Update server.js

server.js

@@ -2,16 +2,15 @@ const express = require("express");
 const Tesseract = require("tesseract.js");
 const fs = require("fs");
 const path = require("path");
-const { fromPath } = require("pdf2pic");
-// gunakan ImageMagick mode agar gm tidak cari binary gm sendiri
-const gm = require('gm').subClass({ imageMagick: true });
-
+const PDF2Pic = require("pdf2pic");
+const { PDFDocument } = require("pdf-lib"); // untuk hitung jumlah halaman
+const gm = require("gm").subClass({ imageMagick: true });
 
 const app = express();
-app.use(express.json({ limit: '50mb' }));
+app.use(express.json({ limit: "50mb" }));
 
 // Buat folder uploads (dalam /tmp agar bisa ditulis di HuggingFace)
-const uploadDir = path.join('/tmp', 'uploads');
+const uploadDir = path.join("/tmp", "uploads");
 if (!fs.existsSync(uploadDir)) fs.mkdirSync(uploadDir, { recursive: true });
 
 app.post("/ocr", async (req, res) => {
@@ -23,33 +22,43 @@ app.post("/ocr", async (req, res) => {
     const pdfPath = path.join(uploadDir, `file_${Date.now()}.pdf`);
     fs.writeFileSync(pdfPath, Buffer.from(file, "base64"));
 
-    // Convert PDF → PNG (halaman pertama)
-    const options = {
+    // Hitung jumlah halaman PDF
+    const pdfBuffer = fs.readFileSync(pdfPath);
+    const pdfDoc = await PDFDocument.load(pdfBuffer);
+    const totalPages = pdfDoc.getPageCount();
+
+    // Konfigurasi pdf2pic
+    const pdf2pic = new PDF2Pic({
       density: 150,
       saveFilename: "page",
       savePath: uploadDir,
       format: "png",
       width: 1024,
-      height: 1024
-    };
-    const storeAsImage = fromPath(pdfPath, options);
-    const pageToConvertAsImage = 1;
-    const result = await storeAsImage(pageToConvertAsImage);
+      height: 1024,
+    });
+
+    let allText = "";
+
+    // Loop semua halaman PDF
+    for (let page = 1; page <= totalPages; page++) {
+      const result = await pdf2pic.convert(pdfPath, page);
+      const { data: { text } } = await Tesseract.recognize(result.path, "ind+eng");
+      allText += `\n\n--- Halaman ${page} ---\n${text}`;
 
-    // OCR dengan Tesseract
-    const { data: { text } } = await Tesseract.recognize(result.path, "ind+eng");
+      // Hapus file gambar sementara
+      if (fs.existsSync(result.path)) fs.unlinkSync(result.path);
+    }
 
-    // Hapus file sementara
+    // Hapus file PDF sementara
     fs.unlinkSync(pdfPath);
-    fs.unlinkSync(result.path);
 
-    res.json({ success: true, text });
+    res.json({ success: true, text: allText });
   } catch (error) {
     res.status(500).json({ success: false, error: error.message });
   }
 });
 
 const PORT = process.env.PORT || 7860;
-app.listen(PORT, '0.0.0.0', () => {
+app.listen(PORT, "0.0.0.0", () => {
   console.log(`🚀 OCR server running on http://0.0.0.0:${PORT}`);
 });