Update utils.py

utils.py

@@ -23,6 +23,7 @@ def analyze_company_data(company_name: str) -> Dict[str, Any]:
         news_extractor = NewsExtractor()
         sentiment_analyzer = SentimentAnalyzer()
         text_summarizer = TextSummarizer()
+        comparative_analyzer = ComparativeAnalyzer()
         
         # Get news articles
         articles = news_extractor.search_news(company_name)
@@ -50,20 +51,20 @@ def analyze_company_data(company_name: str) -> Dict[str, Any]:
             
             # Analyze fine-grained sentiment
             try:
-                # Financial sentiment
-                if 'financial' in sentiment_analyzer.fine_grained_models:
-                    financial_sentiment = sentiment_analyzer.fine_grained_models['financial'](article['content'])[0]
-                    article['financial_sentiment'] = financial_sentiment['label']
+                fine_grained_results = sentiment_analyzer._get_fine_grained_sentiment(article['content'])
+                article['fine_grained_sentiment'] = fine_grained_results
                 
-                # Emotional sentiment
-                if 'emotion' in sentiment_analyzer.fine_grained_models:
-                    emotional_sentiment = sentiment_analyzer.fine_grained_models['emotion'](article['content'])[0]
-                    article['emotional_sentiment'] = emotional_sentiment['label']
+                # Add sentiment indices
+                sentiment_indices = sentiment_analyzer._calculate_sentiment_indices(fine_grained_results)
+                article['sentiment_indices'] = sentiment_indices
+                
+                # Add entities and sentiment targets
+                entities = sentiment_analyzer._extract_entities(article['content'])
+                article['entities'] = entities
+                
+                sentiment_targets = sentiment_analyzer._extract_sentiment_targets(article['content'], entities)
+                article['sentiment_targets'] = sentiment_targets
                 
-                # ESG sentiment
-                if 'esg' in sentiment_analyzer.fine_grained_models:
-                    esg_sentiment = sentiment_analyzer.fine_grained_models['esg'](article['content'])[0]
-                    article['esg_sentiment'] = esg_sentiment['label']
             except Exception as e:
                 print(f"Error in fine-grained sentiment analysis: {str(e)}")
             
@@ -84,13 +85,27 @@ def analyze_company_data(company_name: str) -> Dict[str, Any]:
             # Pad shorter arrays with 'neutral' to match the longest array
             sentiment_scores[source].extend(['neutral'] * (max_length - len(sentiment_scores[source])))
         
-        return {
+        # Get comparative analysis
+        comparative_analysis = comparative_analyzer.analyze_coverage(processed_articles, company_name)
+        
+        # Combine all results
+        result = {
             "articles": processed_articles,
-            "comparative_sentiment_score": sentiment_scores,
+            "comparative_sentiment_score": {
+                "sentiment_distribution": comparative_analysis.get("sentiment_distribution", {}),
+                "sentiment_indices": comparative_analysis.get("sentiment_indices", {}),
+                "source_distribution": comparative_analysis.get("source_distribution", {}),
+                "common_topics": comparative_analysis.get("common_topics", []),
+                "coverage_differences": comparative_analysis.get("coverage_differences", []),
+                "total_articles": len(processed_articles)
+            },
             "final_sentiment_analysis": overall_sentiment,
+            "ensemble_info": sentiment_analyzer._get_ensemble_sentiment("\n".join([a['content'] for a in processed_articles])),
             "audio_path": None
         }
         
+        return result
+        
     except Exception as e:
         print(f"Error analyzing company data: {str(e)}")
         return {