Function extractAssets

extractAssets(
input: DocumentInput,
baseUrl?: string | URL | null,
): AssetsMetadata
Extract assets metadata from HTML.
Parameters
- input: DocumentInput
  Parsed HTML document or raw HTML string
- OptionalbaseUrl: string | URL | null
  Optional base URL for resolving relative URLs
Returns AssetsMetadata
Assets metadata object with categorized URLs
Remarks
Extracts all external assets referenced in the document, organized by type. All URLs are normalized to absolute format based on the document's base URL.

The extractor finds assets from:
- Images: <img>, <picture>, srcset, OpenGraph meta tags
- Stylesheets: <link rel="stylesheet">
- Scripts: <script src>
- Fonts: CSS @font-face and url() with font extensions
- Media: <video>, <audio>, <source>, <track>
- Manifests: <link rel="manifest">
- Preloads: <link rel="preload"> and <link rel="prefetch">
- Connection hints: <link rel="dns-prefetch"> and <link rel="preconnect">
Example
```
// With parsed document (recommended for multiple extractions)
const doc = parseHTML(htmlString);
const assets = extractAssets(doc, 'https://example.com');

// Or directly with HTML string
const assets = extractAssets(htmlString, 'https://example.com');
```
- Defined in src/metadata/assets/extract.ts:55