From fcb8fb8c90ea0d8933264d855b9d342efdc01339 Mon Sep 17 00:00:00 2001
From: Rene Fichtmueller <renefichtmueller@MacStudio-von-Rene-8.local>
Date: Tue, 21 Apr 2026 06:36:12 +0200
Subject: [PATCH] fix(scraper): fall through to img fallback when og:image is
 generic/logo

Previously: if og:image existed (even as a Dell logo URL), page.evaluate() returned
early and the img fallback was never tried. Now: meta tags are extracted first, then
isGenericImage() is checked in Node.js, and the img fallback runs if meta image is null
or generic. This allows vendors like Dell (og:image = logo) to still get product images
via the DOM fallback.
---
 .../src/scrapers/switch-image-playwright.ts   | 54 +++++++++++--------
 1 file changed, 31 insertions(+), 23 deletions(-)

diff --git a/packages/scraper/src/scrapers/switch-image-playwright.ts b/packages/scraper/src/scrapers/switch-image-playwright.ts
index 89aa17e..fba0399 100644
--- a/packages/scraper/src/scrapers/switch-image-playwright.ts
+++ b/packages/scraper/src/scrapers/switch-image-playwright.ts
@@ -260,31 +260,39 @@ export async function fetchSwitchImagesPlaywright(targetVendorSlug?: string): Pr
         // Wait for page to settle (JS rendering)
         await page.waitForLoadState("networkidle", { timeout: 20_000 }).catch(() => {});
 
-        // Extract og:image
-        const imageUrl: string | null = await page.evaluate(() => {
-          const ogMeta = document.querySelector<HTMLMetaElement>('meta[property="og:image"]');
-          if (ogMeta?.content) return ogMeta.content;
-
-          const twitterMeta = document.querySelector<HTMLMetaElement>('meta[name="twitter:image"]');
-          if (twitterMeta?.content) return twitterMeta.content;
-
-          // Fallback: largest visible image that isn't a UI element.
-          // Deliberately broad — isGenericImage() will filter hero/banner/logo images afterward.
-          const imgs = Array.from(document.querySelectorAll<HTMLImageElement>("img"));
-          const skipPattern = /\/flags?\/|\/icons?\/|\/avatars?\/|social[-_]icon|favicon|spinner|loading/i;
-          const candidate = imgs
-            .filter((img) => {
-              const src = img.src || img.getAttribute("data-src") || "";
-              return src.startsWith("http") &&
-                /\.(jpg|jpeg|png|webp)/i.test(src) &&
-                img.naturalWidth >= 200 &&
-                img.naturalHeight >= 150 &&
-                !skipPattern.test(src);
-            })
-            .sort((a, b) => (b.naturalWidth * b.naturalHeight) - (a.naturalWidth * a.naturalHeight))[0];
-          return candidate?.src ?? null;
+        // Extract og:image / twitter:image meta tags.
+        // We DON'T filter generics here — we filter outside so the img fallback can still run
+        // even when og:image exists but is a logo/brand image (e.g. Dell, HPE).
+        const metaImageUrl: string | null = await page.evaluate(() => {
+          const og = document.querySelector<HTMLMetaElement>('meta[property="og:image"]')?.content;
+          if (og) return og;
+          const tw = document.querySelector<HTMLMetaElement>('meta[name="twitter:image"]')?.content;
+          return tw ?? null;
         });
 
+        // Use meta image if it passes the generic filter; otherwise fall through to img fallback.
+        let imageUrl: string | null = (metaImageUrl && !isGenericImage(metaImageUrl)) ? metaImageUrl : null;
+
+        if (!imageUrl) {
+          // Img fallback: largest visible image that isn't a UI element.
+          // Deliberately broad — isGenericImage() will filter hero/banner/logo images afterward.
+          imageUrl = await page.evaluate(() => {
+            const imgs = Array.from(document.querySelectorAll<HTMLImageElement>("img"));
+            const skipPattern = /\/flags?\/|\/icons?\/|\/avatars?\/|social[-_]icon|favicon|spinner|loading/i;
+            const candidate = imgs
+              .filter((img) => {
+                const src = img.src || img.getAttribute("data-src") || "";
+                return src.startsWith("http") &&
+                  /\.(jpg|jpeg|png|webp)/i.test(src) &&
+                  img.naturalWidth >= 200 &&
+                  img.naturalHeight >= 150 &&
+                  !skipPattern.test(src);
+              })
+              .sort((a, b) => (b.naturalWidth * b.naturalHeight) - (a.naturalWidth * a.naturalHeight))[0];
+            return candidate?.src ?? null;
+          });
+        }
+
         if (!imageUrl || isGenericImage(imageUrl)) {
           console.log(`  [MISS] ${data.vendorName} ${data.model} — no product image (${imageUrl?.slice(0, 60) ?? "null"})`);
           missed++;