chapter-organizer/WebApp/Services/MarkdownHelper.cs

using Markdig;
using System.Text.RegularExpressions;

namespace WebApp.Services;

/// <summary>
/// Helper class for rendering markdown to HTML.
/// </summary>
public static class MarkdownHelper
{
    private static readonly MarkdownPipeline _pipeline = new MarkdownPipelineBuilder()
        .UseAdvancedExtensions()
        .Build();

    // Compiled regex patterns for stripping markdown
    private static readonly Regex MarkdownLinkRegex = 
        new(@"\[([^\]]+)\]\([^\)]+\)", RegexOptions.Compiled);
    private static readonly Regex BoldRegex = 
        new(@"\*\*([^\*]+)\*\*", RegexOptions.Compiled);
    private static readonly Regex ItalicRegex = 
        new(@"\*([^\*]+)\*", RegexOptions.Compiled);
    private static readonly Regex InlineCodeRegex = 
        new(@"`([^`]+)`", RegexOptions.Compiled);
    private static readonly Regex HeaderRegex = 
        new(@"#+\s+", RegexOptions.Compiled);
    private static readonly Regex NewlineRegex = 
        new(@"\n+", RegexOptions.Compiled);
    private static readonly Regex HtmlTagRegex = 
        new(@"<[^>]+>", RegexOptions.Compiled);

    /// <summary>
    /// Converts markdown text to HTML.
    /// </summary>
    /// <param name="markdown">The markdown text to convert.</param>
    /// <returns>HTML string ready for rendering.</returns>
    public static string ToHtml(string? markdown)
    {
        if (string.IsNullOrWhiteSpace(markdown))
            return string.Empty;

        var html = Markdown.ToHtml(markdown, _pipeline);
        
        // Add target="_blank" and rel="noopener noreferrer" to all links
        html = Regex.Replace(html, @"<a\s+([^>]*?)href=""([^""]*?)""([^>]*?)>", 
            match =>
            {
                var beforeHref = match.Groups[1].Value;
                var href = match.Groups[2].Value;
                var afterHref = match.Groups[3].Value;
                
                // Check if target is already present
                if (Regex.IsMatch(beforeHref + afterHref, @"target\s*=", RegexOptions.IgnoreCase))
                {
                    return match.Value; // Return unchanged if target already exists
                }
                
                // Add target="_blank" and rel="noopener noreferrer"
                return $"<a {beforeHref}href=\"{href}\" target=\"_blank\" rel=\"noopener noreferrer\"{afterHref}>";
            }, 
            RegexOptions.IgnoreCase);
        
        return html;
    }

    /// <summary>
    /// Strips markdown formatting from content and returns plain text for preview.
    /// </summary>
    /// <param name="content">The markdown content to strip.</param>
    /// <param name="maxLength">Maximum length of the preview. If 0 or negative, no truncation is performed.</param>
    /// <returns>Plain text preview with markdown formatting removed.</returns>
    public static string StripMarkdownPreview(string? content, int maxLength = 0)
    {
        if (string.IsNullOrWhiteSpace(content))
            return string.Empty;

        // Strip markdown formatting for preview using compiled regex
        var plainText = MarkdownLinkRegex.Replace(content, "$1"); // Replace markdown links with link text
        plainText = BoldRegex.Replace(plainText, "$1"); // Remove bold
        plainText = ItalicRegex.Replace(plainText, "$1"); // Remove italic
        plainText = InlineCodeRegex.Replace(plainText, "$1"); // Remove inline code
        plainText = HeaderRegex.Replace(plainText, ""); // Remove headers
        plainText = NewlineRegex.Replace(plainText, " "); // Replace newlines with spaces
        plainText = HtmlTagRegex.Replace(plainText, ""); // Remove HTML tags

        plainText = plainText.Trim();

        // Truncate if maxLength is specified and content exceeds it
        if (maxLength > 0 && plainText.Length > maxLength)
        {
            return plainText.Substring(0, maxLength - 3) + "...";
        }

        return plainText;
    }
}