Update app.py

app.py

@@ -2,127 +2,234 @@ import gradio as gr
 import requests
 from bs4 import BeautifulSoup
 import pandas as pd
-import plotly.express as px
 from langchain_google_genai import ChatGoogleGenerativeAI
 from langchain_core.messages import HumanMessage
 import os
 import re
 
-# All the function definitions are correct and do not need to change.
-# ... (fetch_html, analyze_onpage_seo, analyze_tech_stack, generate_ai_summary) ...
-def fetch_html(url):
+# --- 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 ---
+
+def fetch_html(url: str) -> str | None:
+    """Fetches HTML content from a URL with a browser-like user-agent."""
     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()
+        response.raise_for_status()  # Raise an exception for bad status codes (4xx or 5xx)
         return response.text
     except requests.RequestException as e:
         print(f"Error fetching {url}: {e}")
         return None
 
-def analyze_onpage_seo(soup):
-    title = soup.find('title').get_text(strip=True) if soup.find('title') else "N/A"
-    description = soup.find('meta', attrs={'name': 'description'})['content'] if soup.find('meta', attrs={'name': 'description'}) else "N/A"
-    headings = {'h1': [], 'h2': [], 'h3': []}
-    for h_tag in ['h1', 'h2', 'h3']:
+def analyze_onpage_seo(soup: BeautifulSoup) -> dict:
+    """Analyzes the on-page SEO elements of a webpage."""
+    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().split())
-    return {"title": title, "description": description, "headings": headings, "word_count": word_count}
+            
+    word_count = len(soup.get_text(separator=' ', strip=True).split())
+    
+    return {
+        "title": title,
+        "description": description,
+        "headings": headings,
+        "word_count": word_count
+    }
 
-def analyze_tech_stack(soup, html):
+def analyze_tech_stack(html: str) -> list[str]:
+    """Performs a basic analysis of the technologies used on the page."""
     tech = set()
     if "react.js" in html or 'data-reactroot' in html: tech.add("React")
     if "vue.js" in html: tech.add("Vue.js")
-    if "angular.js" in html: tech.add("Angular")
     if "wp-content" in html: tech.add("WordPress")
     if "gtag('config'" in html: tech.add("Google Analytics (GA4)")
     if "GTM-" in html: tech.add("Google Tag Manager")
-    if soup.find('link', href=lambda x: x and 'tailwind' in x): tech.add("Tailwind CSS")
+    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 generate_ai_summary(url, seo_data, tech_data):
+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()
+
+    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."},
+            ]
+        }
+    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": [], "ads": []}
+
+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 "ERROR: `GEMINI_API_KEY` is not set in the Space secrets. The AI summary cannot be generated. Please ask the Space owner to add it."
+        return "⚠️ **AI Summary Unavailable:** The `GEMINI_API_KEY` is not set in the Space secrets. Please ask the Space owner to add it."
+
     try:
         llm = ChatGoogleGenerativeAI(model="gemini-1.5-flash", google_api_key=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']}'**."
+
         prompt = f"""
-        You are a world-class marketing strategist. Analyze the following data for the website {url} and provide a concise, actionable summary.
-        ## On-Page SEO Analysis:
-        - **Title:** {seo_data['title']}
-        - **Meta Description:** {seo_data['description']}
-        - **H1 Headings:** {', '.join(seo_data['headings']['h1'])}
-        - **Page Word Count:** {seo_data['word_count']}
-        ## Technology Stack:
-        - {', '.join(tech_data)}
-        ## Your Analysis:
-        Provide a 3-bullet point summary covering:
-        1.  **Their Primary Goal:** What is this page trying to achieve based on its language and structure?
-        2.  **Their Target Audience:** Who are they talking to?
-        3.  **A Key Strategic Insight:** What is one clever thing they are doing, or one major missed opportunity?
+        As a world-class marketing strategist, analyze the data for the website `{url}` and provide a concise, actionable summary in markdown format.
+
+        **On-Page Focus:** Their primary H1 heading is "{seo_data['headings']['h1'][0] if seo_data['headings']['h1'] else 'N/A'}".
+        **Paid Strategy:** {ads_summary}
+
+        Based on this, provide your **Strategic Teardown**:
+        - **🎯 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?
         """
         response = llm.invoke([HumanMessage(content=prompt)])
         return response.content
     except Exception as e:
-        return f"Could not generate AI summary. The API call failed: {e}"
+        return f"⚠️ **AI Summary Failed:** The API call could not be completed. Error: {e}"
+
+# --- Main Orchestrator ---
 
-def competitor_teardown(url):
+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
+    yield list(outputs.values())
+
+    # --- 1. Data Fetching ---
     if not url.startswith(('http://', 'https://')):
         url = 'https://' + url
-    yield "Fetching website...", None, None, None, None, gr.Button("Analyzing...", interactive=False)
+    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`."
+        outputs["btn"] = gr.Button("Analyze", interactive=True)
+        yield list(outputs.values())
+        return
+    domain = domain_match.group(1)
+
     html = fetch_html(url)
     if not html:
-        yield "Failed to fetch URL. Please check the address and try again.", None, None, None, None, gr.Button("Analyze", interactive=True)
+        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')
-    yield "Analyzing SEO & Tech...", None, None, None, None, gr.Button("Analyzing...", interactive=False)
+
+    # --- 2. Run All Analyses ---
     seo_data = analyze_onpage_seo(soup)
-    tech_data = analyze_tech_stack(soup, html)
-    yield "Generating AI Summary...", None, None, None, None, gr.Button("Analyzing...", interactive=False)
-    ai_summary = generate_ai_summary(url, seo_data, tech_data)
-    seo_md = f"""
-    ### 👑 SEO & Content
+    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'])}
+    | **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
     """
-    tech_md = "### stack Tech Stack\n\n" + "\n".join([f"- `{t}`" for t in tech_data])
-    yield ai_summary, seo_md, tech_md, "Analysis Complete", "More features coming soon!", gr.Button("Analyze", interactive=True)
+    
+    outputs["tech"] = "### ⚙️ Technology Stack\n\n" + "\n".join([f"- `{t}`" for t in tech_data])
 
-# --- Gradio UI Definition (Unchanged) ---
+    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"
+        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 ---
 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.*")
+
     with gr.Row():
         url_input = gr.Textbox(label="Enter Competitor URL", placeholder="e.g., notion.so or mailchimp.com", scale=4)
         submit_btn = gr.Button("Analyze", variant="primary", scale=1)
-    with gr.Tabs():
-        with gr.TabItem("🧠 AI Summary"):
-            summary_output = gr.Markdown("Your AI-powered strategic summary will appear here.")
-        with gr.TabItem("🔍 On-Page SEO"):
+
+    with gr.Tabs() as tabs:
+        with gr.TabItem("🧠 AI Summary", id=0):
+            summary_output = gr.Markdown()
+        with gr.TabItem("🔍 On-Page SEO", id=1):
             seo_output = gr.Markdown()
-        with gr.TabItem("⚙️ Tech Stack"):
+        with gr.TabItem("⚙️ Tech Stack", id=2):
             tech_output = gr.Markdown()
-        with gr.TabItem("📢 Ads & Keywords (Coming Soon)"):
+        with gr.TabItem("📢 Ads & Keywords", id=3):
             ads_output = gr.Markdown()
-        with gr.TabItem("Social Presence (Coming Soon)"):
-            social_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=[summary_output, seo_output, tech_output, ads_output, social_output, submit_btn]
+        outputs=outputs_list
     )
+    
     gr.Markdown("---")
-    gr.Markdown("### Ready for More? \n Get unlimited reports, save projects, and export to PDF with our Pro plan. \n **[🚀 Launching Soon - Sign up on Gumroad!](https://gumroad.com/)**")
+    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/)**")
 
-# --- THE FIX: Launch the app within a main block ---
-# This tells the Python interpreter that this is the main program to run
-# and it should wait here, keeping the server alive.
 if __name__ == "__main__":
     demo.launch()