新增 Introduce 功能, 将其从 Search 分离 (#25)

* 修复了style和script标签无法去除的问题

* 添加了图片展示 & 统一了引号格式

* 添加了图片展示 & 统一了引号格式

* 统一引号格式并异步处理函数

* 新增了Introduce功能

* 新增了Introduce功能并将其从Search中分离

nonebot_plugin_marshoai/tools/marshoai-meogirl/__init__.py

@@ -1,10 +1,16 @@
+
 from . import mg_Info
 from . import mg_Search
+from . import mg_Introduce
 
 # meogirl
-async def meogirl():
+async def meogirl () :
     return mg_Info.meogirl()
 
 # Search
-async def search(msg : str, num : int = 3):
+async def search (msg : str, num : int = 3) :
     return str(await mg_Search.search(msg, num))
+
+# Show
+async def introduce (msg : str) :
+    return str(await mg_Introduce.introduce(msg))

nonebot_plugin_marshoai/tools/marshoai-meogirl/mg_Introduce.py

@@ -0,0 +1,76 @@
+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 introduce (msg : str) :
+    logger.info(f"介绍 : \"{msg}\" ...")
+    result = ""
+
+    url = "https://mzh.moegirl.org.cn/" + 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}")
+
+    soup = BeautifulSoup(response.text, "html.parser")
+
+    # 正常页
+    if response.status_code == 200 :
+        """
+            萌娘百科页面结构
+            div#mw-content-text
+            └── div#404search           # 空白页面出现
+            └── div.mw-parser-output    # 正常页面
+                └── div, p, table ...   # 大量的解释项
+        """
+        result += msg + "\n"
+
+        img = soup.find("img", class_="infobox-image")
+        if img:
+            result += f"![ {msg} ]( {img["src"]} ) \n"
+
+        div = soup.find("div", class_="mw-parser-output")
+        if div:
+            p_tags = div.find_all("p")
+            num = 0
+            for p_tag in p_tags:
+                p = str(p_tag)
+                p = re.sub(r"<script.*?</script>|<style.*?</style>", "", p, flags=re.DOTALL)
+                p = re.sub(r"<.*?>", "", p, flags=re.DOTALL)
+                p = re.sub(r"\[.*?]", "", p, flags=re.DOTALL)
+
+                if p != "":
+                    result += str(p)
+
+                    num += 1
+                    if num >= 20:
+                        break
+        return result
+
+    # 空白页
+    elif response.status_code == 404 :
+        logger.info(f"未找到\"{msg}\", 进行搜索")
+
+        from . import mg_Search
+        context = await mg_Search.search(msg, 1)
+        keyword = re.search(r".*?\n", context, flags = re.DOTALL).group()[: -1]
+
+        logger.success(f"搜索完成, 打开\"{keyword}\"")
+        return await introduce(keyword)
+
+    # 搜索失败
+    elif response.status_code == 301 :
+        return f"未找到{msg}"
+
+    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/mg_Search.py

@@ -1,6 +1,5 @@
 from nonebot.log import logger
 
-import re
 import httpx
 import urllib.parse
 from bs4 import BeautifulSoup
@@ -9,12 +8,12 @@ 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 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):
+async def search (msg : str, num : int) :
-    logger.info(f"搜索 : \"{msg}\"")
+    logger.info(f"搜索 : \"{msg}\" ...")
     result = ""
 
     url = "https://mzh.moegirl.org.cn/index.php?search=" + urllib.parse.quote_plus(msg)
@@ -22,7 +21,7 @@ async def search(msg : str, num : int):
     logger.success(f"连接\"{url}\"完成, 状态码 : {response.status_code}")
 
     # 正常搜索
-    if response.status_code == 200:
+    if response.status_code == 200 :
         """
             萌娘百科搜索页面结构
             div.searchresults
@@ -39,74 +38,39 @@ async def search(msg : str, num : int):
 
         # 检测ul.mw-search-results, 是否有结果
         ul_tag = soup.find("ul", class_ = "mw-search-results")
-        if ul_tag:
+        if ul_tag :
             li_tags = ul_tag.find_all("li")
-            for li_tag in li_tags:
+            for li_tag in li_tags :
 
                 div_heading = li_tag.find("div", class_ = "mw-search-result-heading")
-                if div_heading:
+                if div_heading :
                     a_tag = div_heading.find("a")
                     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:
+                if div_result :
                     content = str(div_result).replace("<div class=\"searchresult\">", "").replace("</div>", "")
                     content = content.replace("<span class=\"searchmatch\">", "").replace("</span>", "")
                     result += content + "\n"
 
                 num -= 1
-                if num == 0:
+                if num == 0 :
                     break
             return result
 
         # 无ul.mw-search-results, 无结果
-        else:
+        else :
             logger.info("无结果")
             return "无结果"
 
     # 重定向
-    elif response.status_code == 302:
+    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"))
+        from . import mg_Introduce
-        soup = BeautifulSoup(response.text, "html.parser")
+        return await mg_Introduce.introduce(msg)
-        logger.success("重定向成功")
-        num = 0
 
-        """
+    else :
-            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
-                └── ...
-        """
-
-        result += msg + "\n"
-        img = soup.find("img", class_="infobox-image")
-        if img:
-            logger.info(f"照片{img["src"]}")
-            result += f"![ {msg} ]( {img["src"]} ) \n"
-
-        div = soup.find("div", class_="mw-parser-output")
-        if div:
-            p_tags = div.find_all("p")
-            for p_tag in p_tags:
-                p = str(p_tag)
-                p = re.sub(r"<script.*?</script>|<style.*?</style>", "", p, flags=re.DOTALL)
-                p = re.sub(r"<.*?>", "", p, flags = re.DOTALL)
-                p = re.sub(r"\[.*?]", "", p, flags = re.DOTALL)
-                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

@@ -7,18 +7,18 @@
 		}
 	},
 	{
-		"type" : "function",
+		"type": "function",
-		"function" : {
+		"function": {
-			"name" : "marshoai-meogirl__search",
+			"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)",
+			"description": "查找/搜索 某角色/事物 (使用萌娘百科)",
-			"parameters" : {
+			"parameters": {
-				"type" : "object",
+				"type": "object",
-				"properties" : {
+				"properties": {
-					"msg" : {
+					"msg": {
 						"type": "string",
 						"description": "搜索关键词"
 					},
-					"num" : {
+					"num": {
 						"type": "integer",
 						"description": "数据显示条数, 默认3, 可留空"
 					}
@@ -28,5 +28,24 @@
 				"msg"
 			]
 		}
+	},
+	{
+		"type" : "function",
+		"function" : {
+			"name" : "marshoai-meogirl__introduce",
+			"description" : "介绍/展示 某角色/事物 (使用萌娘百科)",
+			"parameters" : {
+				"type" : "object",
+				"properties" : {
+					"msg" : {
+						"type": "string",
+						"description": "关键词"
+					}
+				}
+			},
+			"required": [
+                "msg"
+            ]
+		}
 	}
 ]