Merge pull request #22 from Twisuki/main

* new file:   nonebot_plugin_marshoai/tools/marshoai-meogirl/__init__.py
	new file:   nonebot_plugin_marshoai/tools/marshoai-meogirl/mg_Info.py
	new file:   nonebot_plugin_marshoai/tools/marshoai-meogirl/tools.json

* 接入萌娘百科搜索

* 改成异步了, 改完彻底不能用了

* 修改, 但搜索功能完全无法使用了

* 修复完成

nonebot_plugin_marshoai/tools/marshoai-meogirl/__init__.py

@@ -0,0 +1,10 @@
+from . import mg_Info
+from . import mg_Search
+
+# meogirl
+async def meogirl():
+    return mg_Info.meogirl()
+
+# Search
+async def search(msg : str, num : int = 3):
+    return str(await mg_Search.search(msg, num))

nonebot_plugin_marshoai/tools/marshoai-meogirl/mg_Info.py

@@ -0,0 +1,4 @@
+
+# Meogirl
+def meogirl():
+    return "Meogirl指的是\"萌娘百科\"(https://zh.moegirl.org.cn/ , 简称\"萌百\"), 是一个\"万物皆可萌的百科全书！\"; 同时, MarshoTools也配有\"Meogirl\"插件, 可调用萌百的api"

nonebot_plugin_marshoai/tools/marshoai-meogirl/mg_Search.py

@@ -0,0 +1,103 @@
+from nonebot.log import logger
+
+import re
+import httpx
+import urllib.parse
+from bs4 import BeautifulSoup
+
+headers = {
+    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/100.0.4896.127 Safari/537.36'
+}
+
+async def get_async_data (url):
+    async with httpx.AsyncClient(timeout = None) as client:
+        return await client.get(url, headers = headers)
+
+async def search(msg : str, num : int):
+    logger.info(f"搜索 : \"{msg}\"")
+    result = ""
+
+    url = "https://mzh.moegirl.org.cn/index.php?search=" + urllib.parse.quote_plus(msg)
+    response = await get_async_data(url)
+    logger.success(f"连接{url}完成, 状态码 : {response.status_code}")
+
+    # 正常搜索
+    if response.status_code == 200:
+        """
+            萌娘百科搜索页面结构
+            div.searchresults
+            └── p ...
+            └── ul.mw-search-results                        # 若无, 证明无搜索结果
+                └── li                                      # 一个搜索结果
+                    └── div.mw-search-result-heading > a    # 标题
+                    └── div.mw-searchresult                 # 内容
+                    └── div.mw-search-result-data
+                └── li ...
+                └── li ...
+        """
+        soup = BeautifulSoup(response.text, 'html.parser')
+
+        # 检测ul.mw-search-results, 是否有结果
+        if soup.find('ul', class_='mw-search-results'):
+            ul_tag = soup.select('ul.mw-search-results')[0]
+            li_tags = ul_tag.select('li')
+            for li_tag in li_tags:
+
+                div_heading = li_tag.select('div.mw-search-result-heading')[0]
+                if div_heading:
+                    a_tag = div_heading.select('a')[0]
+                    result += a_tag['title'] + "\n"
+                    logger.info(f"搜索到 : \"{a_tag['title']}\"")
+
+                div_result = li_tag.find('div', class_='searchresult')
+                if div_result:
+                    content = str(div_result).replace('<div class=\"searchresult\">', '').replace('</div>', '')
+                    content = content.replace('<span class=\"searchmatch\">', '').replace('</span>', '')
+                    result += content + "\n\n"
+
+                num -= 1
+                if num == 0:
+                    break
+            return result
+
+        # 无ul.mw-search-results, 无结果
+        else:
+            logger.info("无结果")
+            return "无结果"
+
+    # 重定向
+    elif response.status_code == 302:
+        logger.info(f"\"{msg}\"已被重定向至\"{response.headers.get('location')}\"")
+        # 读取重定向结果
+        response = await get_async_data(response.headers.get('location'))
+        soup = BeautifulSoup(response.text, 'html.parser')
+        logger.success("重定向成功")
+        num = 0
+
+        """
+            萌娘百科重定向介绍页面结构
+            div#mw-content-text
+            └── div.mw-parser-output    # 介绍页面
+                └── ....
+                └── p ?                 # 可能存在的空p
+                └── p                   # 人物介绍
+                └── ...
+        """
+        if soup.find('div', class_='mw-parser-output'):
+            div = soup.find('div', class_='mw-parser-output')
+            p_tags = div.select('p')
+            for p_tag in p_tags:
+                p = str(p_tag)
+                p = re.sub(r'<.*?>', '', p)
+                if p != '':
+                    result += str(p)
+
+                    num += 1
+                    if num >= 5:
+                        break
+            return result
+
+    # 状态码非200或302
+    else:
+        logger.error(f"网络错误, 状态码 : {response.status_code}")
+        return f"网络错误, 状态码 : {response.status_code}"

nonebot_plugin_marshoai/tools/marshoai-meogirl/tools.json

@@ -0,0 +1,32 @@
+[
+	{
+		"type" : "function",
+		"function" : {
+			"name" : "marshoai-meogirl__meogirl",
+			"description" : "介绍Meogirl"
+		}
+	},
+	{
+		"type" : "function",
+		"function" : {
+			"name" : "marshoai-meogirl__search",
+			"description" : "在萌娘百科中搜索(仅用户指定在萌娘百科中搜索才调用此函数)",
+			"parameters" : {
+				"type" : "object",
+				"properties" : {
+					"msg" : {
+						"type": "string",
+						"description": "搜索关键词"
+					},
+					"num" : {
+						"type": "integer",
+						"description": "数据显示条数, 默认3, 可留空"
+					}
+				}
+			},
+			"required": [
+                "msg"
+            ]
+		}
+	}
+]