From 1e431145dddcb09ecf0db30bc7843666d37a0a61 Mon Sep 17 00:00:00 2001
From: CTO <cto@paperclip.ing>
Date: Fri, 10 Apr 2026 20:36:26 +0000
Subject: [PATCH] fix: migrate pdf-parse from v1 to v2 API to resolve DOMMatrix
 error

The old v1 API (`pdfParse(buffer)`) triggered DOMMatrix dependency via
pdfjs-dist canvas rendering path. The v2 API (`new PDFParse({ data })` +
`getText()`) uses a text-only code path that works in Node.js without
DOM/canvas polyfills.

Updated all three call sites:
- src/lib/documents/index.ts (generic document extraction)
- src/app/api/norms/parse/route.ts (norm PDF parsing)
- src/lib/contracts/index.ts (contract text extraction)
- src/types/pdf-parse.d.ts (updated type declarations for v2)

Co-Authored-By: Paperclip <noreply@paperclip.ing>
---
 src/app/api/norms/parse/route.ts |  6 ++++--
 src/lib/contracts/index.ts       |  7 ++++---
 src/lib/documents/index.ts       |  6 ++++--
 src/types/pdf-parse.d.ts         | 21 +++++++++++++--------
 4 files changed, 25 insertions(+), 15 deletions(-)

diff --git a/src/app/api/norms/parse/route.ts b/src/app/api/norms/parse/route.ts
index 48ffe3f..f53a539 100644
--- a/src/app/api/norms/parse/route.ts
+++ b/src/app/api/norms/parse/route.ts
@@ -40,8 +40,10 @@ Antworte NUR mit einem JSON-Array. Kein erklaerener Text, kein Markdown, nur das
 ]`;
 
 async function extractTextFromPdf(buffer: Buffer): Promise<string> {
-  const pdfParse = (await import('pdf-parse')).default;
-  const data = await pdfParse(buffer);
+  const { PDFParse } = await import('pdf-parse');
+  const parser = new PDFParse({ data: buffer });
+  const data = await parser.getText();
+  await parser.destroy();
   return data.text;
 }
 
diff --git a/src/lib/contracts/index.ts b/src/lib/contracts/index.ts
index b17f339..2b075eb 100644
--- a/src/lib/contracts/index.ts
+++ b/src/lib/contracts/index.ts
@@ -137,9 +137,10 @@ export async function extractDocumentText(tenantId: string, documentId: string):
   let text: string;
 
   if (doc.mimeType === 'application/pdf') {
-    // Dynamic import for pdf-parse (optional dependency)
-    const pdfParse = (await import('pdf-parse')).default;
-    const pdfData = await pdfParse(fileBuffer);
+    const { PDFParse } = await import('pdf-parse');
+    const parser = new PDFParse({ data: fileBuffer });
+    const pdfData = await parser.getText();
+    await parser.destroy();
     text = pdfData.text;
   } else {
     // DOCX — use mammoth for extraction
diff --git a/src/lib/documents/index.ts b/src/lib/documents/index.ts
index 878e560..96bc0e3 100644
--- a/src/lib/documents/index.ts
+++ b/src/lib/documents/index.ts
@@ -133,10 +133,12 @@ export async function extractDocumentText(tenantId: string, documentId: string):
     let text: string;
 
     if (doc.mimeType === 'application/pdf') {
-      const pdfParse = (await import('pdf-parse')).default;
+      const { PDFParse } = await import('pdf-parse');
       let pdfData;
       try {
-        pdfData = await pdfParse(fileBuffer);
+        const parser = new PDFParse({ data: fileBuffer });
+        pdfData = await parser.getText();
+        await parser.destroy();
       } catch (pdfErr) {
         const pdfMessage = pdfErr instanceof Error ? pdfErr.message : String(pdfErr);
         if (pdfMessage.includes('encrypted') || pdfMessage.includes('password')) {
diff --git a/src/types/pdf-parse.d.ts b/src/types/pdf-parse.d.ts
index 12e5dc7..12553c1 100644
--- a/src/types/pdf-parse.d.ts
+++ b/src/types/pdf-parse.d.ts
@@ -1,14 +1,19 @@
 declare module 'pdf-parse' {
-  interface PdfData {
-    numpages: number;
-    numrender: number;
-    info: Record<string, unknown>;
-    metadata: Record<string, unknown>;
+  interface TextResult {
     text: string;
-    version: string;
+    total: number;
+    pages: Array<{ page: number; text: string }>;
   }
 
-  function pdfParse(dataBuffer: Buffer, options?: Record<string, unknown>): Promise<PdfData>;
+  interface PDFParseOptions {
+    data?: Buffer | ArrayBuffer | Uint8Array;
+    url?: string;
+  }
 
-  export default pdfParse;
+  export class PDFParse {
+    constructor(options: PDFParseOptions);
+    getText(options?: { partial?: number[] }): Promise<TextResult>;
+    getInfo(options?: { parsePageInfo?: boolean }): Promise<Record<string, unknown>>;
+    destroy(): Promise<void>;
+  }
 }