Configuration Options

/**
 * Configuration options for HTML to Markdown conversion
 */
interface NodeHtmlMarkdownOptions {
  /**
   * Use native window DOMParser when available (browser environment)
   * @default false
   */
  preferNativeParser: boolean;

  /**
   * Code block fence delimiter
   * @default "```"
   */
  codeFence: string;

  /**
   * Bullet marker for unordered lists
   * @default "*"
   */
  bulletMarker: string;

  /**
   * Style for code blocks
   * @default "fenced"
   */
  codeBlockStyle: "indented" | "fenced";

  /**
   * Emphasis (italic) delimiter
   * @default "_"
   */
  emDelimiter: string;

  /**
   * Strong (bold) delimiter
   * @default "**"
   */
  strongDelimiter: string;

  /**
   * Strikethrough delimiter
   * @default "~~"
   */
  strikeDelimiter: string;

  /**
   * HTML elements to ignore (ignores inner text, does not parse children)
   */
  readonly ignore?: string[];

  /**
   * HTML elements to treat as blocks (surrounded with blank lines)
   */
  readonly blockElements?: string[];

  /**
   * Maximum consecutive newlines allowed in output
   * @default 3
   */
  maxConsecutiveNewlines: number;

  /**
   * Line start escape pattern for markdown special characters
   * Overrides default escape settings - consider using textReplace instead
   */
  lineStartEscape: readonly [pattern: RegExp, replacement: string];

  /**
   * Global escape pattern for markdown special characters
   * Overrides default escape settings - consider using textReplace instead
   */
  globalEscape: readonly [pattern: RegExp, replacement: string];

  /**
   * User-defined text replacement patterns
   * Applied to text content retrieved from HTML nodes
   */
  textReplace?: (readonly [pattern: RegExp, replacement: string])[];

  /**
   * Keep images with data: URI (can be up to 1MB each)
   * @default false
   * @example <img src="data:image/gif;base64,R0lGODlhEAAQAMQAAORHHOVSK......0o/">
   */
  keepDataImages?: boolean;

  /**
   * Use link reference definitions at bottom of document
   * @default undefined
   * @example
   * Input: Click <a href="/url1">here</a>. Or <a href="/url2">here</a>.
   * Output: Click [here][1]. Or [here][2].
   * 
   * [1]: /url1
   * [2]: /url2
   */
  useLinkReferenceDefinitions?: boolean;

  /**
   * Wrap URL text in < > instead of []() syntax when link text matches URL
   * @default true
   * @example
   * Input: <a href="https://google.com">https://google.com</a>
   * Output: <https://google.com> (when true)
   * Output: [https://google.com](https://google.com) (when false)
   */
  useInlineLinks?: boolean;
}

import { NodeHtmlMarkdown, NodeHtmlMarkdownOptions } from "node-html-markdown";

// Configure basic formatting options
const options: Partial<NodeHtmlMarkdownOptions> = {
  bulletMarker: "-",
  strongDelimiter: "__",
  emDelimiter: "*",
  codeFence: "~~~"
};

const nhm = new NodeHtmlMarkdown(options);

const html = `
  <ul>
    <li><strong>Bold</strong> and <em>italic</em> text</li>
  </ul>
  <pre><code class="language-javascript">console.log("hello");</code></pre>
`;

const markdown = nhm.translate(html);
console.log(markdown);
// Output:
// - __Bold__ and *italic* text
// 
// ~~~javascript
// console.log("hello");
// ~~~

import { NodeHtmlMarkdown } from "node-html-markdown";

// Fenced code blocks (default)
const fencedNhm = new NodeHtmlMarkdown({
  codeBlockStyle: "fenced",
  codeFence: "```"
});

// Indented code blocks
const indentedNhm = new NodeHtmlMarkdown({
  codeBlockStyle: "indented"
});

const codeHtml = '<pre><code>const x = 42;</code></pre>';

console.log("Fenced:", fencedNhm.translate(codeHtml));
// Output: ```\nconst x = 42;\n```

console.log("Indented:", indentedNhm.translate(codeHtml));
// Output:     const x = 42;

import { NodeHtmlMarkdown } from "node-html-markdown";

const options = {
  // Ignore script and style elements completely
  ignore: ["script", "style", "nav"],
  // Treat custom elements as block elements
  blockElements: ["section", "article", "aside"]
};

const nhm = new NodeHtmlMarkdown(options);

const html = `
  <article>
    <h1>Title</h1>
    <nav>Navigation menu</nav>
    <script>alert("hello");</script>
    <p>Content</p>
  </article>
`;

const result = nhm.translate(html);
console.log(result);
// Output:
// # Title
// 
// Content
// (nav and script elements are ignored)

import { NodeHtmlMarkdown } from "node-html-markdown";

const options = {
  textReplace: [
    // Replace smart quotes with regular quotes
    [/[""]/g, '"'],
    [/['']/g, "'"],
    // Replace em dashes with regular dashes
    [/—/g, "--"],
    // Remove excessive whitespace
    [/\s{3,}/g, " "]
  ]
};

const nhm = new NodeHtmlMarkdown(options);

const html = '<p>"Smart quotes" and em—dash here   with spaces</p>';
const result = nhm.translate(html);
console.log(result);
// Output: "Smart quotes" and em--dash here with spaces

import { NodeHtmlMarkdown } from "node-html-markdown";

const options = {
  useLinkReferenceDefinitions: true
};

const nhm = new NodeHtmlMarkdown(options);

const html = `
  <p>Visit <a href="https://example.com">our website</a> or 
  <a href="https://github.com">GitHub</a>. Also check 
  <a href="https://example.com">our website</a> again.</p>
`;

const result = nhm.translate(html);
console.log(result);
// Output:
// Visit [our website][1] or [GitHub][2]. Also check [our website][1] again.
//
// [1]: https://example.com
// [2]: https://github.com

import { NodeHtmlMarkdown } from "node-html-markdown";

// Keep data URI images (default: false)
const keepDataNhm = new NodeHtmlMarkdown({
  keepDataImages: true
});

// Remove data URI images (default behavior)
const removeDataNhm = new NodeHtmlMarkdown({
  keepDataImages: false
});

const html = '<img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7" alt="1x1 pixel">';

console.log("Keep data images:", keepDataNhm.translate(html));
// Output: ![1x1 pixel](data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7)

console.log("Remove data images:", removeDataNhm.translate(html));
// Output: ![1x1 pixel]()

import { NodeHtmlMarkdown } from "node-html-markdown";

const options = {
  maxConsecutiveNewlines: 2 // Limit to 2 consecutive newlines
};

const nhm = new NodeHtmlMarkdown(options);

const html = `
  <h1>Title</h1>
  
  
  
  
  <p>Paragraph after many line breaks</p>
`;

const result = nhm.translate(html);
console.log(result);
// Output:
// # Title
// 
// Paragraph after many line breaks
// (excessive newlines are reduced to maxConsecutiveNewlines)

import { NodeHtmlMarkdown } from "node-html-markdown";

const customOptions: Partial<NodeHtmlMarkdownOptions> = {
  // Formatting preferences
  bulletMarker: "+",
  strongDelimiter: "__",
  emDelimiter: "*",
  strikeDelimiter: "~",
  codeFence: "```",
  codeBlockStyle: "fenced",
  
  // Content control
  maxConsecutiveNewlines: 2,
  keepDataImages: false,
  useInlineLinks: false,
  useLinkReferenceDefinitions: true,
  
  // Element behavior
  ignore: ["script", "style", "noscript"],
  blockElements: ["section", "article", "aside", "details"],
  
  // Text processing
  textReplace: [
    [/\u00A0/g, " "], // Replace non-breaking spaces
    [/[""]/g, '"'],   // Replace smart quotes
    [/['']/g, "'"],   // Replace smart apostrophes
  ]
};

const nhm = new NodeHtmlMarkdown(customOptions);

// Now use with all custom settings applied
const result = nhm.translate("<strong>Bold</strong> text with "smart quotes"");
console.log(result); // __Bold__ text with "smart quotes"