ingest: suppress Google-news shared placeholder image (googleusercontent) -> null -> category cover fallback

Root cause: Google News RSS items carry no enclosures; pipeline fell back to og:image from the news.google.com interstitial, which serves ONE shared Google-branded tile (lh3.googleusercontent.com/J6_coFbo...) for every article. 601 rows shared that URL (maple 243 / finance 358). New image-guard blocks *.googleusercontent.com images at ingest; 16/16 regression tests; prod DBs migrated tile->NULL; images rebuilt technews:10 finance:4 maple-brief:11 and deployed.
2026-08-18 21:30:23 -04:00
parent 86a77b4907
commit 1e7e951c91
4 changed files with 172 additions and 1 deletions
@@ -0,0 +1,99 @@
+/**
+ * Placeholder-image guard tests (offline — no network).
+ *
+ * Contract under test (src/lib/ingest/image-guard.ts):
+ *  - the ONE shared Google-branded tile observed in production (exact URL
+ *    from the 2026-08-18 finance/maple cleanup, lh3.googleusercontent.com
+ *    /J6_coFbogxh...) is ALWAYS blocked, in every variant (resize suffixes,
+ *    host rotation lh2/lh3/lh4, case, extra query strings)
+ *  - real publisher article photos on googleusercontent-lookalike or normal
+ *    publisher hosts are NEVER blocked (must-pass: false positives here
+ *    would strip legitimate images)
+ *
+ * Runner: `tsx tests/image-guard.test.ts` -> exit 1 on failure.
+ */
+import assert from 'node:assert/strict';
+
+import { isPlaceholderImage } from '@/lib/ingest/image-guard';
+
+let passed = 0;
+let failed = 0;
+
+function check(name: string, fn: () => void): void {
+  try {
+    fn();
+    passed += 1;
+    console.log(`ok ${passed} - ${name}`);
+  } catch (e) {
+    failed += 1;
+    console.error(`not ok ${passed + failed} - ${name}`);
+    console.error(String((e as Error).stack ?? e));
+  }
+}
+
+// The exact shared tile URL stored on 601 production rows (2026-08-18).
+const TILE =
+  'https://lh3.googleusercontent.com/J6_coFbogxhRI9iM864NL_liGXvsQp2Aups' +
+  'Kei7z0cNNfDvGUmWUy20nuUhkREQyrpY4bEeIBuc=s0-w300';
+
+check('MUST-BLOCK: exact prod shared tile url', () => {
+  assert.equal(isPlaceholderImage(TILE).blocked, true);
+});
+check('MUST-BLOCK: -rw resize variant', () => {
+  assert.equal(isPlaceholderImage(`${TILE}-rw`).blocked, true);
+});
+check('MUST-BLOCK: lh4 host rotation', () => {
+  assert.equal(isPlaceholderImage('https://lh4.googleusercontent.com/xyz123=s900').blocked, true);
+});
+check('MUST-BLOCK: lh2 host rotation', () => {
+  assert.equal(isPlaceholderImage('https://lh2.googleusercontent.com/xyz123').blocked, true);
+});
+check('MUST-BLOCK: uppercase host', () => {
+  assert.equal(isPlaceholderImage('https://LH3.GOOGLEUSERCONTENT.COM/J6_coFbo=s0').blocked, true);
+});
+check('MUST-BLOCK: extra query string / fragment', () => {
+  assert.equal(isPlaceholderImage(TILE + '&dummy=1#f').blocked, true);
+});
+check('MUST-BLOCK: unknown key (not the known tile) still blocked by host rule', () => {
+  assert.equal(
+    isPlaceholderImage('https://lh3.googleusercontent.com/someOtherKeyWq9aB=s64').blocked,
+    true,
+  );
+});
+check('MUST-BLOCK: whitespace-wrapped url', () => {
+  assert.equal(isPlaceholderImage(`  ${TILE}  `).blocked, true);
+});
+
+check('MUST-PASS: wired photo (publisher CDN)', () => {
+  assert.equal(
+    isPlaceholderImage('https://media.wired.com/photos/6a84a2c1/191:100/w_1280/c_limit/x.jpg').blocked,
+    false,
+  );
+});
+check('MUST-PASS: cnbcfm photo (publisher CDN)', () => {
+  assert.equal(isPlaceholderImage('https://media.cnbcfm.com/i/2026/08/robo.jpg').blocked, false);
+});
+check('MUST-PASS: google.com logo asset (NOT googleusercontent)', () => {
+  assert.equal(isPlaceholderImage('https://www.google.com/logos/2026/xx512.png').blocked, false);
+});
+check('MUST-PASS: example.com path segment merely containing the host string', () => {
+  assert.equal(
+    isPlaceholderImage('https://cdn.example.com/googleusercontent.com/img.png').blocked,
+    false,
+  );
+});
+check('MUST-PASS: null', () => {
+  assert.equal(isPlaceholderImage(null).blocked, false);
+});
+check('MUST-PASS: undefined', () => {
+  assert.equal(isPlaceholderImage(undefined).blocked, false);
+});
+check('MUST-PASS: empty string', () => {
+  assert.equal(isPlaceholderImage('').blocked, false);
+});
+check('MUST-PASS: not a URL', () => {
+  assert.equal(isPlaceholderImage('not a url').blocked, false);
+});
+
+console.log(`\nimage-guard: ${passed} passed, ${failed} failed`);
+if (failed > 0) process.exit(1);