From 81209125a10b564858fbdb72db4c060bf18fef89 Mon Sep 17 00:00:00 2001
From: myrmidex <myrmidex@myrmidex.net>
Date: Sun, 26 Apr 2026 14:52:40 +0200
Subject: [PATCH] 8 - Add UrlService with host extraction method

---
 app/Services/UrlService.php            |  35 +++++++++
 tests/Unit/Services/UrlServiceTest.php | 101 +++++++++++++++++++++++++
 2 files changed, 136 insertions(+)
 create mode 100644 app/Services/UrlService.php
 create mode 100644 tests/Unit/Services/UrlServiceTest.php

diff --git a/app/Services/UrlService.php b/app/Services/UrlService.php
new file mode 100644
index 0000000..287f8fa
--- /dev/null
+++ b/app/Services/UrlService.php
@@ -0,0 +1,35 @@
+<?php
+
+declare(strict_types=1);
+
+namespace App\Services;
+
+use Illuminate\Support\Uri;
+use InvalidArgumentException;
+
+class UrlService
+{
+    public function host(string $url): string
+    {
+        $uri = Uri::of($url);
+
+        $scheme = $uri->scheme();
+        if ($scheme === null || $scheme === '') {
+            throw new InvalidArgumentException("URL has no scheme: {$url}");
+        }
+        if (! in_array($scheme, ['http', 'https'], true)) {
+            throw new InvalidArgumentException("Invalid URL scheme: {$scheme}");
+        }
+
+        $host = $uri->host();
+        if ($host === null || $host === '') {
+            throw new InvalidArgumentException("URL has no host: {$url}");
+        }
+
+        if (filter_var(trim($host, '[]'), FILTER_VALIDATE_IP) !== false) {
+            throw new InvalidArgumentException("IP literal hosts not allowed: {$host}");
+        }
+
+        return strtolower($host);
+    }
+}
diff --git a/tests/Unit/Services/UrlServiceTest.php b/tests/Unit/Services/UrlServiceTest.php
new file mode 100644
index 0000000..0632766
--- /dev/null
+++ b/tests/Unit/Services/UrlServiceTest.php
@@ -0,0 +1,101 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Tests\Unit\Services;
+
+use App\Services\UrlService;
+use PHPUnit\Framework\Attributes\DataProvider;
+use Tests\TestCase;
+
+class UrlServiceTest extends TestCase
+{
+    private UrlService $service;
+
+    protected function setUp(): void
+    {
+        parent::setUp();
+
+        $this->service = new UrlService;
+    }
+
+    // -------------------------------------------------------------------------
+    // Happy path — simple URL
+    // -------------------------------------------------------------------------
+
+    public function test_extracts_host_from_simple_url(): void
+    {
+        $this->assertSame('example.com', $this->service->host('https://example.com'));
+    }
+
+    // -------------------------------------------------------------------------
+    // Path, query string, and fragment are ignored
+    // -------------------------------------------------------------------------
+
+    #[DataProvider('urlsWithNoise')]
+    public function test_extracts_host_ignoring_path_query_and_fragment(string $url, string $expectedHost): void
+    {
+        $this->assertSame($expectedHost, $this->service->host($url));
+    }
+
+    public static function urlsWithNoise(): array
+    {
+        return [
+            'path only' => ['https://example.com/some/path', 'example.com'],
+            'path and query' => ['https://example.com/page?q=hello&lang=en', 'example.com'],
+            'path, query, fragment' => ['https://example.com/page?q=1#section', 'example.com'],
+            'http scheme with path' => ['http://news.ycombinator.com/item?id=42', 'news.ycombinator.com'],
+        ];
+    }
+
+    // -------------------------------------------------------------------------
+    // Port number is stripped from the host
+    // -------------------------------------------------------------------------
+
+    public function test_strips_port_from_host(): void
+    {
+        $this->assertSame('example.com', $this->service->host('https://example.com:8080/path'));
+    }
+
+    // -------------------------------------------------------------------------
+    // Host is always returned as lowercase
+    // -------------------------------------------------------------------------
+
+    public function test_lowercases_host(): void
+    {
+        $this->assertSame('example.com', $this->service->host('https://EXAMPLE.COM/path'));
+    }
+
+    // -------------------------------------------------------------------------
+    // Throws on malformed, disallowed, or IP-literal input
+    // -------------------------------------------------------------------------
+
+    #[DataProvider('invalidInputs')]
+    public function test_throws_on_invalid_input(string $url): void
+    {
+        $this->expectException(\InvalidArgumentException::class);
+
+        $this->service->host($url);
+    }
+
+    public static function invalidInputs(): array
+    {
+        return [
+            // malformed / missing structure
+            'empty string' => [''],
+            'no scheme' => ['example.com/path'],
+            'scheme only' => ['https://'],
+            'bare string' => ['not a url at all'],
+
+            // disallowed schemes
+            'javascript scheme' => ['javascript:alert(1)'],
+            'ftp scheme' => ['ftp://example.com'],
+            'data scheme' => ['data:text/html,<h1>hi</h1>'],
+
+            // IP literals — not valid page-URL hosts for Trove's purposes
+            'ipv4 literal' => ['https://192.168.1.1/path'],
+            'ipv6 literal' => ['https://[::1]/path'],
+            'ipv4 without path' => ['http://10.0.0.1'],
+        ];
+    }
+}