Update.

requirements.txt

@@ -7,3 +7,5 @@ aiohttp==3.9.1
 feedparser==6.0.10
 websockets==12.0
 trafilatura==1.6.2
+vaderSentiment
+

routers.py

@@ -4,6 +4,7 @@ from fastapi.templating import Jinja2Templates
 import dataset
 import json
 import aiohttp
+import sentiment
 import feedparser
 import asyncio
 from datetime import datetime
@@ -356,15 +357,17 @@ async def websocket_sync(websocket: WebSocket):
                         'last_synchronized': datetime.now().isoformat()
                     }
                     
+                    
                     existing = articles_table.find_one(guid=article_data['guid'])
                     if not existing:
                         new_articles.append(article_data)
                         articles_count += 1
-                    
+                    article_data['sentiment'] = json.dumps(sentiment.analyze(entry.get('description', '') or entry.get('summary', '')))
                     articles_table.upsert(article_data, ['guid'])
 
                     # Index the article to ChromaDB
                     doc_content = f"{article_data.get('title', '')}\n{article_data.get('description', '')}"
+
                     metadata = {key: str(value) for key, value in article_data.items() if key != 'content'} # Exclude large content from metadata
                     chroma_collection.upsert(
                         documents=[doc_content],
@@ -490,6 +493,7 @@ async def search_articles(
             for i, doc_id in enumerate(results['ids'][0]):
                 res = results['metadatas'][0][i]
                 res['distance'] = results['distances'][0][i]
+                res['sentiment'] = sentiment.analyze(res.get('description', '') or res.get('content', '') or res.get('title', ''))
                 formatted_results.append(res)
                  
         return JSONResponse(content={"results": formatted_results})
@@ -565,6 +569,8 @@ async def newspaper_latest(request: Request):
             for article in articles:
                 for key, value in article.items():
                     article[key] = str(value).strip().replace('  ', '')
+            
+                article['sentiment'] = sentiment.analyze(article.get('description', '') or article.get('content', '') or res.get('title', ''))
             return templates.TemplateResponse("newspaper_view.html", {
                 "request": request,
                 "newspaper": first_newspaper,

sentiment.py

@@ -0,0 +1,35 @@
+import json
+from vaderSentiment.vaderSentiment import SentimentIntensityAnalyzer
+
+def analyze_sentiment_vader(text, analyzer):
+    """
+    Analyzes text using VADER and returns a dictionary with the results.
+
+    Args:
+        text (str): The text content to analyze.
+        analyzer (SentimentIntensityAnalyzer): An instantiated VADER analyzer.
+
+    Returns:
+        dict: A dictionary containing the sentiment classification, compound score,
+              and detailed scores (positive, neutral, negative).
+    """
+    scores = analyzer.polarity_scores(text)
+    compound_score = scores['compound']
+
+    if compound_score >= 0.05:
+        sentiment = 'Positive'
+    elif compound_score <= -0.05:
+        sentiment = 'Negative'
+    else:
+        sentiment = 'Neutral'
+
+    return {
+        'sentiment': sentiment,
+        'score': compound_score,
+        'details': scores  
+    }
+
+vader_analyzer = SentimentIntensityAnalyzer()
+
+def analyze(content):
+    return analyze_sentiment_vader(content, vader_analyzer)

templates/newspaper_view.html

@@ -164,6 +164,7 @@
                 <h2 class="article-title">
                     <a href="{{ article.link }}" target="_blank">{{ article.title }}</a>
                 </h2>
+                <input type="hidden" name="sentiment" value="{{ article.sentiment }}">
                 <div class="article-meta">
                     <span class="article-source">{{ article.feed_name }}</span>
                     {% if article.author %}