SEO: host-aware sitemap.xml + dynamische robots.txt je Domain

Sitemap pro Host (1h-Cache): Hub listet Landing + Rechts-/Info-Seiten, Spokes zusaetzlich Kategorien (lokalisierte Slugs), oeffentliche Newsrooms und alle im Archiv sichtbaren PMs des Portals (explizit ohne PortalScope, forPortalValues). Suche und Marketing-Platzhalterseiten bleiben draussen. Statische public/robots.txt entfernt (haette die Route verschattet); robots.txt jetzt dynamisch mit Sitemap-Verweis und Disallow fuer /admin, /login, /register, Suche. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-03 12:34:12 +00:00 · 2026-07-03 12:34:12 +00:00 · e9643e620a
commit e9643e620a
parent b855cbf7c0
5 changed files with 251 additions and 2 deletions
--- a/app/Http/Controllers/Web/SitemapController.php
+++ b/app/Http/Controllers/Web/SitemapController.php
@ -0,0 +1,126 @@
+<?php
+
+namespace App\Http\Controllers\Web;
+
+use App\Http\Controllers\Controller;
+use App\Models\Category;
+use App\Models\Company;
+use App\Models\PressRelease;
+use App\Scopes\PortalScope;
+use App\Support\WebPortalContext;
+use Illuminate\Http\Request;
+use Illuminate\Http\Response;
+use Illuminate\Support\Facades\Cache;
+
+/**
+ * Host-aware XML-Sitemap: Der Hub listet Landing- und Rechts-/Info-Seiten,
+ * die Spokes (presseecho, businessportal24) zusätzlich Kategorien, Newsrooms
+ * und alle im Archiv sichtbaren Pressemitteilungen ihres Portals. Marketing-
+ * Platzhalterseiten (team, presse, karriere, partner) und die Suche bleiben
+ * bewusst draußen (R2.4/R3.2). Pro Host eine Stunde gecacht.
+ */
+class SitemapController extends Controller
+{
+    /** @var list<string> */
+    private const HUB_PAGES = [
+        'faq', 'kontakt', 'ueber-uns', 'hilfe', 'api', 'kuendigung',
+        'impressum', 'datenschutz', 'agb', 'cookies',
+    ];
+
+    /** @var list<string> */
+    private const PORTAL_PAGES = [
+        'preise', 'veroeffentlichen', 'newsrooms', 'faq', 'kontakt',
+        'ueber-uns', 'hilfe', 'api', 'impressum', 'datenschutz', 'agb', 'cookies',
+    ];
+
+    public function __invoke(Request $request): Response
+    {
+        $host = $request->getHost();
+        $base = rtrim($request->getSchemeAndHttpHost(), '/');
+        $isHub = str_contains($host, 'pressekonto');
+
+        $xml = Cache::remember('web.sitemap.'.$host, 3600, function () use ($base, $isHub): string {
+            $urls = $isHub ? $this->hubUrls($base) : $this->portalUrls($base);
+
+            return view('web.sitemap', ['urls' => $urls])->render();
+        });
+
+        return response($xml, 200, ['Content-Type' => 'application/xml; charset=UTF-8']);
+    }
+
+    /**
+     * @return list<array{loc: string, lastmod: ?string}>
+     */
+    private function hubUrls(string $base): array
+    {
+        $urls = [['loc' => $base.'/', 'lastmod' => null]];
+
+        foreach (self::HUB_PAGES as $page) {
+            $urls[] = ['loc' => $base.'/de/'.$page, 'lastmod' => null];
+        }
+
+        return $urls;
+    }
+
+    /**
+     * @return list<array{loc: string, lastmod: ?string}>
+     */
+    private function portalUrls(string $base): array
+    {
+        $context = WebPortalContext::fromRequest();
+
+        $urls = [['loc' => $base.'/de', 'lastmod' => null]];
+
+        foreach (self::PORTAL_PAGES as $page) {
+            $urls[] = ['loc' => $base.'/de/'.$page, 'lastmod' => null];
+        }
+
+        // Kategorien über die lokalisierten Slugs (CategoryTranslation).
+        $categorySlugs = Category::query()
+            ->where('is_active', true)
+            ->whereIn('portal', $context->portalValues)
+            ->with('translations')
+            ->get()
+            ->map(fn (Category $category) => $category->translations->firstWhere('locale', 'de')?->slug)
+            ->filter()
+            ->unique();
+
+        foreach ($categorySlugs as $slug) {
+            $urls[] = ['loc' => $base.'/de/category/'.$slug, 'lastmod' => null];
+        }
+
+        // Öffentliche Newsrooms: aktive Firmen mit mindestens einer im Archiv
+        // sichtbaren PM des Portals (gleiche Regel wie NewsroomsPageData).
+        $companies = Company::query()
+            ->where('is_active', true)
+            ->whereHas('pressReleases', fn ($query) => $query
+                ->withoutGlobalScope(PortalScope::class)
+                ->visibleInArchive()
+                ->forPortalValues($context->portalValues))
+            ->get(['slug', 'updated_at']);
+
+        foreach ($companies as $company) {
+            $urls[] = [
+                'loc' => $base.'/de/newsroom/'.$company->slug,
+                'lastmod' => $company->updated_at?->toAtomString(),
+            ];
+        }
+
+        // Alle im Archiv sichtbaren Pressemitteilungen des Portals.
+        $releases = PressRelease::query()
+            ->withoutGlobalScope(PortalScope::class)
+            ->visibleInArchive()
+            ->forPortalValues($context->portalValues)
+            ->orderByDesc('published_at')
+            ->get(['slug', 'published_at', 'updated_at']);
+
+        foreach ($releases as $release) {
+            $urls[] = [
+                'loc' => $base.'/de/press-release/'.$release->slug,
+                'lastmod' => ($release->updated_at ?? $release->published_at)?->toAtomString(),
+            ];
+        }
+
+        return $urls;
+    }
+}
--- a/public/robots.txt
+++ b/public/robots.txt
@ -1,2 +0,0 @@
-User-agent: *
-Disallow:
--- a/resources/views/web/sitemap.blade.php
+++ b/resources/views/web/sitemap.blade.php
@ -0,0 +1,11 @@
+{!! '<?xml version="1.0" encoding="UTF-8"?>' !!}
+<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
+@foreach ($urls as $url)
+    <url>
+        <loc>{{ $url['loc'] }}</loc>
+@if (! empty($url['lastmod']))
+        <lastmod>{{ $url['lastmod'] }}</lastmod>
+@endif
+    </url>
+@endforeach
+</urlset>
--- a/routes/web.php
+++ b/routes/web.php
@ -9,6 +9,7 @@ use App\Http\Controllers\Web\NewsroomDetailController;
 use App\Http\Controllers\Web\NewsroomsController;
 use App\Http\Controllers\Web\PreiseController;
 use App\Http\Controllers\Web\ReleaseDetailController;
+use App\Http\Controllers\Web\SitemapController;
 use App\Http\Controllers\Web\SucheController;
 use App\Http\Controllers\Web\TerminationController;
 use App\Http\Controllers\Web\VeroeffentlichenController;
@ -218,6 +219,31 @@ Route::prefix('{edition}')->whereIn('edition', SetEdition::EDITIONS)->group(func
    Route::view('cookies', 'web.cookies')->name('cookies');
 });

+// ---------------------------------------------------------------------------
+// SEO: Sitemap + robots.txt, host-aware (Hub vs. Spokes). Die frühere
+// statische public/robots.txt wurde entfernt — sie hätte diese Routen
+// verschattet (Webserver liefert existierende Dateien direkt aus).
+// ---------------------------------------------------------------------------
+Route::get('/sitemap.xml', SitemapController::class)->name('sitemap');
+
+Route::get('/robots.txt', function () {
+    $base = rtrim(request()->getSchemeAndHttpHost(), '/');
+
+    $lines = [
+        'User-agent: *',
+        'Disallow: /admin',
+        'Disallow: /login',
+        'Disallow: /register',
+        'Disallow: /de/suche',
+        'Disallow: /en/suche',
+        '',
+        'Sitemap: '.$base.'/sitemap.xml',
+        '',
+    ];
+
+    return response(implode("\n", $lines), 200, ['Content-Type' => 'text/plain; charset=UTF-8']);
+})->name('robots');
+
 // ---------------------------------------------------------------------------
 // Sprach-/Edition-unabhängige Routen (API-Doku).
 // ---------------------------------------------------------------------------
--- a/tests/Feature/Web/SitemapRobotsTest.php
+++ b/tests/Feature/Web/SitemapRobotsTest.php
@ -0,0 +1,88 @@
+<?php
+
+use App\Enums\Portal;
+use App\Models\Category;
+use App\Models\Company;
+use App\Models\PressRelease;
+use Tests\TestCase;
+
+test('robots.txt is served per host with the matching sitemap reference', function () {
+    /** @var TestCase $this */
+    $this->get('https://presseecho.test/robots.txt')
+        ->assertSuccessful()
+        ->assertHeader('Content-Type', 'text/plain; charset=UTF-8')
+        ->assertSee('Disallow: /admin', false)
+        ->assertSee('Sitemap: https://presseecho.test/sitemap.xml', false);
+
+    $this->get('https://pressekonto.test/robots.txt')
+        ->assertSuccessful()
+        ->assertSee('Sitemap: https://pressekonto.test/sitemap.xml', false);
+});
+
+test('the spoke sitemap lists static pages, categories, newsrooms and archive releases of its portal', function () {
+    /** @var TestCase $this */
+    $category = Category::factory()->create(['portal' => Portal::Presseecho, 'is_active' => true]);
+    $category->translations()->create(['locale' => 'de', 'name' => 'Technologie', 'slug' => 'technologie']);
+
+    $company = Company::factory()->presseecho()->create(['is_active' => true]);
+
+    PressRelease::factory()
+        ->published()
+        ->forPortal(Portal::Presseecho)
+        ->for($category)
+        ->for($company)
+        ->create(['slug' => 'ki-report', 'published_at' => now()->subDay()]);
+
+    // Fremd-Portal-PM darf nicht in der presseecho-Sitemap auftauchen.
+    $bpCategory = Category::factory()->create(['portal' => Portal::Businessportal24]);
+    PressRelease::factory()
+        ->published()
+        ->forPortal(Portal::Businessportal24)
+        ->for($bpCategory)
+        ->for(Company::factory()->businessportal24()->create())
+        ->create(['slug' => 'bp24-only', 'published_at' => now()->subDay()]);
+
+    $response = $this->get('https://presseecho.test/sitemap.xml')->assertSuccessful();
+
+    expect($response->headers->get('Content-Type'))->toContain('application/xml');
+
+    expect($response->getContent())
+        ->toContain('<loc>https://presseecho.test/de</loc>')
+        ->toContain('<loc>https://presseecho.test/de/preise</loc>')
+        ->toContain('<loc>https://presseecho.test/de/impressum</loc>')
+        ->toContain('<loc>https://presseecho.test/de/category/technologie</loc>')
+        ->toContain('<loc>https://presseecho.test/de/newsroom/'.$company->slug.'</loc>')
+        ->toContain('<loc>https://presseecho.test/de/press-release/ki-report</loc>')
+        ->not->toContain('bp24-only')
+        // Suche und Marketing-Platzhalterseiten bleiben draußen.
+        ->not->toContain('/de/suche')
+        ->not->toContain('/de/team');
+});
+
+test('the hub sitemap lists the landing and legal pages but no archive content', function () {
+    /** @var TestCase $this */
+    PressRelease::factory()
+        ->published()
+        ->forPortal(Portal::Presseecho)
+        ->for(Category::factory()->create(['portal' => Portal::Presseecho]))
+        ->for(Company::factory()->presseecho()->create())
+        ->create(['slug' => 'nicht-im-hub', 'published_at' => now()->subDay()]);
+
+    $response = $this->get('https://pressekonto.test/sitemap.xml')->assertSuccessful();
+
+    expect($response->getContent())
+        ->toContain('<loc>https://pressekonto.test/</loc>')
+        ->toContain('<loc>https://pressekonto.test/de/impressum</loc>')
+        ->toContain('<loc>https://pressekonto.test/de/kuendigung</loc>')
+        ->not->toContain('press-release')
+        ->not->toContain('newsroom');
+});
+
+test('marketing pages carry brand-aware titles instead of the hardcoded legacy brand', function () {
+    /** @var TestCase $this */
+    $response = $this->get('https://presseecho.test/de/team')->assertSuccessful();
+
+    expect($response->getContent())
+        ->toContain('<title>Unser Team – presseecho</title>')
+        ->not->toContain('Business Portal 24');
+});