web: Entitaeten im JSON-LD verknuepft, News-Sitemap und llms.txt (Roadmap 2.1, 2.2, 2.7)

- App\Support\StructuredData: Portal-Organization als Publisher mit @id
  und Logo (ImageObject), Newsroom-@id; NewsArticle.author auf der
  Meldung und Organization auf der Newsroom-Seite beschreiben jetzt
  dieselbe Entitaet. Startseiten nutzen denselben Publisher.
- /sitemap-news.xml je Spoke (NewsSitemapController): im Archiv
  sichtbare Meldungen der letzten 48 h, max. 1.000, news:publication
  mit Portalname und Sprache; 15 min gecacht, mit den anderen Sitemaps
  geleert; in robots.txt der Spokes. Hub liefert leere Liste.
- /llms.txt je Host (LlmsTxtController): sachliche Beschreibung,
  URL-Aufbau, aktive Kategorien, Seiten fuer Unternehmen, Rechtliches;
  Hub verweist auf beide Portale. Keine Aussagen ueber KI-Nennungen.
- Protokoll Teil 5 (Block 2) inkl. Betreiber-Schritten fuer das
  Publisher Center und Befund zu 2.5/E5; Roadmap 2.1/2.7 abgehakt.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

tests/Feature/Web/NewsSitemapTest.php

@@ -0,0 +1,74 @@
+<?php
+
+use App\Enums\Portal;
+use App\Models\Category;
+use App\Models\Company;
+use App\Models\PressRelease;
+use App\Support\WebCaches;
+use Illuminate\Support\Facades\Cache;
+use Tests\TestCase;
+
+/**
+ * Google-News-Sitemap je Spoke (Roadmap 2.2): nur Meldungen der letzten
+ * 48 Stunden, mit Publikationsname und Sprache.
+ */
+function newsSitemapRelease(Portal $portal, string $slug, array $attributes = []): PressRelease
+{
+    return PressRelease::factory()
+        ->published()
+        ->forPortal($portal)
+        ->for(Category::factory()->create(['portal' => $portal]))
+        ->for(Company::factory()->businessportal24()->create())
+        ->create(['slug' => $slug, ...$attributes]);
+}
+
+test('the news sitemap lists only releases of the last 48 hours of the own portal', function () {
+    /** @var TestCase $this */
+    newsSitemapRelease(Portal::Businessportal24, 'frisch', ['title' => 'Frische Meldung & mehr', 'published_at' => now()->subHours(3)]);
+    newsSitemapRelease(Portal::Both, 'beide-portale', ['published_at' => now()->subHours(20)]);
+    newsSitemapRelease(Portal::Businessportal24, 'zu-alt', ['published_at' => now()->subDays(3)]);
+    newsSitemapRelease(Portal::Presseecho, 'fremdes-portal', ['published_at' => now()->subHour()]);
+    newsSitemapRelease(Portal::Businessportal24, 'terminiert', ['published_at' => now()->addHour()]);
+
+    $response = $this->get('https://businessportal24.test/sitemap-news.xml')->assertSuccessful();
+
+    expect($response->headers->get('Content-Type'))->toContain('application/xml');
+
+    $xml = (string) $response->getContent();
+    expect($xml)
+        ->toContain('xmlns:news="http://www.google.com/schemas/sitemap-news/0.9"')
+        ->toContain('<loc>https://businessportal24.test/de/press-release/frisch</loc>')
+        ->toContain('<news:name>businessportal24</news:name>')
+        ->toContain('<news:language>de</news:language>')
+        ->toContain('<news:title>Frische Meldung &amp; mehr</news:title>')
+        ->toContain('<loc>https://businessportal24.test/de/press-release/beide-portale</loc>');
+
+    $this->assertStringNotContainsString('zu-alt', $xml);
+    $this->assertStringNotContainsString('fremdes-portal', $xml);
+    $this->assertStringNotContainsString('terminiert', $xml);
+});
+
+test('the hub serves an empty news sitemap and does not advertise it in robots.txt', function () {
+    /** @var TestCase $this */
+    newsSitemapRelease(Portal::Presseecho, 'nicht-im-hub', ['published_at' => now()->subHour()]);
+
+    $xml = (string) $this->get('https://pressekonto.test/sitemap-news.xml')->assertSuccessful()->getContent();
+    expect($xml)->toContain('<urlset');
+    $this->assertStringNotContainsString('<url>', $xml);
+
+    $robots = (string) $this->get('https://pressekonto.test/robots.txt')->getContent();
+    $this->assertStringNotContainsString('sitemap-news.xml', $robots);
+
+    $this->get('https://presseecho.test/robots.txt')
+        ->assertSee('Sitemap: https://presseecho.test/sitemap-news.xml', false);
+});
+
+test('the news sitemap is cached per host and cleared with the other sitemaps', function () {
+    /** @var TestCase $this */
+    $this->get('https://presseecho.test/sitemap-news.xml')->assertSuccessful();
+    expect(Cache::has('web.sitemap-news.presseecho.test'))->toBeTrue();
+
+    WebCaches::forgetSitemaps();
+
+    expect(Cache::has('web.sitemap-news.presseecho.test'))->toBeFalse();
+});