Add API to support upload local files

README.md

@@ -134,6 +134,26 @@ curl -X 'POST' http://127.0.0.1:8000/service/batch_evaluate/response
 }
 ```
 
+3. 上传
+
+支持通过API的方式上传本地文件，并支持指定不同的faiss_path，每次发送API请求会返回一个task_id，之后可以通过task_id来查看文件上传状态（processing、completed、failed）。
+
+- **（1）上传（upload_local_data）**
+
+```bash
+curl -X 'POST' http://127.0.0.1:8000/service/upload_local_data -H 'Content-Type: multipart/form-data' -F 'file=@local_path/PAI.txt' -F 'faiss_path=localdata/storage'
+
+# Return: {"task_id": "2c1e557733764fdb9fefa063538914da"}
+```
+
+- **（2）查看上传状态（upload_local_data）**
+
+```bash
+curl http://127.0.0.1:8077/service/get_upload_state\?task_id\=2c1e557733764fdb9fefa063538914da
+
+# Return: {"task_id":"2c1e557733764fdb9fefa063538914da","status":"completed"}
+```
+
 ### 独立脚本文件：不依赖于整体服务的启动，可独立运行
 
 1. 向当前索引存储中插入新文件

src/pai_rag/app/api/query.py

@@ -1,6 +1,7 @@
 from typing import Any
-from fastapi import APIRouter, Body, BackgroundTasks
+from fastapi import APIRouter, Body, BackgroundTasks, File, UploadFile, Form
 import uuid
+import os
 from pai_rag.core.rag_service import rag_service
 from pai_rag.app.api.models import (
     RagQuery,
@@ -10,6 +11,7 @@
     LlmResponse,
     DataInput,
 )
+from pai_rag.app.web.view_model import _transform_to_dict
 
 router = APIRouter()
 
@@ -86,3 +88,38 @@ async def batch_evaluate():
         type="all"
     )
     return {"status": 200, "result": eval_results}
+
+
+@router.post("/upload_local_data")
+async def upload_local_data(
+    file: UploadFile = File(),
+    faiss_path: str = Form(),
+    background_tasks: BackgroundTasks = BackgroundTasks(),
+):
+    task_id = uuid.uuid4().hex
+    if not file:
+        return {"message": "No upload file sent"}
+    else:
+        fn = file.filename
+        save_path = f"./localdata/upload_files/{task_id}/"
+        if not os.path.exists(save_path):
+            os.makedirs(save_path, exist_ok=True)
+
+        save_file = os.path.join(save_path, fn)
+
+        f = open(save_file, "wb")
+        data = await file.read()
+        f.write(data)
+        f.close()
+    sessioned_config = rag_service.rag_configuration.get_value()
+    if faiss_path:
+        sessioned_config = rag_service.rag_configuration.get_value().copy()
+        sessioned_config.index.update({"persist_path": faiss_path})
+    rag_service.reload(_transform_to_dict(sessioned_config))
+    background_tasks.add_task(
+        rag_service.add_knowledge_async,
+        task_id=task_id,
+        file_dir=save_path,
+        enable_qa_extraction=False,
+    )
+    return {"task_id": task_id}