add kb sync for notion updates

surfsense_backend/app/agents/new_chat/tools/notion/update_page.py

@@ -218,6 +218,38 @@ def create_update_notion_page_tool(
             logger.info(
                 f"update_page result: {result.get('status')} - {result.get('message', '')}"
             )
+
+            if result.get("status") == "success":
+                from app.services.notion import NotionKBSyncService
+
+                logger.info(f"Updating knowledge base for page {final_page_id}...")
+                kb_service = NotionKBSyncService(db_session)
+                kb_result = await kb_service.sync_after_update(
+                    page_id=final_page_id,
+                    search_space_id=search_space_id,
+                    appended_content=final_content,
+                    user_id=user_id,
+                )
+
+                if kb_result["status"] == "success":
+                    result["message"] = (
+                        f"{result['message']}. Knowledge base updated - your search results now reflect the latest content."
+                    )
+                    logger.info(
+                        f"Knowledge base successfully updated for page {final_page_id}"
+                    )
+                elif kb_result["status"] == "not_indexed":
+                    result["message"] = (
+                        f"{result['message']}. (Note: This page hasn't been indexed yet, so it won't appear in search until the next scheduled indexing.)"
+                    )
+                else:
+                    result["message"] = (
+                        f"{result['message']}. However, knowledge base update failed: {kb_result['message']}. The page will sync on the next scheduled indexing."
+                    )
+                    logger.warning(
+                        f"KB update failed for page {final_page_id}: {kb_result['message']}"
+                    )
+
             return result
 
         except Exception as e:

surfsense_backend/app/services/notion/__init__.py

@@ -1,3 +1,4 @@
+from app.services.notion.kb_sync_service import NotionKBSyncService
 from app.services.notion.tool_metadata_service import (
     NotionAccount,
     NotionPage,
@@ -6,6 +7,7 @@ from app.services.notion.tool_metadata_service import (
 
 __all__ = [
     "NotionAccount",
+    "NotionKBSyncService",
     "NotionPage",
     "NotionToolMetadataService",
 ]

surfsense_backend/app/services/notion/kb_sync_service.py

@@ -0,0 +1,89 @@
+import logging
+from datetime import datetime
+
+from sqlalchemy import delete
+from sqlalchemy.ext.asyncio import AsyncSession
+from sqlalchemy.future import select
+
+from app.config import config
+from app.db import Chunk, Document
+from app.services.llm_service import get_user_long_context_llm
+from app.tasks.connector_indexers.base import get_current_timestamp, safe_set_chunks
+from app.utils.document_converters import (
+    create_document_chunks,
+    generate_document_summary,
+)
+from app.utils.hash_utils import generate_content_hash
+
+logger = logging.getLogger(__name__)
+
+
+class NotionKBSyncService:
+    def __init__(self, db_session: AsyncSession):
+        self.db_session = db_session
+
+    async def sync_after_update(
+        self,
+        page_id: str,
+        search_space_id: int,
+        appended_content: str,
+        user_id: str,
+    ) -> dict:
+        try:
+            result = await self.db_session.execute(
+                select(Document).filter(
+                    Document.search_space_id == search_space_id,
+                    Document.document_metadata["page_id"].astext == page_id,
+                )
+            )
+            document = result.scalars().first()
+
+            if not document:
+                return {"status": "not_indexed"}
+
+            new_content = document.content + "\n\n" + appended_content
+
+            user_llm = await get_user_long_context_llm(
+                self.db_session, user_id, search_space_id
+            )
+
+            if user_llm:
+                document_metadata_for_summary = {
+                    "page_title": document.document_metadata.get("page_title"),
+                    "page_id": page_id,
+                    "document_type": "Notion Page",
+                    "connector_type": "Notion",
+                }
+                summary_content, summary_embedding = await generate_document_summary(
+                    new_content, user_llm, document_metadata_for_summary
+                )
+            else:
+                summary_content = f"Notion Page: {document.document_metadata.get('page_title')}\n\n{new_content[:500]}..."
+                summary_embedding = config.embedding_model_instance.embed(
+                    summary_content
+                )
+
+            await self.db_session.execute(
+                delete(Chunk).where(Chunk.document_id == document.id)
+            )
+
+            chunks = await create_document_chunks(new_content)
+
+            document.content = summary_content
+            document.content_hash = generate_content_hash(new_content)
+            document.embedding = summary_embedding
+            document.document_metadata = {
+                **document.document_metadata,
+                "indexed_at": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
+            }
+            safe_set_chunks(document, chunks)
+            document.updated_at = get_current_timestamp()
+
+            await self.db_session.commit()
+
+            logger.info(f"Successfully synced KB for Notion page {page_id}")
+            return {"status": "success"}
+
+        except Exception as e:
+            logger.error(f"Failed to sync KB for page {page_id}: {e}", exc_info=True)
+            return {"status": "error", "message": str(e)}