Update app.py

app.py

@@ -1,81 +1,75 @@
 import gradio as gr
 import requests
 from bs4 import BeautifulSoup
-from transformers import pipeline, AutoTokenizer
+from transformers import pipeline
 
 # 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):
-    """Enhanced extraction with metadata support"""
+    """Extracts text from a website with error handling"""
     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])
-        
-        full_content = f"{title}\n{meta_desc}\n{text}".strip()
-        
-        return full_content if full_content else "No meaningful content found."
-        
+
+        return text.strip() if text.strip() else "No 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):
-    """Updated function with real-time status"""
+    """Handles website summarization with proper truncation"""
     try:
         extracted_text = scrape_website(url)
-        
+
         if "Error" in extracted_text:
-            return "❌ " + extracted_text
+            return f"❌ {extracted_text}"
 
         if len(extracted_text.split()) < 50:
             return "⚠️ Error: Insufficient content for summarization (minimum 50 words required)"
-        
-        truncated_text = truncate_text(extracted_text)
-        
+
+        # Ensure input is within token limit
+        max_input_tokens = 1024  # Model limit
+        truncated_text = " ".join(extracted_text.split()[:max_input_tokens])
+
+        # Summarization
         summary = summarizer(
             truncated_text,
-            max_length=250,  # Increased summary length
-            min_length=80,    # Ensuring more detailed output
+            max_length=250,  # Extended summary
+            min_length=80,
             do_sample=False
         )
-        
+
+        if not summary or not isinstance(summary, list) or "summary_text" not in summary[0]:
+            return "⚠️ Error: Summarization failed. Try a different website."
+
         return f"## 📝 Summary\n\n{summary[0]['summary_text']}"
-    
+
     except Exception as e:
         return f"⛔ Summarization Error: {str(e)}"
 
-# Custom CSS for better mobile experience
+# Custom CSS for mobile optimization
 css = """
 @media screen and (max-width: 600px) {
-    .container { padding: 10px !important; }
-    .input-box textarea { font-size: 18px !important; }
-    .gr-button { width: 100% !important; }
+    .container {
+        padding: 10px !important;
+    }
+    .input-box textarea {
+        font-size: 16px !important;
+    }
 }
 """
 
-# Mobile-optimized interface with real-time updates
+# Mobile-optimized interface with Blocks API
 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!")
-    
+
     with gr.Row():
         url_input = gr.Textbox(
             label="Website URL",
@@ -84,14 +78,14 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Website Summarizer") as a
             max_lines=1,
             elem_id="input-box"
         )
-    
+
     with gr.Row():
         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"],
@@ -101,24 +95,27 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Website Summarizer") as a
         label="Try these examples:",
         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: ("", "🔄 Ready for input..."),
+        fn=lambda: ("", ""),
         inputs=None,
-        outputs=[url_input, status],
+        outputs=[url_input, output],
         queue=False
     )
 
-# Mobile-friendly deployment
+# Launch the app without broken favicon
 app.launch(
-    server_name="0.0.0.0", 
-    server_port=7860, 
-    favicon_path="https://www.svgrepo.com/show/355037/huggingface.svg"
+    server_name="0.0.0.0",
+    server_port=7860
 )