完成RAG测试

backend/services/data_service.py

@@ -143,7 +143,7 @@ class DataService:
                 and_(self.db.queue.c.task_id == task_id, self.db.queue.c.url == clean_url)
             ).values(status='completed'))
 
-    def search(self, query_text: str, query_vector: list, task_id=None, candidates_num: int = 50):
+    def search(self, query_text: str, query_vector: list, task_id=None, candidates_num: int = 50, vector_weight: float = 0.7):
         # 向量格式清洗
         if hasattr(query_vector, 'tolist'): query_vector = query_vector.tolist()
         if isinstance(query_vector, list) and len(query_vector) > 0 and isinstance(query_vector[0], list):
@@ -151,26 +151,52 @@ class DataService:
 
         results = []
         with self.db.engine.connect() as conn:
+            # 1. 构造 Query 对象 (这是 tsquery 类型)
             keyword_query = func.websearch_to_tsquery('english', query_text)
-            vector_score = (1 - self.db.chunks.c.embedding.cosine_distance(query_vector))
-            keyword_score = func.ts_rank(self.db.chunks.c.content_tsvector, keyword_query)
-            final_score = (vector_score * 0.7 + func.coalesce(keyword_score, 0) * 0.3).label("score")
+            
+            # 计算分数 (逻辑不变)
+            vector_dist = self.db.chunks.c.embedding.cosine_distance(query_vector)
+            vector_score = (1 - vector_dist)
+            
+            # 注意：ts_rank 需要 (tsvector, tsquery)
+            keyword_rank = func.ts_rank(self.db.chunks.c.content_tsvector, keyword_query)
+            keyword_score = func.coalesce(keyword_rank, 0)
+
+            keyword_weight = 1.0 - vector_weight
+            final_score = (vector_score * vector_weight + keyword_score * keyword_weight).label("score")
 
             stmt = select(
                 self.db.chunks.c.task_id, self.db.chunks.c.source_url, self.db.chunks.c.title,
                 self.db.chunks.c.content, self.db.chunks.c.meta_info, final_score
             )
+            
+            # ================= 修复点开始 =================
+            # 只有当 vector_weight 为 0 (纯关键词模式) 时，才强制加 WHERE 过滤
+            if vector_weight == 0:
+                # 错误写法 (SQLAlchemy 自动生成 plainto_tsquery 导致报错):
+                # stmt = stmt.where(keyword_query.match(self.db.chunks.c.content_tsvector))
+                
+                # 正确写法 (直接使用 PG 的 @@ 操作符):
+                # 含义: content_tsvector @@ keyword_query
+                stmt = stmt.where(self.db.chunks.c.content_tsvector.op('@@')(keyword_query))
+            # ================= 修复点结束 =================
+
             if task_id: stmt = stmt.where(self.db.chunks.c.task_id == task_id)
+            
             stmt = stmt.order_by(desc("score")).limit(candidates_num)
             
             try:
                 rows = conn.execute(stmt).fetchall()
                 results = [{"task_id": r[0], "source_url": r[1], "title": r[2], "content": r[3], "meta_info": r[4], "score": float(r[5])} for r in rows]
             except Exception as e:
-                logger.error(f"Hybrid search failed: {e}")
-                return self._fallback_vector_search(query_vector, task_id, candidates_num)
+                # 打印详细错误方便调试
+                logger.error(f"Search failed: {e}") 
+                # 只有混合或向量模式才回退，如果是纯关键词模式报错，回退也没用
+                if vector_weight > 0:
+                    return self._fallback_vector_search(query_vector, task_id, candidates_num)
+                return {"results": [], "msg": "Keyword search failed"}
 
-        return {"results": results, "msg": f"Hybrid found {len(results)}"}
+        return {"results": results, "msg": f"Found {len(results)}"}
 
     def _fallback_vector_search(self, vector, task_id, limit):
         logger.warning("Fallback to pure vector search")

backend/services/llm_service.py

@@ -93,5 +93,28 @@ class LLMService:
             logger.error(f"Rerank Exception: {e}")
             # 降级策略
             return documents[:top_n]
+    def chat(self, prompt: str, system_prompt: str = None, model: str = "qwen-turbo") -> str:
+        """
+        [新增] 通用对话生成接口，用于RAG的最终回答或作为测试裁判(Judge)
+        """
+        messages = []
+        if system_prompt:
+            messages.append({'role': 'system', 'content': system_prompt})
+        messages.append({'role': 'user', 'content': prompt})
 
+        try:
+            resp = dashscope.Generation.call(
+                model=model,
+                messages=messages,
+                result_format='message'
+            )
+            if resp.status_code == HTTPStatus.OK:
+                return resp.output.choices[0].message.content
+            else:
+                logger.error(f"Chat API Error: {resp}")
+                return "Error generating response."
+        except Exception as e:
+            logger.error(f"Chat Exception: {e}")
+            return "Error generating response."
+        
 llm_service = LLMService()