[fix] system prompt fit error

api/app/controllers/chunk_controller.py

@@ -1,8 +1,10 @@
 import os
+import csv
+import io
 from typing import Any, Optional
 import uuid
 
-from fastapi import APIRouter, Depends, HTTPException, status, Query
+from fastapi import APIRouter, Depends, HTTPException, status, Query, UploadFile, File
 from fastapi.encoders import jsonable_encoder
 from sqlalchemy.orm import Session
 
@@ -23,6 +25,7 @@ from app.models.user_model import User
 from app.schemas import chunk_schema
 from app.schemas.response_schema import ApiResponse
 from app.services import knowledge_service, document_service, file_service, knowledgeshare_service
+from app.services.file_storage_service import FileStorageService, get_file_storage_service, generate_kb_file_key
 from app.services.model_service import ModelApiKeyService
 
 # Obtain a dedicated API logger
@@ -271,6 +274,9 @@ async def create_chunk(
         "sort_id": sort_id,
         "status": 1,
     }
+    # QA chunk: 注入 chunk_type/question/answer 到 metadata
+    if create_data.is_qa:
+        metadata.update(create_data.qa_metadata)
     chunk = DocumentChunk(page_content=content, metadata=metadata)
     # 3. Segmented vector storage
     vector_service.add_chunks([chunk])
@@ -282,6 +288,187 @@ async def create_chunk(
     return success(data=jsonable_encoder(chunk), msg="Document chunk creation successful")
 
 
+@router.post("/{kb_id}/{document_id}/chunk/batch", response_model=ApiResponse)
+async def create_chunks_batch(
+        kb_id: uuid.UUID,
+        document_id: uuid.UUID,
+        batch_data: chunk_schema.ChunkBatchCreate,
+        db: Session = Depends(get_db),
+        current_user: User = Depends(get_current_user)
+):
+    """
+    Batch create chunks (max 8)
+    """
+    api_logger.info(f"Batch create chunks: kb_id={kb_id}, document_id={document_id}, count={len(batch_data.items)}, username: {current_user.username}")
+
+    if len(batch_data.items) > settings.MAX_CHUNK_BATCH_SIZE:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=f"Batch size exceeds limit: max {settings.MAX_CHUNK_BATCH_SIZE}, got {len(batch_data.items)}"
+        )
+
+    db_knowledge = knowledge_service.get_knowledge_by_id(db, knowledge_id=kb_id, current_user=current_user)
+    if not db_knowledge:
+        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="The knowledge base does not exist or access is denied")
+
+    db_document = db.query(Document).filter(Document.id == document_id).first()
+    if not db_document:
+        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="The document does not exist or you do not have permission to access it")
+
+    vector_service = ElasticSearchVectorFactory().init_vector(knowledge=db_knowledge)
+
+    # Get current max sort_id
+    sort_id = 0
+    total, items = vector_service.search_by_segment(document_id=str(document_id), pagesize=1, page=1, asc=False)
+    if items:
+        sort_id = items[0].metadata["sort_id"]
+
+    chunks = []
+    for create_data in batch_data.items:
+        sort_id += 1
+        doc_id = uuid.uuid4().hex
+        metadata = {
+            "doc_id": doc_id,
+            "file_id": str(db_document.file_id),
+            "file_name": db_document.file_name,
+            "file_created_at": int(db_document.created_at.timestamp() * 1000),
+            "document_id": str(document_id),
+            "knowledge_id": str(kb_id),
+            "sort_id": sort_id,
+            "status": 1,
+        }
+        if create_data.is_qa:
+            metadata.update(create_data.qa_metadata)
+        chunks.append(DocumentChunk(page_content=create_data.chunk_content, metadata=metadata))
+
+    vector_service.add_chunks(chunks)
+
+    db_document.chunk_num += len(chunks)
+    db.commit()
+
+    return success(data=jsonable_encoder(chunks), msg=f"Batch created {len(chunks)} chunks successfully")
+
+
+@router.post("/{kb_id}/import_qa", response_model=ApiResponse)
+async def import_qa_new_doc(
+        kb_id: uuid.UUID,
+        file: UploadFile = File(..., description="CSV 或 Excel 文件（第一行标题跳过，第一列问题，第二列答案）"),
+        db: Session = Depends(get_db),
+        current_user: User = Depends(get_current_user),
+        storage_service: FileStorageService = Depends(get_file_storage_service),
+):
+    """
+    导入 QA 问答对并新建文档（CSV/Excel），异步处理
+    """
+    from app.schemas import file_schema, document_schema
+
+    api_logger.info(f"Import QA (new doc): kb_id={kb_id}, file={file.filename}, username: {current_user.username}")
+
+    # 1. 校验文件格式
+    filename = file.filename or ""
+    if not (filename.endswith(".csv") or filename.endswith(".xlsx") or filename.endswith(".xls")):
+        raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="仅支持 CSV (.csv) 或 Excel (.xlsx) 格式")
+
+    # 2. 校验知识库
+    db_knowledge = knowledge_service.get_knowledge_by_id(db, knowledge_id=kb_id, current_user=current_user)
+    if not db_knowledge:
+        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="知识库不存在或无权访问")
+
+    # 3. 读取文件
+    contents = await file.read()
+    file_size = len(contents)
+    if file_size == 0:
+        raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="文件为空")
+
+    _, file_extension = os.path.splitext(filename)
+    file_ext = file_extension.lower()
+
+    # 4. 创建 File 记录
+    file_data = file_schema.FileCreate(
+        kb_id=kb_id, created_by=current_user.id,
+        parent_id=uuid.UUID("00000000-0000-0000-0000-000000000000"),
+        file_name=filename, file_ext=file_ext, file_size=file_size,
+    )
+    db_file = file_service.create_file(db=db, file=file_data, current_user=current_user)
+
+    # 5. 上传文件到存储后端
+    file_key = generate_kb_file_key(kb_id=kb_id, file_id=db_file.id, file_ext=file_ext)
+    try:
+        await storage_service.storage.upload(file_key=file_key, content=contents, content_type=file.content_type)
+    except Exception as e:
+        api_logger.error(f"Storage upload failed: {e}")
+        raise HTTPException(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, detail=f"文件存储失败: {str(e)}")
+
+    db_file.file_key = file_key
+    db.commit()
+    db.refresh(db_file)
+
+    # 6. 创建 Document 记录（标记为 QA 类型）
+    doc_data = document_schema.DocumentCreate(
+        kb_id=kb_id, created_by=current_user.id, file_id=db_file.id,
+        file_name=filename, file_ext=file_ext, file_size=file_size,
+        file_meta={}, parser_id="qa",
+        parser_config={"doc_type": "qa", "auto_questions": 0}
+    )
+    db_document = document_service.create_document(db=db, document=doc_data, current_user=current_user)
+
+    api_logger.info(f"Created doc for QA import: file_id={db_file.id}, document_id={db_document.id}, file_key={file_key}")
+
+    # 7. 派发异步任务
+    from app.celery_app import celery_app
+    task = celery_app.send_task(
+        "app.core.rag.tasks.import_qa_chunks",
+        args=[str(kb_id), str(db_document.id), filename, contents],
+        queue="qa_import"
+    )
+
+    return success(data={
+        "task_id": task.id,
+        "document_id": str(db_document.id),
+        "file_id": str(db_file.id),
+    }, msg="QA 导入任务已提交，后台处理中")
+
+
+@router.post("/{kb_id}/{document_id}/import_qa", response_model=ApiResponse)
+async def import_qa_chunks(
+        kb_id: uuid.UUID,
+        document_id: uuid.UUID,
+        file: UploadFile = File(..., description="CSV 或 Excel 文件（第一行标题跳过，第一列问题，第二列答案）"),
+        db: Session = Depends(get_db),
+        current_user: User = Depends(get_current_user)
+):
+    """
+    导入 QA 问答对（CSV/Excel），异步处理
+    """
+    api_logger.info(f"Import QA chunks: kb_id={kb_id}, document_id={document_id}, file={file.filename}, username: {current_user.username}")
+
+    # 1. 校验文件格式
+    filename = file.filename or ""
+    if not (filename.endswith(".csv") or filename.endswith(".xlsx") or filename.endswith(".xls")):
+        raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="仅支持 CSV (.csv) 或 Excel (.xlsx) 格式")
+
+    # 2. 校验知识库和文档
+    db_knowledge = knowledge_service.get_knowledge_by_id(db, knowledge_id=kb_id, current_user=current_user)
+    if not db_knowledge:
+        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="知识库不存在或无权访问")
+
+    db_document = db.query(Document).filter(Document.id == document_id).first()
+    if not db_document:
+        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="文档不存在或无权访问")
+
+    # 3. 读取文件内容，派发异步任务
+    contents = await file.read()
+
+    from app.celery_app import celery_app
+    task = celery_app.send_task(
+        "app.core.rag.tasks.import_qa_chunks",
+        args=[str(kb_id), str(document_id), filename, contents],
+        queue="qa_import"
+    )
+
+    return success(data={"task_id": task.id}, msg="QA 导入任务已提交，后台处理中")
+
+
 @router.get("/{kb_id}/{document_id}/{doc_id}", response_model=ApiResponse)
 async def get_chunk(
         kb_id: uuid.UUID,
@@ -342,6 +529,9 @@ async def update_chunk(
     if total:
         chunk = items[0]
         chunk.page_content = content
+        # QA chunk: 更新 metadata 中的 question/answer
+        if update_data.is_qa:
+            chunk.metadata.update(update_data.qa_metadata)
         vector_service.update_by_segment(chunk)
         return success(data=jsonable_encoder(chunk), msg="The document chunk has been successfully updated")
     else:
@@ -356,6 +546,7 @@ async def delete_chunk(
         kb_id: uuid.UUID,
         document_id: uuid.UUID,
         doc_id: str,
+        force_refresh: bool = Query(False, description="Force Elasticsearch refresh after deletion"),
         db: Session = Depends(get_db),
         current_user: User = Depends(get_current_user)
 ):
@@ -373,7 +564,7 @@ async def delete_chunk(
 
     vector_service = ElasticSearchVectorFactory().init_vector(knowledge=db_knowledge)
     if vector_service.text_exists(doc_id):
-        vector_service.delete_by_ids([doc_id])
+        vector_service.delete_by_ids([doc_id], refresh=force_refresh)
         # 更新 chunk_num
         db_document = db.query(Document).filter(Document.id == document_id).first()
         db_document.chunk_num -= 1

api/app/controllers/service/rag_api_chunk_controller.py

@@ -113,6 +113,33 @@ async def create_chunk(
                                                current_user=current_user)
 
 
+@router.post("/{kb_id}/{document_id}/chunk/batch", response_model=ApiResponse)
+@require_api_key(scopes=["rag"])
+async def create_chunks_batch(
+    kb_id: uuid.UUID,
+    document_id: uuid.UUID,
+    request: Request,
+    api_key_auth: ApiKeyAuth = None,
+    db: Session = Depends(get_db),
+    items: list = Body(..., description="chunk items list"),
+):
+    """
+    Batch create chunks (max 8)
+    """
+    body = await request.json()
+    batch_data = chunk_schema.ChunkBatchCreate(**body)
+    # 0. Obtain the creator of the api key
+    api_key = api_key_service.ApiKeyService.get_api_key(db, api_key_auth.api_key_id, api_key_auth.workspace_id)
+    current_user = api_key.creator
+    current_user.current_workspace_id = api_key_auth.workspace_id
+
+    return await chunk_controller.create_chunks_batch(kb_id=kb_id,
+                                                      document_id=document_id,
+                                                      batch_data=batch_data,
+                                                      db=db,
+                                                      current_user=current_user)
+
+
 @router.get("/{kb_id}/{document_id}/{doc_id}", response_model=ApiResponse)
 @require_api_key(scopes=["rag"])
 async def get_chunk(
@@ -176,6 +203,7 @@ async def delete_chunk(
     request: Request,
     api_key_auth: ApiKeyAuth = None,
     db: Session = Depends(get_db),
+    force_refresh: bool = Query(False, description="Force Elasticsearch refresh after deletion"),
 ):
     """
     delete document chunk
@@ -188,6 +216,7 @@ async def delete_chunk(
     return await chunk_controller.delete_chunk(kb_id=kb_id,
                                                document_id=document_id,
                                                doc_id=doc_id,
+                                               force_refresh=force_refresh,
                                                db=db,
                                                current_user=current_user)
 

api/app/core/config.py

@@ -98,6 +98,7 @@ class Settings:
     # File Upload
     MAX_FILE_SIZE: int = int(os.getenv("MAX_FILE_SIZE", "52428800"))
     MAX_FILE_COUNT: int = int(os.getenv("MAX_FILE_COUNT", "20"))
+    MAX_CHUNK_BATCH_SIZE: int = int(os.getenv("MAX_CHUNK_BATCH_SIZE", "8"))
     FILE_PATH: str = os.getenv("FILE_PATH", "/files")
     FILE_URL_EXPIRES: int = int(os.getenv("FILE_URL_EXPIRES", "3600"))
 

api/app/core/rag/graphrag/general/index.py

@@ -46,7 +46,10 @@ async def run_graphrag(
     start = trio.current_time()
     workspace_id, kb_id, document_id = row["workspace_id"], str(row["kb_id"]), row["document_id"]
     chunks = []
-    for d in settings.retriever.chunk_list(document_id, workspace_id, [kb_id], fields=["page_content", "document_id"], sort_by_position=True):
+    for d in settings.retriever.chunk_list(document_id, workspace_id, [kb_id], fields=["page_content", "document_id", "chunk_type"], sort_by_position=True):
+        # 跳过 QA chunks，只用原文 chunks 构建图谱
+        if d.get("chunk_type") == "qa":
+            continue
         chunks.append(d["page_content"])
 
     with trio.fail_after(max(120, len(chunks) * 60 * 10) if enable_timeout_assertion else 10000000000):
@@ -150,6 +153,9 @@ async def run_graphrag_for_kb(
 
         total, items = vector_service.search_by_segment(document_id=str(document_id), query=None, pagesize=9999, page=1, asc=True)
         for doc in items:
+            # 跳过 QA chunks，只用原文 chunks 构建图谱
+            if (doc.metadata or {}).get("chunk_type") == "qa":
+                continue
             content = doc.page_content
             if num_tokens_from_string(current_chunk + content) < 1024:
                 current_chunk += content

api/app/core/rag/prompts/generator.py

@@ -131,18 +131,52 @@ def keyword_extraction(chat_mdl, content, topn=3):
 
 
 def question_proposal(chat_mdl, content, topn=3):
-    template = PROMPT_JINJA_ENV.from_string(QUESTION_PROMPT_TEMPLATE)
-    rendered_prompt = template.render(content=content, topn=topn)
-
-    msg = [{"role": "system", "content": rendered_prompt}, {"role": "user", "content": "Output: "}]
-    _, msg = message_fit_in(msg, getattr(chat_mdl, 'max_length', 8096))
-    kwd = chat_mdl.chat(rendered_prompt, msg[1:], {"temperature": 0.2})
-    if isinstance(kwd, tuple):
-        kwd = kwd[0]
-    kwd = re.sub(r"^.*</think>", "", kwd, flags=re.DOTALL)
-    if kwd.find("**ERROR**") >= 0:
+    """生成问题（向后兼容，返回纯文本问题列表）"""
+    pairs = qa_proposal(chat_mdl, content, topn)
+    if not pairs:
         return ""
-    return kwd
+    return "\n".join([p["question"] for p in pairs])
+
+
+def qa_proposal(chat_mdl, content, topn=3, custom_prompt=None):
+    """生成 QA 对，返回 [{"question": ..., "answer": ...}, ...]
+    
+    Args:
+        chat_mdl: LLM 模型
+        content: 文本内容
+        topn: 生成 QA 对数量
+        custom_prompt: 自定义 prompt 模板（支持 Jinja2，可用变量: content, topn）
+    """
+    if custom_prompt:
+        template = PROMPT_JINJA_ENV.from_string(custom_prompt)
+        sys_prompt = template.render(topn=topn)
+    else:
+        sys_prompt = QUESTION_PROMPT_TEMPLATE
+    msg = [{"role": "system", "content": sys_prompt}, {"role": "user", "content": content}]
+    _, msg = message_fit_in(msg, getattr(chat_mdl, 'max_length', 8096))
+    raw = chat_mdl.chat(sys_prompt, msg[1:], {"temperature": 0.2})
+    if isinstance(raw, tuple):
+        raw = raw[0]
+    raw = re.sub(r"^.*</think>", "", raw, flags=re.DOTALL)
+    if raw.find("**ERROR**") >= 0:
+        return []
+    return parse_qa_pairs(raw)
+
+
+def parse_qa_pairs(text: str) -> list:
+    """解析 LLM 返回的 QA 对文本，格式: Q: xxx A: xxx"""
+    pairs = []
+    for line in text.strip().split("\n"):
+        line = line.strip()
+        if not line:
+            continue
+        # 匹配 Q: ... A: ... 格式
+        match = re.match(r'^Q:\s*(.+?)\s+A:\s*(.+)$', line, re.IGNORECASE)
+        if match:
+            q, a = match.group(1).strip(), match.group(2).strip()
+            if q and a:
+                pairs.append({"question": q, "answer": a})
+    return pairs
 
 
 def graph_entity_types(chat_mdl, scenario):

api/app/core/rag/prompts/question_prompt.md

@@ -1,19 +1,20 @@
 ## Role
-You are a text analyzer.
+You are a text analyzer and knowledge extraction expert.
 
 ## Task
-Propose {{ topn }} questions about a given piece of text content.
+Generate question-answer pairs from the given text content.
 
 ## Requirements
-- Understand and summarize the text content, and propose the top {{ topn }} important questions.
+- Understand and summarize the text content, then generate up to {{ topn }} important question-answer pairs.
+- Each question-answer pair MUST be on a single line, formatted as: Q: <question> A: <answer>
 - The questions SHOULD NOT have overlapping meanings.
 - The questions SHOULD cover the main content of the text as much as possible.
-- The questions MUST be in the same language as the given piece of text content.
-- One question per line.
-- Output questions ONLY.
-
----
-
-## Text Content
-{{ content }}
+- The answers MUST be concise, accurate, and directly derived from the text content.
+- The answers SHOULD be self-contained and understandable without additional context.
+- Both questions and answers MUST be in the same language as the given text content.
+- If the text is too short or lacks substantive content, generate fewer pairs rather than padding.
+- Output question-answer pairs ONLY, no extra explanation or commentary.
 
+## Example Output
+Q: What is the capital of France? A: The capital of France is Paris.
+Q: When was the Eiffel Tower built? A: The Eiffel Tower was built in 1889.

api/app/core/rag/vdb/elasticsearch/elasticsearch_vector.py

@@ -5,7 +5,7 @@ from typing import Any
 from urllib.parse import urlparse
 
 import requests
-from elasticsearch import Elasticsearch, helpers
+from elasticsearch import Elasticsearch, helpers, NotFoundError
 from elasticsearch.helpers import BulkIndexError
 from packaging.version import parse as parse_version
 # langchain-community
@@ -53,13 +53,30 @@ class ElasticSearchVector(BaseVector):
         return "elasticsearch"
 
     def add_chunks(self, chunks: list[DocumentChunk], **kwargs):
-        # 实现 Elasticsearch 保存向量
-        texts = [chunk.page_content for chunk in chunks]
+        # QA chunks: embedding 只对 question 字段做；source chunks: 不做 embedding
+        texts_for_embedding = []
+        for chunk in chunks:
+            chunk_type = (chunk.metadata or {}).get("chunk_type", "chunk")
+            if chunk_type == "source":
+                # source chunk 不需要向量索引
+                texts_for_embedding.append("")
+            elif chunk_type == "qa":
+                # QA chunk: 用 question 字段做 embedding
+                texts_for_embedding.append((chunk.metadata or {}).get("question", chunk.page_content))
+            else:
+                # 普通 chunk: 用 page_content 做 embedding
+                texts_for_embedding.append(chunk.page_content)
+
         if self.is_multimodal_embedding:
-            # 火山引擎多模态 Embedding
-            embeddings = self.embeddings.embed_batch(texts)
+            embeddings = self.embeddings.embed_batch(texts_for_embedding)
         else:
-            embeddings = self.embeddings.embed_documents(list(texts))
+            embeddings = self.embeddings.embed_documents(texts_for_embedding)
+
+        # source chunk 的向量置空
+        for i, chunk in enumerate(chunks):
+            if (chunk.metadata or {}).get("chunk_type") == "source":
+                embeddings[i] = None
+
         self.create(chunks, embeddings, **kwargs)
 
     def create(self, chunks: list[DocumentChunk], embeddings: list[list[float]], **kwargs):
@@ -72,13 +89,25 @@ class ElasticSearchVector(BaseVector):
         uuids = self._get_uuids(chunks)
         actions = []
         for i, chunk in enumerate(chunks):
+            source = {
+                Field.CONTENT_KEY.value: chunk.page_content,
+                Field.METADATA_KEY.value: chunk.metadata or {},
+                Field.VECTOR.value: embeddings[i] or None
+            }
+            # 写入 QA 相关字段
+            meta = chunk.metadata or {}
+            if meta.get("chunk_type"):
+                source[Field.CHUNK_TYPE.value] = meta["chunk_type"]
+            if meta.get("question"):
+                source[Field.QUESTION.value] = meta["question"]
+            if meta.get("answer"):
+                source[Field.ANSWER.value] = meta["answer"]
+            if meta.get("source_chunk_id"):
+                source[Field.SOURCE_CHUNK_ID.value] = meta["source_chunk_id"]
+
             action = {
                 "_index": self._collection_name,
-                "_source": {
-                    Field.CONTENT_KEY.value: chunk.page_content,
-                    Field.METADATA_KEY.value: chunk.metadata or {},
-                    Field.VECTOR.value: embeddings[i] or None
-                }
+                "_source": source
             }
             actions.append(action)
         # using bulk mode
@@ -113,7 +142,7 @@ class ElasticSearchVector(BaseVector):
 
         return True
 
-    def delete_by_ids(self, ids: list[str]):
+    def delete_by_ids(self, ids: list[str], *, refresh: bool = False):
         if not ids:
             return
         if not self._client.indices.exists(index=self._collection_name):
@@ -134,6 +163,8 @@ class ElasticSearchVector(BaseVector):
             actions = [{"_op_type": "delete", "_index": self._collection_name, "_id": es_id} for es_id in actual_ids]
             try:
                 helpers.bulk(self._client, actions)
+                if refresh:
+                    self._client.indices.refresh(index=self._collection_name)
             except BulkIndexError as e:
                 for error in e.errors:
                     delete_error = error.get('delete', {})
@@ -153,7 +184,7 @@ class ElasticSearchVector(BaseVector):
         else:
             return None
 
-    def delete_by_metadata_field(self, key: str, value: str):
+    def delete_by_metadata_field(self, key: str, value: str, *, refresh: bool = False):
         if not self._client.indices.exists(index=self._collection_name):
             return False
         actual_ids = self.get_ids_by_metadata_field(key, value)
@@ -162,6 +193,8 @@ class ElasticSearchVector(BaseVector):
             actions = [{"_op_type": "delete", "_index": self._collection_name, "_id": es_id} for es_id in actual_ids]
             try:
                 helpers.bulk(self._client, actions)
+                if refresh:
+                    self._client.indices.refresh(index=self._collection_name)
             except BulkIndexError as e:
                 for error in e.errors:
                     delete_error = error.get('delete', {})
@@ -192,6 +225,8 @@ class ElasticSearchVector(BaseVector):
             List of DocumentChunk objects that match the query.
         """
         indices = kwargs.get("indices", self._collection_name)  # Default single index, multiple indexes are also supported, such as "index1, index2, index3"
+        if not self._client.indices.exists(index=indices):
+            return 0, []
 
         # Calculate the start position for the current page
         from_ = pagesize * (page-1)
@@ -226,12 +261,15 @@ class ElasticSearchVector(BaseVector):
             })
 
         # For simplicity, we use from/size here which has a limit (usually up to 10,000).
-        result = self._client.search(
-            index=indices,
-            from_=from_,  # Only use from_ for the first page (simplified)
-            size=pagesize,
-            body=query_str,
-        )
+        try:
+            result = self._client.search(
+                index=indices,
+                from_=from_,  # Only use from_ for the first page (simplified)
+                size=pagesize,
+                body=query_str,
+            )
+        except NotFoundError:
+            return 0, []
 
         if "errors" in result:
             raise ValueError(f"Error during query: {result['errors']}")
@@ -241,10 +279,19 @@ class ElasticSearchVector(BaseVector):
         for res in result["hits"]["hits"]:
             source = res["_source"]
             page_content = source.get(Field.CONTENT_KEY.value)
-            # vector = source.get(Field.VECTOR.value)
             vector = None
             metadata = source.get(Field.METADATA_KEY.value, {})
+            chunk_type = source.get(Field.CHUNK_TYPE.value)
             score = res["_score"]
+
+            # 将 QA 字段注入 metadata 供前端展示
+            if chunk_type:
+                metadata["chunk_type"] = chunk_type
+            if chunk_type == "qa":
+                metadata["question"] = source.get(Field.QUESTION.value, "")
+                metadata["answer"] = source.get(Field.ANSWER.value, "")
+                page_content = f"Q: {metadata['question']}\nA: {metadata['answer']}"
+
             docs_and_scores.append((DocumentChunk(page_content=page_content, vector=vector, metadata=metadata), score))
 
         docs = []
@@ -267,13 +314,18 @@ class ElasticSearchVector(BaseVector):
             List of DocumentChunk objects that match the query.
         """
         indices = kwargs.get("indices", self._collection_name)  # Default single index, multi-index available，etc "index1,index2,index3"
+        if not self._client.indices.exists(index=indices):
+            return 0, []
         query_str = {"query": {"term": {f"{Field.DOC_ID.value}": doc_id}}}
-        result = self._client.search(
-            index=indices,
-            from_=0,  # Only use from_ for the first page (simplified)
-            size=1,
-            body=query_str,
-        )
+        try:
+            result = self._client.search(
+                index=indices,
+                from_=0,  # Only use from_ for the first page (simplified)
+                size=1,
+                body=query_str,
+            )
+        except NotFoundError:
+            return 0, []
         # print(result)
         if "errors" in result:
             raise ValueError(f"Error during query: {result['errors']}")
@@ -308,27 +360,43 @@ class ElasticSearchVector(BaseVector):
         Returns:
             updated count.
         """
-        indices = kwargs.get("indices", self._collection_name)  # Default single index, multi-index available，etc "index1,index2,index3"
-        if self.is_multimodal_embedding:
-            # 火山引擎多模态 Embedding
-            chunk.vector = self.embeddings.embed_text(chunk.page_content)
+        indices = kwargs.get("indices", self._collection_name)
+        chunk_type = (chunk.metadata or {}).get("chunk_type")
+
+        # QA chunk: embedding 基于 question；source chunk: 不更新向量
+        if chunk_type == "source":
+            embed_text = ""
+        elif chunk_type == "qa":
+            embed_text = (chunk.metadata or {}).get("question", chunk.page_content)
         else:
-            chunk.vector = self.embeddings.embed_query(chunk.page_content)
+            embed_text = chunk.page_content
+
+        if chunk_type != "source":
+            if self.is_multimodal_embedding:
+                chunk.vector = self.embeddings.embed_text(embed_text)
+            else:
+                chunk.vector = self.embeddings.embed_query(embed_text)
+
+        script_source = "ctx._source.page_content = params.new_content; ctx._source.vector = params.new_vector;"
+        params = {
+            "new_content": chunk.page_content,
+            "new_vector": chunk.vector if chunk_type != "source" else None
+        }
+
+        # QA chunk: 同时更新 question/answer 字段
+        if chunk_type == "qa":
+            script_source += " ctx._source.question = params.new_question; ctx._source.answer = params.new_answer;"
+            params["new_question"] = (chunk.metadata or {}).get("question", "")
+            params["new_answer"] = (chunk.metadata or {}).get("answer", "")
 
         body = {
             "script": {
-                "source": """
-                        ctx._source.page_content = params.new_content;
-                        ctx._source.vector = params.new_vector;
-                    """,
-                "params": {
-                    "new_content": chunk.page_content,
-                    "new_vector": chunk.vector
-                }
+                "source": script_source,
+                "params": params
             },
             "query": {
                 "term": {
-                    Field.DOC_ID.value: chunk.metadata["doc_id"]  # exact match doc_id
+                    Field.DOC_ID.value: chunk.metadata["doc_id"]
                 }
             }
         }
@@ -336,9 +404,6 @@ class ElasticSearchVector(BaseVector):
             index=indices,
             body=body,
         )
-        # Remove debug printing and use logging instead
-        # print(result)
-        # print(f"Update successful, number of affected documents: {result['updated']}")
         return result['updated']
 
     def change_status_by_document_id(self, document_id: str, status: int, **kwargs) -> str:
@@ -397,11 +462,11 @@ class ElasticSearchVector(BaseVector):
                             }
                         }
                     },
-                    "filter": {  # Add the filter condition of status=1
-                        "term": {
-                            "metadata.status": 1
-                        }
-                    }
+                    "filter": [
+                        {"term": {"metadata.status": 1}},
+                        # 排除 source chunk（仅供 GraphRAG 使用，不参与检索）
+                        {"bool": {"must_not": {"term": {Field.CHUNK_TYPE.value: "source"}}}}
+                    ]
                 }
             }
         # If file_names_filter is passed in, merge the filtering conditions
@@ -415,22 +480,14 @@ class ElasticSearchVector(BaseVector):
                             },
                             "script": {
                                 "source": f"cosineSimilarity(params.query_vector, '{Field.VECTOR.value}') + 1.0",
-                                # The script_score query calculates the cosine similarity between the embedding field of each document and the query vector. The addition of +1.0 is to ensure that the scores returned by the script are non-negative, as the range of cosine similarity is [-1, 1]
                                 "params": {"query_vector": query_vector}
                             }
                         }
                     },
                     "filter": [
-                        {
-                            "term": {
-                                "metadata.status": 1
-                            }
-                        },
-                        {
-                            "terms": {
-                                "metadata.file_name": file_names_filter  # Additional file_name filtering
-                            }
-                        }
+                        {"term": {"metadata.status": 1}},
+                        {"terms": {"metadata.file_name": file_names_filter}},
+                        {"bool": {"must_not": {"term": {Field.CHUNK_TYPE.value: "source"}}}}
                     ],
                 }
             }
@@ -451,8 +508,19 @@ class ElasticSearchVector(BaseVector):
             source = res["_source"]
             page_content = source.get(Field.CONTENT_KEY.value)
             metadata = source.get(Field.METADATA_KEY.value, {})
+            chunk_type = source.get(Field.CHUNK_TYPE.value)
             score = res["_score"]
             score = score / 2  # Normalized [0-1]
+
+            # QA chunk: 返回 Q+A 拼接作为上下文
+            if chunk_type == "qa":
+                question = source.get(Field.QUESTION.value, "")
+                answer = source.get(Field.ANSWER.value, "")
+                page_content = f"Q: {question}\nA: {answer}"
+                metadata["chunk_type"] = "qa"
+                metadata["question"] = question
+                metadata["answer"] = answer
+
             docs_and_scores.append((DocumentChunk(page_content=page_content, metadata=metadata), score))
 
         docs = []
@@ -491,11 +559,10 @@ class ElasticSearchVector(BaseVector):
                         }
                     }
                 },
-                "filter": {  # Add the filter condition of status=1
-                    "term": {
-                        "metadata.status": 1
-                    }
-                }
+                "filter": [
+                    {"term": {"metadata.status": 1}},
+                    {"bool": {"must_not": {"term": {Field.CHUNK_TYPE.value: "source"}}}}
+                ]
             }
         }
 
@@ -512,16 +579,9 @@ class ElasticSearchVector(BaseVector):
                         }
                     },
                     "filter": [
-                        {
-                            "term": {
-                                "metadata.status": 1
-                            }
-                        },
-                        {
-                            "terms": {
-                                "metadata.file_name": file_names_filter  # Additional file_name filtering
-                            }
-                        }
+                        {"term": {"metadata.status": 1}},
+                        {"terms": {"metadata.file_name": file_names_filter}},
+                        {"bool": {"must_not": {"term": {Field.CHUNK_TYPE.value: "source"}}}}
                     ],
                 }
             }
@@ -543,6 +603,17 @@ class ElasticSearchVector(BaseVector):
             source = res["_source"]
             page_content = source.get(Field.CONTENT_KEY.value)
             metadata = source.get(Field.METADATA_KEY.value, {})
+            chunk_type = source.get(Field.CHUNK_TYPE.value)
+
+            # QA chunk: 返回 Q+A 拼接作为上下文
+            if chunk_type == "qa":
+                question = source.get(Field.QUESTION.value, "")
+                answer = source.get(Field.ANSWER.value, "")
+                page_content = f"Q: {question}\nA: {answer}"
+                metadata["chunk_type"] = "qa"
+                metadata["question"] = question
+                metadata["answer"] = answer
+
             # Normalize the score to the [0,1] interval
             normalized_score = res["_score"] / max_score
             docs_and_scores.append((DocumentChunk(page_content=page_content, metadata=metadata), normalized_score))
@@ -652,7 +723,7 @@ class ElasticSearchVector(BaseVector):
                         },
                         Field.VECTOR.value: {
                             "type": "dense_vector",
-                            "dims": len(embeddings[0]),  # Make sure the dimension is correct here,The dimension size of the vector. When index is true, it cannot exceed 1024; when index is false or not specified, it cannot exceed 2048, which can improve retrieval efficiency
+                            "dims": len(next((e for e in embeddings if e is not None), [0]*768)),  # 跳过 None 获取向量维度，fallback 768
                             "index": True,
                             "similarity": "cosine"
                         }

api/app/core/rag/vdb/field.py

@@ -14,3 +14,8 @@ class Field(StrEnum):
     DOCUMENT_ID = "metadata.document_id"
     KNOWLEDGE_ID = "metadata.knowledge_id"
     SORT_ID = "metadata.sort_id"
+    # QA fields
+    CHUNK_TYPE = "chunk_type"  # "chunk" | "source" | "qa"
+    QUESTION = "question"
+    ANSWER = "answer"
+    SOURCE_CHUNK_ID = "source_chunk_id"

api/app/core/rag/vdb/vector_base.py

@@ -27,14 +27,14 @@ class BaseVector(ABC):
         raise NotImplementedError
 
     @abstractmethod
-    def delete_by_ids(self, ids: list[str]):
+    def delete_by_ids(self, ids: list[str], *, refresh: bool = False):
         raise NotImplementedError
 
     def get_ids_by_metadata_field(self, key: str, value: str):
         raise NotImplementedError
 
     @abstractmethod
-    def delete_by_metadata_field(self, key: str, value: str):
+    def delete_by_metadata_field(self, key: str, value: str, *, refresh: bool = False):
         raise NotImplementedError
 
     @abstractmethod

api/app/schemas/chunk_schema.py

@@ -20,13 +20,26 @@ class ChunkCreate(BaseModel):
 
     @property
     def chunk_content(self) -> str:
-        """
-        Get the actual content string regardless of input type
-        """
+        """Get the actual content string regardless of input type"""
         if isinstance(self.content, QAChunk):
-            return f"question: {self.content.question} answer: {self.content.answer}"
+            return self.content.question  # QA 模式下 page_content 存 question
         return self.content
 
+    @property
+    def is_qa(self) -> bool:
+        return isinstance(self.content, QAChunk)
+
+    @property
+    def qa_metadata(self) -> dict:
+        """返回 QA 相关的 metadata 字段"""
+        if isinstance(self.content, QAChunk):
+            return {
+                "chunk_type": "qa",
+                "question": self.content.question,
+                "answer": self.content.answer,
+            }
+        return {}
+
 
 class ChunkUpdate(BaseModel):
     content: Union[str, QAChunk] = Field(
@@ -35,13 +48,26 @@ class ChunkUpdate(BaseModel):
 
     @property
     def chunk_content(self) -> str:
-        """
-        Get the actual content string regardless of input type
-        """
+        """Get the actual content string regardless of input type"""
         if isinstance(self.content, QAChunk):
-            return f"question: {self.content.question} answer: {self.content.answer}"
+            return self.content.question  # QA 模式下 page_content 存 question
         return self.content
 
+    @property
+    def is_qa(self) -> bool:
+        return isinstance(self.content, QAChunk)
+
+    @property
+    def qa_metadata(self) -> dict:
+        """返回 QA 相关的 metadata 字段"""
+        if isinstance(self.content, QAChunk):
+            return {
+                "chunk_type": "qa",
+                "question": self.content.question,
+                "answer": self.content.answer,
+            }
+        return {}
+
 
 class ChunkRetrieve(BaseModel):
     query: str
@@ -51,3 +77,8 @@ class ChunkRetrieve(BaseModel):
     vector_similarity_weight: float | None = Field(None)
     top_k: int | None = Field(None)
     retrieve_type: RetrieveType | None = Field(None)
+
+
+class ChunkBatchCreate(BaseModel):
+    """批量创建 chunk"""
+    items: list[ChunkCreate] = Field(..., min_length=1, description="chunk 列表")

api/app/tasks.py

@@ -30,7 +30,7 @@ from app.core.rag.llm.cv_model import QWenCV
 from app.core.rag.llm.embedding_model import OpenAIEmbed
 from app.core.rag.llm.sequence2txt_model import QWenSeq2txt
 from app.core.rag.models.chunk import DocumentChunk
-from app.core.rag.prompts.generator import question_proposal
+from app.core.rag.prompts.generator import question_proposal, qa_proposal
 from app.core.rag.vdb.elasticsearch.elasticsearch_vector import (
     ElasticSearchVectorFactory,
 )
@@ -311,6 +311,7 @@ def parse_document(file_key: str, document_id: uuid.UUID, file_name: str = ""):
             vector_service.delete_by_metadata_field(key="document_id", value=str(document_id))
             # 2.2 Vectorize and import batch documents
             auto_questions_topn = db_document.parser_config.get("auto_questions", 0)
+            qa_prompt = db_document.parser_config.get("qa_prompt", None)
             chat_model = None
             if auto_questions_topn:
                 chat_model = Base(
@@ -318,62 +319,123 @@ def parse_document(file_key: str, document_id: uuid.UUID, file_name: str = ""):
                     model_name=db_knowledge.llm.api_keys[0].model_name,
                     base_url=db_knowledge.llm.api_keys[0].api_base,
                 )
+                logger.info(f"[QA] LLM model: {db_knowledge.llm.api_keys[0].model_name}, base_url: {db_knowledge.llm.api_keys[0].api_base}")
+                if qa_prompt:
+                    logger.info(f"[QA] Using custom prompt ({len(qa_prompt)} chars)")
 
             # 预先构建所有 batch 的 chunks，保证 sort_id 全局有序
             all_batch_chunks: list[list[DocumentChunk]] = []
 
             if auto_questions_topn:
-                # auto_questions 开启：先并发生成所有 chunk 的问题，再按 batch 分组
-                # 构建 (global_idx, item) 列表
+                # QA 模式（FastGPT 方案）：
+                # 1. 原 chunk 标记为 source（保留供 GraphRAG 使用，不参与检索）
+                # 2. LLM 生成 QA 对，每个 QA 对独立存储为 qa chunk
                 indexed_items = list(enumerate(res))
 
-                def _generate_question(idx_item: tuple[int, dict]) -> tuple[int, str]:
-                    """为单个 chunk 生成问题（带缓存），返回 (global_idx, question_text)"""
+                def _generate_qa(idx_item: tuple[int, dict]) -> tuple[int, list]:
+                    """为单个 chunk 生成 QA 对（带缓存），返回 (global_idx, qa_pairs)"""
                     global_idx, item = idx_item
                     content = item["content_with_weight"]
-                    cached = get_llm_cache(chat_model.model_name, content, "question",
-                                           {"topn": auto_questions_topn})
+                    cache_params = {"topn": auto_questions_topn}
+                    if qa_prompt:
+                        import hashlib
+                        cache_params["prompt_hash"] = hashlib.md5(qa_prompt.encode()).hexdigest()[:8]
+                    cached = get_llm_cache(chat_model.model_name, content, "qa", cache_params)
                     if not cached:
-                        cached = question_proposal(chat_model, content, auto_questions_topn)
-                        set_llm_cache(chat_model.model_name, content, cached, "question",
-                                      {"topn": auto_questions_topn})
-                    return global_idx, cached
+                        logger.info(f"[QA] Cache miss for chunk {global_idx}, calling LLM. cache_params={cache_params}")
+                        try:
+                            pairs = qa_proposal(chat_model, content, auto_questions_topn, custom_prompt=qa_prompt)
+                        except Exception as e:
+                            logger.error(f"[QA] LLM call failed: model={chat_model.model_name}, base_url={getattr(chat_model, 'base_url', 'N/A')}, error={e}")
+                            return global_idx, []
+                        logger.info(f"[QA] Chunk {global_idx} generated {len(pairs)} QA pairs")
+                        # 缓存存 JSON 字符串
+                        set_llm_cache(chat_model.model_name, content, json.dumps(pairs, ensure_ascii=False), "qa",
+                                      cache_params)
+                        return global_idx, pairs
+                    logger.info(f"[QA] Cache hit for chunk {global_idx}, cache_params={cache_params}, cached_type={type(cached).__name__}")
+                    # 从缓存读取：可能是 JSON 字符串或旧格式纯文本
+                    if isinstance(cached, str):
+                        try:
+                            parsed = json.loads(cached)
+                            if isinstance(parsed, list):
+                                logger.info(f"[QA] Chunk {global_idx} loaded {len(parsed)} QA pairs from cache")
+                                return global_idx, parsed
+                        except (json.JSONDecodeError, TypeError):
+                            pass
+                        # 旧缓存格式（纯文本问题），尝试解析
+                        from app.core.rag.prompts.generator import parse_qa_pairs
+                        return global_idx, parse_qa_pairs(cached) if cached else []
+                    return global_idx, cached if isinstance(cached, list) else []
 
-                # 并发调用 LLM 生成问题
-                question_map: dict[int, str] = {}
+                # 并发调用 LLM 生成 QA 对
+                qa_map: dict[int, list] = {}
                 with ThreadPoolExecutor(max_workers=AUTO_QUESTIONS_MAX_WORKERS) as q_executor:
-                    futures = {q_executor.submit(_generate_question, item): item[0]
+                    futures = {q_executor.submit(_generate_qa, item): item[0]
                                for item in indexed_items}
                     for future in futures:
-                        global_idx, cached = future.result()
-                        question_map[global_idx] = cached
+                        global_idx, pairs = future.result()
+                        qa_map[global_idx] = pairs
 
                 progress_lines.append(
-                    f"{datetime.now().strftime('%H:%M:%S')} Auto questions generated for {total_chunks} chunks "
+                    f"{datetime.now().strftime('%H:%M:%S')} QA pairs generated for {total_chunks} chunks "
                     f"(workers={AUTO_QUESTIONS_MAX_WORKERS}).")
 
-                # 按 batch 分组组装 DocumentChunk
-                for batch_start in range(0, total_chunks, EMBEDDING_BATCH_SIZE):
-                    batch_end = min(batch_start + EMBEDDING_BATCH_SIZE, total_chunks)
-                    chunks = []
-                    for global_idx in range(batch_start, batch_end):
-                        item = res[global_idx]
-                        metadata = {
+                # 组装 chunks：source chunks + qa chunks
+                source_chunks = []
+                qa_chunks = []
+                qa_sort_id = 0
+
+                for global_idx in range(total_chunks):
+                    item = res[global_idx]
+                    source_chunk_id = uuid.uuid4().hex
+
+                    # source chunk：保留原文，供 GraphRAG 使用，不参与向量检索
+                    source_meta = {
+                        "doc_id": source_chunk_id,
+                        "file_id": str(db_document.file_id),
+                        "file_name": db_document.file_name,
+                        "file_created_at": int(db_document.created_at.timestamp() * 1000),
+                        "document_id": str(db_document.id),
+                        "knowledge_id": str(db_document.kb_id),
+                        "sort_id": global_idx,
+                        "status": 1,
+                        "chunk_type": "source",
+                    }
+                    source_chunks.append(
+                        DocumentChunk(page_content=item["content_with_weight"], metadata=source_meta))
+
+                    # qa chunks：每个 QA 对独立存储
+                    pairs = qa_map.get(global_idx, [])
+                    for pair in pairs:
+                        qa_meta = {
                             "doc_id": uuid.uuid4().hex,
                             "file_id": str(db_document.file_id),
                             "file_name": db_document.file_name,
                             "file_created_at": int(db_document.created_at.timestamp() * 1000),
                             "document_id": str(db_document.id),
                             "knowledge_id": str(db_document.kb_id),
-                            "sort_id": global_idx,
+                            "sort_id": qa_sort_id,
                             "status": 1,
+                            "chunk_type": "qa",
+                            "question": pair["question"],
+                            "answer": pair["answer"],
+                            "source_chunk_id": source_chunk_id,
                         }
-                        cached = question_map[global_idx]
-                        chunks.append(
-                            DocumentChunk(
-                                page_content=f"question: {cached} answer: {item['content_with_weight']}",
-                                metadata=metadata))
-                    all_batch_chunks.append(chunks)
+                        # page_content 存 question，用于向量索引
+                        qa_chunks.append(
+                            DocumentChunk(page_content=pair["question"], metadata=qa_meta))
+                        qa_sort_id += 1
+
+                # 按 batch 分组（source + qa 一起）
+                all_chunks = source_chunks + qa_chunks
+                for batch_start in range(0, len(all_chunks), EMBEDDING_BATCH_SIZE):
+                    batch_end = min(batch_start + EMBEDDING_BATCH_SIZE, len(all_chunks))
+                    all_batch_chunks.append(all_chunks[batch_start:batch_end])
+
+                progress_lines.append(
+                    f"{datetime.now().strftime('%H:%M:%S')} QA mode: {len(source_chunks)} source chunks + "
+                    f"{len(qa_chunks)} QA chunks prepared.")
             else:
                 # 无 auto_questions：直接构建 chunks
                 for batch_start in range(0, total_chunks, EMBEDDING_BATCH_SIZE):
@@ -635,6 +697,136 @@ def build_graphrag_for_document(document_id: str, knowledge_id: str):
             return f"build_graphrag_for_document '{document_id}' failed: {e}"
 
 
+@celery_app.task(name="app.core.rag.tasks.import_qa_chunks", queue="qa_import")
+def import_qa_chunks(kb_id: str, document_id: str, filename: str, contents: bytes):
+    """
+    异步导入 QA 问答对（CSV/Excel）
+    
+    文件格式：第一行标题（跳过），第一列问题，第二列答案
+    """
+    import csv as csv_module
+    import io
+
+    db = None
+    try:
+        from app.db import get_db_context
+        with get_db_context() as db:
+            db_document = db.query(Document).filter(Document.id == uuid.UUID(document_id)).first()
+            db_knowledge = db.query(Knowledge).filter(Knowledge.id == uuid.UUID(kb_id)).first()
+            if not db_document or not db_knowledge:
+                logger.error(f"[ImportQA] document={document_id} or knowledge={kb_id} not found")
+                return {"error": "document or knowledge not found", "imported": 0}
+
+            # 1. 解析文件
+            qa_pairs = []
+            failed_rows = []
+
+            if filename.endswith(".csv"):
+                try:
+                    text = contents.decode("utf-8-sig")
+                except UnicodeDecodeError:
+                    text = contents.decode("gbk", errors="ignore")
+
+                sniffer = csv_module.Sniffer()
+                try:
+                    dialect = sniffer.sniff(text[:2048])
+                    delimiter = dialect.delimiter
+                except csv_module.Error:
+                    delimiter = "," if "," in text[:500] else "\t"
+
+                reader = csv_module.reader(io.StringIO(text), delimiter=delimiter)
+                for i, row in enumerate(reader):
+                    if i == 0:
+                        continue
+                    if len(row) >= 2 and row[0].strip() and row[1].strip():
+                        qa_pairs.append({"question": row[0].strip(), "answer": row[1].strip()})
+                    elif len(row) >= 1 and row[0].strip():
+                        failed_rows.append(i + 1)
+
+            elif filename.endswith(".xlsx") or filename.endswith(".xls"):
+                try:
+                    import openpyxl
+                    wb = openpyxl.load_workbook(io.BytesIO(contents), read_only=True)
+                    for sheet in wb.worksheets:
+                        for i, row in enumerate(sheet.iter_rows(values_only=True)):
+                            if i == 0:
+                                continue
+                            if len(row) >= 2 and row[0] and row[1]:
+                                q = str(row[0]).strip()
+                                a = str(row[1]).strip()
+                                if q and a:
+                                    qa_pairs.append({"question": q, "answer": a})
+                            elif len(row) >= 1 and row[0]:
+                                failed_rows.append(i + 1)
+                    wb.close()
+                except Exception as e:
+                    logger.error(f"[ImportQA] Excel parse failed: {e}")
+                    return {"error": f"Excel parse failed: {e}", "imported": 0}
+
+            if not qa_pairs:
+                logger.warning(f"[ImportQA] No valid QA pairs found in {filename}")
+                return {"error": "No valid QA pairs found", "imported": 0}
+
+            logger.info(f"[ImportQA] Parsed {len(qa_pairs)} QA pairs from {filename}, failed_rows={failed_rows}")
+
+            # 2. 写入 ES
+            vector_service = ElasticSearchVectorFactory().init_vector(knowledge=db_knowledge)
+
+            sort_id = 0
+            total, items = vector_service.search_by_segment(document_id=document_id, pagesize=1, page=1, asc=False)
+            if items:
+                sort_id = items[0].metadata["sort_id"]
+
+            chunks = []
+            for pair in qa_pairs:
+                sort_id += 1
+                doc_id = uuid.uuid4().hex
+                metadata = {
+                    "doc_id": doc_id,
+                    "file_id": str(db_document.file_id),
+                    "file_name": db_document.file_name,
+                    "file_created_at": int(db_document.created_at.timestamp() * 1000),
+                    "document_id": document_id,
+                    "knowledge_id": kb_id,
+                    "sort_id": sort_id,
+                    "status": 1,
+                    "chunk_type": "qa",
+                    "question": pair["question"],
+                    "answer": pair["answer"],
+                }
+                chunks.append(DocumentChunk(page_content=pair["question"], metadata=metadata))
+
+            batch_size = 50
+            for i in range(0, len(chunks), batch_size):
+                batch = chunks[i:i + batch_size]
+                vector_service.add_chunks(batch)
+
+            # 3. 更新 chunk_num 和 progress
+            db_document.chunk_num += len(chunks)
+            db_document.progress = 1.0
+            db_document.progress_msg = f"QA 导入完成: {len(chunks)} 条"
+            db.commit()
+
+            result = {"imported": len(chunks), "failed_rows": failed_rows}
+            logger.info(f"[ImportQA] Done: imported={len(chunks)}, failed={len(failed_rows)}")
+            return result
+
+    except Exception as e:
+        logger.error(f"[ImportQA] Failed: {e}", exc_info=True)
+        # 尝试更新文档状态为失败
+        try:
+            from app.db import get_db_context
+            with get_db_context() as err_db:
+                doc = err_db.query(Document).filter(Document.id == uuid.UUID(document_id)).first()
+                if doc:
+                    doc.progress = -1.0
+                    doc.progress_msg = f"QA 导入失败: {str(e)[:200]}"
+                    err_db.commit()
+        except Exception:
+            pass
+        return {"error": str(e), "imported": 0}
+
+
 @celery_app.task(name="app.core.rag.tasks.sync_knowledge_for_kb")
 def sync_knowledge_for_kb(kb_id: uuid.UUID):
     """