Update scraper.js

scraper.js

@@ -1,14 +1,14 @@
 const puppeteer = require('puppeteer-core');
 
-// --- PINTEREST LOGIC ---
+// --- PINTEREST SCRAPER ---
 async function scrapePinterest(browser, query) {
     const page = await browser.newPage();
     await page.setUserAgent('Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/121.0.0.0 Safari/537.36');
-
+    
     const searchUrl = `https://www.pinterest.com/search/pins/?q=${encodeURIComponent(query)}&rs=typed`;
     await page.goto(searchUrl, { waitUntil: 'networkidle2', timeout: 30000 });
 
-    // Scroll to find more pins
+    // Scroll to load more
     await page.evaluate(async () => {
         for (let i = 0; i < 3; i++) {
             window.scrollBy(0, window.innerHeight * 2);
@@ -36,65 +36,68 @@ async function scrapePinterest(browser, query) {
                 const hashMatch = rawUrl.match(/([a-f0-9]{32})/);
                 if (hashMatch) {
                     const h = hashMatch[1];
-                    // Print 720p URL
+                    // Output 720p version
                     process.stdout.write(`https://v1.pinimg.com/videos/mc/720p/${h.substring(0,2)}/${h.substring(2,4)}/${h.substring(4,6)}/${h}.mp4\n`);
                 }
             }
         } catch (e) {} 
         await pPage.close();
     }
+    await page.close();
 }
 
-// --- PEXELS LOGIC (FALLBACK) ---
+// --- PEXELS SCRAPER (FALLBACK) ---
 async function scrapePexels(browser, query, orientation) {
     const page = await browser.newPage();
     await page.setUserAgent('Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/121.0.0.0 Safari/537.36');
 
-    // Convert "Any" to empty, otherwise use portrait/landscape
-    let orientationParam = "";
-    if (orientation.toLowerCase() === "portrait") orientationParam = "&orientation=portrait";
-    if (orientation.toLowerCase() === "landscape") orientationParam = "&orientation=landscape";
+    // Pexels allows orientation filtering in the URL!
+    let pexelsUrl = `https://www.pexels.com/search/videos/${encodeURIComponent(query)}/`;
+    if (orientation === 'Portrait') pexelsUrl += '?orientation=portrait';
+    if (orientation === 'Landscape') pexelsUrl += '?orientation=landscape';
 
-    const searchUrl = `https://www.pexels.com/search/videos/${encodeURIComponent(query)}/?size=medium${orientationParam}`;
-    
     console.error(`Fallback: Searching Pexels for ${orientation}...`);
-    await page.goto(searchUrl, { waitUntil: 'networkidle2', timeout: 30000 });
+    await page.goto(pexelsUrl, { waitUntil: 'networkidle2', timeout: 30000 });
 
-    // Pexels puts direct MP4 links in <source> tags inside <article>
+    // Extract video links directly from the source tags
     const videoUrls = await page.evaluate(() => {
-        const sources = Array.from(document.querySelectorAll('article video source[type="video/mp4"]'));
+        const sources = Array.from(document.querySelectorAll('source[type="video/mp4"]'));
         return sources.map(s => s.src);
     });
 
-    // Remove duplicates and print
+    // Clean and print URLs
     const uniqueUrls = [...new Set(videoUrls)];
-    uniqueUrls.forEach(url => process.stdout.write(url + "\n"));
+    uniqueUrls.forEach(url => {
+        // Remove query params to get direct link
+        const cleanUrl = url.split('?')[0]; 
+        process.stdout.write(`${cleanUrl}\n`);
+    });
+
+    await page.close();
 }
 
 // --- MAIN CONTROLLER ---
 (async () => {
     const args = process.argv.slice(2);
-    const platform = args[0]; // 'pinterest' or 'pexels'
+    const mode = args[0]; // 'pinterest' or 'pexels'
     const query = args[1];
-    const orientation = args[2] || "Any";
+    const orientation = args[2] || 'Any';
 
-    let browser;
-    try {
-        browser = await puppeteer.launch({
-            executablePath: '/usr/bin/chromium',
-            headless: 'new',
-            args: ['--no-sandbox', '--disable-setuid-sandbox', '--disable-dev-shm-usage']
-        });
+    const browser = await puppeteer.launch({
+        executablePath: '/usr/bin/chromium',
+        headless: 'new',
+        args: ['--no-sandbox', '--disable-setuid-sandbox', '--disable-dev-shm-usage']
+    });
 
-        if (platform === 'pinterest') {
+    try {
+        if (mode === 'pinterest') {
             await scrapePinterest(browser, query);
-        } else if (platform === 'pexels') {
+        } else if (mode === 'pexels') {
             await scrapePexels(browser, query, orientation);
         }
-
-    } catch (error) {
-        console.error(error.message);
+    } catch (e) {
+        console.error(e);
     } finally {
-        if (browser) await browser.close();
+        await browser.close();
     }
 })();