassVar[dict[str, int]] = {"name": 28, "title": 30, "text": 24, "extra": 24}
+    FONT_SIZES: ClassVar[dict[str, int]] = {"name": 28, "title": 30, "text": 24, "extra": 24, "indicator": 60}
     """字体大小"""
-    LINE_HEIGHTS: ClassVar[dict[str, int]] = {"name": 32, "title": 36, "text": 28, "extra": 28}
+    LINE_HEIGHTS: ClassVar[dict[str, int]] = {"name": 32, "title": 36, "text": 28, "extra": 28, "indicator": 68}
     """行高"""
 
     RESOURCES_DIR: ClassVar[Path] = Path(__file__).parent / "resources"
@@ -228,6 +277,7 @@ class CommonRenderer(ImageRenderer):
             title_font=font_infos["title"],
             text_font=font_infos["text"],
             extra_font=font_infos["extra"],
+            indicator_font=font_infos["indicator"],
         )
 
         logger.success(f"加载字体「{self.font_path.name}」成功")
@@ -859,11 +909,11 @@ class CommonRenderer(ImageRenderer):
         if section.alt_text:
             y_pos += self.SECTION_SPACING  # 图片和alt文本之间的间距
             # 计算文本居中位置
-            bbox = self.fontset.extra_font.font.getbbox(section.alt_text)
-            text_width = bbox[2] - bbox[0]
+            extra_font_info = self.fontset.extra_font
+            text_width = extra_font_info.get_text_width(section.alt_text)
             text_x = self.PADDING + (content_width - text_width) // 2
-            draw.text((text_x, y_pos), section.alt_text, fill=self.EXTRA_COLOR, font=self.fontset.extra_font.font)
-            y_pos += self.fontset.extra_font.line_height
+            draw.text((text_x, y_pos), section.alt_text, fill=self.EXTRA_COLOR, font=extra_font_info.font)
+            y_pos += extra_font_info.line_height
 
         return y_pos + self.SECTION_SPACING
 
@@ -985,19 +1035,14 @@ class CommonRenderer(ImageRenderer):
 
         # 绘制+N文字
         text = f"+{count}"
-        # 使用更大的字体
-        font_size = min(img_width, img_height) // 4
-        font = ImageFont.truetype(self.font_path, font_size)
-
+        font_info = self.fontset.indicator_font
         # 计算文字位置（居中）
-        bbox = font.getbbox(text)
-        text_width = bbox[2] - bbox[0]
-        text_height = bbox[3] - bbox[1]
+        text_width = font_info.get_text_width(text)
         text_x = img_x + (img_width - text_width) // 2
-        text_y = img_y + (img_height - text_height) // 2
+        text_y = img_y + (img_height - font_info.line_height) // 2
 
-        # 绘制白色文字
-        draw.text((text_x, text_y), text, fill=(255, 255, 255, 255), font=font)
+        # 绘制50%透明白色文字
+        draw.text((text_x, text_y), text, fill=(255, 255, 255), font=font_info.font)
 
     def _draw_rounded_rectangle(
         self, image: Image.Image, bbox: tuple[int, int, int, int], fill_color: tuple[int, int, int], radius: int = 8
@@ -1056,81 +1101,14 @@ class CommonRenderer(ImageRenderer):
         lines = []
         paragraphs = text.split("\n")
 
-        # 字符宽度缓存
-        char_width_cache = {}
-
-        def get_char_width(char: str) -> int:
-            """获取字符宽度，使用缓存优化"""
-            if char in char_width_cache:
-                return char_width_cache[char]
-
-            bbox = font_info.font.getbbox(char)
-            width = int(bbox[2] - bbox[0])
-            char_width_cache[char] = width
-            return width
-
-        def is_cjk_char(char: str) -> bool:
-            """判断是否为中日韩字符"""
-            return "\u4e00" <= char <= "\u9fff"
-
-        def is_ascii_char(char: str) -> bool:
-            """判断是否为ASCII字符"""
-            return ord(char) < 128
-
         def is_punctuation(char: str) -> bool:
-            """判断是否为标点符号"""
+            """判断是否为不能为行首的标点符号"""
             # 中文标点符号
-            chinese_punctuation = "，。！？；：、''（）【】《》〈〉「」『』〔〕〖〗〘〙〚〛…—·"
+            chinese_punctuation = "，。！？；：、）】》〉」』〕〗〙〛…—·"
             # 英文标点符号
-            english_punctuation = ",.;:!?()[]{}'\"-"
-            # Unicode 标点符号类别
-            import unicodedata
-
-            return (
-                char in chinese_punctuation or char in english_punctuation or unicodedata.category(char).startswith("P")
-            )
-
-        def get_text_width_fast(text: str) -> int:
-            """快速计算文本宽度"""
-            if not text:
-                return 0
-
-            total_width = 0
-            for char in text:
-                if is_cjk_char(char):
-                    total_width += font_info.cjk_width
-                elif is_ascii_char(char):
-                    total_width += font_info.ascii_width
-                else:
-                    total_width += get_char_width(char)
-            return total_width
-
-        def find_break_point(text: str) -> int:
-            """找到合适的断点位置，避免标点符号在行首"""
-            if not text:
-                return 0
-
-            # 从后往前找断点
-            for i in range(len(text) - 1, 0, -1):
-                char = text[i]
-
-                # 优先在空格处断行
-                if char == " ":
-                    return i
-
-                # 对于中文，可以在任意字符处断行
-                if is_cjk_char(char):
-                    return i
+            english_punctuation = ",.;:!?)]}"
 
-                # 对于标点符号，不能在行首，需要跳过
-                if is_punctuation(char):
-                    continue
-
-                # 其他字符可以作为断点
-                return i
-
-            # 如果找不到合适的断点，在中间位置断行
-            return max(1, len(text) // 2)
+            return char in chinese_punctuation or char in english_punctuation
 
         for paragraph in paragraphs:
             if not paragraph:
@@ -1138,51 +1116,41 @@ class CommonRenderer(ImageRenderer):
                 continue
 
             current_line = ""
+            current_line_width = 0
             remaining_text = paragraph
 
             while remaining_text:
+                next_char = remaining_text[0]
+                char_width = font_info.get_char_width_fast(next_char)
+
                 # 如果当前行为空，直接添加字符
                 if not current_line:
-                    current_line = remaining_text[0]
+                    current_line = next_char
+                    current_line_width = char_width
                     remaining_text = remaining_text[1:]
                     continue
 
-                # 测试添加下一个字符
-                test_line = current_line + remaining_text[0]
-                test_width = get_text_width_fast(test_line)
+                # 如果是标点符号，直接添加到当前行（标点符号不应该单独成行）
+                if is_punctuation(next_char):
+                    current_line += next_char
+                    current_line_width += char_width
+                    remaining_text = remaining_text[1:]
+                    continue
+
+                # 测试添加下一个字符后的宽度
+                test_width = current_line_width + char_width
 
                 if test_width <= max_width:
                     # 宽度合适，继续添加
-                    current_line = test_line
+                    current_line += next_char
+                    current_line_width = test_width
                     remaining_text = remaining_text[1:]
                 else:
                     # 宽度超限，需要断行
-                    if len(current_line) == 1:
-                        # 单个字符就超宽，强制添加
-                        lines.append(current_line)
-                        current_line = remaining_text[0]
-                        remaining_text = remaining_text[1:]
-                    else:
-                        # 尝试找到合适的断点
-                        break_point = find_break_point(current_line)
-
-                        # 保存当前行
-                        lines.append(current_line[:break_point].rstrip())
-
-                        # 开始新行，跳过行首的标点符号
-                        current_line = current_line[break_point:].lstrip()
-
-                        # 如果新行以标点符号开头，将其移到上一行
-                        while current_line and is_punctuation(current_line[0]):
-                            if lines:
-                                lines[-1] += current_line[0]
-                                current_line = current_line[1:]
-                            else:
-                                break
-
-                        if not current_line:
-                            current_line = remaining_text[0]
-                            remaining_text = remaining_text[1:]
+                    lines.append(current_line)
+                    current_line = next_char
+                    current_line_width = char_width
+                    remaining_text = remaining_text[1:]
 
             # 保存最后一行
             if current_line:

nonebot_plugin_parser-2.0.9/src/nonebot_plugin_parser/parsers/xiaohongshu.py

@@ -1,170 +0,0 @@
-import json
-import re
-from typing import ClassVar
-from typing_extensions import override
-from urllib.parse import parse_qs, urlparse
-
-import httpx
-import msgspec
-
-from ..exception import ParseException
-from .base import BaseParser, Platform
-
-
-class XiaoHongShuParser(BaseParser):
-    # 平台信息
-    platform: ClassVar[Platform] = Platform(name="xiaohongshu", display_name="小红书")
-
-    # URL 正则表达式模式（keyword, pattern）
-    patterns: ClassVar[list[tuple[str, str]]] = [
-        ("xiaohongshu.com", r"https?://(?:www\.)?xiaohongshu\.com/[A-Za-z0-9._?%&+=/#@-]*"),
-        ("xhslink.com", r"https?://xhslink\.com/[A-Za-z0-9._?%&+=/#@-]*"),
-    ]
-
-    def __init__(self):
-        super().__init__()
-        extra_headers = {
-            "accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,"
-            "application/signed-exchange;v=b3;q=0.9",
-        }
-        self.headers.update(extra_headers)
-
-    @override
-    async def parse(self, matched: re.Match[str]):
-        """解析 URL 获取内容信息并下载资源
-
-        Args:
-            matched: 正则表达式匹配对象，由平台对应的模式匹配得到
-
-        Returns:
-            ParseResult: 解析结果
-
-        Raises:
-            ParseException: 解析失败时抛出
-        """
-        # 从匹配对象中获取原始URL
-        url = matched.group(0)
-        # 处理 xhslink 短链
-        if "xhslink" in url:
-            url = await self.get_redirect_url(url, self.headers)
-        # ?: 非捕获组
-        pattern = r"(?:/explore/|/discovery/item/|source=note&noteId=)(\w+)"
-        match_result = re.search(pattern, url)
-        if not match_result:
-            raise ParseException("小红书分享链接不完整")
-        xhs_id = match_result.group(1)
-        # 解析 URL 参数
-        parsed_url = urlparse(url)
-        params = parse_qs(parsed_url.query)
-        # 提取 xsec_source 和 xsec_token
-        xsec_source = params.get("xsec_source", [None])[0] or "pc_feed"
-        xsec_token = params.get("xsec_token", [None])[0]
-
-        # 构造完整 URL
-        url = f"https://www.xiaohongshu.com/explore/{xhs_id}?xsec_source={xsec_source}&xsec_token={xsec_token}"
-        async with httpx.AsyncClient(headers=self.headers, timeout=self.timeout) as client:
-            response = await client.get(url)
-            html = response.text
-
-        pattern = r"window.__INITIAL_STATE__=(.*?)</script>"
-        match_result = re.search(pattern, html)
-        if not match_result:
-            raise ParseException("小红书分享链接失效或内容已删除")
-
-        json_str = match_result.group(1)
-        json_str = json_str.replace("undefined", "null")
-
-        json_obj = json.loads(json_str)
-
-        note_data = json_obj["note"]["noteDetailMap"][xhs_id]["note"]
-        note_detail = msgspec.convert(note_data, type=NoteDetail)
-
-        # 使用新的简洁构建方式
-        contents = []
-
-        # 添加视频内容
-        if video_url := note_detail.video_url:
-            # 使用第一张图片作为封面
-            cover_url = note_detail.image_urls[0] if note_detail.image_urls else None
-            contents.append(self.create_video_content(video_url, cover_url))
-
-        # 添加图片内容
-        elif image_urls := note_detail.image_urls:
-            contents.extend(self.create_image_contents(image_urls))
-
-        # 构建作者
-        author = self.create_author(note_detail.nickname, note_detail.avatar_url)
-
-        return self.result(
-            title=note_detail.title_desc,
-            author=author,
-            contents=contents,
-        )
-
-
-from msgspec import Struct, field
-
-
-class Image(Struct):
-    urlDefault: str
-
-
-class Stream(Struct):
-    h264: list[dict] | None = None
-    h265: list[dict] | None = None
-    av1: list[dict] | None = None
-    h266: list[dict] | None = None
-
-
-class Media(Struct):
-    stream: Stream
-
-
-class Video(Struct):
-    media: Media
-
-
-class User(Struct):
-    nickname: str
-    avatar: str
-
-
-class NoteDetail(Struct):
-    type: str
-    title: str
-    desc: str
-    user: User
-    imageList: list[Image] = field(default_factory=list)
-    video: Video | None = None
-
-    @property
-    def nickname(self) -> str:
-        return self.user.nickname
-
-    @property
-    def avatar_url(self) -> str:
-        return self.user.avatar
-
-    @property
-    def title_desc(self) -> str:
-        return f"{self.title}\n{self.desc}".strip()
-
-    @property
-    def image_urls(self) -> list[str]:
-        return [item.urlDefault for item in self.imageList]
-
-    @property
-    def video_url(self) -> str | None:
-        if self.type != "video" or not self.video:
-            return None
-        stream = self.video.media.stream
-
-        if stream.h264:
-            return stream.h264[0]["masterUrl"]
-        elif stream.h265:
-            return stream.h265[0]["masterUrl"]
-        elif stream.av1:
-            return stream.av1[0]["masterUrl"]
-        elif stream.h266:
-            return stream.h266[0]["masterUrl"]
-        return None