fix: Weaviate 连接优化 - 缓存 embedding、超时、UI 状态显示

app.py

@@ -48,42 +48,75 @@ CLARE_RUN_PATH = "Clare_Run.png"
 CLARE_READING_PATH = "Clare_reading.png"  # 确保存在
 
 # ================== Weaviate 直连 / GenAICoursesDB 检索 ==================
-def _retrieve_from_weaviate(question: str, top_k: int = 5) -> str:
-    """直接连接 Weaviate Cloud 检索 GENAI 课程。Embedding 需与索引一致（all-MiniLM-L6-v2）。"""
-    if not USE_WEAVIATE_DIRECT or len(question.strip()) < 5:
-        return ""
-    try:
-        import weaviate
-        from weaviate.classes.init import Auth
-        from llama_index.core import Settings, VectorStoreIndex
-        from llama_index.embeddings.huggingface import HuggingFaceEmbedding
-        from llama_index.vector_stores.weaviate import WeaviateVectorStore
+_WEAVIATE_EMBED_MODEL = None  # 缓存，避免每次请求都加载 sentence-transformers
+
 
-        # 必须与 GenAICoursesDB 建索引时一致
-        Settings.embed_model = HuggingFaceEmbedding(
+def _get_weaviate_embed_model():
+    """懒加载并缓存 embedding 模型（与建索引时一致）。"""
+    global _WEAVIATE_EMBED_MODEL
+    if _WEAVIATE_EMBED_MODEL is None:
+        from llama_index.embeddings.huggingface import HuggingFaceEmbedding
+        _WEAVIATE_EMBED_MODEL = HuggingFaceEmbedding(
             model_name="sentence-transformers/all-MiniLM-L6-v2"
         )
-        client = weaviate.connect_to_weaviate_cloud(
-            cluster_url=WEAVIATE_URL,
-            auth_credentials=Auth.api_key(WEAVIATE_API_KEY),
-        )
+    return _WEAVIATE_EMBED_MODEL
+
+
+def _retrieve_from_weaviate(question: str, top_k: int = 5, timeout_sec: float = 25.0) -> str:
+    """直接连接 Weaviate Cloud 检索 GENAI 课程。带超时，避免 HF Space 阻塞。"""
+    if not USE_WEAVIATE_DIRECT or len(question.strip()) < 5:
+        return ""
+    import concurrent.futures
+
+    def _call():
         try:
-            if not client.is_ready():
-                return ""
-            vs = WeaviateVectorStore(
-                weaviate_client=client,
-                index_name=WEAVIATE_COLLECTION,
+            import weaviate
+            from weaviate.classes.init import Auth
+            from llama_index.core import Settings, VectorStoreIndex
+            from llama_index.vector_stores.weaviate import WeaviateVectorStore
+
+            Settings.embed_model = _get_weaviate_embed_model()
+            client = weaviate.connect_to_weaviate_cloud(
+                cluster_url=WEAVIATE_URL,
+                auth_credentials=Auth.api_key(WEAVIATE_API_KEY),
             )
-            index = VectorStoreIndex.from_vector_store(vs)
-            nodes = index.as_retriever(similarity_top_k=top_k).retrieve(question)
-            return "\n\n---\n\n".join(n.get_content() for n in nodes) if nodes else ""
-        finally:
-            client.close()
-    except Exception as e:
-        print(f"[weaviate] retrieve failed: {repr(e)}")
+            try:
+                if not client.is_ready():
+                    return ""
+                vs = WeaviateVectorStore(
+                    weaviate_client=client,
+                    index_name=WEAVIATE_COLLECTION,
+                )
+                index = VectorStoreIndex.from_vector_store(vs)
+                nodes = index.as_retriever(similarity_top_k=top_k).retrieve(question)
+                return "\n\n---\n\n".join(n.get_content() for n in nodes) if nodes else ""
+            finally:
+                client.close()
+        except Exception as e:
+            print(f"[weaviate] retrieve failed: {repr(e)}")
+            return ""
+
+    try:
+        with concurrent.futures.ThreadPoolExecutor(max_workers=1) as ex:
+            fut = ex.submit(_call)
+            return fut.result(timeout=timeout_sec)
+    except concurrent.futures.TimeoutError:
+        print(f"[weaviate] retrieve timeout after {timeout_sec}s")
         return ""
 
 
+def get_weaviate_status_text() -> str:
+    """用于 UI 显示的 Weaviate 连接状态。"""
+    if not USE_WEAVIATE_DIRECT:
+        return (
+            "⚠️ **Weaviate**: 未配置\n\n"
+            "在 Space 页点击 **Settings → Repository secrets → New secret** 添加：\n"
+            "- `WEAVIATE_URL` = `https://iydyvd4wqnekotfiftma.c0.us-west3.gcp.weaviate.cloud`\n"
+            "- `WEAVIATE_API_KEY` = 你的 Weaviate API Key"
+        )
+    return "✅ **Weaviate**: 已配置 — GENAI 课程检索已启用"
+
+
 def _retrieve_from_genai_courses(question: str, top_k: int = 5, timeout_sec: float = 25.0) -> str:
     """调用 GenAICoursesDB Space 的 retrieve 接口（Weaviate 未配置时回退）。"""
     if not GENAI_COURSES_SPACE or len(question.strip()) < 5:
@@ -538,6 +571,8 @@ with gr.Blocks(
                 interactive=False,
             )
 
+            gr.Markdown(get_weaviate_status_text())
+
             with gr.Accordion(
                 "User Guide", open=True, elem_classes="main-user-guide"
             ):
@@ -1077,6 +1112,10 @@ with gr.Blocks(
             if USE_WEAVIATE_DIRECT:
                 course_chunks = _retrieve_from_weaviate(message)
                 course_source = "Weaviate Cloud (GENAI COURSES)"
+                if course_chunks:
+                    print(f"[ClareVoice] Weaviate 检索成功, 约 {len(course_chunks)} 字符")
+                else:
+                    print("[ClareVoice] Weaviate 检索无结果 (object count 可能为 0，请运行 build_weaviate_index.py)")
             elif GENAI_COURSES_SPACE:
                 course_chunks = _retrieve_from_genai_courses(message)
                 course_source = "GenAICoursesDB"
@@ -1090,6 +1129,16 @@ with gr.Blocks(
                         "_rag_score": 1.0,
                     }
                 ]
+            elif USE_WEAVIATE_DIRECT and course_source:
+                # 已配置 Weaviate 但无结果，在 References 中说明
+                rag_used_chunks = list(rag_used_chunks or []) + [
+                    {
+                        "source_file": course_source,
+                        "section": "已尝试检索，无结果（请先运行 build_weaviate_index.py 上传索引）",
+                        "text": "",
+                        "_rag_score": 0.0,
+                    }
+                ]
         else:
             rag_context_text, rag_used_chunks = "", []
 

config.py

@@ -1,6 +1,13 @@
 # config.py
 import os
 from typing import List, Dict
+
+# 确保加载 .env（本地开发）；HF Space 的 Secrets 在启动时已注入
+try:
+    from dotenv import load_dotenv
+    load_dotenv()
+except Exception:
+    pass
 from openai import OpenAI
 from langchain_openai import ChatOpenAI, OpenAIEmbeddings
 
@@ -43,6 +50,12 @@ USE_WEAVIATE_DIRECT = bool(WEAVIATE_URL and WEAVIATE_API_KEY)
 if USE_WEAVIATE_DIRECT and not WEAVIATE_URL.startswith("http"):
     WEAVIATE_URL = "https://" + WEAVIATE_URL
 
+# 启动时打印，便于排查
+if USE_WEAVIATE_DIRECT:
+    print("[ClareVoice] Weaviate 直连: 已配置, collection=", WEAVIATE_COLLECTION)
+else:
+    print("[ClareVoice] Weaviate 直连: 未配置 (在 Secrets 添加 WEAVIATE_URL, WEAVIATE_API_KEY)")
+
 # ---------- 默认 GenAI 课程大纲 ----------
 DEFAULT_COURSE_TOPICS: List[str] = [
     "Week 0 – Welcome & What is Generative AI; course outcomes LO1–LO5.",