✨ 实现 Post.content 相关扩展协议 (#553)

* ✨ `post` 新增 `get_content()` 方法

* 🚨 make linter happy

* 💄 auto fix by pre-commit hooks

* 🐛 fix: 调整函数使用

* 💄 auto fix by pre-commit hooks

* ✨ 转用函数处理文本

* 💄 auto fix by pre-commit hooks

* 🔨 使用`Dict`存储`content_handlers`

* 💄 auto fix by pre-commit hooks

* 🎨 fix

* :arts: 简化函数使用

* 🐛 移除`Theme`的过时参数

* 🗑️ 复用 `self.plain_content`

* 💄 auto fix by pre-commit hooks

* ✨ 注册式装饰器写法

* 💄 auto fix by pre-commit hooks

* 🐛 fix

* 💄 auto fix by pre-commit hooks

* :feat: 通用纯文本处理函数

* 💄 auto fix by pre-commit hooks

* :downgrade: 复用`==`处理标题

* 🎨 简化（？）写法

* ✅ 测试修复

* ♻️ via ContentSupport extensions

* 🐛 fix test

* 💄 auto fix by pre-commit hooks

* 🐛 for clean text

* 🐛 fix

* 💄 auto fix by pre-commit hooks

* fix: for xml

* 💄 auto fix by pre-commit hooks

* chore: art

* 💄 auto fix by pre-commit hooks

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

nonebot_bison/platform/arknights.py

@@ -1,3 +1,5 @@
+import re
+import html
 from typing import Any
 from functools import partial
 
@@ -10,6 +12,7 @@ from nonebot.compat import type_validate_python
 from ..post import Post
 from ..utils import Site
 from ..types import Target, RawPost, Category
+from ..post.protocol import HTMLContentSupport
 from .platform import NewMessage, StatusChange
 
 
@@ -58,6 +61,39 @@ class ArknightsSite(Site):
     schedule_setting = {"seconds": 30}
 
 
+class ArknightsPost(Post, HTMLContentSupport):
+    def _cleantext(self, text: str, old_split="\n", new_split="\n") -> str:
+        """清理文本：去掉所有多余的空格和换行"""
+        lines = text.strip().split(old_split)
+        cleaned_lines = [line.strip() for line in lines if line != ""]
+        return new_split.join(cleaned_lines)
+
+    async def get_html_content(self) -> str:
+        return self.content
+
+    async def get_plain_content(self) -> str:
+        content = html.unescape(self.content)  # 转义HTML特殊字符
+        content = re.sub(
+            r'\<p style="text-align:center;"\>(.*?)\<strong\>(.*?)\<span style=(.*?)\>(.*?)\<\/span\>(.*?)\<\/strong\>(.*?)<\/p\>',  # noqa: E501
+            r"==\4==\n",
+            content,
+            flags=re.DOTALL,
+        )  # 去“标题型”p
+        content = re.sub(
+            r'\<p style="text-align:(left|right);"?\>(.*?)\<\/p\>',
+            r"\2\n",
+            content,
+            flags=re.DOTALL,
+        )  # 去左右对齐的p
+        content = re.sub(r"\<p\>(.*?)\</p\>", r"\1\n", content, flags=re.DOTALL)  # 去普通p
+        content = re.sub(r'\<a href="(.*?)" target="_blank">(.*?)\<\/a\>', r"\1", content, flags=re.DOTALL)  # 去a
+        content = re.sub(r"<br/>", "\n", content)  # 去br
+        content = re.sub(r"\<strong\>(.*?)\</strong\>", r"\1", content)  # 去strong
+        content = re.sub(r'<span style="color:(#.*?)">(.*?)</span>', r"\2", content)  # 去color
+        content = re.sub(r'<div class="media-wrap image-wrap">(.*?)</div>', "", content)  # 去img
+        return self._cleantext(content)
+
+
 class Arknights(NewMessage):
     categories = {1: "游戏公告"}
     platform_name = "arknights"
@@ -108,7 +144,7 @@ class Arknights(NewMessage):
             # 只有一张图片
             title = title_escape(data.title)
 
-        return Post(
+        return ArknightsPost(
             self,
             content=data.content,
             title=title,
@@ -205,7 +241,7 @@ class MonsterSiren(NewMessage):
         text = f'{raw_post["title"]}\n{soup.text.strip()}'
         return Post(
             self,
-            text,
+            content=text,
             images=imgs,
             url=url,
             nickname="塞壬唱片新闻",
@@ -246,7 +282,7 @@ class TerraHistoricusComic(NewMessage):
         url = f'https://terra-historicus.hypergryph.com/comic/{raw_post["comicCid"]}/episode/{raw_post["episodeCid"]}'
         return Post(
             self,
-            raw_post["subtitle"],
+            content=raw_post["subtitle"],
             title=f'{raw_post["title"]} - {raw_post["episodeShortTitle"]}',
             images=[raw_post["coverUrl"]],
             url=url,

nonebot_bison/platform/bilibili/platforms.py

@@ -465,7 +465,7 @@ class Bilibililive(StatusChange):
         target_name = f"{raw_post.uname} {raw_post.area_name}"
         return Post(
             self,
-            "",
+            content="",
             title=title,
             url=url,
             images=list(pic),
@@ -550,7 +550,7 @@ class BilibiliBangumi(StatusChange):
         title = lastest_episode["share_copy"]
         return Post(
             self,
-            content,
+            content=content,
             title=title,
             url=url,
             images=list(pic),

nonebot_bison/platform/ff14.py

@@ -44,4 +44,4 @@ class FF14(NewMessage):
         title = raw_post["Title"]
         text = raw_post["Summary"]
         url = raw_post["Author"]
-        return Post(self, text, title=title, url=url, nickname="最终幻想XIV官方公告")
+        return Post(self, content=text, title=title, url=url, nickname="最终幻想XIV官方公告")

nonebot_bison/platform/ncm.py

@@ -69,7 +69,7 @@ class NcmArtist(NewMessage):
         target_name = raw_post["artist"]["name"]
         pics = [raw_post["picUrl"]]
         url = "https://music.163.com/#/album?id={}".format(raw_post["id"])
-        return Post(self, text, url=url, images=pics, nickname=target_name)
+        return Post(self, content=text, url=url, images=pics, nickname=target_name)
 
 
 class NcmRadio(NewMessage):
@@ -130,4 +130,4 @@ class NcmRadio(NewMessage):
         target_name = raw_post["radio"]["name"]
         pics = [raw_post["coverUrl"]]
         url = "https://music.163.com/#/program/{}".format(raw_post["id"])
-        return Post(self, text, url=url, images=pics, nickname=target_name)
+        return Post(self, content=text, url=url, images=pics, nickname=target_name)

nonebot_bison/platform/rss.py

@@ -9,7 +9,7 @@ from bs4 import BeautifulSoup as bs
 from ..post import Post
 from .platform import NewMessage
 from ..types import Target, RawPost
-from ..utils import Site, text_similarity
+from ..utils import Site, text_fletten, text_similarity
 
 
 class RssSite(Site):
@@ -18,6 +18,23 @@ class RssSite(Site):
     schedule_setting = {"seconds": 30}
 
 
+class RssPost(Post):
+
+    async def get_plain_content(self) -> str:
+        soup = bs(self.content, "html.parser")
+
+        for img in soup.find_all("img"):
+            img.replace_with("[图片]")
+
+        for br in soup.find_all("br"):
+            br.replace_with("\n")
+
+        for p in soup.find_all("p"):
+            p.insert_after("\n")
+
+        return text_fletten(soup.get_text())
+
+
 class Rss(NewMessage):
     categories = {}
     enable_tag = False
@@ -72,9 +89,9 @@ class Rss(NewMessage):
             for media in raw_post["media_content"]:
                 if media.get("medium") == "image" and media.get("url"):
                     pics.append(media.get("url"))
-        return Post(
+        return RssPost(
             self,
-            desc,
+            content=desc,
             title=title,
             url=raw_post.link,
             images=pics,

nonebot_bison/platform/weibo.py

@@ -183,7 +183,13 @@ class Weibo(NewMessage):
                 res.raise_for_status()
                 pics.append(res.content)
         detail_url = f"https://weibo.com/{info['user']['id']}/{info['bid']}"
-        return Post(self, parsed_text, url=detail_url, images=pics, nickname=info["user"]["screen_name"])
+        return Post(
+            self,
+            content=parsed_text,
+            url=detail_url,
+            images=pics,
+            nickname=info["user"]["screen_name"],
+        )
 
     async def parse(self, raw_post: RawPost) -> Post:
         info = raw_post["mblog"]