Merge tag v2.7.33 into develop

Author: roadiz-ci

composer.json

@@ -11,7 +11,8 @@
     },
     "require-dev": {
         "phpstan/phpstan": "^2.1.36",
-        "phpstan/phpdoc-parser": "<2"
+        "phpstan/phpdoc-parser": "<2",
+        "phpunit/phpunit": "^9.6"
     },
     "license": "MIT",
     "authors": [
@@ -27,6 +28,11 @@
             "RZ\\Roadiz\\Markdown\\": "src/"
         }
     },
+    "autoload-dev": {
+        "psr-4": {
+            "RZ\\Roadiz\\Markdown\\Tests\\": "tests/"
+        }
+    },
     "extra": {
         "branch-alias": {
             "dev-master": "2.7.x-dev",

src/CommonMark.php

@@ -59,4 +59,25 @@ public function line(?string $markdown = null): string
 
         return $html;
     }
+
+    #[\Override]
+    public function strip(?string $markdown = null): ?string
+    {
+        if (!is_string($markdown)) {
+            return null;
+        }
+        /*
+         * Strip Markdown syntax
+         */
+        $markdown = $this->textExtra($markdown);
+        // replace BR with space to avoid merged words.
+        $markdown = str_replace(['<br>', '<br />', '<br/>'], ' ', $markdown);
+        $markdown = strip_tags($markdown);
+        /*
+         * Remove control characters (including DEL).
+         */
+        $markdown = preg_replace('/[\x00-\x1F\x7F]/', '', $markdown);
+
+        return $markdown;
+    }
 }

src/MarkdownInterface.php

@@ -26,4 +26,9 @@ public function textExtra(?string $markdown = null, bool $allowHtml = false): st
      * Convert Markdown to HTML using only inline HTML elements.
      */
     public function line(?string $markdown = null): string;
+
+    /**
+     * Convert Markdown then strip tags, control characters and replace br with whitespace.
+     */
+    public function strip(?string $markdown = null): ?string;
 }

src/Twig/MarkdownExtension.php

@@ -23,6 +23,8 @@ public function getFilters(): array
             new TwigFilter('inline_markdown', $this->inlineMarkdown(...), ['is_safe' => ['html']]),
             new TwigFilter('markdownExtra', $this->markdownExtra(...), ['is_safe' => ['html']]),
             new TwigFilter('markdown_extra', $this->markdownExtra(...), ['is_safe' => ['html']]),
+            new TwigFilter('strip_markdown', $this->strip(...)),
+            new TwigFilter('stripMarkdown', $this->strip(...)),
         ];
     }
 
@@ -62,4 +64,13 @@ public function markdownExtra(?string $input, bool $allowHtml = false): string
 
         return $this->markdown->textExtra($input, $allowHtml);
     }
+
+    public function strip(?string $input): string
+    {
+        if (null === $input) {
+            return '';
+        }
+
+        return $this->markdown->strip($input) ?? '';
+    }
 }

tests/CommonMarkTest.php

@@ -0,0 +1,58 @@
+<?php
+
+declare(strict_types=1);
+
+namespace RZ\Roadiz\Markdown\Tests;
+
+use League\CommonMark\Environment\Environment;
+use League\CommonMark\Extension\CommonMark\CommonMarkCoreExtension;
+use League\CommonMark\MarkdownConverter;
+use PHPUnit\Framework\TestCase;
+use RZ\Roadiz\Markdown\CommonMark;
+
+final class CommonMarkTest extends TestCase
+{
+    private function createCommonMark(): CommonMark
+    {
+        $environment = new Environment(['html_input' => 'strip']);
+        $environment->addExtension(new CommonMarkCoreExtension());
+        $converter = new MarkdownConverter($environment);
+
+        // strip() only relies on the textExtra converter, but the constructor
+        // requires all five converters.
+        return new CommonMark($converter, $converter, $converter, $converter, $converter);
+    }
+
+    /**
+     * @dataProvider stripProvider
+     */
+    public function testStrip(?string $input, ?string $expected): void
+    {
+        $this->assertSame($expected, $this->createCommonMark()->strip($input));
+    }
+
+    /**
+     * @return array<array{0: ?string, 1: ?string}>
+     */
+    public static function stripProvider(): array
+    {
+        return [
+            'null returns null' => [null, null],
+            'plain text is untouched' => ['Hello world', 'Hello world'],
+            'markdown emphasis is stripped' => ['**Hello** _world_', 'Hello world'],
+            'headings are stripped' => ['# Title', 'Title'],
+            'links keep their label only' => ['[Roadiz](https://roadiz.io)', 'Roadiz'],
+            'hard line break becomes a space' => ["Line1  \nLine2", 'Line1 Line2'],
+            'control characters are removed' => ["a\x07b", 'ab'],
+            'DEL character is removed' => ["a\x7Fb", 'ab'],
+        ];
+    }
+
+    public function testStripRemovesHtmlTags(): void
+    {
+        $result = $this->createCommonMark()->strip('Some **bold** text with a [link](https://roadiz.io).');
+
+        $this->assertStringNotContainsString('<', (string) $result);
+        $this->assertStringNotContainsString('>', (string) $result);
+    }
+}