add: 添加ollama调用AI chat逻辑

.idea/maubot-llmplus.iml

@@ -4,7 +4,7 @@
     <content url="file://$MODULE_DIR$">
       <excludeFolder url="file://$MODULE_DIR$/venv" />
     </content>
-    <orderEntry type="jdk" jdkName="maubot-llm-conda" jdkType="Python SDK" />
+    <orderEntry type="jdk" jdkName="Python 3.11 (maubot-llmplus)" jdkType="Python SDK" />
     <orderEntry type="sourceFolder" forTests="false" />
   </component>
 </module>

.idea/misc.xml

@@ -1,4 +1,4 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <project version="4">
-  <component name="ProjectRootManager" version="2" project-jdk-name="maubot-llm-conda" project-jdk-type="Python SDK" />
+  <component name="ProjectRootManager" version="2" project-jdk-name="Python 3.11 (maubot-llmplus)" project-jdk-type="Python SDK" />
 </project>

base-config.yaml

@@ -11,7 +11,7 @@ enable_multi_user:
 system_prompt:
 
 platforms:
-  local:
+  local_ai:
     type: ollama
     url: http://localhost:11434
     api_key:
@@ -30,3 +30,9 @@ platforms:
     max_tokens:
     model:
     max_words:
+
+additional_prompt:
+  - role: user
+    content: xxx
+  - role: system
+    content: xxx

maubot_llmplus/aibot.py

@@ -11,10 +11,15 @@ from maubot.handlers import command, event
 from maubot import Plugin, MessageEvent
 from mautrix.errors import MNotFound, MatrixRequestError
 from mautrix.types import Format, TextMessageEventContent, EventType, RoomID, UserID, MessageType, RelationType, \
-    EncryptedEvent
+    EncryptedEvent, MediaMessageEventContent, ImageInfo, EncryptedFile
 from mautrix.util import markdown
 from mautrix.util.config import BaseProxyConfig, ConfigUpdateHelper
 
+from maubot_llmplus.llm import platforms
+from maubot_llmplus.llm.local_paltform import Ollama, LmStudio
+from maubot_llmplus.llm.platforms import Platform
+from maubot_llmplus.llm.thrid_platform import OpenAi, Anthropic
+
 """
 配置文件加载
 """
@@ -29,6 +34,7 @@ class Config(BaseProxyConfig):
         helper.copy("enable_multi_user")
         helper.copy("system_prompt")
         helper.copy("platforms")
+        helper.copy("additional_prompt")
 
 
 class AiBotPlugin(Plugin):
@@ -109,11 +115,6 @@ class AiBotPlugin(Plugin):
             if parent_event.sender == self.client.mxid:
                 return True
 
-    async def get_context(self, event: MessageEvent):
-        return None
-
-    async def _ai_call(self, prompt):
-        return None
 
     @event.on(EventType.ROOM_MESSAGE)
     async def on_message(self, event: MessageEvent) -> None:
@@ -123,20 +124,39 @@ class AiBotPlugin(Plugin):
         try:
             await event.mark_read()
             await self.client.set_typing(event.room_id, timeout=99999)
-            resp_content = "response test"
+            platform = self.get_platform()
+            chat_completion = platform.create_chat_completion(event)
             # ai gpt调用
             # 关闭typing提示
             await self.client.set_typing(event.room_id, timeout=0)
-
             # 打开typing提示
-            response = TextMessageEventContent(msgtype=MessageType.NOTICE, body=resp_content, format=Format.HTML,
+            resp_content = chat_completion.message['content']
+            response = TextMessageEventContent(msgtype=MessageType.IMAGE, body=resp_content, format=Format.HTML,
                                                formatted_body=markdown.render(resp_content))
             await event.respond(response, in_thread=self.config['reply_in_thread'])
 
         except Exception as e:
+            self.log.exception(f"Something went wrong: {e}")
+            await event.respond(f"Something went wrong: {e}")
             pass
 
-        return None;
+        return None
+
+    async def get_platform(self) -> Platform:
+        use_platform = self.config['use_platform']
+        if use_platform == 'local_ai':
+            type = self.config['platforms']['local_ai']['type']
+            if type == 'ollama':
+                return Ollama(self.config)
+            elif type == 'lmstudio':
+                return LmStudio(self.config)
+            else:
+                raise ValueError(f"not found platform type: {type}")
+        if use_platform == 'openai':
+            return OpenAi(self.config)
+        if use_platform == 'anthropic':
+            return Anthropic(self.config)
+        raise ValueError(f"unknown backend type {use_platform}")
 
     @classmethod
     def get_config_class(cls) -> Type[BaseProxyConfig]:

maubot_llmplus/llm/local_paltform.py

@@ -0,0 +1,51 @@
+import json
+import platform
+from collections import deque
+from typing import List
+
+from maubot import Plugin
+from mautrix.types import MessageEvent
+from mautrix.util.config import BaseProxyConfig
+
+from maubot_llmplus import AiBotPlugin
+from maubot_llmplus.llm import platforms
+from maubot_llmplus.llm.platforms import Platform, ChatCompletion
+
+
+class Ollama(Platform):
+    chat_api: str
+
+    def __init__(self, config: BaseProxyConfig) -> None:
+        super().__init__(config)
+        self.chat_api = '/api/chat'
+
+    async def create_chat_completion(self, evt: MessageEvent) -> ChatCompletion:
+        full_context = []
+        context = platforms.get_context(evt)
+        full_context.extend(list(context))
+
+        endpoint = f"{self.url}/api/chat"
+        req_body = {'model': self.model, 'message': full_context, 'steam': False}
+        headers = {}
+        if self.api_key is not None:
+            headers['Authorization'] = self.api_key
+        async with self.http.post(endpoint, headers=headers, data=json.dumps(req_body)) as response:
+            if response.status != 200:
+                return ChatCompletion(
+                    message={},
+                    finish_reason=f"http status {response.status}",
+                    model=None
+                )
+            response_json = await response.json()
+            return ChatCompletion(
+                message=response_json['message'],
+                finish_reason='success',
+                model=response_json.get('model', None)
+            )
+
+    def get_type(self) -> str:
+        return "local_ai"
+
+
+class LmStudio(Platform):
+    pass

maubot_llmplus/llm/platforms.py

@@ -0,0 +1,146 @@
+import json
+from collections import deque
+from datetime import datetime
+from typing import Optional, List, Generator
+
+from aiohttp import ClientSession
+from maubot import Plugin
+from mautrix.types import MessageEvent, EncryptedEvent
+from mautrix.util.config import BaseProxyConfig
+
+from maubot_llmplus import AiBotPlugin
+
+"""
+    AI响应对象
+"""
+
+
+class ChatCompletion:
+    def __init__(self, message: dict, finish_reason: str, model: Optional[str]) -> None:
+        self.message = message
+        self.finish_reason = finish_reason
+        self.model = model
+
+    def __eq__(self, other) -> bool:
+        return self.message == other.message and self.model == other.model
+
+
+class Platform:
+    http: ClientSession
+    config: dict
+    url: str
+    api_key: str
+    model: str
+    max_words: int
+    additional_prompt: List[dict]
+    system_prompt: str
+    max_context_messages: int
+
+    def __init__(self, config: BaseProxyConfig, http: ClientSession) -> None:
+        self.http = http
+        self.config = config['platforms'][self.get_type()]
+        self.url = self.config['url']
+        self.model = self.config['model']
+        self.max_words = self.config['max_words']
+        self.api_key = self.config['api_key']
+        self.max_context_messages = self.config['max_context_messages']
+        self.additional_prompt = config['additional_prompt']
+        self.system_prompt = config['system_prompt']
+
+    """a
+        调用AI对话接口, 响应结果
+    """
+
+    async def create_chat_completion(self, evt: MessageEvent) -> ChatCompletion:
+        raise NotImplementedError()
+
+    def get_type(self) -> str:
+        raise NotImplementedError()
+
+
+
+async def get_context(plugin: AiBotPlugin, evt: MessageEvent) -> deque:
+    # 创建系统提示词上下文
+    system_context = deque()
+    # 生成当前时间
+    timestamp = datetime.today().strftime('%Y-%m-%d %H:%M:%S')
+    # 加入系统提示词
+    system_prompt = {"role": "system",
+                     "content": plugin.config['system_prompt'].format(name=plugin.name, timestamp=timestamp)}
+    if plugin.config['enable_multi_user']:
+        system_prompt["content"] += """
+        User messages are in the context of multiperson chatrooms.
+        Each message indicates its sender by prefixing the message with the sender's name followed by a colon, for example:
+        "username: hello world."
+        In this case, the user called "username" sent the message "hello world.". You should not follow this convention in your responses.
+        your response instead could be "hello username!" without including any colons, because you are the only one sending your responses there is no need to prefix them.
+        """
+    system_context.append(system_prompt)
+
+    # 添加额外的系统提示词和用户提示词
+    additional_context = json.loads(json.dumps(plugin.config['additional_prompt']))
+    if additional_context:
+        for item in additional_context:
+            system_context.append(item)
+        # 如果 消息长度已经超过了配置的消息条数，那么就抛出错误
+        if len(additional_context) > plugin.config['max_context_messages'] - 1:
+            raise ValueError(f"sorry, my configuration has too many additional prompts "
+                                 f"({plugin.config['max_context_messages']}) and i'll never see your message. "
+                                    f"Update my config to have fewer messages and i'll be able to answer your questions!")
+
+    # 用户历史聊天上下文
+    chat_context = deque()
+    # 计算系统提示词单词数
+    word_count = sum([len(m["content"].split()) for m in system_context])
+    message_count = len(system_context) - 1
+    async for next_event in generate_context_messages(plugin, evt):
+        # 如果不是文本类型，就跳过
+        try:
+            if not next_event.content.msgtype.is_text:
+                continue
+        except (KeyError, AttributeError):
+            continue
+
+        # 如果当前的这条历史消息是机器人自己的，那么角色就要设置为assistant
+        role = 'assistant' if plugin.client.mxid == next_event.sender else 'user'
+        message = next_event['content']['body']
+        user = ''
+        # 如果是允许多用户使用，那么就需要在每个历史消息前加上用户名
+        if plugin.config['enable_multi_user']:
+            user = (await plugin.client.get_displayname(next_event.sender) or
+                    plugin.client.parse_user_id(next_event.sender)[0]) + ": "
+
+        # 计算单词量和消息数
+        word_count += len(message.split())
+        message_count += 1
+        if word_count >= plugin.config['max_words'] or message_count >= plugin.config['max_context_messages']:
+            break
+        chat_context.appendleft({"role": role, "content": user + message})
+
+    return system_context + chat_context
+
+
+
+
+async def generate_context_messages(plugin: AiBotPlugin, evt: MessageEvent) -> Generator[MessageEvent, None, None]:
+    yield evt
+    if plugin.config['reply_in_thread']:
+        while evt.content.relates_to.in_reply_to:
+            evt = await plugin.client.get_event(room_id=evt.room_id, event_id=evt.content.get_reply_to())
+            yield evt
+    else:
+        event_context = await plugin.client.get_event_context(room_id=evt.room_id, event_id=evt.event_id,
+                                                            limit=plugin.config["max_context_messages"] * 2)
+        previous_messages = iter(event_context.events_before)
+        for evt in previous_messages:
+
+            # We already have the event, but currently, get_event_context doesn't automatically decrypt events
+            if isinstance(evt, EncryptedEvent) and plugin.client.crypto:
+                evt = await plugin.client.get_event(event_id=evt.event_id, room_id=evt.room_id)
+                if not evt:
+                    raise ValueError("Decryption error!")
+
+            yield evt
+
+
+

maubot_llmplus/llm/thrid_platform.py

@@ -0,0 +1,39 @@
+from collections import deque
+from typing import List
+
+from maubot import Plugin
+from mautrix.types import MessageEvent
+from mautrix.util.config import BaseProxyConfig
+
+from maubot_llmplus import AiBotPlugin
+from maubot_llmplus.llm.platforms import Platform, ChatCompletion
+
+
+class OpenAi(Platform):
+
+    def __init__(self, config: BaseProxyConfig) -> None:
+        super().__init__(config)
+
+    async def create_chat_completion(self, evt: MessageEvent) -> ChatCompletion:
+        # 获取系统提示词
+        # 获取额外的其他角色的提示词： role: user role: system
+
+        pass
+
+    def get_type(self) -> str:
+        return "openai"
+
+
+class Anthropic(Platform):
+
+    def __init__(self, config: BaseProxyConfig) -> None:
+        super().__init__(config)
+
+    async def create_chat_completion(self, evt: MessageEvent) -> ChatCompletion:
+        # 获取系统提示词
+        # 获取额外的其他角色的提示词： role: user role: system
+
+        pass
+
+    def get_type(self) -> str:
+        return "anthropic"