Merge pull request #81 from felinae98/feat/custom-post

添加 CustomPost

docs/dev/README.md

@@ -38,7 +38,10 @@ Nonebot 项目使用了全异步的处理方式，所以你需要对异步，Pyt
 
 ## 基本概念
 
-- `nonebot_bison.post.Post`: 可以理解为推送内容，其中包含需要发送的文字，图片，链接，平台信息等
+- `nonebot_bison.post`: 可以理解为推送内容，其中包含需要发送的文字，图片，链接，平台信息等，分为：
+  - `nonebot_bison.post.Post`: 简单的推送内容格式，需要发送的内容由 bison 处理
+  - `nonebot_bison.post.CustomPost`: 基于 markdown 语法的，自由度较高的推送内容格式
+  - 详细的介绍可参见[生成 bison 的推送文本](#生成bison的推送文本)
 - `nonebot_bison.types.RawPost`: 从站点/平台中爬到的单条信息
 - `nonebot_bison.types.Target`: 目标账号，Bilibili，微博等社交媒体中的账号
 - `nonebot_bison.types.Category`: 信息分类，例如视频，动态，图文，文章等
@@ -211,3 +214,46 @@ class Weibo(NewMessage):
       #将需要bot推送的RawPost处理成正式推送的Post
       ...
 ```
+
+## 生成 bison 的推送文本
+
+### 什么是`nonebot_bison.post`
+
+可以认为`nonebot_bison.post`是最终要交付给 bison 推送到群内的内容，经过`parse`函数处理过后的报文应该返回属于`nonebot_bison.post`下的某个类  
+目前 bison 所支持的类有：
+
+- `nonebot_bison.post.Post`
+- `nonebot_bison.post.CustomPost`
+
+### 什么是`nonebot_bison.post.Post`
+
+Post 类存在参数`text`与`pics`，分别对应接收文本与图片类消息，需要注意的是`pics`接收的是一个列表 List，列表中的值可以为 url 或者 bytes。  
+Post 会将`text`与`pics`分为若干条消息进行分别发送  
+可选参数:  
+使用`compress`参数将所有消息压缩为一条进行发送。  
+使用`extra_msg`可以携带额外的消息进行发送  
+使用`override_use_pic`参数可以无视全局配置中的 bison_use_pic 配置进行强制指定  
+可参考[Post 的用法](https://github.com/felinae98/nonebot-bison/blob/v0.5.4/src/plugins/nonebot_bison/platform/arknights.py#L227)
+
+### 什么是`nonebot_bison.post.CustomPost`
+
+CustomPost 类能接受的消息为[`List[MessageSegment]`](https://github.com/botuniverse/onebot-11/blob/master/message/array.md#%E6%B6%88%E6%81%AF%E6%AE%B5)  
+::: tip
+
+消息段（Message Segment 或 Segment）  
+表示聊天消息的一个部分，在一些平台上，聊天消息支持图文混排，其中就会有多个消息段，分别表示每个图片和每段文字。
+:::
+准确来说，CustomPost 只支持使用 MessageSegment 内的`text`和`image`类型，CustomPost 会将 List 中的每个`text`类型元素理解为一个单行的 text 文本，
+当然，markdown 语法可以在每个`text`类型元素使用，但如果这样，在不开启`bison_use_pic`**全局配置项** 的情况下，bison 会将写在 text 类型元素里的 markdown 语法按原样推送，不会解析。  
+对于上述情况，建议开启 CustomPost 的`override_use_pic`选项，这样 CustomPost 只会发送经过 markdown 语法渲染好的图片，而非文本消息。  
+CustomPost 的可选参数及作用与上文中的[Post](#什么是nonebot-bison-post-post)一致。
+::: details CustomPost 例子
+
+```python
+  async def parse(self, raw_post:RawPost) -> str:
+    #假定传入的raw_post为List[MessageSegment]
+    #do something...
+    return CustomPost(message_segments=raw_post, only_pic=True)
+```
+
+:::

src/plugins/nonebot_bison/post/__init__.py

@@ -0,0 +1,3 @@
+from .post import Post
+
+__all__ = ["Post"]

src/plugins/nonebot_bison/post/abstract_post.py

@@ -0,0 +1,55 @@
+from abc import abstractmethod
+from dataclasses import dataclass, field
+from functools import reduce
+from typing import Optional
+
+from nonebot.adapters.onebot.v11.message import Message, MessageSegment
+
+from ..plugin_config import plugin_config
+
+
+@dataclass
+class BasePost:
+    @abstractmethod
+    async def generate_text_messages(self) -> list[MessageSegment]:
+        "Generate Message list from this instance"
+        ...
+
+    @abstractmethod
+    async def generate_pic_messages(self) -> list[MessageSegment]:
+        "Generate Message list from this instance with `use_pic`"
+        ...
+
+
+@dataclass
+class OptionalMixin:
+    # Because of https://stackoverflow.com/questions/51575931/class-inheritance-in-python-3-7-dataclasses
+
+    override_use_pic: Optional[bool] = None
+    compress: bool = False
+    extra_msg: list[Message] = field(default_factory=list)
+
+    def _use_pic(self):
+        if not self.override_use_pic is None:
+            return self.override_use_pic
+        return plugin_config.bison_use_pic
+
+
+@dataclass
+class AbstractPost(OptionalMixin, BasePost):
+    async def generate_messages(self) -> list[Message]:
+        if self._use_pic():
+            msg_segments = await self.generate_pic_messages()
+        else:
+            msg_segments = await self.generate_text_messages()
+        if msg_segments:
+            if self.compress:
+                msgs = [reduce(lambda x, y: x.append(y), msg_segments, Message())]
+            else:
+                msgs = list(
+                    map(lambda msg_segment: Message([msg_segment]), msg_segments)
+                )
+        else:
+            msgs = []
+        msgs.extend(self.extra_msg)
+        return msgs

src/plugins/nonebot_bison/post/custom_post.py

@@ -0,0 +1,74 @@
+from dataclasses import dataclass, field
+from pathlib import Path
+from typing import Optional
+
+from nonebot.adapters.onebot.v11.message import Message, MessageSegment
+from nonebot.log import logger
+from nonebot.plugin import require
+
+from .abstract_post import AbstractPost, BasePost
+
+
+@dataclass
+class _CustomPost(BasePost):
+
+    message_segments: list[MessageSegment] = field(default_factory=list)
+    css_path: Optional[str] = None  # 模板文件所用css路径
+
+    async def generate_text_messages(self) -> list[MessageSegment]:
+        return self.message_segments
+
+    async def generate_pic_messages(self) -> list[MessageSegment]:
+        require("nonebot_plugin_htmlrender")
+        from nonebot_plugin_htmlrender import md_to_pic
+
+        pic_bytes = await md_to_pic(md=self._generate_md(), css_path=self.css_path)
+        return [MessageSegment.image(pic_bytes)]
+
+    def _generate_md(self) -> str:
+        md = ""
+
+        for message_segment in self.message_segments:
+            if message_segment.type == "text":
+                md += "{}<br>".format(message_segment.data.get("text", ""))
+            elif message_segment.type == "image":
+                # 先尝试获取file的值，没有再尝试获取url的值，都没有则为空
+                pic_res = message_segment.data.get("file") or message_segment.data.get(
+                    "url", ""
+                )
+                if not pic_res:
+                    logger.warning("无法获取到图片资源:MessageSegment.image中file/url字段均为空")
+                else:
+                    md += "![Image]({})\n".format(pic_res)
+            else:
+                logger.warning("custom_post不支持处理类型:{}".format(message_segment.type))
+                continue
+
+        return md
+
+
+@dataclass
+class CustomPost(_CustomPost, AbstractPost):
+    """基于 markdown 语法的，自由度较高的推送内容格式
+
+    简介:
+    支持处理text/image两种MessageSegment,
+    通过将text/image转换成对应的markdown语法以生成markdown文本。
+    理论上text类型中可以直接使用markdown语法,例如`##第一章`。
+    但会导致不启用`override_use_pic`时, 发送不会被渲染的纯文本消息。
+    图片渲染最终由htmlrender执行。
+
+    注意:
+    每一个MessageSegment元素都会被解释为单独的一行
+
+    可选参数:
+    `override_use_pic`:是否覆盖`bison_use_pic`全局配置
+    `compress`:将所有消息压缩为一条进行发送
+    `extra_msg`:需要附带发送的额外消息
+
+    成员函数:
+    `generate_text_messages()`:负责生成文本消息
+    `generate_pic_messages()`:负责生成图片消息
+    """
+
+    pass

src/plugins/nonebot_bison/post/post.py

@@ -7,28 +7,21 @@ from nonebot.adapters.onebot.v11.message import Message, MessageSegment
 from nonebot.log import logger
 from PIL import Image
 
-from .plugin_config import plugin_config
+from ..utils import http_client, parse_text
-from .utils import http_client, parse_text
+from .abstract_post import AbstractPost, BasePost, OptionalMixin
 
 
 @dataclass
-class Post:
+class _Post(BasePost):
 
     target_type: str
     text: str
     url: Optional[str] = None
     target_name: Optional[str] = None
-    compress: bool = False
-    override_use_pic: Optional[bool] = None
     pics: list[Union[str, bytes]] = field(default_factory=list)
-    extra_msg: list[Message] = field(default_factory=list)
 
-    _message: Optional[list[Message]] = None
+    _message: Optional[list[MessageSegment]] = None
-
+    _pic_message: Optional[list[MessageSegment]] = None
-    def _use_pic(self):
-        if not self.override_use_pic is None:
-            return self.override_use_pic
-        return plugin_config.bison_use_pic
 
     async def _pic_url_to_image(self, data: Union[str, bytes]) -> Image.Image:
         pic_buffer = BytesIO()
@@ -106,42 +99,49 @@ class Post:
         self.pics = self.pics[matrix[0] * matrix[1] :]
         self.pics.insert(0, target_io.getvalue())
 
-    async def generate_messages(self) -> list[Message]:
+    async def generate_text_messages(self) -> list[MessageSegment]:
+
         if self._message is None:
             await self._pic_merge()
             msg_segments: list[MessageSegment] = []
             text = ""
             if self.text:
-                if self._use_pic():
-                    text += "{}".format(self.text)
-                else:
                 text += "{}".format(
                     self.text if len(self.text) < 500 else self.text[:500] + "..."
                 )
-            text += "\n来源: {}".format(self.target_type)
+            if text:
+                text += "\n"
+            text += "来源: {}".format(self.target_type)
             if self.target_name:
                 text += " {}".format(self.target_name)
-            if self._use_pic():
-                msg_segments.append(await parse_text(text))
-                if not self.target_type == "rss" and self.url:
-                    msg_segments.append(MessageSegment.text(self.url))
-            else:
             if self.url:
                 text += " \n详情: {}".format(self.url)
             msg_segments.append(MessageSegment.text(text))
             for pic in self.pics:
                 msg_segments.append(MessageSegment.image(pic))
-            if self.compress:
+            self._message = msg_segments
-                msgs = [reduce(lambda x, y: x.append(y), msg_segments, Message())]
-            else:
-                msgs = list(
-                    map(lambda msg_segment: Message([msg_segment]), msg_segments)
-                )
-            msgs.extend(self.extra_msg)
-            self._message = msgs
-        assert len(self._message) > 0, f"message list empty, {self}"
         return self._message
 
+    async def generate_pic_messages(self) -> list[MessageSegment]:
+
+        if self._pic_message is None:
+            await self._pic_merge()
+            msg_segments: list[MessageSegment] = []
+            text = ""
+            if self.text:
+                text += "{}".format(self.text)
+                text += "\n"
+            text += "来源: {}".format(self.target_type)
+            if self.target_name:
+                text += " {}".format(self.target_name)
+            msg_segments.append(await parse_text(text))
+            if not self.target_type == "rss" and self.url:
+                msg_segments.append(MessageSegment.text(self.url))
+            for pic in self.pics:
+                msg_segments.append(MessageSegment.image(pic))
+            self._pic_message = msg_segments
+        return self._pic_message
+
     def __str__(self):
         return "type: {}\nfrom: {}\ntext: {}\nurl: {}\npic: {}".format(
             self.target_type,
@@ -157,3 +157,8 @@ class Post:
                 )
             ),
         )
+
+
+@dataclass
+class Post(AbstractPost, _Post):
+    pass

src/plugins/nonebot_bison/post/templates/custom_post.css

@@ -0,0 +1,112 @@
+@charset "utf-8";
+
+/**
+ * markdown.css
+ *
+ * This program is free software: you can redistribute it and/or modify it under
+ * the terms of the GNU Lesser General Public License as published by the Free
+ * Software Foundation, either version 3 of the License, or (at your option) any
+ * later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+ * FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU Lesser General Public License
+ * along with this program. If not, see http://gnu.org/licenses/lgpl.txt.
+ *
+ * @project      Weblog and Open Source Projects of Florian Wolters
+ * @version      GIT: $Id$
+ * @package      xhtml-css
+ * @author       Florian Wolters <florian.wolters.85@googlemail.com>
+ * @copyright    2012 Florian Wolters
+ * @cssdoc       version 1.0-pre
+ * @license      http://gnu.org/licenses/lgpl.txt GNU Lesser General Public License
+ * @link         http://github.com/FlorianWolters/jekyll-bootstrap-theme
+ * @media        all
+ * @valid        true
+ */
+
+body {
+    font-family: Helvetica, Arial, Freesans, clean, sans-serif;
+    padding: 1em;
+    margin: auto;
+    max-width: 42em;
+    background: #fefefe;
+}
+
+h1, h2, h3, h4, h5, h6 {
+    font-weight: bold;
+}
+
+h1 {
+    color: #000000;
+    font-size: 28px;
+}
+
+h2 {
+    border-bottom: 1px solid #CCCCCC;
+    color: #000000;
+    font-size: 24px;
+}
+
+h3 {
+    font-size: 18px;
+}
+
+h4 {
+    font-size: 16px;
+}
+
+h5 {
+    font-size: 14px;
+}
+
+h6 {
+    color: #777777;
+    background-color: inherit;
+    font-size: 14px;
+}
+
+hr {
+    height: 0.2em;
+    border: 0;
+    color: #CCCCCC;
+    background-color: #CCCCCC;
+}
+
+p, blockquote, ul, ol, dl, li, table, pre {
+    margin: 15px 0;
+}
+
+code, pre {
+    border-radius: 3px;
+    background-color: #F8F8F8;
+    color: inherit;
+}
+
+code {
+    border: 1px solid #EAEAEA;
+    margin: 0 2px;
+    padding: 0 5px;
+}
+
+pre {
+    border: 1px solid #CCCCCC;
+    line-height: 1.25em;
+    overflow: auto;
+    padding: 6px 10px;
+}
+
+pre>code {
+    border: 0;
+    margin: 0;
+    padding: 0;
+}
+
+a, a:visited {
+    color: #4183C4;
+    background-color: inherit;
+    text-decoration: none;
+}

tests/test_custom_post.py

@@ -0,0 +1,80 @@
+import base64
+import hashlib
+import platform
+from io import UnsupportedOperation
+from pathlib import Path
+
+import pytest
+import respx
+from httpx import Response
+from nonebot.adapters.onebot.v11.message import MessageSegment
+from nonebug.app import App
+
+
+@pytest.fixture
+def ms_list():
+    msg_segments: list[MessageSegment] = []
+    msg_segments.append(MessageSegment.text("【Zc】每早合约日替攻略！"))
+    msg_segments.append(
+        MessageSegment.image(
+            file="http://i0.hdslb.com/bfs/live/new_room_cover/cf7d4d3b2f336c6dba299644c3af952c5db82612.jpg",
+            cache=0,
+        )
+    )
+    msg_segments.append(MessageSegment.text("来源: Bilibili直播 魔法Zc目录"))
+    msg_segments.append(MessageSegment.text("详情: https://live.bilibili.com/3044248"))
+
+    return msg_segments
+
+
+@pytest.fixture
+def pic_hash():
+    platform_name = platform.system()
+    if platform_name == "Windows":
+        return "58723fdc24b473b6dbd8ec8cbc3b7e46160c83df"
+    elif platform_name == "Linux":
+        return "4d540798108762df76de34f7bdbc667dada6b5cb"
+    elif platform_name == "Darwin":
+        return "a482bf8317d56e5ddc71437584343ace29ff545c"
+    else:
+        raise UnsupportedOperation(f"未支持的平台{platform_name}")
+
+
+@pytest.fixture
+def expect_md():
+    return "【Zc】每早合约日替攻略！<br>![Image](http://i0.hdslb.com/bfs/live/new_room_cover/cf7d4d3b2f336c6dba299644c3af952c5db82612.jpg)\n来源: Bilibili直播 魔法Zc目录<br>详情: https://live.bilibili.com/3044248<br>"
+
+
+def test_gene_md(app: App, expect_md, ms_list):
+    from nonebot_bison.post.custom_post import CustomPost
+
+    cp = CustomPost(message_segments=ms_list)
+    cp_md = cp._generate_md()
+    assert cp_md == expect_md
+
+
+@respx.mock
+@pytest.mark.asyncio
+async def test_gene_pic(app: App, ms_list, pic_hash):
+    from nonebot_bison.post.custom_post import CustomPost
+
+    pic_router = respx.get(
+        "http://i0.hdslb.com/bfs/live/new_room_cover/cf7d4d3b2f336c6dba299644c3af952c5db82612.jpg"
+    )
+
+    pic_path = Path(__file__).parent / "platforms" / "static" / "custom_post_pic.jpg"
+    with open(pic_path, mode="rb") as f:
+        mock_pic = f.read()
+
+    pic_router.mock(return_value=Response(200, stream=mock_pic))
+
+    cp = CustomPost(message_segments=ms_list)
+    cp_pic_bytes: list[MessageSegment] = await cp.generate_pic_messages()
+
+    pure_b64 = base64.b64decode(
+        cp_pic_bytes[0].data.get("file").replace("base64://", "")
+    )
+    sha1obj = hashlib.sha1()
+    sha1obj.update(pure_b64)
+    sha1hash = sha1obj.hexdigest()
+    assert sha1hash == pic_hash