refactor: 重构openai包基础组件架构

pkg/openai/requester/api.py

@@ -0,0 +1,31 @@
+from __future__ import annotations
+
+import abc
+import typing
+
+from ...core import app
+from ...core import entities as core_entities
+from .. import entities as llm_entities
+from ..session import entities as session_entities
+
+class LLMAPIRequester(metaclass=abc.ABCMeta):
+    """LLM API请求器
+    """
+
+    ap: app.Application
+
+    def __init__(self, ap: app.Application):
+        self.ap = ap
+
+    async def initialize(self):
+        pass
+
+    @abc.abstractmethod
+    async def request(
+        self,
+        query: core_entities.Query,
+        conversation: session_entities.Conversation,
+    ) -> typing.AsyncGenerator[llm_entities.Message, None]:
+        """请求
+        """
+        raise NotImplementedError

pkg/openai/requester/apis/chatcmpl.py

@@ -0,0 +1,32 @@
+from __future__ import annotations
+
+import asyncio
+import typing
+
+import openai
+
+from .. import api
+from ....core import entities as core_entities
+from ... import entities as llm_entities
+from ...session import entities as session_entities
+
+
+class OpenAIChatCompletion(api.LLMAPIRequester):
+
+    client: openai.Client
+
+    async def initialize(self):
+        self.client = openai.Client(
+            base_url=self.ap.cfg_mgr.data['openai_config']['reverse_proxy'],
+            timeout=self.ap.cfg_mgr.data['process_message_timeout']
+        )
+
+    async def request(self, query: core_entities.Query, conversation: session_entities.Conversation) -> typing.AsyncGenerator[llm_entities.Message, None]:
+        """请求
+        """
+        await asyncio.sleep(10)
+
+        yield llm_entities.Message(
+            role=llm_entities.MessageRole.ASSISTANT,
+            content="hello"
+        )

pkg/openai/requester/entities.py

@@ -0,0 +1,23 @@
+import typing
+
+import pydantic
+
+from . import api
+from . import token
+
+
+class LLMModelInfo(pydantic.BaseModel):
+    """模型"""
+
+    name: str
+
+    provider: str
+
+    token_mgr: token.TokenManager
+
+    requester: api.LLMAPIRequester
+
+    function_call_supported: typing.Optional[bool] = False
+
+    class Config:
+        arbitrary_types_allowed = True

pkg/openai/requester/modelmgr.py

@@ -0,0 +1,40 @@
+from __future__ import annotations
+
+from . import entities
+from ...core import app
+
+from .apis import chatcmpl
+from . import token
+
+
+class ModelManager:
+
+    ap: app.Application
+
+    model_list: list[entities.LLMModelInfo]
+    
+    def __init__(self, ap: app.Application):
+        self.ap = ap
+        self.model_list = []
+
+    async def initialize(self):
+        openai_chat_completion = chatcmpl.OpenAIChatCompletion(self.ap)
+        openai_token_mgr = token.TokenManager(self.ap, self.ap.cfg_mgr.data['openai_config']['api_key'].values())
+
+        self.model_list.append(
+            entities.LLMModelInfo(
+                name="gpt-3.5-turbo",
+                provider="openai",
+                token_mgr=openai_token_mgr,
+                requester=openai_chat_completion,
+                function_call_supported=True
+            )
+        )
+
+    async def get_model_by_name(self, name: str) -> entities.LLMModelInfo:
+        """通过名称获取模型
+        """
+        for model in self.model_list:
+            if model.name == name:
+                return model
+        raise ValueError(f"Model {name} not found")

pkg/openai/requester/token.py

@@ -0,0 +1,25 @@
+from __future__ import annotations
+
+import typing
+
+import pydantic
+
+
+class TokenManager():
+
+    provider: str
+
+    tokens: list[str]
+
+    using_token_index: typing.Optional[int] = 0
+
+    def __init__(self, provider: str, tokens: list[str]):
+        self.provider = provider
+        self.tokens = tokens
+        self.using_token_index = 0
+
+    def get_token(self) -> str:
+        return self.tokens[self.using_token_index]
+    
+    def next_token(self):
+        self.using_token_index = (self.using_token_index + 1) % len(self.tokens)