Update app.py

app.py

@@ -1,27 +1,55 @@
 import gradio as gr
 import requests
+import socket
 from bs4 import BeautifulSoup
+from urllib.parse import urlparse
 from google import genai
 
 # -----------------------------
-# Helper: Fetch & clean webpage
+# URL Normalization
 # -----------------------------
-def fetch_website_text(url):
-    headers = {"User-Agent": "Mozilla/5.0"}
-    response = requests.get(url, headers=headers, timeout=10)
+def normalize_url(url: str) -> str:
+    parsed = urlparse(url)
+    if not parsed.scheme:
+        return "https://" + url
+    return url
+
+
+# -----------------------------
+# Fetch & Parse Website (Hardened)
+# -----------------------------
+def fetch_website_text(url: str) -> str:
+    socket.setdefaulttimeout(10)
+
+    headers = {
+        "User-Agent": (
+            "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
+            "AppleWebKit/537.36 (KHTML, like Gecko) "
+            "Chrome/121.0 Safari/537.36"
+        ),
+        "Accept": "text/html,application/xhtml+xml",
+        "Accept-Language": "en-US,en;q=0.9",
+    }
+
+    response = requests.get(
+        url,
+        headers=headers,
+        timeout=10,
+        allow_redirects=True,
+    )
     response.raise_for_status()
 
     soup = BeautifulSoup(response.text, "html.parser")
 
-    # Remove scripts & styles
+    # Remove noisy tags
     for tag in soup(["script", "style", "noscript"]):
         tag.decompose()
 
-    title = soup.title.string if soup.title else ""
+    title = soup.title.string.strip() if soup.title else ""
     h1 = soup.find("h1").get_text(strip=True) if soup.find("h1") else ""
 
     body_text = " ".join(soup.stripped_strings)
-    body_text = body_text[:8000]  # keep token usage sane
+    body_text = body_text[:8000]  # token safety
 
     return f"""
 PAGE TITLE:
@@ -34,17 +62,41 @@ VISIBLE CONTENT:
 {body_text}
 """
 
+
 # -----------------------------
-# Gemini analysis
+# Safe Wrapper (Never Crash)
+# -----------------------------
+def fetch_website_text_safe(url: str) -> str:
+    try:
+        return fetch_website_text(url)
+    except Exception as e:
+        return f"""
+⚠️ Unable to fully fetch website content.
+
+Error:
+{str(e)}
+
+Fallback:
+Analyze based on URL structure, homepage intent, and general best practices.
+"""
+
+
+# -----------------------------
+# Gemini Analysis
 # -----------------------------
 def analyze_website(api_key, url, industry, goal):
     if not api_key:
         return "❌ Please enter your Gemini API key."
 
+    if not url:
+        return "❌ Please enter a website URL."
+
     try:
+        url = normalize_url(url)
+
         client = genai.Client(api_key=api_key)
 
-        website_text = fetch_website_text(url)
+        website_text = fetch_website_text_safe(url)
 
         prompt = f"""
 You are an AI consultant helping small businesses improve their websites.
@@ -85,53 +137,66 @@ Website content:
 
         response = client.models.generate_content(
             model="gemini-1.5-flash",
-            contents=prompt
+            contents=prompt,
         )
 
         return response.text
 
     except Exception as e:
-        return f"❌ Error: {str(e)}"
+        return f"❌ Error during analysis: {str(e)}"
+
 
 # -----------------------------
 # Gradio UI
 # -----------------------------
 with gr.Blocks(title="AI Website Review Tool") as demo:
     gr.Markdown("## 🔍 AI Website Review Tool")
-    gr.Markdown("Analyze a website and receive practical, business-focused recommendations.")
+    gr.Markdown(
+        "Analyze any website and receive practical, business-focused recommendations."
+    )
 
-    with gr.Row():
-        api_key = gr.Textbox(
-            label="Gemini API Key",
-            placeholder="Paste your Gemini API key here",
-            type="password"
-        )
+    api_key = gr.Textbox(
+        label="Gemini API Key",
+        placeholder="Paste your Gemini API key here",
+        type="password",
+    )
 
     url = gr.Textbox(
         label="Website URL",
-        placeholder="https://example.com"
+        placeholder="https://example.com",
     )
 
     industry = gr.Dropdown(
         label="Industry",
-        choices=["General SMB", "Law Firm", "Hospitality", "Healthcare", "Real Estate"],
-        value="General SMB"
+        choices=[
+            "General SMB",
+            "Law Firm",
+            "Hospitality",
+            "Healthcare",
+            "Real Estate",
+        ],
+        value="General SMB",
     )
 
     goal = gr.Dropdown(
         label="Primary Website Goal",
-        choices=["Generate leads", "Sell services", "Build credibility", "Educate visitors"],
-        value="Generate leads"
+        choices=[
+            "Generate leads",
+            "Sell services",
+            "Build credibility",
+            "Educate visitors",
+        ],
+        value="Generate leads",
     )
 
     analyze_btn = gr.Button("Analyze Website")
 
-    output = gr.Markdown(label="Analysis Results")
+    output = gr.Markdown()
 
     analyze_btn.click(
         fn=analyze_website,
         inputs=[api_key, url, industry, goal],
-        outputs=output
+        outputs=output,
     )
 
 demo.launch()