flutter-rimworld-modman/lib/format_converter.dart

import 'dart:math' as math;

/// Utility class to convert mixed format content (BBCode, Markdown, and HTML) to HTML
class FormatConverter {
  /// Converts mixed format text (BBCode, Markdown, HTML) to pure HTML
  static String toHtml(String content) {
    if (content.isEmpty) return '';

    // First, normalize line endings and escape any literal backslashes that aren't already escaped
    String result = content.replaceAll('\r\n', '\n');

    // Handle BBCode format
    result = _convertBBCodeToHtml(result);

    // Handle Markdown format
    result = _convertMarkdownToHtml(result);

    // Sanitize HTML
    result = _sanitizeHtml(result);

    // Wrap the final content in a container with styles
    result =
        '<div style="line-height: 1.5; word-wrap: break-word;">$result</div>';

    return result;
  }

  /// Converts BBCode to HTML
  static String _convertBBCodeToHtml(String bbcode) {
    String result = bbcode;

    // Fix unclosed tags - RimWorld descriptions often have unclosed BBCode tags
    final List<String> tagTypes = [
      'b',
      'i',
      'color',
      'size',
      'url',
      'code',
      'quote',
      'list',
      'table',
      'tr',
      'td',
    ];
    for (final tag in tagTypes) {
      final openCount = '[$tag'.allMatches(result).length;
      final closeCount = '[/$tag]'.allMatches(result).length;
      if (openCount > closeCount) {
        result = result + '[/$tag]' * (openCount - closeCount);
      }
    }

    // URLs
    // [url=http://example.com]text[/url] -> <a href="http://example.com">text</a>
    result = RegExp(
      r'\[url=([^\]]+)\](.*?)\[/url\]',
      dotAll: true,
    ).allMatches(result).fold(result, (prev, match) {
      final url = match.group(1);
      final text = match.group(2);
      return prev.replaceFirst(
        match.group(0)!,
        '<a href="$url" target="_blank">$text</a>',
      );
    });

    // Simple URL [url]http://example.com[/url] -> <a href="http://example.com">http://example.com</a>
    result = result.replaceAllMapped(
      RegExp(r'\[url\](.*?)\[/url\]', dotAll: true),
      (match) =>
          '<a href="${match.group(1)}" target="_blank">${match.group(1)}</a>',
    );

    // Bold
    result = result
        .replaceAll('[b]', '<strong>')
        .replaceAll('[/b]', '</strong>');

    // Italic
    result = result.replaceAll('[i]', '<em>').replaceAll('[/i]', '</em>');

    // Headers
    result = result.replaceAllMapped(
      RegExp(r'\[h1\](.*?)\[/h1\]', dotAll: true),
      (match) =>
          '<h1 style="margin-top: 16px; margin-bottom: 8px;">${match.group(1)?.trim()}</h1>',
    );
    result = result.replaceAllMapped(
      RegExp(r'\[h2\](.*?)\[/h2\]', dotAll: true),
      (match) =>
          '<h2 style="margin-top: 12px; margin-bottom: 6px;">${match.group(1)?.trim()}</h2>',
    );
    result = result.replaceAllMapped(
      RegExp(r'\[h3\](.*?)\[/h3\]', dotAll: true),
      (match) =>
          '<h3 style="margin-top: 10px; margin-bottom: 4px;">${match.group(1)?.trim()}</h3>',
    );

    // Lists
    result = result
        .replaceAll(
          '[list]',
          '<ul style="padding-left: 20px; margin-top: 8px; margin-bottom: 8px;">',
        )
        .replaceAll('[/list]', '</ul>');

    // List items
    result = result.replaceAllMapped(
      RegExp(r'\[\*\](.*?)(?=\[\*\]|\[/list\]|$)', dotAll: true),
      (match) {
        final content = match.group(1)?.trim() ?? '';
        return '<li style="margin-bottom: 4px;">$content</li>';
      },
    );

    // Color
    result = result.replaceAllMapped(
      RegExp(r'\[color=([^\]]+)\](.*?)\[/color\]', dotAll: true),
      (match) {
        final color = match.group(1) ?? '';
        final content = match.group(2) ?? '';
        if (content.trim().isEmpty) return '';
        return '<span style="color:$color">$content</span>';
      },
    );

    // Images
    result = result.replaceAllMapped(
      RegExp(r'\[img\](.*?)\[/img\]', dotAll: true),
      (match) =>
          '<img src="${match.group(1)}" alt="Image" style="max-width: 100%;" />',
    );

    // Image with size
    result = result.replaceAllMapped(
      RegExp(r'\[img[^\]]*width=(\d+)[^\]]*\](.*?)\[/img\]', dotAll: true),
      (match) {
        final width = match.group(1) ?? '';
        final url = match.group(2) ?? '';
        return '<img src="$url" alt="Image" width="$width" style="max-width: 100%;" />';
      },
    );

    // Tables
    result = result
        .replaceAll(
          '[table]',
          '<table border="1" style="border-collapse: collapse; width: 100%; margin: 10px 0;">',
        )
        .replaceAll('[/table]', '</table>');
    result = result.replaceAll('[tr]', '<tr>').replaceAll('[/tr]', '</tr>');
    result = result
        .replaceAll('[td]', '<td style="padding: 8px;">')
        .replaceAll('[/td]', '</td>');

    // Size
    result = result.replaceAllMapped(
      RegExp(r'\[size=([^\]]+)\](.*?)\[/size\]', dotAll: true),
      (match) {
        final size = match.group(1) ?? '';
        final content = match.group(2) ?? '';
        return '<span style="font-size:${size}px">$content</span>';
      },
    );

    // Code
    result = result
        .replaceAll(
          '[code]',
          '<pre style="background-color: rgba(0,0,0,0.1); padding: 8px; border-radius: 4px; overflow-x: auto;"><code>',
        )
        .replaceAll('[/code]', '</code></pre>');

    // Quote
    result = result.replaceAllMapped(
      RegExp(r'\[quote\](.*?)\[/quote\]', dotAll: true),
      (match) {
        final content = match.group(1)?.trim() ?? '';
        if (content.isEmpty) return '';
        return '<blockquote style="border-left: 4px solid rgba(128,128,128,0.5); padding-left: 10px; margin: 10px 0; color: rgba(255,255,255,0.8);">$content</blockquote>';
      },
    );

    // Handle any remaining custom BBCode tags
    result = result.replaceAllMapped(
      RegExp(r'\[([a-zA-Z0-9_]+)(?:=[^\]]+)?\](.*?)\[/\1\]', dotAll: true),
      (match) => match.group(2) ?? '',
    );

    // Handle RimWorld-specific patterns
    // [h1] without closing tag is common
    result = result.replaceAllMapped(
      RegExp(r'\[h1\]([^\[]+)'),
      (match) =>
          '<h1 style="margin-top: 16px; margin-bottom: 8px;">${match.group(1)?.trim()}</h1>',
    );

    return result;
  }

  /// Converts Markdown to HTML
  static String _convertMarkdownToHtml(String markdown) {
    String result = markdown;

    // Headers
    // Convert # Header to <h1>Header</h1>
    result = result.replaceAllMapped(
      RegExp(r'^#\s+(.*?)$', multiLine: true),
      (match) =>
          '<h1 style="margin-top: 16px; margin-bottom: 8px;">${match.group(1)?.trim()}</h1>',
    );

    // Convert ## Header to <h2>Header</h2>
    result = result.replaceAllMapped(
      RegExp(r'^##\s+(.*?)$', multiLine: true),
      (match) =>
          '<h2 style="margin-top: 12px; margin-bottom: 6px;">${match.group(1)?.trim()}</h2>',
    );

    // Convert ### Header to <h3>Header</h3>
    result = result.replaceAllMapped(
      RegExp(r'^###\s+(.*?)$', multiLine: true),
      (match) =>
          '<h3 style="margin-top: 10px; margin-bottom: 4px;">${match.group(1)?.trim()}</h3>',
    );

    // Bold - **text** to <strong>text</strong>
    result = result.replaceAllMapped(
      RegExp(r'\*\*(.*?)\*\*'),
      (match) => '<strong>${match.group(1)}</strong>',
    );

    // Italic - *text* or _text_ to <em>text</em>
    result = result.replaceAllMapped(
      RegExp(r'\*(.*?)\*|_(.*?)_'),
      (match) => '<em>${match.group(1) ?? match.group(2)}</em>',
    );

    // Inline code - `code` to <code>code</code>
    result = result.replaceAllMapped(
      RegExp(r'`(.*?)`'),
      (match) =>
          '<code style="background-color: rgba(0,0,0,0.1); padding: 2px 4px; border-radius: 3px;">${match.group(1)}</code>',
    );

    // Links - [text](url) to <a href="url">text</a>
    result = result.replaceAllMapped(
      RegExp(r'\[(.*?)\]\((.*?)\)'),
      (match) =>
          '<a href="${match.group(2)}" target="_blank">${match.group(1)}</a>',
    );

    // Images - ![alt](url) to <img src="url" alt="alt" />
    result = result.replaceAllMapped(
      RegExp(r'!\[(.*?)\]\((.*?)\)'),
      (match) =>
          '<img src="${match.group(2)}" alt="${match.group(1)}" style="max-width: 100%;" />',
    );

    // Lists - Convert Markdown bullet lists to HTML lists
    // This is a simple implementation and might not handle all cases
    result = result.replaceAllMapped(
      RegExp(r'^(\s*)\*\s+(.*?)$', multiLine: true),
      (match) => '<li style="margin-bottom: 4px;">${match.group(2)}</li>',
    );

    // Wrap adjacent list items in <ul> tags (simple approach)
    result = result.replaceAll('</li>\n<li>', '</li><li>');
    result = result.replaceAll(
      '<li>',
      '<ul style="padding-left: 20px; margin-top: 8px; margin-bottom: 8px;"><li>',
    );
    result = result.replaceAll('</li>', '</li></ul>');

    // Remove duplicated </ul><ul> tags
    result = result.replaceAll(
      '</ul><ul style="padding-left: 20px; margin-top: 8px; margin-bottom: 8px;">',
      '',
    );

    // Paragraphs - Convert newlines to <br>, but skipping where tags already exist
    result = result.replaceAllMapped(
      RegExp(r'(?<!>)\n(?!<)'),
      (match) => '<br />',
    );

    return result;
  }

  /// Performs basic sanitization and fixes for the HTML
  static String _sanitizeHtml(String html) {
    // Remove potentially dangerous elements and attributes
    final String result = html
        // Remove any script tags
        .replaceAll(RegExp(r'<script.*?>.*?</script>', dotAll: true), '')
        // Remove on* event handlers
        .replaceAll(RegExp(r'\son\w+=".*?"'), '')
        // Ensure newlines are converted to <br /> if not already handled
        .replaceAll(RegExp(r'(?<!>)\n(?!<)'), '<br />');

    // Fix double paragraph or break issues
    return result
        .replaceAll('<br /><br />', '<br />')
        .replaceAll('<br></br>', '<br />')
        .replaceAll('<p></p>', '');
  }
}