import { LiteParseConfig } from '@llamaindex/liteparse'; export interface LiteParseHTMLOptions { addPageNumbers?: boolean; addCitation?: boolean; liteParseOptions?: Partial; } /** * Converts a PDF (URL or ArrayBuffer) into HTML using LiteParse's spatial text * extraction, mirroring the return shape of `convertPDFToHTML`. * @param pdfURLOrBuffer - URL to a PDF file or buffer from fs.readFile * @param options.addPageNumbers default=false - Adds `[n]` markers at each page boundary * @param options.addCitation default=true - Populates `title`/`author` from PDF metadata * @param options.liteParseOptions - Passed through to the `LiteParse` constructor; * defaults to `{ ocrEnabled: false, ocrFailureFatal: false }` (use detectPdfNeedsOcr * to decide when a document is worth re-parsing with `ocrEnabled: true`) * @returns `{ html, title, author, format: "pdf" }`, or `{ error }` on failure * @category Extract */ export declare function convertPDFToHTMLWithLiteParse(pdfURLOrBuffer: any, options?: LiteParseHTMLOptions): Promise<{ error: any; author?: undefined; title?: undefined; html?: undefined; format?: undefined; } | { author: string; title: string; html: any; format: string; error?: undefined; }>;