new chunking showing code

app.py

@@ -21,8 +21,9 @@ def create_chunks_display_html(chunk_info):
     for i, chunk in enumerate(chunk_info):
         bg_color = "#f8f9fa" if i % 2 == 0 else "#e9ecef"
         
-        # Get section display info similar to format_context_for_llm
-        section_display = format_section_for_display(chunk)
+        # Get section display info
+        section_display = get_section_display(chunk)
+        formatted_content = get_formatted_content(chunk)
         
         html += f"""
         <div style='background-color: {bg_color}; padding: 10px; margin: 5px 0; border-radius: 5px; border-left: 4px solid #007bff; color: black;'>
@@ -30,7 +31,7 @@ def create_chunks_display_html(chunk_info):
             <strong style='color: black;'>Раздел:</strong> <span style='color: black;'>{section_display}</span><br>
             <strong style='color: black;'>Содержание:</strong><br>
             <div style='background-color: white; padding: 8px; margin-top: 5px; border-radius: 3px; font-family: monospace; font-size: 12px; color: black; max-height: 200px; overflow-y: auto;'>
-                {chunk['chunk_text']}
+                {formatted_content}
             </div>
         </div>
         """
@@ -38,13 +39,10 @@ def create_chunks_display_html(chunk_info):
     html += "</div>"
     return html
 
-def format_section_for_display(chunk):
+def get_section_display(chunk):
+    """Get section display for the 'Раздел' field - without 'пункт' prefix"""
     section_path = chunk.get('section_path', '')
     section_id = chunk.get('section_id', 'unknown')
-    section_text = chunk.get('section_text', '')
-    level = chunk.get('level', '')
-    parent_section = chunk.get('parent_section', '')
-    parent_title = chunk.get('parent_title', '')
     doc_type = chunk.get('type', 'text')
     
     if doc_type == 'table' and chunk.get('table_number'):
@@ -59,23 +57,46 @@ def format_section_for_display(chunk):
             image_num = f"№{image_num}"
         return f"рисунок {image_num}"
     
+    # For text documents, return just the section_path or section_id without "пункт"
     if section_path:
-        if level in ['subsection', 'sub_subsection', 'sub_sub_subsection'] and parent_section and parent_title:
-            return f"пункт {section_path} в разделе {parent_section} ({parent_title})"
-        elif section_text:
-            return f"пункт {section_path} ({section_text})"
-        else:
-            return f"пункт {section_path}"
+        return section_path
     elif section_id and section_id != 'unknown':
-        if level in ['subsection', 'sub_subsection', 'sub_sub_subsection'] and parent_section and parent_title:
-            return f"пункт {section_id} в разделе {parent_section} ({parent_title})"
-        elif section_text:
-            return f"пункт {section_id} ({section_text})"
-        else:
-            return f"пункт {section_id}"
+        return section_id
     
     return section_id
 
+def get_formatted_content(chunk):
+    """Format the content with proper section context"""
+    document_id = chunk.get('document_id', 'unknown')
+    section_path = chunk.get('section_path', '')
+    section_id = chunk.get('section_id', 'unknown')
+    parent_section = chunk.get('parent_section', '')
+    level = chunk.get('level', '')
+    chunk_text = chunk.get('chunk_text', '')
+    doc_type = chunk.get('type', 'text')
+    
+    if doc_type == 'table':
+        table_num = chunk.get('table_number', 'unknown')
+        if not str(table_num).startswith('№'):
+            table_num = f"№{table_num}"
+        return f"В таблице {table_num} документа {document_id}: {chunk_text}"
+    
+    if doc_type == 'image':
+        image_num = chunk.get('image_number', 'unknown')
+        if not str(image_num).startswith('№'):
+            image_num = f"№{image_num}"
+        return f"В рисунке {image_num} документа {document_id}: {chunk_text}"
+    
+    # For text documents
+    if level in ['subsection', 'sub_subsection', 'sub_sub_subsection'] and parent_section:
+        # For subsections: В разделе X в документе Y, пункт X.X content
+        current_section = section_path if section_path else section_id
+        return f"В разделе {parent_section} в документе {document_id}, пункт {current_section} {chunk_text}"
+    else:
+        # For main sections: В разделе X в документе Y пункт X content  
+        current_section = section_path if section_path else section_id
+        return f"В разделе {current_section} в документе {document_id} пункт {current_section} {chunk_text}"
+
 
 def initialize_system(repo_id, hf_token, download_dir, chunks_filename=None, 
                      json_files_dir=None, table_data_dir=None, image_data_dir=None, 

utils.py

@@ -142,9 +142,9 @@ def generate_sources_html(nodes, chunks_df=None):
         
         # Add section information
         if section_path:
-            sources_by_doc[key]['sections'].add(f"пункт {section_path}")
+            sources_by_doc[key]['sections'].add(f"{section_path}")
         elif section_id and section_id != 'unknown':
-            sources_by_doc[key]['sections'].add(f"пункт {section_id}")
+            sources_by_doc[key]['sections'].add(f"{section_id}")
     
     # Generate HTML for each unique source
     for source_info in sources_by_doc.values():