Merge branch 'master' into wechatmp

.gitignore

@@ -13,6 +13,7 @@ plugins.json
 itchat.pkl
 *.log
 user_datas.pkl
+chatgpt_tool_hub/
 plugins/**/
 !plugins/bdunit
 !plugins/dungeon
@@ -22,4 +23,5 @@ plugins/**/
 !plugins/banwords
 !plugins/banwords/**/
 !plugins/hello
-!plugins/role
+!plugins/role
+!plugins/keyword

bot/chatgpt/chat_gpt_bot.py

@@ -142,7 +142,7 @@ class ChatGPTBot(Bot, OpenAIImage):
                 logger.warn("[CHATGPT] RateLimitError: {}".format(e))
                 result["content"] = "提问太快啦，请休息一下再问我吧"
                 if need_retry:
-                    time.sleep(5)
+                    time.sleep(20)
             elif isinstance(e, openai.error.Timeout):
                 logger.warn("[CHATGPT] Timeout: {}".format(e))
                 result["content"] = "我没有收到你的消息"

bot/openai/open_ai_bot.py

@@ -114,7 +114,7 @@ class OpenAIBot(Bot, OpenAIImage):
                 logger.warn("[OPEN_AI] RateLimitError: {}".format(e))
                 result["content"] = "提问太快啦，请休息一下再问我吧"
                 if need_retry:
-                    time.sleep(5)
+                    time.sleep(20)
             elif isinstance(e, openai.error.Timeout):
                 logger.warn("[OPEN_AI] Timeout: {}".format(e))
                 result["content"] = "我没有收到你的消息"

bridge/context.py

@@ -8,6 +8,8 @@ class ContextType(Enum):
     VOICE = 2  # 音频消息
     IMAGE = 3  # 图片消息
     IMAGE_CREATE = 10  # 创建图片命令
+    JOIN_GROUP = 20  # 加入群聊
+    PATPAT = 21  # 拍了拍
 
     def __str__(self):
         return self.name

channel/wechat/wechat_channel.py

@@ -26,20 +26,25 @@ from lib.itchat.content import *
 from plugins import *
 
 
-@itchat.msg_register([TEXT, VOICE, PICTURE])
+@itchat.msg_register([TEXT, VOICE, PICTURE, NOTE])
 def handler_single_msg(msg):
-    # logger.debug("handler_single_msg: {}".format(msg))
-    if msg["Type"] == PICTURE and msg["MsgType"] == 47:
+    try:
+        cmsg = WeChatMessage(msg, False)
+    except NotImplementedError as e:
+        logger.debug("[WX]single message {} skipped: {}".format(msg["MsgId"], e))
         return None
-    WechatChannel().handle_single(WeChatMessage(msg))
+    WechatChannel().handle_single(cmsg)
     return None
 
 
-@itchat.msg_register([TEXT, VOICE, PICTURE], isGroupChat=True)
+@itchat.msg_register([TEXT, VOICE, PICTURE, NOTE], isGroupChat=True)
 def handler_group_msg(msg):
-    if msg["Type"] == PICTURE and msg["MsgType"] == 47:
+    try:
+        cmsg = WeChatMessage(msg, True)
+    except NotImplementedError as e:
+        logger.debug("[WX]group message {} skipped: {}".format(msg["MsgId"], e))
         return None
-    WechatChannel().handle_group(WeChatMessage(msg, True))
+    WechatChannel().handle_group(cmsg)
     return None
 
 
@@ -165,12 +170,16 @@ class WechatChannel(ChatChannel):
             logger.debug("[WX]receive voice msg: {}".format(cmsg.content))
         elif cmsg.ctype == ContextType.IMAGE:
             logger.debug("[WX]receive image msg: {}".format(cmsg.content))
-        else:
+        elif cmsg.ctype == ContextType.PATPAT:
+            logger.debug("[WX]receive patpat msg: {}".format(cmsg.content))
+        elif cmsg.ctype == ContextType.TEXT:
             logger.debug(
                 "[WX]receive text msg: {}, cmsg={}".format(
                     json.dumps(cmsg._rawmsg, ensure_ascii=False), cmsg
                 )
             )
+        else:
+            logger.debug("[WX]receive msg: {}, cmsg={}".format(cmsg.content, cmsg))
         context = self._compose_context(
             cmsg.ctype, cmsg.content, isgroup=False, msg=cmsg
         )
@@ -186,9 +195,13 @@ class WechatChannel(ChatChannel):
             logger.debug("[WX]receive voice for group msg: {}".format(cmsg.content))
         elif cmsg.ctype == ContextType.IMAGE:
             logger.debug("[WX]receive image for group msg: {}".format(cmsg.content))
-        else:
+        elif cmsg.ctype in [ContextType.JOIN_GROUP, ContextType.PATPAT]:
+            logger.debug("[WX]receive note msg: {}".format(cmsg.content))
+        elif cmsg.ctype == ContextType.TEXT:
             # logger.debug("[WX]receive group msg: {}, cmsg={}".format(json.dumps(cmsg._rawmsg, ensure_ascii=False), cmsg))
             pass
+        else:
+            logger.debug("[WX]receive group msg: {}".format(cmsg.content))
         context = self._compose_context(
             cmsg.ctype, cmsg.content, isgroup=True, msg=cmsg
         )

channel/wechat/wechat_message.py

@@ -1,3 +1,5 @@
+import re
+
 from bridge.context import ContextType
 from channel.chat_message import ChatMessage
 from common.log import logger
@@ -24,9 +26,37 @@ class WeChatMessage(ChatMessage):
             self.ctype = ContextType.IMAGE
             self.content = TmpDir().path() + itchat_msg["FileName"]  # content直接存临时目录路径
             self._prepare_fn = lambda: itchat_msg.download(self.content)
+        elif itchat_msg["Type"] == NOTE and itchat_msg["MsgType"] == 10000:
+            if is_group and (
+                "加入群聊" in itchat_msg["Content"] or "加入了群聊" in itchat_msg["Content"]
+            ):
+                self.ctype = ContextType.JOIN_GROUP
+                self.content = itchat_msg["Content"]
+                # 这里只能得到nickname， actual_user_id还是机器人的id
+                if "加入了群聊" in itchat_msg["Content"]:
+                    self.actual_user_nickname = re.findall(
+                        r"\"(.*?)\"", itchat_msg["Content"]
+                    )[-1]
+                elif "加入群聊" in itchat_msg["Content"]:
+                    self.actual_user_nickname = re.findall(
+                        r"\"(.*?)\"", itchat_msg["Content"]
+                    )[0]
+            elif "拍了拍我" in itchat_msg["Content"]:
+                self.ctype = ContextType.PATPAT
+                self.content = itchat_msg["Content"]
+                if is_group:
+                    self.actual_user_nickname = re.findall(
+                        r"\"(.*?)\"", itchat_msg["Content"]
+                    )[0]
+            else:
+                raise NotImplementedError(
+                    "Unsupported note message: " + itchat_msg["Content"]
+                )
         else:
             raise NotImplementedError(
-                "Unsupported message type: {}".format(itchat_msg["Type"])
+                "Unsupported message type: Type:{} MsgType:{}".format(
+                    itchat_msg["Type"], itchat_msg["MsgType"]
+                )
             )
 
         self.from_user_id = itchat_msg["FromUserName"]
@@ -58,4 +88,5 @@ class WeChatMessage(ChatMessage):
         if self.is_group:
             self.is_at = itchat_msg["IsAt"]
             self.actual_user_id = itchat_msg["ActualUserName"]
-            self.actual_user_nickname = itchat_msg["ActualNickName"]
+            if self.ctype not in [ContextType.JOIN_GROUP, ContextType.PATPAT]:
+                self.actual_user_nickname = itchat_msg["ActualNickName"]

plugins/banwords/banwords.py

@@ -102,4 +102,4 @@ class Banwords(Plugin):
                 return
 
     def get_help_text(self, **kwargs):
-        return Banwords.desc
+        return "过滤消息中的敏感词。"

plugins/hello/hello.py

@@ -23,7 +23,27 @@ class Hello(Plugin):
         logger.info("[Hello] inited")
 
     def on_handle_context(self, e_context: EventContext):
-        if e_context["context"].type != ContextType.TEXT:
+        if e_context["context"].type not in [
+            ContextType.TEXT,
+            ContextType.JOIN_GROUP,
+            ContextType.PATPAT,
+        ]:
+            return
+
+        if e_context["context"].type == ContextType.JOIN_GROUP:
+            e_context["context"].type = ContextType.TEXT
+            msg: ChatMessage = e_context["context"]["msg"]
+            e_context[
+                "context"
+            ].content = f'请你随机使用一种风格说一句问候语来欢迎新用户"{msg.actual_user_nickname}"加入群聊。'
+            e_context.action = EventAction.CONTINUE  # 事件继续，交付给下个插件或默认逻辑
+            return
+
+        if e_context["context"].type == ContextType.PATPAT:
+            e_context["context"].type = ContextType.TEXT
+            msg: ChatMessage = e_context["context"]["msg"]
+            e_context["context"].content = f"请你随机使用一种风格介绍你自己，并告诉用户输入#help可以查看帮助信息。"
+            e_context.action = EventAction.CONTINUE  # 事件继续，交付给下个插件或默认逻辑
             return
 
         content = e_context["context"].content

plugins/keyword/README.md

@@ -0,0 +1,13 @@
+# 目的
+关键字匹配并回复
+
+# 试用场景
+目前是在微信公众号下面使用过。
+
+# 使用步骤
+1. 复制 `config.json.template` 为 `config.json`
+2. 在关键字 `keyword` 新增需要关键字匹配的内容
+3. 重启程序做验证
+
+# 验证结果
+![结果](test-keyword.png)

plugins/keyword/__init__.py

@@ -0,0 +1 @@
+from .keyword import *

plugins/keyword/config.json.template

@@ -0,0 +1,5 @@
+{
+  "keyword": {
+      "关键字匹配": "测试成功"
+  }
+}

plugins/keyword/keyword.py

@@ -0,0 +1,67 @@
+# encoding:utf-8
+
+import json
+import os
+
+import plugins
+from bridge.context import ContextType
+from bridge.reply import Reply, ReplyType
+from common.log import logger
+from plugins import *
+
+
+@plugins.register(
+    name="Keyword",
+    desire_priority=900,
+    hidden=True,
+    desc="关键词匹配过滤",
+    version="0.1",
+    author="fengyege.top",
+)
+class Keyword(Plugin):
+    def __init__(self):
+        super().__init__()
+        try:
+            curdir = os.path.dirname(__file__)
+            config_path = os.path.join(curdir, "config.json")
+            conf = None
+            if not os.path.exists(config_path):
+                logger.debug(f"[keyword]不存在配置文件{config_path}")
+                conf = {"keyword": {}}
+                with open(config_path, "w", encoding="utf-8") as f:
+                    json.dump(conf, f, indent=4)
+            else:
+                logger.debug(f"[keyword]加载配置文件{config_path}")
+                with open(config_path, "r", encoding="utf-8") as f:
+                    conf = json.load(f)
+            # 加载关键词
+            self.keyword = conf["keyword"]
+
+            logger.info("[keyword] {}".format(self.keyword))
+            self.handlers[Event.ON_HANDLE_CONTEXT] = self.on_handle_context
+            logger.info("[keyword] inited.")
+        except Exception as e:
+            logger.warn(
+                "[keyword] init failed, ignore or see https://github.com/zhayujie/chatgpt-on-wechat/tree/master/plugins/keyword ."
+            )
+            raise e
+
+    def on_handle_context(self, e_context: EventContext):
+        if e_context["context"].type != ContextType.TEXT:
+            return
+
+        content = e_context["context"].content.strip()
+        logger.debug("[keyword] on_handle_context. content: %s" % content)
+        if content in self.keyword:
+            logger.debug(f"[keyword] 匹配到关键字【{content}】")
+            reply_text = self.keyword[content]
+
+            reply = Reply()
+            reply.type = ReplyType.TEXT
+            reply.content = reply_text
+            e_context["reply"] = reply
+            e_context.action = EventAction.BREAK_PASS  # 事件结束，并跳过处理context的默认逻辑
+
+    def get_help_text(self, **kwargs):
+        help_text = "关键词过滤"
+        return help_text

plugins/tool/README.md

@@ -9,9 +9,21 @@
 ### 1. python
 ###### python解释器，使用它来解释执行python指令，可以配合你想要chatgpt生成的代码输出结果或执行事务
 
-### 2. url-get
+### 2. 访问网页的工具汇总(默认url-get)
+
+#### 2.1 url-get
 ###### 往往用来获取某个网站具体内容，结果可能会被反爬策略影响
 
+#### 2.2 browser
+###### 浏览器，功能与2.1类似，但能更好模拟，不会被识别为爬虫影响获取网站内容
+
+> 注1：url-get默认配置、browser需额外配置，browser依赖google-chrome，你需要提前安装好
+
+> 注2：browser默认使用summary tool 分段总结长文本信息，tokens可能会大量消耗！
+
+这是debian端安装google-chrome教程，其他系统请执行查找
+> https://www.linuxjournal.com/content/how-can-you-install-google-browser-debian
+
 ### 3. terminal
 ###### 在你运行的电脑里执行shell命令，可以配合你想要chatgpt生成的代码使用，给予自然语言控制手段
 
@@ -38,47 +50,83 @@
 ### 5. wikipedia
 ###### 可以回答你想要知道确切的人事物
 
-### 6. news *
+### 6. 新闻类工具
+
+#### 6.1. news-api *
 ###### 从全球 80,000 多个信息源中获取当前和历史新闻文章
 
-### 7. morning-news *
+#### 6.2. morning-news *
 ###### 每日60秒早报，每天凌晨一点更新，本工具使用了[alapi-每日60秒早报](https://alapi.cn/api/view/93)
 
 > 该tool每天返回内容相同
 
-### 8. bing-search *
+#### 6.3. finance-news 
+###### 获取实时的金融财政新闻
+
+> 该工具需要解决browser tool 的google-chrome依赖安装
+
+### 7. bing-search *
 ###### bing搜索引擎，从此你不用再烦恼搜索要用哪些关键词
 
-### 9. wolfram-alpha *
+### 8. wolfram-alpha *
 ###### 知识搜索引擎、科学问答系统，常用于专业学科计算
 
-### 10. google-search *
+### 9. google-search *
 ###### google搜索引擎，申请流程较bing-search繁琐
 
-###### 注1：带*工具需要获取api-key才能使用，部分工具需要外网支持  
+
+### 10. arxiv(dev 开发中)
+###### 用于查找论文
+
+
+### 11. debug(dev 开发中，目前没有接入wechat)
+###### 当bot遇到无法确定的信息时，将会向你寻求帮助的工具
+
+
+### 12. summary
+###### 总结工具，该工具必须输入一个本地文件的绝对路径
+
+> 该工具目前是和其他工具配合使用，暂未测试单独使用效果
+
+
+### 13. image2text
+###### 将图片转换成文字，底层调用imageCaption模型，该工具必须输入一个本地文件的绝对路径
+
+
+### 14. searxng-search *
+###### 一个私有化的搜索引擎工具
+
+> 安装教程：https://docs.searxng.org/admin/installation.html
+
+---
+
+###### 注1：带*工具需要获取api-key才能使用(在config.json内的kwargs添加项)，部分工具需要外网支持  
 #### [申请方法](https://github.com/goldfishh/chatgpt-tool-hub/blob/master/docs/apply_optional_tool.md)
 
 ## config.json 配置说明
 ###### 默认工具无需配置，其它工具需手动配置，一个例子：
 ```json
 {
-  "tools": ["wikipedia"],  // 填入你想用到的额外工具名
+  "tools": ["wikipedia", "你想要添加的其他工具"],  // 填入你想用到的额外工具名
   "kwargs": {
-      "request_timeout": 60,  // openai接口超时时间
+      "debug": true, // 当你遇到问题求助时，需要配置
+      "request_timeout": 120,  // openai接口超时时间
       "no_default": false,  // 是否不使用默认的4个工具
-      "OPTIONAL_API_NAME": "OPTIONAL_API_KEY"  // 带*工具需要申请api-key，在这里填入，api_name参考前述`申请方法`
+      // 带*工具需要申请api-key，在这里填入，api_name参考前述`申请方法`
   }
 }
 
 ```
 注：config.json文件非必须，未创建仍可使用本tool；带*工具需在kwargs填入对应api-key键值对  
-- `tools`：本插件初始化时加载的工具, 目前可选集：["wikipedia", "wolfram-alpha", "bing-search", "google-search", "news", "morning-news"] & 默认工具，除wikipedia工具之外均需要申请api-key
+- `tools`：本插件初始化时加载的工具, 目前可选集：["wikipedia", "wolfram-alpha", "bing-search", "google-search", "news"] & 默认工具，除wikipedia工具之外均需要申请api-key
 - `kwargs`：工具执行时的配置，一般在这里存放**api-key**，或环境配置
+  - `debug`: 输出chatgpt-tool-hub额外信息用于调试
   - `request_timeout`: 访问openai接口的超时时间，默认与wechat-on-chatgpt配置一致，可单独配置
   - `no_default`: 用于配置默认加载4个工具的行为，如果为true则仅使用tools列表工具，不加载默认工具
   - `top_k_results`: 控制所有有关搜索的工具返回条目数，数字越高则参考信息越多，但无用信息可能干扰判断，该值一般为2
   - `model_name`: 用于控制tool插件底层使用的llm模型，目前暂未测试3.5以外的模型，一般保持默认
 
+---
 
 ## 备注
 - 强烈建议申请搜索工具搭配使用，推荐bing-search

plugins/tool/tool.py

@@ -1,7 +1,7 @@
 import json
 import os
 
-from chatgpt_tool_hub.apps import load_app
+from chatgpt_tool_hub.apps import AppFactory
 from chatgpt_tool_hub.apps.app import App
 from chatgpt_tool_hub.tools.all_tool_list import get_all_tool_names
 
@@ -18,7 +18,7 @@ from plugins import *
 @plugins.register(
     name="tool",
     desc="Arming your ChatGPT bot with various tools",
-    version="0.3",
+    version="0.4",
     author="goldfishh",
     desire_priority=0,
 )
@@ -131,17 +131,17 @@ class Tool(Plugin):
 
     def _build_tool_kwargs(self, kwargs: dict):
         tool_model_name = kwargs.get("model_name")
+        request_timeout = kwargs.get("request_timeout")
 
         return {
+            "debug": kwargs.get("debug", False),
             "openai_api_key": conf().get("open_ai_api_key", ""),
             "proxy": conf().get("proxy", ""),
-            "request_timeout": str(conf().get("request_timeout", 60)),
+            "request_timeout": request_timeout if request_timeout else conf().get("request_timeout", 120),
             # note: 目前tool暂未对其他模型测试，但这里仍对配置来源做了优先级区分，一般插件配置可覆盖全局配置
-            "model_name": tool_model_name
-            if tool_model_name
-            else conf().get("model", "gpt-3.5-turbo"),
+            "model_name": tool_model_name if tool_model_name else conf().get("model", "gpt-3.5-turbo"),
             "no_default": kwargs.get("no_default", False),
-            "top_k_results": kwargs.get("top_k_results", 2),
+            "top_k_results": kwargs.get("top_k_results", 3),
             # for news tool
             "news_api_key": kwargs.get("news_api_key", ""),
             # for bing-search tool
@@ -157,8 +157,6 @@ class Tool(Plugin):
             "zaobao_api_key": kwargs.get("zaobao_api_key", ""),
             # for visual_dl tool
             "cuda_device": kwargs.get("cuda_device", "cpu"),
-            # for browser tool
-            "phantomjs_exec_path": kwargs.get("phantomjs_exec_path", ""),
         }
 
     def _filter_tool_list(self, tool_list: list):
@@ -172,11 +170,12 @@ class Tool(Plugin):
 
     def _reset_app(self) -> App:
         tool_config = self._read_json()
+        app_kwargs = self._build_tool_kwargs(tool_config.get("kwargs", {}))
+
+        app = AppFactory()
+        app.init_env(**app_kwargs)
 
         # filter not support tool
         tool_list = self._filter_tool_list(tool_config.get("tools", []))
 
-        return load_app(
-            tools_list=tool_list,
-            **self._build_tool_kwargs(tool_config.get("kwargs", {})),
-        )
+        return app.create_app(tools_list=tool_list, **app_kwargs)

requirements-optional.txt

@@ -20,5 +20,6 @@ pysilk_mod>=1.6.0 # needed by send voice
 web.py
 
 # chatgpt-tool-hub plugin
+
 --extra-index-url https://pypi.python.org/simple
-chatgpt_tool_hub>=0.3.9
+chatgpt_tool_hub>=0.4.1

voice/azure/azure_voice.py

@@ -66,7 +66,11 @@ class AzureVoice(Voice):
             )
             reply = Reply(ReplyType.TEXT, result.text)
         else:
-            logger.error("[Azure] voiceToText error, result={}".format(result))
+            logger.error(
+                "[Azure] voiceToText error, result={}, canceldetails={}".format(
+                    result, result.cancellation_details
+                )
+            )
             reply = Reply(ReplyType.ERROR, "抱歉，语音识别失败")
         return reply
 
@@ -83,6 +87,10 @@ class AzureVoice(Voice):
             )
             reply = Reply(ReplyType.VOICE, fileName)
         else:
-            logger.error("[Azure] textToVoice error, result={}".format(result))
+            logger.error(
+                "[Azure] textToVoice error, result={}, canceldetails={}".format(
+                    result, result.cancellation_details
+                )
+            )
             reply = Reply(ReplyType.ERROR, "抱歉，语音合成失败")
         return reply