add data viewer

app/law_document_chunker.py

@@ -286,7 +286,17 @@ class LawDocumentChunker:
             chunk_stack.append((metadata.id, current_level, current_level_value))
             logger.debug(f"[CHUNKER] Created final chunk: {metadata.id[:8]}... Level: {current_level}, Parent: {current_parent}")
         
-        logger.info(f"[CHUNKER] Created {len(chunks)} chunks from document")
+        # Debug: Kiểm tra kết quả
+        root_count = sum(1 for chunk in chunks if chunk.cha is None)
+        logger.info(f"[CHUNKER] Created {len(chunks)} chunks, {root_count} root chunks")
+        
+        # Debug: Log chi tiết từng chunk
+        for i, chunk in enumerate(chunks[:10]):  # Log 10 chunks đầu tiên
+            logger.debug(f"[CHUNKER] Chunk {i+1}: {chunk.content[:100]}... -> Parent: {chunk.cha}")
+        
+        if len(chunks) > 10:
+            logger.debug(f"[CHUNKER] ... and {len(chunks) - 10} more chunks")
+        
         return chunks
     
     def _find_parent_for_level(self, chunk_stack: List[Tuple[str, str, Optional[str]]], 
@@ -308,6 +318,8 @@ class LawDocumentChunker:
         logger.info(f"[CHUNKER] Creating embeddings and storing {len(chunks)} chunks")
         
         success_count = 0
+        failed_count = 0
+        
         for i, chunk in enumerate(chunks, 1):
             try:
                 # Tạo embedding - TẠM THỜI COMMENT LẠI ĐỂ TEST
@@ -333,15 +345,18 @@ class LawDocumentChunker:
                 success = self.supabase_client.store_document_chunk(chunk_dict)
                 if success:
                     success_count += 1
-                    logger.info(f"[CHUNKER] Stored chunk {i}/{len(chunks)}: {chunk.id[:8]}...")
+                    if i % 100 == 0:  # Log mỗi 100 chunks
+                        logger.info(f"[CHUNKER] Stored chunk {i}/{len(chunks)}: {chunk.id[:8]}...")
                 else:
+                    failed_count += 1
                     logger.error(f"[CHUNKER] Failed to store chunk {chunk.id}")
                 
             except Exception as e:
+                failed_count += 1
                 logger.error(f"[CHUNKER] Error processing chunk {chunk.id}: {e}")
                 continue
         
-        logger.info(f"[CHUNKER] Successfully processed {success_count}/{len(chunks)} chunks")
+        logger.info(f"[CHUNKER] Successfully processed {success_count}/{len(chunks)} chunks, {failed_count} failed")
         return success_count
 
     async def _store_chunks_to_supabase(self, chunk_data: List[Dict]) -> bool:

app/supabase_db.py

@@ -192,26 +192,15 @@ class SupabaseClient:
         try:
             logger.info("[SUPABASE] Fetching all document chunks")
             
-            all_chunks = []
-            page_size = 1000
-            offset = 0
+            # Thử lấy tất cả records một lần
+            response = self.client.table('document_chunks').select('*').execute()
             
-            while True:
-                # Lấy từng page
-                response = self.client.table('document_chunks').select('*').range(offset, offset + page_size - 1).execute()
-                
-                if not response.data:
-                    break
-                
-                all_chunks.extend(response.data)
-                offset += page_size
-                
-                # Nếu số records ít hơn page_size, đã hết
-                if len(response.data) < page_size:
-                    break
-            
-            logger.info(f"[SUPABASE] Successfully fetched {len(all_chunks)} document chunks")
-            return all_chunks
+            if response.data:
+                logger.info(f"[SUPABASE] Successfully fetched {len(response.data)} document chunks")
+                return response.data
+            else:
+                logger.warning("[SUPABASE] No document chunks found")
+                return []
                 
         except Exception as e:
             logger.error(f"[SUPABASE] Error fetching document chunks: {e}")