Meziantou.Framework.HtmlToMarkdown 2.0.12

Meziantou.Framework.HtmlToMarkdown

A .NET library to convert HTML fragments to Markdown.

Usage

Basic conversion

HtmlToMarkdown and its options are in the Meziantou.Framework namespace.

var html = "<h1>Title</h1><p>Hello <strong>world</strong></p>";
var markdown = HtmlToMarkdown.Convert(html);
Assert.Equal("""
    # Title

    Hello **world**
    """, markdown);

Configure conversion options

var html = "<h1>Title</h1><p>Line 1<br>Line 2</p>";

var markdown = HtmlToMarkdown.Convert(html, new HtmlToMarkdownOptions
{
    HeadingStyle = HeadingStyle.Setext,
    EmphasisMarker = EmphasisMarker.Underscore,
    CodeBlockStyle = CodeBlockStyle.Fenced,
    CodeBlockFenceCharacter = '~',
    UnorderedListMarker = '*',
    ThematicBreak = "***",
    LineBreakStyle = LineBreakStyle.Backslash,
    UseSimplePunctuation = true,
    EmojiShortcodeMode = EmojiShortcodeMode.GitHub,
    UnknownElementHandling = UnknownElementHandling.StripKeepContent,
});
Assert.Equal("""
    Title
    =====

    Line 1\
    Line 2
    """, markdown);

Replace SmartyPants punctuation with simple ASCII

You can opt in to convert smart punctuation in regular text nodes:

var markdown = HtmlToMarkdown.Convert(
    "<p>“Hello” ‘Hello’ — – … « »</p>",
    new HtmlToMarkdownOptions { UseSimplePunctuation = true });
Assert.Equal("""
    "Hello" 'Hello' \-\-\- \-\- ... \<\< \>\>
    """, markdown);

Emoji shortcode replacement

var markdown = HtmlToMarkdown.Convert("<p>I ❤️ Markdown</p>", new HtmlToMarkdownOptions
{
    EmojiShortcodeMode = EmojiShortcodeMode.GitHub,
});
Assert.Equal("I :heart: Markdown", markdown);

var unicodeMarkdown = HtmlToMarkdown.Convert("<p>I ❤️ Markdown</p>", new HtmlToMarkdownOptions
{
    EmojiShortcodeMode = EmojiShortcodeMode.Unicode,
});
Assert.Equal("I :red_heart: Markdown", unicodeMarkdown);

Supported HTML elements

The converter supports common Markdown-related elements, including:

• Headings (h1-h6)
• Paragraphs and line breaks (p, br)
• Emphasis (strong, b, em, i, del, s, strike)
• Links and images (a, img)
• Lists (ul, ol, li) including task list checkboxes (input type="checkbox")
• Blockquotes (blockquote)
• Code (code, pre) with fenced or indented blocks
• Tables (table, thead, tbody, tfoot, tr, th, td) with alignment support
• Definition lists (dl, dt, dd)

script, style, and noscript elements are always stripped, including when they appear inside an element that is passed through as raw HTML.

Unknown elements are handled using UnknownElementHandling:

• PassThrough (default): keep raw HTML
• Strip: remove the element and its content
• StripKeepContent: remove the element but keep converted child content