Update app.py

app.py

@@ -1,45 +1,118 @@
 import gradio as gr
+import requests
+from bs4 import BeautifulSoup
 import spaces
 
 # ============================================================
-# STEP 1: Add the required @spaces.GPU decorator
-# This satisfies Hugging Face's startup requirement
+# REQUIRED: @spaces.GPU decorator for Hugging Face
 # ============================================================
 
 @spaces.GPU(duration=1)
 def noop_gpu_function():
-    """
-    This dummy function satisfies the Hugging Face Space
-    startup requirement. It doesn't actually use GPU.
-    """
+    """Dummy function to satisfy Hugging Face startup"""
     pass
 
 # ============================================================
-# STEP 2: Your main function (the "Hello World")
+# SCRAPER FUNCTION
 # ============================================================
 
-def greet(name):
-    """Simple greeting function"""
-    return f"Hello {name}! Your scraper is working! 🎉"
+def scrape_website(url, keyword, max_results=5):
+    """Scrape a website and extract links containing the keyword"""
+    results = []
+    
+    # Ensure URL has protocol
+    if not url.startswith(('http://', 'https://')):
+        url = 'https://' + url
+    
+    try:
+        # Fetch the webpage
+        headers = {
+            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36'
+        }
+        response = requests.get(url, headers=headers, timeout=30)
+        response.raise_for_status()
+        
+        # Parse HTML
+        soup = BeautifulSoup(response.text, 'html.parser')
+        
+        # Remove unwanted elements
+        for element in soup(['script', 'style', 'nav', 'footer', 'header']):
+            element.decompose()
+        
+        # Find all links
+        links = soup.find_all('a', href=True)
+        
+        keyword_lower = keyword.lower()
+        for link in links:
+            try:
+                href = link.get('href', '')
+                text = link.get_text(strip=True)
+                
+                # Skip invalid links
+                if not href or href.startswith('#') or href.startswith('javascript:') or not text:
+                    continue
+                
+                # Check if keyword is in the link text or URL
+                if keyword_lower in text.lower() or keyword_lower in href.lower():
+                    results.append({
+                        'title': text[:100],
+                        'url': href,
+                        'source': url
+                    })
+                    
+                if len(results) >= max_results:
+                    break
+                    
+            except Exception as e:
+                continue
+        
+        # Format results for display
+        if results:
+            output = f"✅ **Found {len(results)} results for '{keyword}':**\n\n"
+            for i, item in enumerate(results, 1):
+                output += f"**{i}. {item['title']}**\n"
+                output += f"   🔗 {item['url']}\n\n"
+            return output
+        else:
+            return f"⚠️ No results found for '{keyword}' at {url}"
+            
+    except requests.exceptions.RequestException as e:
+        return f"❌ Network error: {str(e)}"
+    except Exception as e:
+        return f"❌ Error: {str(e)}"
 
 # ============================================================
-# STEP 3: Create the Gradio Interface
+# GRADIO INTERFACE
 # ============================================================
 
 demo = gr.Interface(
-    fn=greet,
-    inputs=gr.Textbox(
-        label="Your Name",
-        placeholder="Enter your name...",
-        value="World"
-    ),
-    outputs=gr.Textbox(label="Greeting"),
-    title="🕷️ Market Scraper - Test",
-    description="This is a test to verify your Space is working."
+    fn=scrape_website,
+    inputs=[
+        gr.Textbox(
+            label="URL",
+            placeholder="https://www.pinterest.com/search/pins/?q=planner",
+            value="https://www.pinterest.com/search/pins/?q=planner"
+        ),
+        gr.Textbox(
+            label="Keyword",
+            placeholder="planner, notion, template, etc.",
+            value="planner"
+        ),
+        gr.Slider(
+            minimum=1,
+            maximum=20,
+            value=5,
+            step=1,
+            label="Max Results"
+        )
+    ],
+    outputs=gr.Markdown(label="Results"),
+    title="🕷️ Market Scraper",
+    description="Scrape websites to find content related to your market research keywords."
 )
 
 # ============================================================
-# STEP 4: Launch (MUST be at module level, NOT inside if __name__)
+# LAUNCH (MUST be at module level)
 # ============================================================
 
 demo.launch(server_name="0.0.0.0", server_port=7860)