feat: implement checksum-based duplicate detection for split email attachments

paperless-backend/src/email/email-import.service.ts

@@ -16,6 +16,7 @@ import { PdfService } from '../preprocessing/pdf.service';
 import * as path from 'path';
 import * as os from 'os';
 import * as fs from 'fs/promises';
+import * as crypto from 'crypto';
 
 @Injectable()
 export class EmailImportService {
@@ -154,6 +155,25 @@ export class EmailImportService {
     }
   }
 
+  // --- Checksum Check for Split Documents ---
+  async checkSplitChecksum(attachmentId: number, pages: { start: number; end: number }): Promise<boolean> {
+    const content = await this.contentRepo.findOne({ where: { AttachmentEntityId: attachmentId } });
+    if (!content) return false;
+
+    const pdfDoc = await PDFDocument.load(content.Content1, { ignoreEncryption: true });
+    const total = pdfDoc.getPageCount();
+    const startIdx = Math.max(1, pages.start) - 1;
+    const endIdx = Math.min(pages.end === 999 ? total : pages.end, total) - 1;
+
+    const sliced = await PDFDocument.create();
+    const indices = Array.from({ length: endIdx - startIdx + 1 }, (_, i) => startIdx + i);
+    const copied = await sliced.copyPages(pdfDoc, indices);
+    copied.forEach(p => sliced.addPage(p));
+
+    const checksum = crypto.createHash('md5').update(Buffer.from(await sliced.save())).digest('hex');
+    return this.paperlessService.checksumExists(checksum);
+  }
+
   // --- Print Preview ---
   async generatePrintPdf(attachmentId: number, barcodeData: any): Promise<Buffer> {
     const content = await this.contentRepo.findOne({ where: { AttachmentEntityId: attachmentId } });