Create scraper.js

scraper.js

@@ -0,0 +1,76 @@
+const puppeteer = require('puppeteer-core');
+
+async function getPinterestVideo(query) {
+    // 1. Launch Browser using the System Chromium installed via packages.txt
+    const browser = await puppeteer.launch({
+        executablePath: '/usr/bin/chromium', 
+        headless: 'new',
+        args: [
+            '--no-sandbox', 
+            '--disable-setuid-sandbox',
+            '--disable-dev-shm-usage',
+            '--window-size=1920,1080'
+        ]
+    });
+
+    try {
+        const page = await browser.newPage();
+        await page.setUserAgent('Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36');
+        
+        // 2. Search Pinterest
+        const searchUrl = `https://www.pinterest.com/search/pins/?q=${encodeURIComponent(query)}&rs=typed`;
+        await page.goto(searchUrl, { waitUntil: 'networkidle2', timeout: 20000 });
+
+        // 3. Find Pin URLs
+        const pinUrls = await page.evaluate(() => {
+            return Array.from(document.querySelectorAll('div[data-test-id="pinWrapper"] a'))
+                .map(a => a.href)
+                .filter(href => href.includes('/pin/'))
+                .slice(0, 10); // Limit to top 10 to be fast
+        });
+
+        // 4. Loop through pins to find a video
+        for (const pinUrl of pinUrls) {
+            try {
+                const pPage = await browser.newPage();
+                await pPage.goto(pinUrl, { waitUntil: 'domcontentloaded', timeout: 10000 });
+                const content = await pPage.content();
+                await pPage.close();
+
+                // Regex to find video file
+                const match = content.match(/https:\/\/v1\.pinimg\.com\/videos\/mc\/[^\s"']+/);
+                
+                if (match) {
+                    let rawUrl = match[0].replace(/\\u002F/g, '/');
+                    // Convert to 720p URL if possible
+                    if(rawUrl.includes('/720p/')) {
+                         // It's already high quality
+                    } else {
+                        // Attempt to reconstruct high quality URL logic
+                        const hashMatch = rawUrl.match(/([a-f0-9]{32})/);
+                        if (hashMatch) {
+                            const h = hashMatch[1];
+                            rawUrl = `https://v1.pinimg.com/videos/mc/720p/${h.substring(0,2)}/${h.substring(2,4)}/${h.substring(4,6)}/${h}.mp4`;
+                        }
+                    }
+                    
+                    console.log(rawUrl); // <--- PRINT ONLY THE URL
+                    await browser.close();
+                    return;
+                }
+            } catch (e) {
+                // Ignore individual pin errors
+            }
+        }
+        console.log("ERROR: No video found");
+    } catch (error) {
+        console.log("ERROR: " + error.message);
+    } finally {
+        if(browser) await browser.close();
+    }
+}
+
+// Get args from Python
+const args = process.argv.slice(2);
+const query = args[0] || "cats video";
+getPinterestVideo(query);