Update seo_analyzer_ui.py

seo_analyzer_ui.py

@@ -1,6 +1,7 @@
 """
-SEO Analyzer UI with Auto Ad Generator Tab
+SEO Analyzer UI using Gradio, Web Crawler, and OpenAI
 """
+
 import gradio as gr
 import logging
 import json
@@ -18,9 +19,6 @@ from concurrent.futures import ThreadPoolExecutor
 from datetime import datetime
 import tempfile
 
-from bs4 import BeautifulSoup
-import requests
-
 from crawler import Crawler
 from frontier import URLFrontier
 from models import URL, Page
@@ -30,6 +28,7 @@ from dotenv import load_dotenv, find_dotenv
 
 load_dotenv(find_dotenv())
 
+# Check if we're in deployment mode (e.g., Hugging Face Spaces)
 IS_DEPLOYMENT = os.getenv('DEPLOYMENT', 'false').lower() == 'true'
 
 # Custom CSS for better styling
@@ -39,26 +38,22 @@ CUSTOM_CSS = """
     margin: auto;
     padding: 20px;
 }
-
 .header {
     text-align: center;
     margin-bottom: 2rem;
 }
-
 .header h1 {
     color: #2d3748;
     font-size: 2.5rem;
     font-weight: 700;
     margin-bottom: 1rem;
 }
-
 .header p {
     color: #4a5568;
     font-size: 1.1rem;
     max-width: 800px;
     margin: 0 auto;
 }
-
 .input-section {
     background: #f7fafc;
     border-radius: 12px;
@@ -66,7 +61,6 @@ CUSTOM_CSS = """
     margin-bottom: 24px;
     box-shadow: 0 2px 4px rgba(0,0,0,0.1);
 }
-
 .analysis-section {
     background: white;
     border-radius: 12px;
@@ -74,7 +68,6 @@ CUSTOM_CSS = """
     margin-top: 24px;
     box-shadow: 0 2px 4px rgba(0,0,0,0.1);
 }
-
 .log-section {
     font-family: monospace;
     background: #1a202c;
@@ -83,7 +76,6 @@ CUSTOM_CSS = """
     border-radius: 8px;
     margin-top: 24px;
 }
-
 /* Custom styling for inputs */
 .input-container {
     background: white;
@@ -91,14 +83,12 @@ CUSTOM_CSS = """
     border-radius: 8px;
     margin-bottom: 16px;
 }
-
 /* Custom styling for the slider */
 .slider-container {
     padding: 12px;
     background: white;
     border-radius: 8px;
 }
-
 /* Custom styling for buttons */
 .primary-button {
     background: #4299e1 !important;
@@ -108,46 +98,38 @@ CUSTOM_CSS = """
     font-weight: 600 !important;
     transition: all 0.3s ease !important;
 }
-
 .primary-button:hover {
     background: #3182ce !important;
     transform: translateY(-1px) !important;
 }
-
 /* Markdown output styling */
 .markdown-output {
     font-family: system-ui, -apple-system, sans-serif;
     line-height: 1.6;
 }
-
 .markdown-output h1 {
     color: #2d3748;
     border-bottom: 2px solid #e2e8f0;
     padding-bottom: 0.5rem;
 }
-
 .markdown-output h2 {
     color: #4a5568;
     margin-top: 2rem;
 }
-
 .markdown-output h3 {
     color: #718096;
     margin-top: 1.5rem;
 }
-
 /* Progress bar styling */
 .progress-bar {
     height: 8px !important;
     border-radius: 4px !important;
     background: #ebf8ff !important;
 }
-
 .progress-bar-fill {
     background: #4299e1 !important;
     border-radius: 4px !important;
 }
-
 /* Add some spacing between sections */
 .gap {
     margin: 2rem 0;
@@ -434,10 +416,8 @@ class SEOAnalyzer:
             # Format the results
             formatted_analysis = f"""
 # SEO Analysis Report for {domain}
-
 ## Overall Analysis
 {overall_analysis}
-
 ## Page-Specific Analyses
 """
             for page_analysis in page_analyses:
@@ -494,16 +474,13 @@ class SEOAnalyzer:
             # Create analysis prompt
             prompt = f"""
 You are an expert SEO consultant. Analyze this website's SEO based on the crawled data:
-
 {json.dumps(site_overview, indent=2)}
-
 Provide a comprehensive SEO analysis including:
 1. Overall site structure and navigation
 2. Common SEO issues across pages
 3. Content quality and optimization
 4. Technical SEO recommendations
 5. Priority improvements
-
 Format your response in Markdown.
 """
             
@@ -533,17 +510,14 @@ Format your response in Markdown.
                 # Create page analysis prompt
                 prompt = f"""
 Analyze this page's SEO:
-
 URL: {page['url']}
 Metadata: {json.dumps(page['metadata'], indent=2)}
-
 Provide specific recommendations for:
 1. Title and meta description
 2. Heading structure
 3. Content optimization
 4. Internal linking
 5. Technical improvements
-
 Format your response in Markdown.
 """
                 
@@ -609,7 +583,6 @@ def create_ui() -> gr.Interface:
     # Create markdown content for the about section
     about_markdown = """
     # 🔍 SEO Analyzer Pro
-
     Analyze your website's SEO performance using advanced crawling and AI technology.
     
     ### Features:
@@ -695,284 +668,15 @@ def create_ui() -> gr.Interface:
 
     return iface
 
-# ----- SEO Analyzer UI (as a function) -----
-def seo_analyzer_ui():
-    def analyze(url: str, api_key: str, max_pages: int, progress: gr.Progress = gr.Progress()) -> Tuple[str, str]:
-        try:
-            analyzer = SEOAnalyzer(api_key)
-            analysis, _, logs = analyzer.analyze_website(url, max_pages, progress)
-            log_output = ""
-            while not analyzer.log_queue.empty():
-                try:
-                    log_output += analyzer.log_queue.get_nowait() + "\n"
-                except queue.Empty:
-                    break
-            progress(1.0, "Analysis complete")
-            return analysis, log_output
-        except Exception as e:
-            error_msg = f"Error: {str(e)}"
-            logger.error(error_msg)
-            return error_msg, error_msg
-
-    about_markdown = """
-    # 🔍 SEO Analyzer Pro
-    Analyze your website's SEO performance using advanced crawling and AI technology.
-    ...
-    """
-    with gr.Blocks() as seo_tab:
-        gr.Markdown(about_markdown)
-        with gr.Row():
-            with gr.Column(scale=2):
-                with gr.Group(elem_classes="input-section"):
-                    gr.Markdown("### 📝 Enter Website Details")
-                    url_input = gr.Textbox(
-                        label="Website URL",
-                        placeholder="https://example.com",
-                        elem_classes="input-container",
-                        info="Enter the full URL of the website you want to analyze (e.g., https://example.com)"
-                    )
-                    api_key = gr.Textbox(
-                        label="OpenAI API Key",
-                        placeholder="sk-...",
-                        type="password",
-                        elem_classes="input-container",
-                        info="Your OpenAI API key is required for AI-powered analysis. Keep this secure!"
-                    )
-                    max_pages = gr.Slider(
-                        minimum=1,
-                        maximum=50,
-                        value=10,
-                        step=1,
-                        label="Maximum Pages to Crawl",
-                        elem_classes="slider-container",
-                        info="Choose how many pages to analyze. More pages = more comprehensive analysis but takes longer"
-                    )
-                    analyze_btn = gr.Button(
-                        "🔍 Analyze Website",
-                        elem_classes="primary-button"
-                    )
-
-        with gr.Row():
-            with gr.Column():
-                with gr.Group(elem_classes="analysis-section"):
-                    gr.Markdown("### 📊 Analysis Results")
-                    analysis_output = gr.Markdown(
-                        label="SEO Analysis",
-                        elem_classes="markdown-output"
-                    )
-        with gr.Row():
-            with gr.Column():
-                with gr.Group(elem_classes="log-section"):
-                    gr.Markdown("### 📋 Process Logs")
-                    logs_output = gr.Textbox(
-                        label="Logs",
-                        lines=10,
-                        elem_classes="log-output"
-                    )
-        analyze_btn.click(
-            fn=analyze,
-            inputs=[url_input, api_key, max_pages],
-            outputs=[analysis_output, logs_output],
-        )
-    return seo_tab
-
-# ---- Auto Ad Generator UI as a function ----
-def auto_ad_generator_ui():
-    openai.api_key = os.getenv("OPENAI_API_KEY")
-
-    def extract_text_from_url(url):
-        try:
-            resp = requests.get(url, timeout=30, headers={
-                "User-Agent": "Mozilla/5.0 (compatible; Bot/1.0)"
-            })
-            soup = BeautifulSoup(resp.content, "html.parser")
-            candidates = soup.find_all(['h1','h2','h3','h4','p','span','li'])
-            text = ' '.join([c.get_text(strip=True) for c in candidates])
-            text = text[:4000]
-            if len(text) < 100:
-                raise ValueError("Could not extract enough content (site may require JavaScript). Please enter keywords manually.")
-            return text
-        except Exception as e:
-            raise ValueError(f"URL extraction error: {e}")
-
-    def extract_keywords(text):
-        prompt = f"""
-        Extract up to 10 concise, relevant SEO keywords suitable for an automotive advertisement from the following content:
-        {text}
-        Keywords:
-        """
-        response = openai.ChatCompletion.create(
-            model="gpt-4",
-            messages=[{"role": "user", "content": prompt}],
-            temperature=0.6,
-            max_tokens=100
-        )
-        output = response.choices[0].message.content.strip()
-        if ',' in output:
-            keywords = output.split(',')
-        else:
-            keywords = output.split('\n')
-        return [kw.strip() for kw in keywords if kw.strip()]
-
-    def generate_ad_copy(platform, keywords):
-        prompt = f"""
-        Create a compelling, SEO-optimized {platform} ad using these keywords: {', '.join(keywords)}. 
-        Include a clear and enticing call-to-action.
-        """
-        response = openai.ChatCompletion.create(
-            model="gpt-4",
-            messages=[{"role": "user", "content": prompt}],
-            temperature=0.7,
-            max_tokens=300
-        )
-        return response.choices[0].message.content.strip()
-
-    def generate_ad_image(keywords):
-        kw_str = ", ".join(keywords)
-        image_prompt = (
-            f"High-quality, photorealistic automotive ad photo of a luxury car. "
-            f"Clean background, professional lighting, stylish dealership setting. "
-            f"Keywords: {kw_str}. Room for text overlay, wide format, visually appealing."
-        )
-        response = openai.Image.create(
-            prompt=image_prompt,
-            n=1,
-            size="512x512"
-        )
-        image_url = response["data"][0]["url"]
-        img_data = requests.get(image_url).content
-        img_file = "generated_ad_image.png"
-        with open(img_file, "wb") as f:
-            f.write(img_data)
-        return img_file
-
-    def platform_html(platform, ad_text):
-        if platform == "Facebook":
-            color = "#1877F2"
-            icon = "🌐"
-        elif platform == "Instagram":
-            color = "linear-gradient(90deg, #f58529 0%, #dd2a7b 50%, #8134af 100%)"
-            icon = "📸"
-        elif platform == "X (Twitter)":
-            color = "#14171A"
-            icon = "🐦"
-        else: # Google Search
-            color = "#4285F4"
-            icon = "🔍"
-
-        if platform == "Instagram":
-            content = f"""
-            <div style="background: {color}; padding: 2px; border-radius: 12px; margin-bottom:16px;">
-                <div style="background: white; color: #333; padding: 18px 20px; border-radius: 10px;">
-                    <span style="font-size: 1.5em;">{icon} <b>{platform}</b></span>
-                    <div style="margin-top: 12px; font-size: 1.1em; line-height:1.6;">{ad_text}</div>
-                </div>
-            </div>
-            """
-        else:
-            content = f"""
-            <div style="background: {color}; color: white; padding: 18px 20px; border-radius: 12px; margin-bottom:16px; min-height: 120px;">
-                <span style="font-size: 1.5em;">{icon} <b>{platform}</b></span>
-                <div style="margin-top: 12px; font-size: 1.1em; line-height:1.6;">{ad_text}</div>
-            </div>
-            """
-        return content
-
-    def main_workflow(input_mode, url_or_keywords):
-        error = None
-        keywords = []
-        ad_copies = {}
-        image_path = None
-
-        if input_mode == "URL":
-            try:
-                text = extract_text_from_url(url_or_keywords)
-                keywords = extract_keywords(text)
-            except Exception as e:
-                return None, None, None, f"{e}"
-        else:
-            keywords = [kw.strip() for kw in url_or_keywords.split(",") if kw.strip()]
-            if not keywords:
-                return None, None, None, "Please provide at least one keyword."
-        # Generate ad copies
-        platforms = ["Facebook", "Instagram", "X (Twitter)", "Google Search"]
-        for platform in platforms:
-            ad_copies[platform] = generate_ad_copy(platform, keywords)
-        # Generate image
-        try:
-            image_path = generate_ad_image(keywords)
-        except Exception as e:
-            error = f"Image generation error: {e}"
-
-        # Save ads to txt
-        output_txt = "generated_ads.txt"
-        with open(output_txt, "w", encoding="utf-8") as f:
-            for platform, content in ad_copies.items():
-                f.write(f"--- {platform} Ad Copy ---\n{content}\n\n")
-        return keywords, ad_copies, image_path, error
-
-    def run_space(input_mode, url, keywords):
-        url_or_keywords = url if input_mode == "URL" else keywords
-        keywords, ad_copies, image_path, error = main_workflow(input_mode, url_or_keywords)
-        ad_previews = ""
-        if ad_copies:
-            for platform, ad in ad_copies.items():
-                ad_previews += platform_html(platform, ad)
-        return (
-            keywords,
-            ad_previews,
-            image_path,
-            "generated_ads.txt" if ad_copies else None,
-            error
-        )
-
-    with gr.Blocks() as ad_tab:
-        gr.Markdown("# 🚗 Auto Ad Generator\nPaste a car listing URL **or** enter your own keywords, then preview AI-generated ads for each social media platform, plus an auto-generated image!")
-        input_mode = gr.Radio(["URL", "Keywords"], value="URL", label="Input Type")
-        url_input = gr.Textbox(label="Listing URL", placeholder="https://www.cars.com/listing/...", visible=True)
-        kw_input = gr.Textbox(label="Manual Keywords (comma separated)", placeholder="e.g. BMW, used car, sunroof", visible=False)
-        submit_btn = gr.Button("Generate Ads")
-
-        gr.Markdown("## Keywords")
-        kw_out = gr.JSON(label="Extracted/Provided Keywords")
-
-        gr.Markdown("## Ad Copy Previews")
-        ad_out = gr.HTML(label="Ad Copy Preview")
-
-        gr.Markdown("## Generated Ad Image")
-        img_out = gr.Image(label="Generated Ad Image", type="filepath")
-
-        gr.Markdown("## Download Ad Copies")
-        file_out = gr.File(label="Download TXT")
-
-        err_out = gr.Textbox(label="Errors", interactive=False)
-
-        def show_hide_fields(choice):
-            return (
-                gr.update(visible=choice == "URL"),
-                gr.update(visible=choice == "Keywords"),
-            )
-
-        input_mode.change(show_hide_fields, input_mode, [url_input, kw_input])
-
-        submit_btn.click(
-            run_space,
-            inputs=[input_mode, url_input, kw_input],
-            outputs=[kw_out, ad_out, img_out, file_out, err_out]
-        )
-    return ad_tab
-
-# ---- Main App: Two Tabs ----
 if __name__ == "__main__":
+    # Create base storage directory if it doesn't exist
     os.makedirs(config.STORAGE_PATH, exist_ok=True)
-    with gr.Blocks(css=CUSTOM_CSS) as demo:
-        with gr.Tab("SEO Analyzer"):
-            seo_analyzer_ui()
-        with gr.Tab("Auto Ad Generator"):
-            auto_ad_generator_ui()
-    demo.launch(
+    
+    # Create and launch UI
+    ui = create_ui()
+    ui.launch(
         share=False,
         server_name="0.0.0.0",
         show_api=False,
         show_error=True,
-    )
+    )