HTML to Markdown Converter

Converting HTML DOM Trees into Clean Markdown

Migrating rich text content between content management systems (WordPress, Drupal, Notion), web scrapers, and developer documentation platforms often requires converting complex HTML markup into clean, human-readable GitHub Flavored Markdown (GFM).

A robust HTML to Markdown converter parses the HTML into a Document Object Model (DOM) tree, traverses elements recursively, and translates semantic tags into corresponding markdown equivalents:

• <h1> through <h6> become # through ######.
• <strong> and <b> become **bold**.
• <em> and <i> become *italic*.
• <a href="..."> becomes [text](url).
• <table> structures are translated into GFM pipe-delimited tables.
• <pre><code> blocks become fenced code blocks with language identifiers.

Our HTML to Markdown Converter strips non-semantic wrapper tags (<div>, <span>), cleans redundant inline CSS styles, and emits clean markdown ready for Git repositories and static site generators.

Programmatic HTML-to-Markdown Conversion in Node.js & Python

// Node.js using Turndown library
import TurndownService from 'turndown';
import { gfm } from 'turndown-plugin-gfm';

const turndown = new TurndownService({ headingStyle: 'atx', codeBlockStyle: 'fenced' });
turndown.use(gfm); // Enable tables and task lists

const html = '<h2>Title</h2><p>Check <a href="https://doitquick.tools">DoItQuick</a>.</p>';
const markdown = turndown.turndown(html);
console.log(markdown);
// Output: ## Title\n\nCheck [DoItQuick](https://doitquick.tools).

# Python 3 using markdownify
from markdownify import markdownify as md

html = '<h1>Example</h1><p>This is <b>bold</b>.</p>'
print(md(html, heading_style="ATX"))

Cleaning Messy WYSIWYG & Word HTML

Pasting text from Microsoft Word or Google Docs into rich text editors generates hundreds of lines of proprietary XML namespaces (mso-bidi-*), nested spans, and font wrappers. Our parser strips these non-standard styles automatically, distilling messy pastes down to clean, portable markdown tokens.

Table & Code Block Preservation

HTML tables (<table><thead>...) are parsed and formatted into clean GFM pipe tables with proper column alignment indicators (| :--- | :---: | ---: |). Code snippets inside <pre><code class="language-js"> preserve their syntax highlighting hints.

CommonMark AST Traversal: Block Elements vs Inline Formatting Nodes

Translating HTML markup into clean, readable Markdown syntax requires structural parsing through an Abstract Syntax Tree (AST). Rather than using fragile regular expression replacements that break on nested elements, our converter parses HTML strings into a DOM tree and traverses nodes according to strict CommonMark and GitHub Flavored Markdown (GFM) specifications.

Block-level elements (headings, blockquotes, unordered/ordered lists, code fences, table structures) dictate document hierarchy and mandate double-newline boundary separation. Inline elements (anchor tags, emphasis, strong text, inline code snippets) are converted while stripping extraneous inline styles, tracking attributes, and empty decorative span wrappers.

Edge Case Handling: Deeply Nested Lists & Blockquote Quotations

A common challenge in markdown compilation is converting deeply nested HTML lists containing ordered and unordered elements simultaneously. While standard browsers render nested lists gracefully, markdown parsers require strict 2-space or 4-space indentation per nesting tier. Our parser normalizes list hierarchies and ensures that consecutive blockquotes preserve their directional attribution symbols (> and >>) without leaking unclosed HTML tags.

Preserving Semantic Structure During HTML to Markdown Conversion

Converting rich web documents into lightweight Markdown requires distinguishing structural semantic nodes from transient presentation wrappers. Modern web pages wrap content inside deeply nested <div> and <section> containers with extensive CSS utility classes. Our parser isolates semantic headings (<h1> through <h6>), lists, blockquotes, and code blocks, translating layout semantics into pure CommonMark text while pruning noise attributes.

Task Lists & Checkbox Formats: Modern technical READMEs and project wikis utilize Markdown task list syntax (- [ ] and - [x]). Our converter automatically identifies standard HTML checkbox inputs and converts them into interactive GFM task list markers.

Frequently Asked Questions