/** Isomorphic content-extraction helpers shared by the REST handlers (server + * any downstream runtime). `defuddle` and `linkedom` are Workers-compatible * (both verified 2026-09-01 via source-code audit — zero `node:` imports in * the paths this module uses). */ /** * Parse HTML with linkedom and extract content with Defuddle. Dynamic-imports * both libraries because they are heavy (linkedom alone is ~1MB) and not always * needed for every REST request. */ export declare function extractWithDefuddle(rawHtml: string, pageUrl: string, markdown: boolean): Promise; /** Extract the standard set of metadata fields from a Defuddle result. */ export declare function metadataFromDefuddle(result: Awaited>): Record; /** * Run the standard format-extraction matrix used by `/v1/content` and * `/v1/scrape`. Given a raw HTML string + page URL + a way to fetch the page's * innerText + a list of requested formats, returns the filled-in `content` map * and optional `metadata` object. */ export declare function extractFormats(rawHtml: string, pageUrl: string, innerText: () => Promise, formats: ReadonlyArray<"html" | "text" | "markdown" | "readability">): Promise<{ content: Record; metadata: Record | undefined; }>; //# sourceMappingURL=helpers.d.ts.map