修改配置和response的细节

backend/main.py

@@ -1,7 +1,8 @@
 # backend/main.py
 from fastapi import FastAPI, APIRouter, BackgroundTasks
-from .service import crawler_sql_service
-from .workflow import workflow
+# 确保导入路径与你的文件名一致，如果文件名是 workflow.py 则用 workflow
+from .services.crawler_sql_service import crawler_sql_service
+from .services.automated_crawler import workflow 
 from .schemas import (
     RegisterRequest, PendingRequest, SaveResultsRequest, AddUrlsRequest, SearchRequest,
     AutoMapRequest, AutoProcessRequest, TextSearchRequest
@@ -10,6 +11,11 @@ from .utils import make_response
 
 app = FastAPI(title="Wiki Crawler API")
 
+# ==========================================
+# 工具函数
+# ==========================================
+
+
 # ==========================================
 # V1 Router: 原始的底层接口 (Manual Control)
 # ==========================================
@@ -18,8 +24,10 @@ router_v1 = APIRouter()
 @router_v1.post("/register")
 async def register(req: RegisterRequest):
     try:
-        data = crawler_sql_service.register_task(req.url)
-        return make_response(1, "Success", data)
+        # Service 返回: {'task_id': 1, 'is_new_task': True, 'msg': '...'}
+        res = crawler_sql_service.register_task(req.url)
+        # 使用 pop 将 msg 提取出来作为响应的 msg，剩下的作为 data
+        return make_response(1, res.pop("msg", "Success"), res)
     except Exception as e:
         return make_response(0, str(e))
 
@@ -27,44 +35,43 @@ async def register(req: RegisterRequest):
 async def add_urls(req: AddUrlsRequest):
     try:
         urls = req.urls_obj["urls"]
-        data = crawler_sql_service.add_urls(req.task_id, urls=urls)
-        return make_response(1, "Success", data)
+        res = crawler_sql_service.add_urls(req.task_id, urls=urls)
+        return make_response(1, res.pop("msg", "Success"), res)
     except Exception as e:
         return make_response(0, str(e))
 
 @router_v1.post("/pending_urls")
 async def pending_urls(req: PendingRequest):
     try:
-        data = crawler_sql_service.get_pending_urls(req.task_id, req.limit)
-        msg = "Success" if data["urls"] else "Queue Empty"
-        return make_response(1, msg, data)
+        res = crawler_sql_service.get_pending_urls(req.task_id, req.limit)
+        # 即使队列为空，Service 也会返回 msg="Queue is empty"
+        return make_response(1, res.pop("msg", "Success"), res)
     except Exception as e:
         return make_response(0, str(e))
 
 @router_v1.post("/save_results")
 async def save_results(req: SaveResultsRequest):
     try:
-        data = crawler_sql_service.save_results(req.task_id, req.results)
-        return make_response(1, "Success", data)
+        res = crawler_sql_service.save_results(req.task_id, req.results)
+        return make_response(1, res.pop("msg", "Success"), res)
     except Exception as e:
         return make_response(0, str(e))
 
 @router_v1.post("/search")
 async def search_v1(req: SearchRequest):
-    """V1 搜索：需要客户端自己传向量"""
+    """V1 搜索：客户端手动传向量"""
     try:
         vector = req.query_embedding['vector']
-        # 注意：这里需要确认你数据库的向量维度。TextEmbedding V3 可能是 1024，V2 是 1536。
-        # 请根据你的 PGVector 设置进行匹配。
         if not vector:
             return make_response(2, "Vector is empty", None)
 
-        data = crawler_sql_service.search_knowledge(
+        # Service 现在返回 {'results': [...], 'msg': 'Found ...'}
+        res = crawler_sql_service.search_knowledge(
             query_embedding=vector,
             task_id=req.task_id,
             limit=req.limit
         )
-        return make_response(1, "Search Done", data)
+        return make_response(1, res.pop("msg", "Search Done"), res)
     except Exception as e:
         return make_response(0, str(e))
 
@@ -75,16 +82,14 @@ async def search_v1(req: SearchRequest):
 router_v2 = APIRouter()
 
 @router_v2.post("/auto/map")
-async def auto_map(req: AutoMapRequest, background_tasks: BackgroundTasks):
+async def auto_map(req: AutoMapRequest):
     """
-    [异步] 输入首页 URL，自动调用 Firecrawl Map 并入库
+    [同步] 输入首页 URL，自动调用 Firecrawl Map 并入库
     """
-    # 也可以放入 background_tasks，但 map 通常比较快，这里演示同步返回任务ID
     try:
-        # 为了不阻塞主线程，如果 map 很慢，建议放入 background_tasks
-        # 这里为了能立刻看到 task_id，先同步调用 (Firecrawl Map 比较快)
-        data = workflow.map_and_ingest(req.url)
-        return make_response(1, "Mapping Started", data)
+        # Workflow 返回: {'task_id':..., 'msg': 'Task mapped...', ...}
+        res = workflow.map_and_ingest(req.url)
+        return make_response(1, res.pop("msg", "Mapping Started"), res)
     except Exception as e:
         return make_response(0, str(e))
 
@@ -93,9 +98,14 @@ async def auto_process(req: AutoProcessRequest, background_tasks: BackgroundTask
     """
     [异步] 触发后台任务：消费队列 -> 抓取 -> Embedding -> 入库
     """
-    # 将耗时操作放入后台任务
-    background_tasks.add_task(workflow.process_task_queue, req.task_id, req.batch_size)
-    return make_response(1, "Processing started in background", {"task_id": req.task_id})
+    try:
+        # 将耗时操作放入后台任务
+        background_tasks.add_task(workflow.process_task_queue, req.task_id, req.batch_size)
+        
+        # 因为是后台任务，无法立即获取 Service 的返回值 msg，只能返回通用消息
+        return make_response(1, "Background processing started", {"task_id": req.task_id})
+    except Exception as e:
+        return make_response(0, str(e))
 
 @router_v2.post("/search")
 async def search_v2(req: TextSearchRequest):
@@ -103,8 +113,9 @@ async def search_v2(req: TextSearchRequest):
     [智能] 输入自然语言文本 -> 后端转向量 -> 搜索
     """
     try:
-        data = workflow.search_with_embedding(req.query, req.task_id, req.limit)
-        return make_response(1, "Search Success", data)
+        # Workflow 返回 {'results': [...], 'msg': '...'}
+        res = workflow.search_with_embedding(req.query, req.task_id, req.limit)
+        return make_response(1, res.pop("msg", "Search Success"), res)
     except Exception as e:
         return make_response(0, f"Search Failed: {str(e)}")