Update app.py

app.py

@@ -1,69 +1,77 @@
 import gradio as gr
 import requests
 from bs4 import BeautifulSoup
-from transformers import pipeline
+from transformers import pipeline, AutoTokenizer
 
 # Load summarization pipeline
 summarizer = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6")
+tokenizer = AutoTokenizer.from_pretrained("sshleifer/distilbart-cnn-12-6")
 
 def scrape_website(url):
-    """Extracts text from a website with error handling"""
+    """Enhanced extraction with metadata support"""
     try:
         headers = {'User-Agent': 'Mozilla/5.0'}
         response = requests.get(url, headers=headers, timeout=10)
         response.raise_for_status()
         
         soup = BeautifulSoup(response.text, "html.parser")
+        
+        # Extract title and meta description
+        title = soup.title.string.strip() if soup.title else ""
+        meta_desc = soup.find("meta", attrs={"name": "description"})
+        meta_desc = meta_desc["content"].strip() if meta_desc else ""
+
+        # Extract main text content
         text_elements = soup.find_all(['p', 'article', 'main', 'section'])
         text = " ".join([e.get_text(strip=True, separator=' ') for e in text_elements])
-        return text.strip() if text.strip() else "No content found"
+        
+        full_content = f"{title}\n{meta_desc}\n{text}".strip()
+        
+        return full_content if full_content else "No meaningful content found."
         
     except Exception as e:
         return f"Scraping Error: {str(e)}"
 
+def truncate_text(text, max_tokens=1024):
+    """Properly truncates text at the token level."""
+    tokens = tokenizer.tokenize(text)
+    return tokenizer.convert_tokens_to_string(tokens[:max_tokens])
+
 def summarize_website(url):
-    """Handles the full summarization pipeline"""
+    """Updated function with real-time status"""
     try:
-        with gr.Column(variant="panel"):
-            gr.Markdown("## ⚡ Processing...")
-            
-            extracted_text = scrape_website(url)
-            
-            if "Error" in extracted_text:
-                return f"❌ {extracted_text}"
-                
-            if len(extracted_text.split()) < 50:
-                return "⚠️ Error: Insufficient content for summarization (minimum 50 words required)"
-                
-            max_input_length = 1000
-            truncated_text = extracted_text[:max_input_length]
-            
-            summary = summarizer(
-                truncated_text,
-                max_length=200,
-                min_length=50,
-                do_sample=False,
-                truncation=True
-            )
-            
-            return f"## 📝 Summary\n\n{summary[0]['summary_text']}"
-            
+        extracted_text = scrape_website(url)
+        
+        if "Error" in extracted_text:
+            return "❌ " + extracted_text
+
+        if len(extracted_text.split()) < 50:
+            return "⚠️ Error: Insufficient content for summarization (minimum 50 words required)"
+        
+        truncated_text = truncate_text(extracted_text)
+        
+        summary = summarizer(
+            truncated_text,
+            max_length=250,  # Increased summary length
+            min_length=80,    # Ensuring more detailed output
+            do_sample=False
+        )
+        
+        return f"## 📝 Summary\n\n{summary[0]['summary_text']}"
+    
     except Exception as e:
         return f"⛔ Summarization Error: {str(e)}"
 
-# Custom CSS for mobile optimization
+# Custom CSS for better mobile experience
 css = """
 @media screen and (max-width: 600px) {
-    .container {
-        padding: 10px !important;
-    }
-    .input-box textarea {
-        font-size: 16px !important;
-    }
+    .container { padding: 10px !important; }
+    .input-box textarea { font-size: 18px !important; }
+    .gr-button { width: 100% !important; }
 }
 """
 
-# Mobile-optimized interface with Blocks API
+# Mobile-optimized interface with real-time updates
 with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Website Summarizer") as app:
     gr.Markdown("# 🌐 AI Website Summarizer")
     gr.Markdown("Paste any website URL below to get an instant AI-powered summary!")
@@ -81,9 +89,9 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Website Summarizer") as a
         submit_btn = gr.Button("Generate Summary 🚀", variant="primary")
         clear_btn = gr.Button("Clear 🔄")
     
+    status = gr.Markdown("🔄 Ready for input...", elem_id="status-msg")
     output = gr.Markdown()
     
-    # Example section
     gr.Examples(
         examples=[
             ["https://en.wikipedia.org/wiki/Large_language_model"],
@@ -94,27 +102,23 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Website Summarizer") as a
         examples_per_page=2
     )
     
-    # Progress indicator
-    progress = gr.Textbox(visible=False)
-    
-    # Event handlers
     submit_btn.click(
         fn=summarize_website,
         inputs=url_input,
-        outputs=output,
+        outputs=[output],
         api_name="summarize"
     )
     
     clear_btn.click(
-        fn=lambda: ("", ""),
+        fn=lambda: ("", "🔄 Ready for input..."),
         inputs=None,
-        outputs=[url_input, output],
+        outputs=[url_input, status],
         queue=False
     )
 
-# Mobile-friendly configuration
+# Mobile-friendly deployment
 app.launch(
-    server_name="0.0.0.0",
-    server_port=7860,
+    server_name="0.0.0.0", 
+    server_port=7860, 
     favicon_path="https://www.svgrepo.com/show/355037/huggingface.svg"
-)
+)