z3l.FastHtmlParser.HtmlToMarkdown
1.6.1
dotnet add package z3l.FastHtmlParser.HtmlToMarkdown --version 1.6.1
NuGet\Install-Package z3l.FastHtmlParser.HtmlToMarkdown -Version 1.6.1
<PackageReference Include="z3l.FastHtmlParser.HtmlToMarkdown" Version="1.6.1" />
<PackageVersion Include="z3l.FastHtmlParser.HtmlToMarkdown" Version="1.6.1" />
<PackageReference Include="z3l.FastHtmlParser.HtmlToMarkdown" />
paket add z3l.FastHtmlParser.HtmlToMarkdown --version 1.6.1
#r "nuget: z3l.FastHtmlParser.HtmlToMarkdown, 1.6.1"
#:package z3l.FastHtmlParser.HtmlToMarkdown@1.6.1
#addin nuget:?package=z3l.FastHtmlParser.HtmlToMarkdown&version=1.6.1
#tool nuget:?package=z3l.FastHtmlParser.HtmlToMarkdown&version=1.6.1
z3l.FastHtmlParser.HtmlToMarkdown
Fast HTML to Markdown for .NET 9 and .NET 10: for a language model, a search index, a note to keep.
Website: https://htmltomarkdown.fasthtmlparser.com (try it on any page) · Source: https://github.com/romanti3172/z3l.FastHtmlParser
10–20x faster than ReverseMarkdown and Html2Markdown on typical pages, with 37–62x less memory.
| books.toscrape.com, 51 KB: HTML → Markdown | Mean | Allocated |
|---|---|---|
| z3l.FastHtmlParser.HtmlToMarkdown | 114 μs | 39 KB |
| Html2Markdown | 1,751 μs (15x) | 1,627 KB (42x) |
| ReverseMarkdown | 2,195 μs (19x) | 2,240 KB (58x) |
| html-to-markdown (a Rust library, through its .NET binding) | 3,675 μs (32x) | 122 KB (3x) |
All four pages and what is compared: benchmarks.
Install
dotnet add package z3l.FastHtmlParser.HtmlToMarkdown
Use
using z3l.FastHtmlParser;
string markdown = HtmlToMarkdown.Convert(html);
// with the address of the page the links and images are absolute
string page = HtmlToMarkdown.Convert(html, "https://example.com/blog/post");
// only the text and its structure: less to read for a language model
string text = HtmlToMarkdown.Convert(html, options: new MarkdownOptions { Links = false, Images = false });
<h2>Title</h2><p>Some <b>bold</b> text</p> gives ## Title, an empty line and Some **bold** text.
The result is never null: HTML that is null, empty or shows nothing gives an empty string.
A part of the page
The package is z3l.FastHtmlParser with one more method: the parser comes with it, with its CSS selectors and XPath. Parse the page and take the Markdown of what you need.
var doc = new HtmlParser(html, "https://example.com/blog/post");
string? article = doc.CssFirst("article").Markdown(); // the article without the menu and the footer
string? specs = doc.CssFirst("table.specs").Markdown(); // "| Name | Value |\n| --- | --- |\n| ..."
string? title = doc.Find(HtmlTag.H1).Text();
Already using z3l.FastHtmlParser? You do not need this package: Markdown() is a part of the parser, and
HtmlToMarkdown.Convert(html, url) is new HtmlParser(html, url).Markdown().
What is converted
- Headings, paragraphs, line breaks, emphasis, strikethrough, quotes, rules, nested lists (numbered from their
start, task lists with checkboxes), definition lists, code spans, code blocks with their language (class="language-…"), links and images with their titles. - A table of data becomes a GitHub table:
colspanandrowspankeep the columns in place, and a table without a heading row gets an empty one. A table that only lays out a page (a table inside it, a single column,role="presentation") is written as the blocks in its cells. - Only what a browser shows:
head,script,style,noscript,template, comments,svgand form controls are left out. - Links and images are absolute when the page URL is given. An image that a script loads on scroll is taken from
data-src; a link without an address or with a script is written as its text. - Text gets a backslash only where Markdown would read it as its own:
snake_case,2 * 3,[1]andAT&Tstay as they are;*text*, or a line that starts with#,-or1., is escaped. - The Markdown renders as the page reads. Emphasis that Markdown would not take for emphasis (
**Note:**text,文**「字」**文) gets an empty comment next to it, two lists in a row get different markers, a link around several blocks becomes a link in each of them. - One pass over the parsed tags, no object per tag, linear time on any markup. Lines end with LF; the result has no line break at its start or end.
On every test page the Markdown, rendered back to HTML by Markdig, has the same text, links, images, headings and list items as the page itself.
License
Free for individuals and organizations with annual revenue below USD 1,000,000 and fewer than 100 employees, including commercial use. Organizations with annual revenue of USD 1,000,000 or more, or with 100 or more employees, need a commercial license: a one-time payment of USD 1,000 per organization, perpetual, all versions. The same license covers every z3l library: you pay once. See LICENSE.md for the full terms and pricing to buy a commercial license.
| Product | Versions Compatible and additional computed target framework versions. |
|---|---|
| .NET | net9.0 is compatible. net9.0-android was computed. net9.0-browser was computed. net9.0-ios was computed. net9.0-maccatalyst was computed. net9.0-macos was computed. net9.0-tvos was computed. net9.0-windows was computed. net10.0 is compatible. net10.0-android was computed. net10.0-browser was computed. net10.0-ios was computed. net10.0-maccatalyst was computed. net10.0-macos was computed. net10.0-tvos was computed. net10.0-windows was computed. |
-
net10.0
- z3l.FastHtmlParser (>= 1.6.1)
-
net9.0
- z3l.FastHtmlParser (>= 1.6.1)
NuGet packages
This package is not used by any NuGet packages.
GitHub repositories
This package is not used by any popular GitHub repositories.
| Version | Downloads | Last Updated |
|---|---|---|
| 1.6.1 | 0 | 11/21/2026 |