feat: citation UI for web links & many UI/UX improvements

docs/chinese-llm-setup.md

@@ -24,7 +24,7 @@ SurfSense 现已支持以下国产 LLM：
 
 1. 登录 SurfSense Dashboard
 2. 进入 **Settings** → **API Keys** (或 **LLM Configurations**)
-3. 点击 **Add New Configuration**
+3. 点击 **Add LLM Model**
 4. 从 **Provider** 下拉菜单中选择你的国产 LLM 提供商
 5. 填写必填字段（见下方各提供商详细配置）
 6. 点击 **Save**

surfsense_backend/alembic/versions/111_add_prompts_table.py

@@ -42,7 +42,9 @@ def upgrade() -> None:
             )
         """)
         op.execute("CREATE INDEX ix_prompts_user_id ON prompts (user_id)")
-        op.execute("CREATE INDEX ix_prompts_search_space_id ON prompts (search_space_id)")
+        op.execute(
+            "CREATE INDEX ix_prompts_search_space_id ON prompts (search_space_id)"
+        )
 
 
 def downgrade() -> None:

surfsense_backend/app/agents/new_chat/tools/onedrive/create_file.py

@@ -81,7 +81,8 @@ async def create_onedrive_file(
                 select(SearchSourceConnector).filter(
                     SearchSourceConnector.search_space_id == search_space_id,
                     SearchSourceConnector.user_id == user_id,
-                    SearchSourceConnector.connector_type == SearchSourceConnectorType.ONEDRIVE_CONNECTOR,
+                    SearchSourceConnector.connector_type
+                    == SearchSourceConnectorType.ONEDRIVE_CONNECTOR,
                 )
             )
             connectors = result.scalars().all()
@@ -95,12 +96,14 @@ async def create_onedrive_file(
             accounts = []
             for c in connectors:
                 cfg = c.config or {}
-                accounts.append({
-                    "id": c.id,
-                    "name": c.name,
-                    "user_email": cfg.get("user_email"),
-                    "auth_expired": cfg.get("auth_expired", False),
-                })
+                accounts.append(
+                    {
+                        "id": c.id,
+                        "name": c.name,
+                        "user_email": cfg.get("user_email"),
+                        "auth_expired": cfg.get("auth_expired", False),
+                    }
+                )
 
             if all(a.get("auth_expired") for a in accounts):
                 return {
@@ -119,16 +122,22 @@ async def create_onedrive_file(
                     client = OneDriveClient(session=db_session, connector_id=cid)
                     items, err = await client.list_children("root")
                     if err:
-                        logger.warning("Failed to list folders for connector %s: %s", cid, err)
+                        logger.warning(
+                            "Failed to list folders for connector %s: %s", cid, err
+                        )
                         parent_folders[cid] = []
                     else:
                         parent_folders[cid] = [
                             {"folder_id": item["id"], "name": item["name"]}
                             for item in items
-                            if item.get("folder") is not None and item.get("id") and item.get("name")
+                            if item.get("folder") is not None
+                            and item.get("id")
+                            and item.get("name")
                         ]
                 except Exception:
-                    logger.warning("Error fetching folders for connector %s", cid, exc_info=True)
+                    logger.warning(
+                        "Error fetching folders for connector %s", cid, exc_info=True
+                    )
                     parent_folders[cid] = []
 
             context: dict[str, Any] = {
@@ -152,8 +161,12 @@ async def create_onedrive_file(
                 }
             )
 
-            decisions_raw = approval.get("decisions", []) if isinstance(approval, dict) else []
-            decisions = decisions_raw if isinstance(decisions_raw, list) else [decisions_raw]
+            decisions_raw = (
+                approval.get("decisions", []) if isinstance(approval, dict) else []
+            )
+            decisions = (
+                decisions_raw if isinstance(decisions_raw, list) else [decisions_raw]
+            )
             decisions = [d for d in decisions if isinstance(d, dict)]
             if not decisions:
                 return {"status": "error", "message": "No approval decision received"}
@@ -192,15 +205,19 @@ async def create_onedrive_file(
                         SearchSourceConnector.id == final_connector_id,
                         SearchSourceConnector.search_space_id == search_space_id,
                         SearchSourceConnector.user_id == user_id,
-                        SearchSourceConnector.connector_type == SearchSourceConnectorType.ONEDRIVE_CONNECTOR,
+                        SearchSourceConnector.connector_type
+                        == SearchSourceConnectorType.ONEDRIVE_CONNECTOR,
                     )
                 )
                 connector = result.scalars().first()
             else:
                 connector = connectors[0]
 
             if not connector:
-                return {"status": "error", "message": "Selected OneDrive connector is invalid."}
+                return {
+                    "status": "error",
+                    "message": "Selected OneDrive connector is invalid.",
+                }
 
             docx_bytes = _markdown_to_docx(final_content or "")
 
@@ -212,7 +229,9 @@ async def create_onedrive_file(
                 mime_type=DOCX_MIME,
             )
 
-            logger.info(f"OneDrive file created: id={created.get('id')}, name={created.get('name')}")
+            logger.info(
+                f"OneDrive file created: id={created.get('id')}, name={created.get('name')}"
+            )
 
             kb_message_suffix = ""
             try:

surfsense_backend/app/agents/new_chat/tools/onedrive/trash_file.py

@@ -52,10 +52,15 @@ async def delete_onedrive_file(
             - If status is "not_found", relay the exact message to the user and ask them
               to verify the file name or check if it has been indexed.
         """
-        logger.info(f"delete_onedrive_file called: file_name='{file_name}', delete_from_kb={delete_from_kb}")
+        logger.info(
+            f"delete_onedrive_file called: file_name='{file_name}', delete_from_kb={delete_from_kb}"
+        )
 
         if db_session is None or search_space_id is None or user_id is None:
-            return {"status": "error", "message": "OneDrive tool not properly configured."}
+            return {
+                "status": "error",
+                "message": "OneDrive tool not properly configured.",
+            }
 
         try:
             doc_result = await db_session.execute(
@@ -89,8 +94,12 @@ async def delete_onedrive_file(
                             Document.search_space_id == search_space_id,
                             Document.document_type == DocumentType.ONEDRIVE_FILE,
                             func.lower(
-                                cast(Document.document_metadata["onedrive_file_name"], String)
-                            ) == func.lower(file_name),
+                                cast(
+                                    Document.document_metadata["onedrive_file_name"],
+                                    String,
+                                )
+                            )
+                            == func.lower(file_name),
                             SearchSourceConnector.user_id == user_id,
                         )
                     )
@@ -110,28 +119,38 @@ async def delete_onedrive_file(
                 }
 
             if not document.connector_id:
-                return {"status": "error", "message": "Document has no associated connector."}
+                return {
+                    "status": "error",
+                    "message": "Document has no associated connector.",
+                }
 
             meta = document.document_metadata or {}
             file_id = meta.get("onedrive_file_id")
             document_id = document.id
 
             if not file_id:
-                return {"status": "error", "message": "File ID is missing. Please re-index the file."}
+                return {
+                    "status": "error",
+                    "message": "File ID is missing. Please re-index the file.",
+                }
 
             conn_result = await db_session.execute(
                 select(SearchSourceConnector).filter(
                     and_(
                         SearchSourceConnector.id == document.connector_id,
                         SearchSourceConnector.search_space_id == search_space_id,
                         SearchSourceConnector.user_id == user_id,
-                        SearchSourceConnector.connector_type == SearchSourceConnectorType.ONEDRIVE_CONNECTOR,
+                        SearchSourceConnector.connector_type
+                        == SearchSourceConnectorType.ONEDRIVE_CONNECTOR,
                     )
                 )
             )
             connector = conn_result.scalars().first()
             if not connector:
-                return {"status": "error", "message": "OneDrive connector not found or access denied."}
+                return {
+                    "status": "error",
+                    "message": "OneDrive connector not found or access denied.",
+                }
 
             cfg = connector.config or {}
             if cfg.get("auth_expired"):
@@ -170,8 +189,12 @@ async def delete_onedrive_file(
                 }
             )
 
-            decisions_raw = approval.get("decisions", []) if isinstance(approval, dict) else []
-            decisions = decisions_raw if isinstance(decisions_raw, list) else [decisions_raw]
+            decisions_raw = (
+                approval.get("decisions", []) if isinstance(approval, dict) else []
+            )
+            decisions = (
+                decisions_raw if isinstance(decisions_raw, list) else [decisions_raw]
+            )
             decisions = [d for d in decisions if isinstance(d, dict)]
             if not decisions:
                 return {"status": "error", "message": "No approval decision received"}
@@ -206,7 +229,8 @@ async def delete_onedrive_file(
                             SearchSourceConnector.id == final_connector_id,
                             SearchSourceConnector.search_space_id == search_space_id,
                             SearchSourceConnector.user_id == user_id,
-                            SearchSourceConnector.connector_type == SearchSourceConnectorType.ONEDRIVE_CONNECTOR,
+                            SearchSourceConnector.connector_type
+                            == SearchSourceConnectorType.ONEDRIVE_CONNECTOR,
                         )
                     )
                 )
@@ -224,10 +248,14 @@ async def delete_onedrive_file(
                 f"Deleting OneDrive file: file_id='{final_file_id}', connector={actual_connector_id}"
             )
 
-            client = OneDriveClient(session=db_session, connector_id=actual_connector_id)
+            client = OneDriveClient(
+                session=db_session, connector_id=actual_connector_id
+            )
             await client.trash_file(final_file_id)
 
-            logger.info(f"OneDrive file deleted (moved to recycle bin): file_id={final_file_id}")
+            logger.info(
+                f"OneDrive file deleted (moved to recycle bin): file_id={final_file_id}"
+            )
 
             trash_result: dict[str, Any] = {
                 "status": "success",
@@ -272,6 +300,9 @@ async def delete_onedrive_file(
             if isinstance(e, GraphInterrupt):
                 raise
             logger.error(f"Error deleting OneDrive file: {e}", exc_info=True)
-            return {"status": "error", "message": "Something went wrong while trashing the file. Please try again."}
+            return {
+                "status": "error",
+                "message": "Something went wrong while trashing the file. Please try again.",
+            }
 
     return delete_onedrive_file

surfsense_backend/app/connectors/onedrive/client.py

@@ -39,7 +39,9 @@ async def _get_valid_token(self) -> str:
 
         cfg = connector.config or {}
         is_encrypted = cfg.get("_token_encrypted", False)
-        token_encryption = TokenEncryption(config.SECRET_KEY) if config.SECRET_KEY else None
+        token_encryption = (
+            TokenEncryption(config.SECRET_KEY) if config.SECRET_KEY else None
+        )
 
         access_token = cfg.get("access_token", "")
         refresh_token = cfg.get("refresh_token")
@@ -206,18 +208,20 @@ async def download_file(self, item_id: str) -> tuple[bytes | None, str | None]:
     async def download_file_to_disk(self, item_id: str, dest_path: str) -> str | None:
         """Stream file content to disk. Returns error message on failure."""
         token = await self._get_valid_token()
-        async with httpx.AsyncClient(follow_redirects=True) as client:
-            async with client.stream(
+        async with (
+            httpx.AsyncClient(follow_redirects=True) as client,
+            client.stream(
                 "GET",
                 f"{GRAPH_API_BASE}/me/drive/items/{item_id}/content",
                 headers={"Authorization": f"Bearer {token}"},
                 timeout=120.0,
-            ) as resp:
-                if resp.status_code != 200:
-                    return f"Download failed: {resp.status_code}"
-                with open(dest_path, "wb") as f:
-                    async for chunk in resp.aiter_bytes(chunk_size=5 * 1024 * 1024):
-                        f.write(chunk)
+            ) as resp,
+        ):
+            if resp.status_code != 200:
+                return f"Download failed: {resp.status_code}"
+            with open(dest_path, "wb") as f:
+                async for chunk in resp.aiter_bytes(chunk_size=5 * 1024 * 1024):
+                    f.write(chunk)
         return None
 
     async def create_file(

surfsense_backend/app/connectors/onedrive/content_extractor.py

@@ -5,6 +5,7 @@
 """
 
 import asyncio
+import contextlib
 import logging
 import os
 import tempfile
@@ -60,7 +61,9 @@ async def download_and_extract_content(
 
     temp_file_path = None
     try:
-        extension = Path(file_name).suffix or get_extension_from_mime(mime_type) or ".bin"
+        extension = (
+            Path(file_name).suffix or get_extension_from_mime(mime_type) or ".bin"
+        )
         with tempfile.NamedTemporaryFile(delete=False, suffix=extension) as tmp:
             temp_file_path = tmp.name
 
@@ -76,10 +79,8 @@ async def download_and_extract_content(
         return None, metadata, str(e)
     finally:
         if temp_file_path and os.path.exists(temp_file_path):
-            try:
+            with contextlib.suppress(Exception):
                 os.unlink(temp_file_path)
-            except Exception:
-                pass
 
 
 async def _parse_file_to_markdown(file_path: str, filename: str) -> str:
@@ -94,9 +95,10 @@ async def _parse_file_to_markdown(file_path: str, filename: str) -> str:
             return f.read()
 
     if lower.endswith((".mp3", ".mp4", ".mpeg", ".mpga", ".m4a", ".wav", ".webm")):
-        from app.config import config as app_config
         from litellm import atranscription
 
+        from app.config import config as app_config
+
         stt_service_type = (
             "local"
             if app_config.STT_SERVICE and app_config.STT_SERVICE.startswith("local/")
@@ -106,9 +108,13 @@ async def _parse_file_to_markdown(file_path: str, filename: str) -> str:
             from app.services.stt_service import stt_service
 
             t0 = time.monotonic()
-            logger.info(f"[local-stt] START file={filename} thread={threading.current_thread().name}")
+            logger.info(
+                f"[local-stt] START file={filename} thread={threading.current_thread().name}"
+            )
             result = await asyncio.to_thread(stt_service.transcribe_file, file_path)
-            logger.info(f"[local-stt] END file={filename} elapsed={time.monotonic() - t0:.2f}s")
+            logger.info(
+                f"[local-stt] END file={filename} elapsed={time.monotonic() - t0:.2f}s"
+            )
             text = result.get("text", "")
         else:
             with open(file_path, "rb") as audio_file:
@@ -150,7 +156,9 @@ async def _parse_file_to_markdown(file_path: str, filename: str) -> str:
             parse_with_llamacloud_retry,
         )
 
-        result = await parse_with_llamacloud_retry(file_path=file_path, estimated_pages=50)
+        result = await parse_with_llamacloud_retry(
+            file_path=file_path, estimated_pages=50
+        )
         markdown_documents = await result.aget_markdown_documents(split_by_page=False)
         if not markdown_documents:
             raise RuntimeError(f"LlamaCloud returned no documents for {filename}")
@@ -161,9 +169,13 @@ async def _parse_file_to_markdown(file_path: str, filename: str) -> str:
 
         converter = DocumentConverter()
         t0 = time.monotonic()
-        logger.info(f"[docling] START file={filename} thread={threading.current_thread().name}")
+        logger.info(
+            f"[docling] START file={filename} thread={threading.current_thread().name}"
+        )
         result = await asyncio.to_thread(converter.convert, file_path)
-        logger.info(f"[docling] END file={filename} elapsed={time.monotonic() - t0:.2f}s")
+        logger.info(
+            f"[docling] END file={filename} elapsed={time.monotonic() - t0:.2f}s"
+        )
         return result.document.export_to_markdown()
 
     raise RuntimeError(f"Unknown ETL_SERVICE: {app_config.ETL_SERVICE}")