How to Split a PDF: Page Ranges, Burst Extraction & Document Decomposition

/**
 * Deconstructs an arbitrary PDF file into a discrete sub-document.
 * Enforces zero telemetry, streaming ArrayBuffer isolation.
 */
import { PDFDocument } from 'pdf-lib';

export async function splitPdfBuffer(
  sourceBuffer: ArrayBuffer,
  pageIndicesToRetain: number[] // 0-based index array
): Promise<Uint8Array> {
  // Step 1: Ingest source document into immutable memory representation
  const sourceDoc = await PDFDocument.load(sourceBuffer, {
    ignoreEncryption: false,
    updateMetadata: false
  });

  // Step 2: Initialize fresh target container
  const targetDoc = await PDFDocument.create();

  // Step 3: Deep copy isolated page tree branches without cross-polluting orphans
  const copiedPages = await targetDoc.copyPages(sourceDoc, pageIndicesToRetain);

  // Step 4: Mount pages into target /Pages catalog
  copiedPages.forEach((page) => targetDoc.addPage(page));

  // Step 5: Serialize with compressed object streams (PDF 1.5+)
  const pdfBytes = await targetDoc.save({
    useObjectStreams: true,
    addDefaultPage: false
  });

  return pdfBytes; // Instant ready-for-download Blob wrapper
}