v.1.29

app.py

@@ -12,7 +12,8 @@ import os
 groq_key = os.environ['groq_key']
 from langchain_openai import ChatOpenAI
 from langchain.prompts import PromptTemplate
-
+from openpyxl import load_workbook
+from openpyxl.utils.dataframe import dataframe_to_rows
 
 
 def fuzzy_deduplicate(df, column, threshold=55):
@@ -528,11 +529,123 @@ def process_file(file_obj):
         logger.error(f"File processing error: {str(e)}")
         raise
 
+def create_output_file(df, uploaded_file):
+    """Create Excel file with multiple sheets from processed DataFrame"""
+    try:
+        wb = load_workbook("sample_file.xlsx")
+        
+        # 1. Update 'Публикации' sheet
+        ws = wb['Публикации']
+        for r_idx, row in enumerate(dataframe_to_rows(df, index=False, header=True), start=1):
+            for c_idx, value in enumerate(row, start=1):
+                ws.cell(row=r_idx, column=c_idx, value=value)
+
+        # 2. Update 'Мониторинг' sheet with events
+        ws = wb['Мониторинг']
+        row_idx = 4
+        events_df = df[df['Event_Type'] != 'Нет'].copy()
+        for _, row in events_df.iterrows():
+            ws.cell(row=row_idx, column=5, value=row['Объект'])
+            ws.cell(row=row_idx, column=6, value=row['Заголовок'])
+            ws.cell(row=row_idx, column=7, value=row['Event_Type'])
+            ws.cell(row=row_idx, column=8, value=row['Event_Summary'])
+            ws.cell(row=row_idx, column=9, value=row['Выдержки из текста'])
+            row_idx += 1
+
+        # 3. Update 'Сводка' sheet
+        ws = wb['Сводка']
+        unique_entities = df['Объект'].unique()
+        entity_stats = []
+        for entity in unique_entities:
+            entity_df = df[df['Объект'] == entity]
+            stats = {
+                'Объект': entity,
+                'Всего': len(entity_df),
+                'Негативные': len(entity_df[entity_df['Sentiment'] == 'Negative']),
+                'Позитивные': len(entity_df[entity_df['Sentiment'] == 'Positive'])
+            }
+            
+            # Get most severe impact for entity
+            negative_df = entity_df[entity_df['Sentiment'] == 'Negative']
+            if len(negative_df) > 0:
+                impacts = negative_df['Impact'].dropna()
+                if len(impacts) > 0:
+                    stats['Impact'] = impacts.iloc[0]
+                else:
+                    stats['Impact'] = 'Неопределенный эффект'
+            else:
+                stats['Impact'] = 'Неопределенный эффект'
+                
+            entity_stats.append(stats)
+        
+        # Sort by number of negative mentions
+        entity_stats = sorted(entity_stats, key=lambda x: x['Негативные'], reverse=True)
+        
+        # Write to sheet
+        row_idx = 4  # Starting row in Сводка sheet
+        for stats in entity_stats:
+            ws.cell(row=row_idx, column=5, value=stats['Объект'])
+            ws.cell(row=row_idx, column=6, value=stats['Всего'])
+            ws.cell(row=row_idx, column=7, value=stats['Негативные'])
+            ws.cell(row=row_idx, column=8, value=stats['Позитивные'])
+            ws.cell(row=row_idx, column=9, value=stats['Impact'])
+            row_idx += 1
+
+        # 4. Update 'Значимые' sheet
+        ws = wb['Значимые']
+        row_idx = 3
+        sentiment_df = df[df['Sentiment'].isin(['Negative', 'Positive'])].copy()
+        for _, row in sentiment_df.iterrows():
+            ws.cell(row=row_idx, column=3, value=row['Объект'])
+            ws.cell(row=row_idx, column=4, value='релевантно')
+            ws.cell(row=row_idx, column=5, value=row['Sentiment'])
+            ws.cell(row=row_idx, column=6, value=row.get('Impact', '-'))
+            ws.cell(row=row_idx, column=7, value=row['Заголовок'])
+            ws.cell(row=row_idx, column=8, value=row['Выдержки из текста'])
+            row_idx += 1
+
+        # 5. Update 'Анализ' sheet
+        ws = wb['Анализ']
+        row_idx = 4
+        negative_df = df[df['Sentiment'] == 'Negative'].copy()
+        for _, row in negative_df.iterrows():
+            ws.cell(row=row_idx, column=5, value=row['Объект'])
+            ws.cell(row=row_idx, column=6, value=row['Заголовок'])
+            ws.cell(row=row_idx, column=7, value="Риск убытка")
+            ws.cell(row=row_idx, column=8, value=row.get('Reasoning', '-'))
+            ws.cell(row=row_idx, column=9, value=row['Выдержки из текста'])
+            row_idx += 1
+
+        # 6. Update 'Тех.приложение' sheet
+        if 'Тех.приложение' not in wb.sheetnames:
+            wb.create_sheet('Тех.приложение')
+        ws = wb['Тех.приложение']
+        
+        tech_cols = ['Объект', 'Заголовок', 'Выдержки из текста', 'Translated', 'Sentiment', 'Impact', 'Reasoning']
+        tech_df = df[tech_cols].copy()
+        
+        for r_idx, row in enumerate(dataframe_to_rows(tech_df, index=False, header=True), start=1):
+            for c_idx, value in enumerate(row, start=1):
+                ws.cell(row=r_idx, column=c_idx, value=value)
+
+        # Save workbook
+        output = io.BytesIO()
+        wb.save(output)
+        output.seek(0)
+        return output
+
+    except Exception as e:
+        logger.error(f"Error creating output file: {str(e)}")
+        logger.error(f"DataFrame shape: {df.shape}")
+        logger.error(f"Available columns: {df.columns.tolist()}")
+        return None
+    
+
 def create_interface():
     control = ProcessControl()
     
     with gr.Blocks(theme=gr.themes.Soft()) as app:
-        gr.Markdown("# AI-анализ мониторинга новостей v.1.28")
+        gr.Markdown("# AI-анализ мониторинга новостей v.1.29")
         
         with gr.Row():
             file_input = gr.File(