Use BBC's clean image variant (cpsprodpb) instead of the branded one

BBC's og:image comes from the "branded_news" CDN path with a "BBC NEWS" logo
baked into the picture (shows as "…EWS" once the hero crops it). The identical
photo is served under "cpsprodpb" with no logo, so rewrite branded_news →
cpsprodpb. Best of both: full-resolution hero, no burned-in branding. Re-enriched
recent briefs so live images swap over. 99 tests pass.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

goodnews/enrich.py

@@ -58,6 +58,19 @@ def _is_generic_image(url: str) -> bool:
     return any(marker in lowered for marker in _GENERIC_IMAGE_MARKERS)
 
 
+def _prefer_unbranded(url: str) -> str:
+    """Swap BBC's logo-branded image variant for its clean one.
+
+    BBC's og:image is served from the "branded_news" CDN path with a "BBC NEWS"
+    logo baked into the picture (it shows as "…EWS" once the hero crops it). The
+    identical photo is served under "cpsprodpb" with no logo, so prefer that — a
+    clean hero at the same full resolution.
+    """
+    if "ichef.bbci.co.uk" in url and "/branded_news/" in url:
+        return url.replace("/branded_news/", "/cpsprodpb/")
+    return url
+
+
 def _attr(tag: bytes, name: bytes) -> bytes | None:
     m = re.search(name + rb"""\s*=\s*["']([^"']*)["']""", tag, re.IGNORECASE)
     return m.group(1) if m else None
@@ -73,9 +86,9 @@ def og_image_from_html(html: bytes) -> str | None:
             if not content:
                 continue
             image = canonicalize_url(content.decode("utf-8", "replace"))
-            # Skip branded/generic share images; keep scanning for a real one.
+            # Skip generic placeholders; keep scanning for a real one.
             if image and not _is_generic_image(image):
-                return image
+                return _prefer_unbranded(image)
     return None
 
 

tests/test_enrich.py

@@ -77,8 +77,9 @@ def test_rejects_generic_share_images():
     assert og_image_from_html(b'<meta property="og:image" content="https://x.com/og-default.jpg">') is None
     # a real article image comes through
     assert og_image_from_html(b'<meta property="og:image" content="https://x.com/real-photo.jpg">') == "https://x.com/real-photo.jpg"
-    # BBC's branded_news path is a REAL photo path, not a placeholder — keep it
+    # BBC's branded_news path is a real photo (logo baked in) — keep it, but swap
-    assert og_image_from_html(b'<meta property="og:image" content="https://ichef.bbci.co.uk/news/1024/branded_news/x.jpg">') == "https://ichef.bbci.co.uk/news/1024/branded_news/x.jpg"
+    # to the clean cpsprodpb variant so the hero isn't branded
+    assert og_image_from_html(b'<meta property="og:image" content="https://ichef.bbci.co.uk/news/1024/branded_news/x.jpg">') == "https://ichef.bbci.co.uk/news/1024/cpsprodpb/x.jpg"
     # if the first og is a generic placeholder but a later one is real, take the real one
     html = (b'<meta property="og:image" content="https://x.com/og-default.jpg">'
             b'<meta property="og:image" content="https://x.com/article.jpg">')