fix: replace pdf-parse with direct pdfjs-dist to fix DOMMatrix error in production

pdf-parse v2 depends on @napi-rs/canvas (native module) which fails in
Next.js standalone Docker builds — native binaries aren't traced/copied
to the standalone output, causing DOMMatrix is not defined at runtime.

Replaced pdf-parse entirely with pdfjs-dist legacy build which works
natively in Node.js without canvas or DOM API dependencies:

- New src/lib/pdf.ts: extractTextFromPdf() using pdfjs-dist/legacy/build
- Worker file explicitly imported so Next.js file tracer includes it
- Updated all call sites: documents, norms/parse, contracts
- Removed pdf-parse from dependencies, added pdfjs-dist directly
- Changed serverExternalPackages from pdf-parse to pdfjs-dist

Verified: build succeeds, both pdf.mjs and pdf.worker.mjs present in
.next/standalone, text extraction works in standalone context.

Co-Authored-By: Paperclip <noreply@paperclip.ing>

next.config.ts

@@ -11,7 +11,7 @@ const commitHash = (() => {
 
 const nextConfig: NextConfig = {
   output: "standalone",
-  serverExternalPackages: ["pdf-parse", "drizzle-orm", "pg"],
+  serverExternalPackages: ["pdfjs-dist", "drizzle-orm", "pg"],
   env: {
     NEXT_PUBLIC_BUILD_HASH: commitHash,
   },

package-lock.json

@@ -17,7 +17,7 @@
         "mammoth": "^1.12.0",
         "next": "16.2.3",
         "next-auth": "^4.24.13",
-        "pdf-parse": "^2.4.5",
+        "pdfjs-dist": "^5.4.296",
         "pg": "^8.20.0",
         "react": "19.2.4",
         "react-dom": "19.2.4"
@@ -2058,6 +2058,7 @@
       "resolved": "https://registry.npmjs.org/@napi-rs/canvas/-/canvas-0.1.80.tgz",
       "integrity": "sha512-DxuT1ClnIPts1kQx8FBmkk4BQDTfI5kIzywAaMjQSXfNnra5UFU9PwurXrl+Je3bJ6BGsp/zmshVVFbCmyI+ww==",
       "license": "MIT",
+      "optional": true,
       "workspaces": [
         "e2e/*"
       ],
@@ -7148,26 +7149,6 @@
       "dev": true,
       "license": "MIT"
     },
-    "node_modules/pdf-parse": {
-      "version": "2.4.5",
-      "resolved": "https://registry.npmjs.org/pdf-parse/-/pdf-parse-2.4.5.tgz",
-      "integrity": "sha512-mHU89HGh7v+4u2ubfnevJ03lmPgQ5WU4CxAVmTSh/sxVTEDYd1er/dKS/A6vg77NX47KTEoihq8jZBLr8Cxuwg==",
-      "license": "Apache-2.0",
-      "dependencies": {
-        "@napi-rs/canvas": "0.1.80",
-        "pdfjs-dist": "5.4.296"
-      },
-      "bin": {
-        "pdf-parse": "bin/cli.mjs"
-      },
-      "engines": {
-        "node": ">=20.16.0 <21 || >=22.3.0"
-      },
-      "funding": {
-        "type": "github",
-        "url": "https://github.com/sponsors/mehmet-kozan"
-      }
-    },
     "node_modules/pdfjs-dist": {
       "version": "5.4.296",
       "resolved": "https://registry.npmjs.org/pdfjs-dist/-/pdfjs-dist-5.4.296.tgz",

package.json

@@ -18,7 +18,7 @@
     "mammoth": "^1.12.0",
     "next": "16.2.3",
     "next-auth": "^4.24.13",
-    "pdf-parse": "^2.4.5",
+    "pdfjs-dist": "^5.4.296",
     "pg": "^8.20.0",
     "react": "19.2.4",
     "react-dom": "19.2.4"

src/app/api/norms/parse/route.ts

@@ -39,12 +39,9 @@ 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');
-  const parser = new PDFParse({ data: buffer });
-  const data = await parser.getText();
-  await parser.destroy();
-  return data.text;
+async function extractTextFromPdfBuffer(buffer: Buffer): Promise<string> {
+  const { extractTextFromPdf } = await import('@/lib/pdf');
+  return extractTextFromPdf(buffer);
 }
 
 const CHUNK_CHAR_LIMIT = 10_000;
@@ -100,7 +97,7 @@ export async function POST(request: Request) {
 
     if (file.type === 'application/pdf' || file.name.endsWith('.pdf')) {
       try {
-        text = await extractTextFromPdf(buffer);
+        text = await extractTextFromPdfBuffer(buffer);
       } catch (err) {
         console.error('PDF parse error:', err);
         return Response.json(

src/lib/contracts/index.ts

@@ -137,11 +137,8 @@ export async function extractDocumentText(tenantId: string, documentId: string):
   let text: string;
 
   if (doc.mimeType === 'application/pdf') {
-    const { PDFParse } = await import('pdf-parse');
-    const parser = new PDFParse({ data: fileBuffer });
-    const pdfData = await parser.getText();
-    await parser.destroy();
-    text = pdfData.text;
+    const { extractTextFromPdf } = await import('@/lib/pdf');
+    text = await extractTextFromPdf(fileBuffer);
   } else {
     // DOCX — use mammoth for extraction
     const mammoth = await import('mammoth');

src/lib/documents/index.ts

@@ -133,12 +133,9 @@ export async function extractDocumentText(tenantId: string, documentId: string):
     let text: string;
 
     if (doc.mimeType === 'application/pdf') {
-      const { PDFParse } = await import('pdf-parse');
-      let pdfData;
+      const { extractTextFromPdf } = await import('@/lib/pdf');
       try {
-        const parser = new PDFParse({ data: fileBuffer });
-        pdfData = await parser.getText();
-        await parser.destroy();
+        text = await extractTextFromPdf(fileBuffer);
       } catch (pdfErr) {
         const pdfMessage = pdfErr instanceof Error ? pdfErr.message : String(pdfErr);
         if (pdfMessage.includes('encrypted') || pdfMessage.includes('password')) {
@@ -146,7 +143,6 @@ export async function extractDocumentText(tenantId: string, documentId: string):
         }
         throw new Error(`PDF konnte nicht gelesen werden: ${pdfMessage}`);
       }
-      text = pdfData.text;
 
       // Detect scanned PDFs with no text layer
       if (!text || text.trim().length === 0) {

src/lib/pdf.ts

@@ -0,0 +1,40 @@
+// PDF text extraction using pdfjs-dist legacy build (Node.js compatible, no canvas/DOMMatrix)
+
+// Force Next.js file tracer to include the worker file in standalone builds
+import 'pdfjs-dist/legacy/build/pdf.worker.mjs';
+
+/**
+ * Extract all text from a PDF buffer.
+ * Uses pdfjs-dist legacy build which works in Node.js without canvas or DOM APIs.
+ */
+export async function extractTextFromPdf(buffer: Buffer): Promise<string> {
+  const pdfjsLib = await import('pdfjs-dist/legacy/build/pdf.mjs');
+
+  // Resolve the worker path at runtime so pdfjs can find it in standalone builds
+  const { createRequire } = await import('module');
+  const require = createRequire(import.meta.url ?? __filename);
+  pdfjsLib.GlobalWorkerOptions.workerSrc = require.resolve(
+    'pdfjs-dist/legacy/build/pdf.worker.mjs',
+  );
+
+  const data = new Uint8Array(buffer);
+  const doc = await pdfjsLib.getDocument({
+    data,
+    useSystemFonts: true,
+    isEvalSupported: false,
+  }).promise;
+
+  const pages: string[] = [];
+  for (let i = 1; i <= doc.numPages; i++) {
+    const page = await doc.getPage(i);
+    const content = await page.getTextContent();
+    const pageText = content.items
+      .filter((item) => 'str' in item)
+      .map((item) => (item as { str: string }).str)
+      .join(' ');
+    pages.push(pageText);
+  }
+
+  doc.destroy();
+  return pages.join('\n');
+}

src/types/pdf-parse.d.ts

@@ -1,19 +1,45 @@
-declare module 'pdf-parse' {
-  interface TextResult {
-    text: string;
-    total: number;
-    pages: Array<{ page: number; text: string }>;
+// Worker module — imported for side-effect (file tracer) only
+declare module 'pdfjs-dist/legacy/build/pdf.worker.mjs' {}
+
+// pdfjs-dist legacy build type shim for dynamic import
+declare module 'pdfjs-dist/legacy/build/pdf.mjs' {
+  export const GlobalWorkerOptions: {
+    workerSrc: string;
+  };
+
+  export function getDocument(params: {
+    data: Uint8Array;
+    useSystemFonts?: boolean;
+    isEvalSupported?: boolean;
+    disableAutoFetch?: boolean;
+  }): { promise: Promise<PDFDocumentProxy> };
+
+  interface PDFDocumentProxy {
+    numPages: number;
+    getPage(pageNumber: number): Promise<PDFPageProxy>;
+    destroy(): void;
   }
 
-  interface PDFParseOptions {
-    data?: Buffer | ArrayBuffer | Uint8Array;
-    url?: string;
+  interface PDFPageProxy {
+    getTextContent(): Promise<TextContent>;
   }
 
-  export class PDFParse {
-    constructor(options: PDFParseOptions);
-    getText(options?: { partial?: number[] }): Promise<TextResult>;
-    getInfo(options?: { parsePageInfo?: boolean }): Promise<Record<string, unknown>>;
-    destroy(): Promise<void>;
+  interface TextContent {
+    items: Array<TextItem | TextMarkedContent>;
+  }
+
+  interface TextItem {
+    str: string;
+    dir: string;
+    width: number;
+    height: number;
+    transform: number[];
+    fontName: string;
+    hasEOL: boolean;
+  }
+
+  interface TextMarkedContent {
+    type: string;
+    id: string;
   }
 }