In-Page Extraction

extractInPage() runs a single page.evaluate() call to extract links, media, and metadata directly inside the browser context. This eliminates the HTML→string→parse round-trip that normally happens when using Cheerio or HTMLRewriter.

When to Use

Use extractInPage when:

• Content is rendered by JavaScript (SPAs)
• You need the browser's resolved URLs (relative→absolute)
• You want to avoid the HTML serialization overhead

Usage

import { extractInPage } from "feedstock";

crawler.setHook("beforeReturnHtml", async (page) => {
  const data = await extractInPage(page);

  console.log(`${data.links.internal.length} internal links`);
  console.log(`${data.links.external.length} external links`);
  console.log(`${data.media.images.length} images`);
  console.log(`Title: ${data.metadata.title}`);
});

What's Extracted

Links

Internal and external links with href, text, title, and baseDomain. URLs are already resolved to absolute by the browser.

Media

Images with src, alt, width, score, and format. Videos and audio sources.

Metadata

title, description, keywords, author, language, Open Graph fields (ogTitle, ogImage, ogUrl, ogType, ogSiteName), canonical, and JSON-LD structured data.

Result Type

interface InPageExtractionResult {
  links: {
    internal: LinkData[];
    external: LinkData[];
  };
  media: {
    images: MediaData[];
    videos: MediaData[];
    audios: MediaData[];
  };
  metadata: Record<string, unknown>;
}