feat(bilibili): 增强哔哩哔哩图文动态解析

src/nonebot_plugin_parser/parsers/bilibili/__init__.py

@@ -48,7 +48,10 @@ async def _parse_short_link(self, searched: Match[str]):
         return await self.parse_with_redirect(url)
 
     @handle("BV", r"^(?P<bvid>BV[0-9a-zA-Z]{10})(?:\s)?(?P<page_num>\d{1,3})?$")
-    @handle("/BV", r"bilibili\.com(?:/video)?/(?P<bvid>BV[0-9a-zA-Z]{10})(?:\?p=(?P<page_num>\d{1,3}))?")
+    @handle(
+        "/BV",
+        r"bilibili\.com(?:/video)?/(?P<bvid>BV[0-9a-zA-Z]{10})(?:\?p=(?P<page_num>\d{1,3}))?",
+    )
     async def _parse_bv(self, searched: Match[str]):
         """解析视频信息"""
         bvid = str(searched.group("bvid"))
@@ -57,7 +60,10 @@ async def _parse_bv(self, searched: Match[str]):
         return await self.parse_video(bvid=bvid, page_num=page_num)
 
     @handle("av", r"^av(?P<avid>\d{6,})(?:\s)?(?P<page_num>\d{1,3})?$")
-    @handle("/av", r"bilibili\.com(?:/video)?/av(?P<avid>\d{6,})(?:\?p=(?P<page_num>\d{1,3}))?")
+    @handle(
+        "/av",
+        r"bilibili\.com(?:/video)?/av(?P<avid>\d{6,})(?:\?p=(?P<page_num>\d{1,3}))?",
+    )
     async def _parse_av(self, searched: Match[str]):
         """解析视频信息"""
         avid = int(searched.group("avid"))
@@ -67,10 +73,11 @@ async def _parse_av(self, searched: Match[str]):
 
     @handle("/dynamic/", r"bilibili\.com/dynamic/(?P<dynamic_id>\d+)")
     @handle("t.bili", r"t\.bilibili\.com/(?P<dynamic_id>\d+)")
+    @handle("/opus/", r"bilibili\.com/opus/(?P<dynamic_id>\d+)")
     async def _parse_dynamic(self, searched: Match[str]):
         """解析动态信息"""
         dynamic_id = int(searched.group("dynamic_id"))
-        return await self.parse_dynamic(dynamic_id)
+        return await self.parse_dynamic_or_opus(dynamic_id)
 
     @handle("live.bili", r"live\.bilibili\.com/(?P<room_id>\d+)")
     async def _parse_live(self, searched: Match[str]):
@@ -88,13 +95,7 @@ async def _parse_favlist(self, searched: Match[str]):
     async def _parse_read(self, searched: Match[str]):
         """解析专栏信息"""
         read_id = int(searched.group("read_id"))
-        return await self.parse_read_with_opus(read_id)
-
-    @handle("/opus/", r"bilibili\.com/opus/(?P<opus_id>\d+)")
-    async def _parse_opus(self, searched: Match[str]):
-        """解析图文动态信息"""
-        opus_id = int(searched.group("opus_id"))
-        return await self.parse_opus(opus_id)
+        return await self.parse_read(read_id)
 
     async def parse_video(
         self,
@@ -167,8 +168,8 @@ async def download_video():
             extra={"info": ai_summary},
         )
 
-    async def parse_dynamic(self, dynamic_id: int):
-        """解析动态信息
+    async def parse_dynamic_or_opus(self, dynamic_id: int):
+        """解析动态和图文信息
 
         Args:
             url (str): 动态链接
@@ -178,6 +179,8 @@ async def parse_dynamic(self, dynamic_id: int):
         from .dynamic import DynamicData
 
         dynamic = Dynamic(dynamic_id, await self.credential)
+        if await dynamic.is_article():
+            return await self._parse_opus_obj(dynamic.turn_to_opus())
         dynamic_info = convert(await dynamic.get_info(), DynamicData).item
 
         author = self.create_author(dynamic_info.name, dynamic_info.avatar)
@@ -205,7 +208,7 @@ async def parse_opus(self, opus_id: int):
         opus = Opus(opus_id, await self.credential)
         return await self._parse_opus_obj(opus)
 
-    async def parse_read_with_opus(self, read_id: int):
+    async def parse_read(self, read_id: int):
         """解析专栏信息, 使用 Opus 接口
 
         Args:

src/nonebot_plugin_parser/parsers/bilibili/dynamic.py

@@ -63,9 +63,10 @@ class OpusContent(Struct):
 class DynamicMajor(Struct):
     """动态主要内容"""
 
-    type: str
+    type: str | None = None
     archive: VideoArchive | None = None
     opus: OpusContent | None = None
+    desc: OpusSummary | None = None
 
     @property
     def title(self) -> str | None:
@@ -81,6 +82,8 @@ def text(self) -> str | None:
             return self.archive.desc
         elif self.type == "MAJOR_TYPE_OPUS" and self.opus:
             return self.opus.summary.text
+        elif self.desc:
+            return self.desc.text
         return None
 
     @property
@@ -126,7 +129,10 @@ def pub_ts(self) -> int:
     def major_info(self) -> dict[str, Any] | None:
         """获取主要内容信息"""
         if self.module_dynamic:
-            return self.module_dynamic.get("major")
+            if major := self.module_dynamic.get("major"):
+                return major
+            # 转发类型动态没有 major
+            return self.module_dynamic
         return None
 
 

src/nonebot_plugin_parser/renders/common.py

@@ -130,7 +130,9 @@ class FontSet:
     def new(cls, font_path: Path):
         font_infos: dict[str, FontInfo] = {}
         for name, size, fill in cls._FONT_INFOS:
-            font = ImageFont.truetype(font_path, size)
+            with open(font_path, "rb") as f:
+                font_bytes = BytesIO(f.read())
+            font = ImageFont.truetype(font_bytes, size, encoding="utf-8")
             height = get_font_height(font)
             font_infos[name] = FontInfo(
                 font=font,
@@ -360,10 +362,23 @@ async def text(
     ) -> int:
         """绘制文本"""
         if emosvg is not None:
-            emosvg.text(ctx.image, xy, lines, font.font, fill=font.fill, line_height=font.line_height)
+            emosvg.text(
+                ctx.image,
+                xy,
+                lines,
+                font.font,
+                fill=font.fill,
+                line_height=font.line_height,
+            )
         else:
             await Apilmoji.text(
-                ctx.image, xy, lines, font.font, fill=font.fill, line_height=font.line_height, source=cls.EMOJI_SOURCE
+                ctx.image,
+                xy,
+                lines,
+                font.font,
+                fill=font.fill,
+                line_height=font.line_height,
+                source=cls.EMOJI_SOURCE,
             )
         return font.line_height * len(lines)
 
@@ -1217,10 +1232,34 @@ def _draw_rounded_rectangle_border(
         draw.rectangle((x2 - width, y1 + radius, x2, y2 - radius), fill=border_color)  # 右
 
         # 绘制四个圆角边框
-        draw.arc((x1, y1, x1 + 2 * radius, y1 + 2 * radius), 180, 270, fill=border_color, width=width)
-        draw.arc((x2 - 2 * radius, y1, x2, y1 + 2 * radius), 270, 360, fill=border_color, width=width)
-        draw.arc((x1, y2 - 2 * radius, x1 + 2 * radius, y2), 90, 180, fill=border_color, width=width)
-        draw.arc((x2 - 2 * radius, y2 - 2 * radius, x2, y2), 0, 90, fill=border_color, width=width)
+        draw.arc(
+            (x1, y1, x1 + 2 * radius, y1 + 2 * radius),
+            180,
+            270,
+            fill=border_color,
+            width=width,
+        )
+        draw.arc(
+            (x2 - 2 * radius, y1, x2, y1 + 2 * radius),
+            270,
+            360,
+            fill=border_color,
+            width=width,
+        )
+        draw.arc(
+            (x1, y2 - 2 * radius, x1 + 2 * radius, y2),
+            90,
+            180,
+            fill=border_color,
+            width=width,
+        )
+        draw.arc(
+            (x2 - 2 * radius, y2 - 2 * radius, x2, y2),
+            0,
+            90,
+            fill=border_color,
+            width=width,
+        )
 
     def _wrap_text(self, text: str, max_width: int, font_info: FontInfo) -> list[str]:
         """使用 emoji.emoji_list 优化的文本自动换行算法，正确处理组合 emoji

tests/parsers/test_bilibili.py

@@ -43,7 +43,7 @@ async def test_read():
     parser = BilibiliParser()
     _, searched = parser.search_url(url)
     read_id = int(searched.group("read_id"))
-    result = await parser.parse_read_with_opus(read_id)
+    result = await parser.parse_read(read_id)
     logger.debug(f"result: {result}")
     assert result.title, "标题为空"
     assert result.author, "作者为空"
@@ -60,58 +60,21 @@ async def test_read():
 
 
 @pytest.mark.asyncio
-async def test_opus():
+async def test_dynamic():
     from nonebot_plugin_parser.parsers import BilibiliParser
 
-    opus_urls = [
+    dynamic_urls = [
+        "https://t.bilibili.com/1120105154190770281",
         "https://www.bilibili.com/opus/998440765151510535",
         "https://www.bilibili.com/opus/1040093151889457152",
     ]
 
     parser = BilibiliParser()
 
-    async def test_parse_opus(opus_url: str) -> None:
-        _, searched = parser.search_url(opus_url)
-        opus_id = int(searched.group("opus_id"))
-        try:
-            result = await parser.parse_opus(opus_id)
-        except Exception as e:
-            pytest.skip(f"{opus_url} | opus 解析失败: {e} (风控)")
-
-        assert result.contents, "内容为空"
-        for content in result.contents:
-            path = await content.get_path()
-            assert path.exists(), "内容不存在"
-
-        assert result.author, "作者为空"
-        avatar_path = await result.author.get_avatar_path()
-        assert avatar_path, "头像不存在"
-        assert avatar_path.exists(), "头像不存在"
-
-        graphics_contents = result.graphics_contents
-        assert graphics_contents, "图文内容为空"
-
-        for graphics_content in graphics_contents:
-            path = await graphics_content.get_path()
-            assert path.exists(), "图文内容不存在"
-
-    await asyncio.gather(*[test_parse_opus(opus_url) for opus_url in opus_urls])
-    logger.success("B站动态解析成功")
-
-
-@pytest.mark.asyncio
-async def test_dynamic():
-    from nonebot_plugin_parser.parsers import BilibiliParser
-
-    dynamic_urls = ["https://t.bilibili.com/1120105154190770281"]
-
-    parser = BilibiliParser()
-
     async def test_parse_dynamic(dynamic_url: str) -> None:
         _, searched = parser.search_url(dynamic_url)
         dynamic_id = int(searched.group("dynamic_id"))
-        result = await parser.parse_dynamic(dynamic_id)
-        assert result.title, "标题为空"
+        result = await parser.parse_dynamic_or_opus(dynamic_id)
         assert result.author, "作者为空"
         avatar_path = await result.author.get_avatar_path()
         assert avatar_path, "头像不存在"