meta fix

Dockers/puppeteer-api/main.py

@@ -305,8 +305,73 @@ async def extract_seo(url: str, x_api_key: Optional[str] = Header(None)):
     try:
         print(f"Extracting SEO from: {decoded_url}")
 
-        # Use the safe_browser_operation function to perform the browser operation
+        # Define the operation to perform with the browser
-        result = await safe_browser_operation(decoded_url, asyncio.run)
+        async def seo_operation(page):
+            try:
+                response = await page.goto(decoded_url, waitUntil='networkidle2', timeout=30000)
+
+                # Extract SEO information
+                seo_data = await page.evaluate('''() => {
+                    const data = {
+                        title: document.title || '',
+                        description: '',
+                        canonical: '',
+                        h1: [],
+                        h2: [],
+                        images: 0,
+                        links: 0
+                    };
+
+                    // Get meta description
+                    const metaDescription = document.querySelector('meta[name="description"]');
+                    if (metaDescription) {
+                        data.description = metaDescription.getAttribute('content') || '';
+                    }
+
+                    // Get canonical link
+                    const canonicalLink = document.querySelector('link[rel="canonical"]');
+                    if (canonicalLink) {
+                        data.canonical = canonicalLink.getAttribute('href') || '';
+                    }
+
+                    // Get h1 tags
+                    document.querySelectorAll('h1').forEach(h1 => {
+                        const text = h1.innerText.trim();
+                        if (text) data.h1.push(text);
+                    });
+
+                    // Get h2 tags
+                    document.querySelectorAll('h2').forEach(h2 => {
+                        const text = h2.innerText.trim();
+                        if (text) data.h2.push(text);
+                    });
+
+                    // Count images
+                    data.images = document.querySelectorAll('img').length;
+
+                    // Count links
+                    data.links = document.querySelectorAll('a').length;
+
+                    return data;
+                }''')
+
+                result = {
+                    "status": "success",
+                    "url": decoded_url,
+                    "seo": seo_data
+                }
+
+                return result
+
+            except Exception as e:
+                print(f"Error during SEO extraction: {e}")
+                return {"status": "error", "url": decoded_url, "error": str(e)}
+
+        # Perform the operation
+        result = await safe_browser_operation(decoded_url, seo_operation)
+
+        # Save to cache
+        save_to_cache(decoded_url, "seo", result)
 
         return result
 
@@ -331,8 +396,63 @@ async def extract_meta_tags(url: str, x_api_key: Optional[str] = Header(None)):
     try:
         print(f"Extracting meta tags from: {decoded_url}")
 
-        # Use the safe_browser_operation function to perform the browser operation
+        # Define the operation to perform with the browser
-        result = await safe_browser_operation(decoded_url, asyncio.run)
+        async def meta_operation(page):
+            try:
+                response = await page.goto(decoded_url, waitUntil='networkidle2', timeout=30000)
+
+                # Extract all meta tags
+                meta_tags = await page.evaluate('''() => {
+                    const metas = Array.from(document.querySelectorAll('meta'));
+                    return metas.map(meta => {
+                        const attributes = {};
+                        Array.from(meta.attributes).forEach(attr => {
+                            attributes[attr.name] = attr.value;
+                        });
+                        return attributes;
+                    });
+                }''')
+
+                # Extract Open Graph tags
+                og_tags = await page.evaluate('''() => {
+                    const ogTags = {};
+                    document.querySelectorAll('meta[property^="og:"]').forEach(tag => {
+                        const property = tag.getAttribute('property');
+                        ogTags[property] = tag.getAttribute('content');
+                    });
+                    return ogTags;
+                }''')
+
+                # Extract Twitter card tags
+                twitter_tags = await page.evaluate('''() => {
+                    const twitterTags = {};
+                    document.querySelectorAll('meta[name^="twitter:"]').forEach(tag => {
+                        const name = tag.getAttribute('name');
+                        twitterTags[name] = tag.getAttribute('content');
+                    });
+                    return twitterTags;
+                }''')
+
+                result = {
+                    "status": "success",
+                    "url": decoded_url,
+                    "meta_tags": meta_tags,
+                    "open_graph": og_tags,
+                    "twitter_card": twitter_tags,
+                    "title": await page.title()
+                }
+
+                return result
+
+            except Exception as e:
+                print(f"Error during meta tag extraction: {e}")
+                return {"status": "error", "url": decoded_url, "error": str(e)}
+
+        # Perform the operation
+        result = await safe_browser_operation(decoded_url, meta_operation)
+
+        # Save to cache
+        save_to_cache(decoded_url, "meta", result)
 
         return result