Merge branch 'main' into arknights

README.md

@@ -30,6 +30,7 @@
 ## 使用方法
 
 ### 使用以及部署
+**!!本项目需要Python3.9及以上**  
 本项目可作为单独插件使用，仅包含订阅相关功能（绝对simple和stupid），也可直接克隆项目进行使用（包含自动同意superuser，自动接受入群邀请等功能）  
 作为插件使用请安装`nonebot-hk-reporter`包，并在`bot.py`中加载`nonebot_hk_reporter`插件；或直接克隆本项目进行使用  
 配置与安装请参考[nonebot2文档](https://v2.nonebot.dev/)
@@ -77,6 +78,12 @@ go-cqhttp镜像可使用`felinae98/go-cqhttp-ffmpeg`（数据目录为`/data`）
 * 通过图片发送文本，防止风控
 * 使用队列限制发送频率
 
+# FAQ
+1. 报错`TypeError: 'type' object is not subscriptable`  
+    本项目使用了Python 3.9的语法，请将Python版本升级到3.9及以上，推荐使用docker部署
+2. bot不理我  
+    请确认自己是群主或者管理员，并且检查`COMMAND_START`环境变量是否设为`[""]`
+
 ## 鸣谢
 * [`go-cqhttp`](https://github.com/Mrs4s/go-cqhttp)：简单又完善的 cqhttp 实现
 * [`NoneBot2`](https://github.com/nonebot/nonebot2)：超好用的开发框架

src/plugins/nonebot_hk_reporter/platform/__init__.py

@@ -1,6 +1,7 @@
 from .bilibili import Bilibili
 from .rss import Rss
 from .weibo import Weibo
+from .wechat import Wechat
 from .utils import check_sub_target
 from .platform import PlatformNoTarget
 from .utils import platform_manager

src/plugins/nonebot_hk_reporter/platform/utils.py

@@ -6,6 +6,7 @@ from .arkninghts import Arknights
 from .weibo import Weibo
 from .bilibili import Bilibili
 from .rss import Rss
+from .wechat import Wechat
 from .platform import PlatformProto
 from ..config import Config
 from ..post import Post
@@ -19,6 +20,7 @@ platform_manager: dict[str, PlatformProto] = {
         'weibo': Weibo(),
         'rss': Rss(),
         'arknights': Arknights()
+        'wechat': Wechat(),
     }
 
 async def fetch_and_send(target_type: str):

src/plugins/nonebot_hk_reporter/platform/wechat.py

@@ -0,0 +1,75 @@
+from datetime import datetime
+import hashlib
+import json
+import re
+from typing import Any, Optional
+
+from bs4 import BeautifulSoup as bs
+import httpx
+
+from ..post import Post
+from ..types import *
+from .platform import Platform
+
+
+class Wechat(Platform):
+
+    categories = {}
+    enable_tag = False
+    platform_name = 'wechat'
+
+    @classmethod
+    def _get_query_url(cls, target: Target):
+        return 'https://weixin.sogou.com/weixin?type=1&s_from=input&query={}&ie=utf8&_sug_=n&_sug_type_='.format(target)
+
+    @classmethod
+    async def _get_target_soup(cls, target: Target) -> Optional[bs]:
+        target_url = cls._get_query_url(target)
+        async with httpx.AsyncClient() as client:
+            res = await client.get(target_url)
+        soup = bs(res.text, 'html.parser')
+        blocks = soup.find(class_='news-list2').find_all('li',recursive=False)
+        for block in blocks:
+            if block.find(string=[target]):
+                return block
+
+    @classmethod
+    async def get_account_name(cls, target: Target) -> Optional[str]:
+        if not (block := await cls._get_target_soup(target)):
+            return None
+        return block.find('p', class_='tit').find('a').text
+
+    async def get_sub_list(self, target: Target) -> list[RawPost]:
+        block = await self._get_target_soup(target)
+        if (last_post_dt := block.find('dt', string='最近文章：')):
+            post = {
+                    'title': last_post_dt.find_parent().find('a').text,
+                    'target': target,
+                    'page_url': self._get_query_url(target),
+                    'name': block.find('p', class_='tit').find('a').text
+                    }
+            return [post]
+        else:
+            return []
+
+    def get_id(self, post: RawPost) -> Any:
+        return post['title']
+
+    def get_date(self, post: RawPost):
+        return None
+
+    def get_tags(self, post: RawPost):
+        return None
+
+    def get_category(self, post: RawPost):
+        return None
+
+    async def parse(self, raw_post: RawPost) -> Post:
+        # TODO get content of post
+        return Post(target_type='wechat',
+                text='{}\n详细内容请自行查看公众号'.format(raw_post['title']),
+                target_name=raw_post['name'],
+                pics=[],
+                url=''
+                )
+

src/plugins/nonebot_hk_reporter/scheduler.py

@@ -22,6 +22,10 @@ async def rss_check():
 async def arknights_check():
     await fetch_and_send('arknights')
 
+@scheduler.scheduled_job('interval', seconds=30)
+async def wechat_check():
+    await fetch_and_send('wechat')
+
 @scheduler.scheduled_job('interval', seconds=1)
 async def _():
     await do_send_msgs()

src/plugins/nonebot_hk_reporter/utils.py

@@ -1,10 +1,11 @@
 import os
 import asyncio
-from typing import Optional
+from typing import Awaitable, Callable, Optional
 import nonebot
 from nonebot import logger
 import base64
 from pyppeteer import launch
+from pyppeteer.page import Page
 from pyppeteer.chromium_downloader import check_chromium, download_chromium
 from html import escape
 from hashlib import sha256
@@ -19,7 +20,7 @@ class Singleton(type):
             cls._instances[cls] = super(Singleton, cls).__call__(*args, **kwargs)
         return cls._instances[cls]
 
-supported_target_type = ('weibo', 'bilibili', 'rss', 'arknights')
+supported_target_type = ('weibo', 'bilibili', 'rss', 'arknights', 'wechat')
 
 if not plugin_config.hk_reporter_use_local and not check_chromium():
     os.environ['PYPPETEER_DOWNLOAD_HOST'] = 'http://npm.taobao.org/mirrors'
@@ -30,14 +31,18 @@ class Render(metaclass=Singleton):
     def __init__(self):
         self.lock = asyncio.Lock()
 
-    async def render(self, url: str, viewport: Optional[dict] = None, target: Optional[str] = None) -> str:
+    async def render(self, url: str, viewport: Optional[dict] = None, target: Optional[str] = None,
+            operation: Optional[Callable[[Page], Awaitable[None]]] = None) -> str:
         async with self.lock:
             if plugin_config.hk_reporter_use_local:
                 browser = await launch(executablePath='/usr/bin/chromium', args=['--no-sandbox'])
             else:
                 browser = await launch(args=['--no-sandbox'])
             page = await browser.newPage()
-            await page.goto(url)
+            if operation:
+                await operation(page)
+            else:
+                await page.goto(url)
             if viewport:
                 await page.setViewport(viewport)
             if target: