Update app.py

app.py

@@ -7,48 +7,34 @@ from langchain_core.messages import HumanMessage
 import os
 import re
 
-# --- Configuration & Initialization ---
-# For deployment on Hugging Face, set GEMINI_API_KEY in the Space's secrets.
-# AHREFS_API_KEY is optional for now, as we will simulate the data.
-AHREFS_API_KEY = os.environ.get("AHREFS_API_KEY")
-
-# --- Core Analysis Functions ---
+# --- Configuration ---
+GEMINI_API_KEY = os.environ.get("GEMINI_API_KEY")
 
+# --- Core Analysis Functions (Unchanged) ---
 def fetch_html(url: str) -> str | None:
-    """Fetches HTML content from a URL with a browser-like user-agent."""
+    # ... (code is the same as before) ...
     try:
-        headers = {
-            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
-        }
+        headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'}
         response = requests.get(url, headers=headers, timeout=10)
-        response.raise_for_status()  # Raise an exception for bad status codes (4xx or 5xx)
+        response.raise_for_status()
         return response.text
-    except requests.RequestException as e:
-        print(f"Error fetching {url}: {e}")
+    except requests.RequestException:
         return None
 
 def analyze_onpage_seo(soup: BeautifulSoup) -> dict:
-    """Analyzes the on-page SEO elements of a webpage."""
+    # ... (code is the same as before) ...
     title = soup.find('title').get_text(strip=True) if soup.find('title') else "Not found"
     description_tag = soup.find('meta', attrs={'name': 'description'})
     description = description_tag['content'] if description_tag and description_tag.has_attr('content') else "Not found"
-    
     headings = {'h1': [], 'h2': []}
     for h_tag in ['h1', 'h2']:
         for tag in soup.find_all(h_tag):
             headings[h_tag].append(tag.get_text(strip=True))
-            
     word_count = len(soup.get_text(separator=' ', strip=True).split())
-    
-    return {
-        "title": title,
-        "description": description,
-        "headings": headings,
-        "word_count": word_count
-    }
+    return {"title": title, "description": description, "headings": headings, "word_count": word_count}
 
 def analyze_tech_stack(html: str) -> list[str]:
-    """Performs a basic analysis of the technologies used on the page."""
+    # ... (code is the same as before) ...
     tech = set()
     if "react.js" in html or 'data-reactroot' in html: tech.add("React")
     if "vue.js" in html: tech.add("Vue.js")
@@ -57,62 +43,32 @@ def analyze_tech_stack(html: str) -> list[str]:
     if "GTM-" in html: tech.add("Google Tag Manager")
     if "tailwind" in html: tech.add("Tailwind CSS")
     if "shopify" in html: tech.add("Shopify")
-        
     return list(tech) if tech else ["Basic HTML/CSS"]
 
 def analyze_ads_and_keywords(domain: str) -> dict:
-    """
-    Simulates fetching paid keywords and ad data from a service like Ahrefs.
-    This provides a 'wow' demo for specific, well-known sites.
-    """
-    # In a real product with a subscription, you would uncomment this block:
-    # if not AHREFS_API_KEY:
-    #     return {"error": "This feature requires a Pro subscription to an SEO data provider."}
-    # url = f"https://api.ahrefs.com/v3/...?target={domain}"
-    # response = requests.get(url, headers={"Authorization": f"Bearer {AHREFS_API_KEY}"})
-    # return response.json()
-
+    # ... (code is the same as before, with simulated data) ...
     print(f"Simulating Ads & Keywords API call for {domain}")
     if "notion" in domain:
-        return {
-            "keywords": [
-                {"keyword": "what is notion", "volume": 65000, "cpc": 0.50},
-                {"keyword": "notion templates", "volume": 45000, "cpc": 1.20},
-                {"keyword": "second brain app", "volume": 12000, "cpc": 2.50},
-                {"keyword": "project management software", "volume": 25000, "cpc": 8.00},
-            ],
-            "ads": [
-                {"title": "Notion – Your All-in-One Workspace", "text": "Organize your life and work. From notes and docs, to projects and wikis, Notion is all you need."},
-                {"title": "The Best Second Brain App | Notion", "text": "Stop juggling tools. Notion combines everything you need to think, write, and plan in one place."},
-            ]
-        }
+        return { "keywords": [{"keyword": "what is notion", "volume": 65000, "cpc": 0.50}, {"keyword": "notion templates", "volume": 45000, "cpc": 1.20}, {"keyword": "second brain app", "volume": 12000, "cpc": 2.50}, {"keyword": "project management software", "volume": 25000, "cpc": 8.00},], "ads": [{"title": "Notion – Your All-in-One Workspace", "text": "Organize your life and work. From notes and docs, to projects and wikis, Notion is all you need."}, {"title": "The Best Second Brain App | Notion", "text": "Stop juggling tools. Notion combines everything you need to think, write, and plan in one place."},] }
     if "mailchimp" in domain:
-        return {
-            "keywords": [
-                {"keyword": "email marketing", "volume": 90500, "cpc": 15.50},
-                {"keyword": "free email marketing tools", "volume": 14800, "cpc": 12.00},
-                {"keyword": "newsletter software", "volume": 8100, "cpc": 9.50},
-            ],
-            "ads": [
-                {"title": "Mailchimp: Marketing & Email", "text": "Grow your business with Mailchimp's All-in-One marketing, automation & email marketing platform."},
-            ]
-        }
+        return { "keywords": [{"keyword": "email marketing", "volume": 90500, "cpc": 15.50}, {"keyword": "free email marketing tools", "volume": 14800, "cpc": 12.00}, {"keyword": "newsletter software", "volume": 8100, "cpc": 9.50},], "ads": [{"title": "Mailchimp: Marketing & Email", "text": "Grow your business with Mailchimp's All-in-One marketing, automation & email marketing platform."},] }
     return {"keywords": [], "ads": []}
 
+# --- AI Summary (Now with a Conversion Trigger) ---
 def generate_ai_summary(url: str, seo_data: dict, ads_data: dict) -> str:
-    """Generates a high-level strategic summary using an LLM."""
-    api_key = os.environ.get("GEMINI_API_KEY")
-    if not api_key:
-        return "⚠️ **AI Summary Unavailable:** The `GEMINI_API_KEY` is not set in the Space secrets. Please ask the Space owner to add it."
+    if not GEMINI_API_KEY:
+        return "⚠️ **AI Summary Unavailable:** The `GEMINI_API_KEY` is not set in the Space secrets."
 
     try:
-        llm = ChatGoogleGenerativeAI(model="gemini-1.5-flash", google_api_key=api_key)
+        llm = ChatGoogleGenerativeAI(model="gemini-1.5-flash", google_api_key=GEMINI_API_KEY)
         
         ads_summary = "They do not appear to be running any significant Google Ads campaigns."
         if ads_data and ads_data.get('keywords'):
             top_keyword = ads_data['keywords'][0]
             ads_summary = f"They are actively running Google Ads, primarily bidding on high-intent keywords like **'{top_keyword['keyword']}'**."
 
+        # *** NEW: CONVERSION TRIGGER IN THE PROMPT ***
+        # We are telling the AI to hint at deeper insights that the Pro version would provide.
         prompt = f"""
         As a world-class marketing strategist, analyze the data for the website `{url}` and provide a concise, actionable summary in markdown format.
 
@@ -123,27 +79,20 @@ def generate_ai_summary(url: str, seo_data: dict, ads_data: dict) -> str:
         - **🎯 Core Marketing Angle:** What is their main value proposition and selling point?
         - **📈 Customer Acquisition Focus:** Based on the data, are they focused more on organic SEO or paid advertising?
         - **💡 One Actionable Insight:** What is one clever tactic they're using, or one key opportunity they are missing?
+
+        **Finally, add a "Go Deeper" section that hints at what a full analysis could uncover, like this example: "Go Deeper: A full Pro analysis could reveal their top-performing ad copy and entire keyword portfolio."**
         """
         response = llm.invoke([HumanMessage(content=prompt)])
         return response.content
     except Exception as e:
         return f"⚠️ **AI Summary Failed:** The API call could not be completed. Error: {e}"
 
-# --- Main Orchestrator ---
-
+# --- Main Orchestrator (Unchanged) ---
 def competitor_teardown(url: str):
-    """The main function that runs the entire analysis pipeline."""
-    # Define the initial state for all outputs
-    outputs = {
-        "summary": " ", "seo": " ", "tech": " ", "ads": " ",
-        "btn": gr.Button("Analyzing...", interactive=False)
-    }
-    # Immediately update the UI to show the "Analyzing..." state
+    # ... (code is the same as before) ...
+    outputs = { "summary": " ", "seo": " ", "tech": " ", "ads": " ", "btn": gr.Button("Analyzing...", interactive=False) }
     yield list(outputs.values())
-
-    # --- 1. Data Fetching ---
-    if not url.startswith(('http://', 'https://')):
-        url = 'https://' + url
+    if not url.startswith(('http://', 'https://')): url = 'https://' + url
     domain_match = re.search(r'https?://(?:www\.)?([^/]+)', url)
     if not domain_match:
         outputs["summary"] = "❌ **Invalid URL:** Please enter a valid website address like `notion.so`."
@@ -151,56 +100,43 @@ def competitor_teardown(url: str):
         yield list(outputs.values())
         return
     domain = domain_match.group(1)
-
     html = fetch_html(url)
     if not html:
         outputs["summary"] = f"❌ **Fetch Failed:** Could not retrieve content from `{url}`. The site may be down or blocking scrapers."
         outputs["btn"] = gr.Button("Analyze", interactive=True)
         yield list(outputs.values())
         return
-    
     soup = BeautifulSoup(html, 'html.parser')
-
-    # --- 2. Run All Analyses ---
     seo_data = analyze_onpage_seo(soup)
     tech_data = analyze_tech_stack(html)
     ads_data = analyze_ads_and_keywords(domain)
     ai_summary = generate_ai_summary(url, seo_data, ads_data)
-
-    # --- 3. Prepare Rich Markdown Outputs ---
     outputs["summary"] = ai_summary
-    
-    outputs["seo"] = f"""
-    ### 👑 SEO & Content Analysis
-    | Metric | Value |
-    | :--- | :--- |
-    | **Page Title** | `{seo_data['title']}` |
-    | **Meta Description** | `{seo_data['description']}` |
-    | **Word Count** | `{seo_data['word_count']:,}` |
-    
-    #### Heading Structure
-    - **H1 Tags ({len(seo_data['headings']['h1'])}):** {', '.join(f'`{h}`' for h in seo_data['headings']['h1']) if seo_data['headings']['h1'] else 'None Found'}
-    - **H2 Tags ({len(seo_data['headings']['h2'])}):** {len(seo_data['headings']['h2'])} found
-    """
-    
+    outputs["seo"] = f"""### 👑 SEO & Content Analysis
+| Metric | Value |
+| :--- | :--- |
+| **Page Title** | `{seo_data['title']}` |
+| **Meta Description** | `{seo_data['description']}` |
+| **Word Count** | `{seo_data['word_count']:,}` |
+#### Heading Structure
+- **H1 Tags ({len(seo_data['headings']['h1'])}):** {', '.join(f'`{h}`' for h in seo_data['headings']['h1']) if seo_data['headings']['h1'] else 'None Found'}
+- **H2 Tags ({len(seo_data['headings']['h2'])}):** {len(seo_data['headings']['h2'])} found
+"""
     outputs["tech"] = "### ⚙️ Technology Stack\n\n" + "\n".join([f"- `{t}`" for t in tech_data])
-
     if ads_data.get("keywords"):
         df = pd.DataFrame(ads_data["keywords"])
         df['cpc'] = df['cpc'].apply(lambda x: f"${x:.2f}")
         ads_md = "### 📢 Paid Ads & Keywords\nThis competitor is actively bidding on Google Search ads. Here are their top keywords:\n\n"
         ads_md += df.to_markdown(index=False)
         ads_md += "\n\n### ✍️ Sample Ad Copy\n\n"
-        for ad in ads_data["ads"]:
-            ads_md += f"**{ad['title']}**\n\n>{ad['text']}\n\n---\n\n"
+        for ad in ads_data["ads"]: ads_md += f"**{ad['title']}**\n\n>{ad['text']}\n\n---\n\n"
         outputs["ads"] = ads_md
     else:
         outputs["ads"] = "### 📢 Paid Ads & Keywords\n\nNo significant paid advertising activity was detected for this domain."
-
     outputs["btn"] = gr.Button("Analyze", interactive=True)
     yield list(outputs.values())
 
-# --- Gradio UI Definition ---
+# --- Gradio UI (Now with your Gumroad link and better conversion copy) ---
 with gr.Blocks(theme=gr.themes.Soft(), css="footer {display: none !important;}") as demo:
     gr.Markdown("# 🕵️ Gumbo Board: The Instant Competitor Teardown")
     gr.Markdown("Enter a competitor's website to get an instant analysis of their online strategy. *Powered by Gumbo (BeautifulSoup) & AI.*")
@@ -219,17 +155,20 @@ with gr.Blocks(theme=gr.themes.Soft(), css="footer {display: none !important;}")
         with gr.TabItem("📢 Ads & Keywords", id=3):
             ads_output = gr.Markdown()
     
-    # Define the list of outputs in the correct order.
     outputs_list = [summary_output, seo_output, tech_output, ads_output, submit_btn]
-
-    submit_btn.click(
-        fn=competitor_teardown,
-        inputs=[url_input],
-        outputs=outputs_list
-    )
+    submit_btn.click(fn=competitor_teardown, inputs=[url_input], outputs=outputs_list)
     
     gr.Markdown("---")
-    gr.Markdown("### Ready for More?\nGet unlimited reports, save projects, and export to PDF with our Pro plan.\n**[🚀 Click Here to Go Pro on Gumroad!](https://gumroad.com/)**")
+    # *** YOUR GUMROAD LINK AND CONVERSION-FOCUSED COPY ***
+    gr.Markdown("""
+    ### You're scratching the surface. Ready to go deeper?
+    The Pro version gives you the unfair advantage:
+    - **✅ Unlimited Reports:** Analyze everyone in your market.
+    - **🔑 Live API Data:** Get real-time Ads & Keyword data for *any* domain.
+    - **📄 PDF Exports:** Create beautiful reports for clients or your team.
+
+    **[🚀 Unlock Gumbo Board Pro Now!](https://gumroadian864.gumroad.com/l/zvgxv)**
+    """)
 
 if __name__ == "__main__":
     demo.launch()