feat: add MCP server for the website chatbot over a mirrored TYPO3 content feed

tests/ContentFeed/HtmlToMarkdownConverterTest.php

@@ -0,0 +1,53 @@
+<?php
+
+declare(strict_types=1);
+
+namespace App\Tests\ContentFeed;
+
+use App\ContentFeed\HtmlToMarkdownConverter;
+use PHPUnit\Framework\TestCase;
+
+class HtmlToMarkdownConverterTest extends TestCase
+{
+    private HtmlToMarkdownConverter $converter;
+
+    protected function setUp(): void
+    {
+        $this->converter = new HtmlToMarkdownConverter();
+    }
+
+    public function testDropsEmbedsMediaAndFormsWithTheirContent(): void
+    {
+        $markdown = $this->converter->convert('<p>Text</p><script>alert(1)</script><style>p{}</style><noscript>No JS</noscript><iframe src="https://youtube.com"></iframe><img src="a.jpg" alt="Bild"><picture><source></picture><svg><text>Icon</text></svg><form><label>Name</label></form>');
+
+        self::assertSame('Text', $markdown);
+    }
+
+    public function testKeepsHeadingsListsTablesAndEmphasis(): void
+    {
+        $markdown = $this->converter->convert('<h2>Leistungen</h2><ul><li><strong>Skipass</strong></li><li><em>Bus</em></li></ul><table><tr><th>A</th><th>B</th></tr><tr><td>1</td><td>2</td></tr></table>');
+
+        self::assertSame("## Leistungen\n\n- **Skipass**\n- *Bus*\n\n| A | B |\n|---|---|\n| 1 | 2 |", $markdown);
+    }
+
+    public function testResolvesRelativeLinksAgainstTheDocumentUrl(): void
+    {
+        $markdown = $this->converter->convert('<a href="/agb">AGB</a> <a href="kontakt">Kontakt</a> <a href="mailto:[email protected]">Mail</a> <a href="https://snowzone.net/">SZ</a>', 'https://www.ep-reisen.de/info/reise');
+
+        self::assertSame('[AGB](https://www.ep-reisen.de/agb) [Kontakt](https://www.ep-reisen.de/info/kontakt) [Mail](mailto:[email protected]) [SZ](https://snowzone.net/)', $markdown);
+    }
+
+    public function testCollapsesTemplateWhitespaceAndDecodesEntities(): void
+    {
+        $markdown = $this->converter->convert("<div>\n   <a href=\"/x\">\n  Ski-\n   Gruppenhäuser\n </a></div>\n\n\n\n<p>Essen&nbsp;&amp;&nbsp;Trinken</p>", 'https://www.ep-reisen.de/');
+
+        self::assertSame("[Ski- Gruppenhäuser](https://www.ep-reisen.de/x)\n\nEssen & Trinken", $markdown);
+    }
+
+    public function testEmptyInputIsEmpty(): void
+    {
+        self::assertSame('', $this->converter->convert(null));
+        self::assertSame('', $this->converter->convert('  '));
+        self::assertSame('', $this->converter->convert('<article><img src="a.jpg"></article>'));
+    }
+}