This page should be picked as English because the Open Graph locale is present and no stronger language signal is available. The body is intentionally long enough so that the page is not treated as thin content and the test stays focused on language extraction only.
The crawler should store both the detected language code and the source that produced it. That makes it easier to debug why a page was assigned to a particular language branch inside the audit.
Tento doplňkový odstavec je zde jen proto, aby stránka měla dostatečnou délku textu a zůstala použitelným izolovaným testem bez nechtěné penalizace za příliš krátký obsah.