top k 200, 50 + max chunk size = 10 000, max chunk row = 40

index_retriever.py

@@ -51,7 +51,7 @@ def create_query_engine(vector_index):
         
         vector_retriever = VectorIndexRetriever(
             index=vector_index, 
-            similarity_top_k=200,
+            similarity_top_k=50,
             similarity_cutoff=0.35
         )
         
@@ -73,7 +73,41 @@ def create_query_engine(vector_index):
         )
         
         log_message("Query engine успешно создан")
+
+
+        all_nodes = list(vector_index.docstore.docs.values())
+        c25_tables = []
+
+        for node_id, node in vector_index.docstore.docs.items():
+            metadata = node.metadata
+            text = node.get_content()
+            
+            # Check if this is a С-25 table
+            if ('С-25' in text or 'C-25' in text or 
+                'С-25' in str(metadata.get('table_title', '')) or
+                'С-25' in str(metadata.get('table_number', ''))):
+                
+                c25_tables.append({
+                    'node_id': node_id,
+                    'doc_id': metadata.get('document_id'),
+                    'table_num': metadata.get('table_number'),
+                    'table_title': metadata.get('table_title', ''),
+                    'text_preview': text[:200]
+                })
+
+        log_message(f"\n{'='*70}")
+        log_message(f"DEBUG: Found {len(c25_tables)} С-25 tables in index:")
+        for t in c25_tables:
+            log_message(f"  • {t['doc_id']} - Table {t['table_num']}")
+            log_message(f"    Title: {t['table_title']}")
+            log_message(f"    Preview: {t['text_preview']}")
+        log_message(f"{'='*70}\n")
+
+
         return query_engine
+    
+
+
         
     except Exception as e:
         log_message(f"Ошибка создания query engine: {str(e)}")

utils.py

@@ -172,28 +172,6 @@ def deduplicate_nodes(nodes):
     
     return unique_nodes
 
-def debug_search_tables(vector_index, search_term="С-25"):
-    """Debug function to find all tables containing a specific term"""
-    all_nodes = list(vector_index.docstore.docs.values())
-    
-    matching = []
-    for node in all_nodes:
-        if node.metadata.get('type') == 'table':
-            text = node.get_content()
-            if search_term in text or search_term in node.metadata.get('table_title', ''):
-                matching.append({
-                    'doc_id': node.metadata.get('document_id'),
-                    'table_num': node.metadata.get('table_number'),
-                    'title': node.metadata.get('table_title', '')[:100]
-                })
-    
-    log_message(f"\n{'='*60}")
-    log_message(f"DEBUG: Found {len(matching)} tables containing '{search_term}'")
-    for m in matching:
-        log_message(f"  • {m['doc_id']} - Table {m['table_num']}: {m['title']}")
-    log_message(f"{'='*60}\n")
-    
-    return matching
 
 def answer_question(question, query_engine, reranker, current_model, chunks_df=None):
     if query_engine is None: