From 80fbe15792de3f06915d41bbb34b3ae5cc3a2d7e Mon Sep 17 00:00:00 2001
From: DEls <davide.els1@outlook.com>
Date: Fri, 18 Sep 2026 19:10:35 +0200
Subject: [PATCH] fix(dn): detect PDF content for page range extraction

---
 .../task-2-report.md                          | 37 +++++++++++++++++++
 src/lib/dn-pipeline/split.test.ts             |  5 +++
 src/lib/dn-pipeline/split.ts                  | 13 +++++--
 3 files changed, 51 insertions(+), 4 deletions(-)

diff --git a/.superpowers/sdd/2026-09-18-hybrid-delivery-note-ingestion/task-2-report.md b/.superpowers/sdd/2026-09-18-hybrid-delivery-note-ingestion/task-2-report.md
index 394abbd..80327ec 100644
--- a/.superpowers/sdd/2026-09-18-hybrid-delivery-note-ingestion/task-2-report.md
+++ b/.superpowers/sdd/2026-09-18-hybrid-delivery-note-ingestion/task-2-report.md
@@ -68,3 +68,40 @@ Exit status: 0.
 - `npm install sharp` reports existing audit vulnerabilities and an unrelated
   `isolated-vm` engine warning on Node 22; neither blocks this task's tests or
   typecheck.
+
+## Fix Round 1
+
+Addressed review findings:
+
+- `extractPageRange()` now detects PDF content using `%PDF-` magic bytes,
+  independent of declared MIME, matching the normalizer and split adapter.
+- Updated `splitPages()` documentation to describe decoding and canonical PNG
+  output instead of pass-through behavior.
+- Added a regression test proving a multi-page PDF declared as `image/jpeg`
+  extracts only the requested range.
+
+RED evidence:
+
+```text
+npx vitest run src/lib/dn-pipeline/split.test.ts
+...
+× split (requires poppler + the fixture) > extracts a page range from a PDF despite an incorrect declared MIME
+  → expected 4 to be 2
+Test Files  1 failed (1)
+Tests  1 failed | 5 passed (6)
+```
+
+GREEN and verification evidence:
+
+```text
+npx vitest run src/lib/dn-pipeline/normalise-source.test.ts src/lib/dn-pipeline/split.test.ts
+Test Files  2 passed (2)
+Tests  12 passed (12)
+
+npm run typecheck
+> mega-infrastructure-dashboard@1.0.0 typecheck
+> tsc --noEmit
+
+git diff --check
+exit status: 0
+```
diff --git a/src/lib/dn-pipeline/split.test.ts b/src/lib/dn-pipeline/split.test.ts
index aeb2c39..0f04dbf 100644
--- a/src/lib/dn-pipeline/split.test.ts
+++ b/src/lib/dn-pipeline/split.test.ts
@@ -31,6 +31,11 @@ describe.skipIf(!existsSync(FIXTURE))("split (requires poppler + the fixture)",
     expect(await pageCount(out, PDF)).toBe(2);
   });
 
+  it("extracts a page range from a PDF despite an incorrect declared MIME", async () => {
+    const out = await extractPageRange(buf, "image/jpeg", 2, 3);
+    expect(await pageCount(out, PDF)).toBe(2);
+  });
+
   it("rejects a range outside the document", async () => {
     await expect(extractPageRange(buf, PDF, 3, 9)).rejects.toThrow("out of range");
   });
diff --git a/src/lib/dn-pipeline/split.ts b/src/lib/dn-pipeline/split.ts
index ba17cb5..d88a27c 100644
--- a/src/lib/dn-pipeline/split.ts
+++ b/src/lib/dn-pipeline/split.ts
@@ -15,6 +15,11 @@ import { promisify } from "node:util";
 import { normaliseSource } from "./normalise-source";
 
 const run = promisify(execFile);
+const PDF_MAGIC = Buffer.from("%PDF-");
+
+function isPdf(buffer: Buffer): boolean {
+  return buffer.subarray(0, PDF_MAGIC.length).equals(PDF_MAGIC);
+}
 
 export interface PageImage {
   page: number;
@@ -23,7 +28,7 @@ export interface PageImage {
 
 /** Page count of a PDF. Images are always one page. */
 export async function pageCount(buffer: Buffer, mime: string): Promise<number> {
-  if (!buffer.subarray(0, 5).equals(Buffer.from("%PDF-"))) return 1;
+  if (!isPdf(buffer)) return 1;
   const dir = await mkdtemp(join(tmpdir(), "dn-count-"));
   try {
     const path = join(dir, "in.pdf");
@@ -38,8 +43,8 @@ export async function pageCount(buffer: Buffer, mime: string): Promise<number> {
 }
 
 /**
- * Render every page to PNG. An image upload passes through untouched as
- * a single page.
+ * Decode and render every page to canonical PNG. An image upload becomes one
+ * decoded, orientation-corrected PNG page.
  */
 export async function splitPages(buffer: Buffer, mime: string): Promise<PageImage[]> {
   const pages = await normaliseSource(buffer, mime);
@@ -57,7 +62,7 @@ export async function extractPageRange(
   pageFrom: number,
   pageTo: number,
 ): Promise<Buffer> {
-  if (mime !== "application/pdf") return buffer;
+  if (!isPdf(buffer)) return buffer;
 
   const src = await PDFDocument.load(buffer);
   const out = await PDFDocument.create();