'commit'

dsRag/Start.py

@@ -1,3 +1,5 @@
+import asyncio
+import json
 import logging
 import os
 import subprocess
@@ -11,8 +13,11 @@ from logging.handlers import RotatingFileHandler
 import fastapi
 import uvicorn
 from fastapi import FastAPI, HTTPException
+from openai import AsyncOpenAI
+from sse_starlette import EventSourceResponse
 from starlette.staticfiles import StaticFiles
 
+from Config import Config
 from Util.ALiYunUtil import ALiYunUtil
 from Util.SearchUtil import *
 
@@ -123,16 +128,54 @@ async def rag(request: fastapi.Request):
 
     return {"data": "没有在知识库中找到相关的信息，无法回答此问题。"}
 
-@app.post("/api/helloWorld", response_model=None)
-async def helloWorld(request: fastapi.Request):
-    async def generate_hello_world():
-        message = "Hello,World,"
-        for char in message:
-            yield f"data: {char}\n\n"
-            import asyncio
-            await asyncio.sleep(0.5)
-    
-    return StreamingResponse(generate_hello_world(), media_type="text/event-stream")
+
+
+@app.post("/api/rag_stream", response_model=None)
+async def rag_stream(request: fastapi.Request):
+    data = await request.json()
+    query = data.get('query', '')
+    query_tags = data.get('tags', [])
+
+    # 调用es进行混合搜索
+    search_results = queryByEs(query, query_tags, logger)
+
+    # 流式调用大模型
+    # 获取StreamingResponse对象
+    return callLLM(request, query, search_results, logger, True)
+
+
+# 与用户交流聊天
+@app.post("/api/helloWorld")
+async def reply():
+    # 初始化异步 OpenAI 客户端
+    client = AsyncOpenAI(
+        api_key=Config.MODEL_API_KEY,
+        base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
+    )
+
+    async def generate_response_stream():
+        try:
+            # 流式调用大模型
+            stream = await client.chat.completions.create(
+                model=Config.MODEL_NAME,
+                messages=[
+                    {"role": "system",
+                     "content": "你是聊天人的好朋友,你认识深刻，知识渊博,不要使用哎呀这样的语气词。聊天的回复内容不要超过150字。"},
+                    {"role": "user", "content": "你是谁?"}
+                ],
+                max_tokens=4000,
+                stream=True  # 启用流式模式
+            )
+
+            # 流式返回模型生成的回复
+            async for chunk in stream:
+                if chunk.choices[0].delta.content:
+                    yield f"data: {json.dumps({'reply': chunk.choices[0].delta.content}, ensure_ascii=False)}\n\n"
+
+        except Exception as e:
+            yield f"data: {json.dumps({'error': str(e)})}\n\n"
+
+    return EventSourceResponse(generate_response_stream())
 
 
 if __name__ == "__main__":

dsRag/Util/ALiYunUtil.py

@@ -27,7 +27,7 @@ class ALiYunUtil:
         except Exception as e:
             return f"发生错误: {str(e)}"
             
-    async def chat_stream(self, prompt, model=None):
+    def chat_stream(self, prompt, model=None):
         """
         与阿里云大模型流式对话
         :param prompt: 用户输入的问题
@@ -35,15 +35,15 @@ class ALiYunUtil:
         :return: 异步生成器，返回模型流式响应
         """
         try:
-            stream = await self.client.chat.completions.create(
+            stream = self.client.chat.completions.create(
                 model=model or self.model_name,
                 messages=[
                     {'role': 'user', 'content': prompt}
                 ],
                 stream=True
             )
-            async for chunk in stream:
+            for chunk in stream:
                 if chunk.choices[0].delta.content:
                     yield chunk.choices[0].delta.content
         except Exception as e:
-            yield f"发生错误: {str(e)}"
+            yield f"发生错误: {str(e)}"

dsRag/Util/SearchUtil.py

@@ -170,6 +170,7 @@ def callLLM(request, query, search_results, logger,streamBack=False):
             async def generate():
                 async for chunk in aliyun_util.chat_stream(prompt):
                     yield f"data: {chunk}\n\n"
+            # data: 发生错误: object Stream can't be used in 'await' expression
             return StreamingResponse(generate(), media_type="text/event-stream")
         else:
             # 一次性返回