From e7405051d10e13a7f12972ce6dfa7d3c1351db1b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Th=C3=A9o=20Poizat?= <hello@zatteo.com>
Date: Mon, 15 Dec 2025 11:13:22 +0100
Subject: [PATCH] fixup! Use html parser instead of regex to clean composer
 input

Improve efficiency and add more tests in convertHtmlContentToTextContent
---
 core/lib/utils/string_convert.dart       | 19 ++-----
 core/test/utils/string_convert_test.dart | 63 ++++++++++++++++++++++++
 2 files changed, 67 insertions(+), 15 deletions(-)

diff --git a/core/lib/utils/string_convert.dart b/core/lib/utils/string_convert.dart
index 595ef1958..fefedf069 100644
--- a/core/lib/utils/string_convert.dart
+++ b/core/lib/utils/string_convert.dart
@@ -204,22 +204,11 @@ class StringConvert {
     // Each paragraph is surrounded by block tags so we add a /n for each block tag
     // Even <br> are surrounded by block tags so we can ignore <br> and treat them
     // as paragraph
-    const blockTags = [
-      'p',
-      'div',
-      'li',
-      'section',
-      'article',
-      'header',
-      'footer',
-      'h1','h2','h3','h4','h5','h6'
-    ];
+    const blockTags = 'p, div, li, section, blockquote, article, header, footer, h1, h2, h3, h4, h5, h6';
 
-    for (final tag in blockTags) {
-      document.querySelectorAll(tag).forEach((element) {
-        element.append(Text('\n'));
-      });
-    }
+    document.querySelectorAll(blockTags).forEach((element) {
+      element.append(Text('\n'));
+    });
 
     final String textContent = document.body?.text ?? '';
 
diff --git a/core/test/utils/string_convert_test.dart b/core/test/utils/string_convert_test.dart
index b72ab12db..d3dde6e70 100644
--- a/core/test/utils/string_convert_test.dart
+++ b/core/test/utils/string_convert_test.dart
@@ -831,5 +831,68 @@ void main() {
 
       expect(result, 'First paragraph\n\nSecond paragraph');
     });
+
+    test('should preserve list items for ul li', () {
+      const htmlContent = '<ul><li>First item</li><li>Second item</li><li>Third item</li></ul>';
+      final result = StringConvert.convertHtmlContentToTextContent(htmlContent);
+
+      expect(result, 'First item\nSecond item\nThird item');
+    });
+
+    test('should preserve list items for ol li', () {
+      const htmlContent = '<ol><li>First item</li><li>Second item</li><li>Third item</li></ol>';
+      final result = StringConvert.convertHtmlContentToTextContent(htmlContent);
+
+      expect(result, 'First item\nSecond item\nThird item');
+    });
+
+    test('should strip inline style about text alignment', () {
+      const htmlContent = '<p>First paragraph</p><p style="text-align: right; ">Aligned right text</p><p>Last paragraph</p>';
+      final result = StringConvert.convertHtmlContentToTextContent(htmlContent);
+
+      expect(result, 'First paragraph\nAligned right text\nLast paragraph');
+    });
+
+    test('should strip inline style about indentation', () {
+      const htmlContent = '<p>First paragraph</p><p style="margin-left: 25px; ">Indented text</p><p>Last paragraph</p>';
+      final result = StringConvert.convertHtmlContentToTextContent(htmlContent);
+
+      expect(result, 'First paragraph\nIndented text\nLast paragraph');
+    });
+
+    test('should decode HTML entities', () {
+      const htmlContent = '<p>Text with &nbsp; and &amp; and &lt; and &gt; and &quot;</p>';
+      final result = StringConvert.convertHtmlContentToTextContent(htmlContent);
+
+      expect(result, 'Text with \u00A0 and & and < and > and "');
+    });
+
+    test('should preserve text with nested inline formatting', () {
+      const htmlContent = '<p>Text with <strong>bold</strong> and <em>italic</em> and <u>underline</u></p>';
+      final result = StringConvert.convertHtmlContentToTextContent(htmlContent);
+
+      expect(result, 'Text with bold and italic and underline');
+    });
+
+    test('should handle empty content', () {
+      const htmlContent = '<p></p><div></div><p></p>';
+      final result = StringConvert.convertHtmlContentToTextContent(htmlContent);
+
+      expect(result, '');
+    });
+
+    test('should preserve header text with line breaks', () {
+      const htmlContent = '<h1>Main Title</h1><p>Content paragraph</p>';
+      final result = StringConvert.convertHtmlContentToTextContent(htmlContent);
+
+      expect(result, 'Main Title\nContent paragraph');
+    });
+
+    test('should preserve link text only', () {
+      const htmlContent = '<p>Click <a href="https://example.com">here</a> for more info</p>';
+      final result = StringConvert.convertHtmlContentToTextContent(htmlContent);
+
+      expect(result, 'Click here for more info');
+    });
   });
 }