kepano · simjnd · Apr 22, 2026
diff --git a/src/extractor-registry.ts b/src/extractor-registry.ts
@@ -22,6 +22,7 @@ import { ThreadsExtractor } from './extractors/threads';
 import { BlueskyExtractor } from './extractors/bluesky';
 import { DiscourseExtractor } from './extractors/discourse';
 import { MediumExtractor } from './extractors/medium';
+import { DuckDuckGoLiteExtractor } from './extractors/duckduckgo-lite';
 
 type ExtractorConstructor = new (document: Document, url: string, schemaOrgData?: any, options?: ExtractorOptions) => BaseExtractor;
 
@@ -191,6 +192,13 @@ export class ExtractorRegistry {
 			extractor: DiscourseExtractor
 		});
 
+		this.register({
+			patterns: [
+				'lite.duckduckgo.com',
+			],
+			extractor: DuckDuckGoLiteExtractor
+		});
+
 		this.register({
 			patterns: [/.*/],
 			extractor: BbcodeDataExtractor

diff --git a/src/extractors/duckduckgo-lite.ts b/src/extractors/duckduckgo-lite.ts
@@ -0,0 +1,228 @@
+import { BaseExtractor, ExtractorOptions } from './_base';
+import { ExtractorResult } from '../types/extractors';
+import { serializeHTML, escapeHtml } from '../utils/dom';
+
+interface SearchResult {
+	title: string;
+	url: string;
+	snippet: string;
+}
+
+interface ZeroClickInfo {
+	header: string;
+	body: string;
+	textContent: string;
+}
+
+export class DuckDuckGoLiteExtractor extends BaseExtractor {
+	private isSearchPage: boolean;
+
+	constructor(document: Document, url: string, schemaOrgData?: any, options?: ExtractorOptions) {
+		super(document, url, schemaOrgData, options);
+		this.isSearchPage = this.detectSearchPage();
+	}
+
+	private detectSearchPage(): boolean {
+		return !!(
+			this.document.querySelector('input.query[name="q"]') &&
+			this.document.querySelector('a.result-link') &&
+			this.document.querySelector('table[border="0"]')
+		);
+	}
+
+	canExtract(): boolean {
+		return this.isSearchPage;
+	}
+
+	extract(): ExtractorResult {
+		const zeroClick = this.extractZeroClickInfo();
+		const results = this.extractResults();
+		const contentHtml = this.buildListingHtml(results, zeroClick);
+		const searchQuery = this.getSearchQuery();
+
+		return {
+			content: contentHtml,
+			contentHtml: contentHtml,
+			extractedContent: {
+				query: searchQuery,
+				resultCount: String(results.length),
+			},
+			variables: {
+				title: searchQuery || 'DuckDuckGo Search',
+				site: 'DuckDuckGo Lite',
+				description: zeroClick?.textContent || (searchQuery
+					? `Search results for "${searchQuery}"`
+					: 'DuckDuckGo Lite search results'),
+			}
+		};
+	}
+
+	// Reads the current query string from the search input
+	private getSearchQuery(): string {
+		const queryInput = this.document.querySelector('input.query[name="q"]');
+		return queryInput?.getAttribute('value') || '';
+	}
+
+	// DuckDuckGo wraps external URLs in a redirect so it can track clicks
+	// Unwraps the real destination from the `uddg` query parameter
+	private unwrapRedirectUrl(href: string): string {
+		try {
+			const url = new URL(href, this.url);
+			const actual = url.searchParams.get('uddg');
+			return actual ? decodeURIComponent(actual) : href;
+		} catch {
+			return href;
+		}
+	}
+
+	// Zero-click info appears in a `table[border="0"]` that does NOT contain
+	// `.result-link` elements. The structure is:
+	//   Row 1: "Zero-click info: <a href="...">Title</a>"
+	//   Row 2: Description text + "More at <q>Source</q>" link
+	//   Row 3: Spacer
+	// The extracted HTML is wrapped in a `<blockquote data-callout="info">`
+	private extractZeroClickInfo(): ZeroClickInfo | null {
+		const tables = Array.from(this.document.querySelectorAll('table[border="0"]'));
+
+		for (const table of tables) {
+			// Skip result tables — zero-click tables have no .result-link
+			if (table.querySelector('a.result-link')) continue;
+
+			const rows = Array.from(table.querySelectorAll('tr'));
+			if (rows.length < 2) continue;
+
+			const firstCell = rows[0].querySelector('td');
+			if (!firstCell) continue;
+
+			const firstText = firstCell.textContent || '';
+			if (!firstText.includes('Zero-click info:')) continue;
+
+			// Header: "Zero-click info: <a>Title</a>"
+			const headerHtml = this.unwrapLinksInHtml(firstCell);
+
+			// Body: description text + "More at <q>Source</q>" link
+			const bodyCell = rows[1].querySelector('td');
+			if (!bodyCell) continue;
+
+			// Clone so we can safely remove the "More at" link
+			const cleanBody = bodyCell.cloneNode(true) as Element;
+			const links = Array.from(cleanBody.querySelectorAll('a'));
+			for (const link of links) {
+				if (link.textContent?.includes('More at')) {
+					link.remove();
+					break;
+				}
+			}
+
+			const bodyHtml = this.unwrapLinksInHtml(cleanBody);
+
+			// Text-only version for the description variable
+			const textContent = cleanBody.textContent?.trim() || '';
+
+			return { header: headerHtml, body: bodyHtml, textContent };
+		}
+
+		return null;
+	}
+
+	// Unwraps DuckDuckGo redirect URLs in all <a> tags within the given element.
+	// Operates on the element's inner HTML string.
+	private unwrapLinksInHtml(el: Element): string {
+		const clone = el.cloneNode(true) as Element;
+		clone.querySelectorAll('a').forEach(a => {
+			const href = a.getAttribute('href') || '';
+			const unwrapped = this.unwrapRedirectUrl(href);
+			if (unwrapped !== href) {
+				a.setAttribute('href', unwrapped);
+			}
+		});
+		return serializeHTML(clone);
+	}
+
+	// Iterates over the top-level result tables and collects each
+	// title / URL / snippet triplet into a structured array.
+	//
+	// The lite layout uses a flat table where each result is
+	// three consecutive rows:
+	//   1. title link
+	//   2. snippet
+	//   3. blank spacer
+	private extractResults(): SearchResult[] {
+		const results: SearchResult[] = [];
+		const tables = Array.from(this.document.querySelectorAll('table[border="0"]'));
+
+		for (const table of tables) {
+			// Skip zero-click info tables (no .result-link)
+			if (!table.querySelector('a.result-link')) continue;
+
+			const rows = Array.from(table.querySelectorAll('tr'));
+			let i = 0;
+
+			while (i < rows.length) {
+				const row = rows[i];
+				const linkEl = row.querySelector('a.result-link');
+
+				if (linkEl) {
+					const title = linkEl.textContent?.trim() || '';
+					const href = linkEl.getAttribute('href') || '';
+					const url = this.unwrapRedirectUrl(href);
+
+					i++;
+					const snippetRow = rows[i];
+					const snippetEl = snippetRow?.querySelector('.result-snippet');
+					const snippet = snippetEl ? serializeHTML(snippetEl) : '';
+
+					// Skip the blank spacer row
+					i++;
+					i++;
+
+					if (title && url) {
+						results.push({ title, url, snippet });
+					}
+				} else {
+					i++;
+				}
+			}
+		}
+
+		return results;
+	}
+
+
+	// Builds the final ordered list HTML. When there are no results a
+	// short "No results" message is returned instead of an empty string.
+	// Zero-click info, when present, is prepended as a blockquote.
+	private buildListingHtml(results: SearchResult[], zeroClick?: ZeroClickInfo | null): string {
+		let html = '';
+
+		if (zeroClick) {
+			html += '<blockquote data-callout="info">';
+			html += `<p>${zeroClick.header}</p>`;
+			if (zeroClick.body) {
+				html += `<p>${zeroClick.body}</p>`;
+			}
+			html += '</blockquote>';
+		}
+
+		if (results.length === 0) {
+			html += '<p>No results found.</p>';
+			return html;
+		}
+
+		const items = results.map(result => {
+			let itemHtml = '<li>';
+			itemHtml += `<a href="${escapeHtml(result.url)}">${escapeHtml(result.title)}</a>`;
+
+			if (result.snippet) {
+				itemHtml += `<p>${result.snippet}</p>`;
+			}
+
+			itemHtml += '</li>';
+			return itemHtml;
+		});
+
+		html += `<ol>${items.join('')}</ol>`;
+
+		return html;
+	}
+}
diff --git a/tests/expected/duckduckgo-lite.md b/tests/expected/duckduckgo-lite.md
@@ -0,0 +1,19 @@
+```json
+{
+  "title": "test query",
+  "author": "",
+  "site": "DuckDuckGo Lite",
+  "published": ""
+}
+```
+
+> Zero-click info: [Gasoline](https://en.wikipedia.org/wiki/Gasoline)
+> 
+> Gasoline or petrol is a petrochemical product characterized as a transparent, yellowish and flammable liquid normally used as a fuel for spark-ignited internal combustion engines.
+
+1. [Example Result Title](https://example.com/test-page)
+	This is a **test** snippet describing the **query** result with some highlighted terms.
+2. [Another Example Page](https://example.org/another-page?foo=bar)
+	More content about **test** and **query** functionality.
+3. [Foo Example Dot Net](https://foo.example.net/page/)
+	A third result with different **query** content here.
diff --git a/tests/fixtures/duckduckgo-lite.html b/tests/fixtures/duckduckgo-lite.html
@@ -0,0 +1,96 @@
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
+<html>
+<!--{"url": "https://lite.duckduckgo.com/lite/?q=test+query"}-->
+<head>
+  <meta http-equiv="content-type" content="text/html; charset=UTF-8">
+  <meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=3.0, user-scalable=1;">
+  <title>test query at DuckDuckGo</title>
+</head>
+<body>
+  <p class='extra'>&nbsp;</p>
+  <div class="header">
+    DuckDuckGo
+  </div>
+  <p class='extra'>&nbsp;</p>
+
+  <form action="/lite/" method="post">
+      <input class='query' type="text" size="40" name="q" value="test query">
+      <input class='submit' type="submit" value="Search">
+  </form>
+
+  <p class='extra'>&nbsp;</p>
+
+  <table border="0">
+    <tr>
+      <td>Zero-click info: <a rel="nofollow" href="https://lite.duckduckgo.com/lite/?q=test+query&amp;uddg=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FGasoline">Gasoline</a></td>
+    </tr>
+    <tr>
+      <td>
+        Gasoline or petrol is a petrochemical product characterized as a transparent, yellowish and flammable liquid normally used as a fuel for spark-ignited internal combustion engines.
+        <a rel="nofollow" href="https://lite.duckduckgo.com/lite/?q=test+query&amp;uddg=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FGasoline">More at <q>Wikipedia</q></a>
+      </td>
+    </tr>
+    <tr>
+      <td>&nbsp;</td>
+    </tr>
+  </table>
+
+  <table border="0">
+    <tr>
+      <td valign="top">
+        1.&nbsp;
+      </td>
+      <td>
+        <a rel="nofollow" href="https://example.com/test-page" class='result-link'>Example Result Title</a>
+      </td>
+    </tr>
+    <tr>
+      <td>&nbsp;&nbsp;&nbsp;</td>
+      <td class='result-snippet'>This is a <b>test</b> snippet describing the <b>query</b> result with some highlighted terms.</td>
+    </tr>
+    <tr>
+      <td>&nbsp;</td>
+      <td>&nbsp;</td>
+    </tr>
+    <tr>
+      <td valign="top">
+        2.&nbsp;
+      </td>
+      <td>
+        <a rel="nofollow" href="https://example.org/another-page?foo=bar" class='result-link'>Another Example Page</a>
+      </td>
+    </tr>
+    <tr>
+      <td>&nbsp;&nbsp;&nbsp;</td>
+      <td class='result-snippet'>
+        More content about <b>test</b> and <b>query</b> functionality.
+      </td>
+    </tr>
+    <tr>
+      <td>&nbsp;</td>
+      <td>&nbsp;</td>
+    </tr>
+    <tr>
+      <td valign="top">
+        3.&nbsp;
+      </td>
+      <td>
+        <a rel="nofollow" href="https://foo.example.net/page/" class='result-link'>Foo Example Dot Net</a>
+      </td>
+    </tr>
+    <tr>
+      <td>&nbsp;&nbsp;&nbsp;</td>
+      <td class='result-snippet'>
+        A third result with different <b>query</b> content here.
+      </td>
+    </tr>
+    <tr>
+      <td>&nbsp;</td>
+      <td>&nbsp;</td>
+    </tr>
+  </table>
+
+  <p class='extra'>&nbsp;</p>
+
+</body>
+</html>