Update server.js

server.js

@@ -2,9 +2,9 @@ const express = require("express");
 const Tesseract = require("tesseract.js");
 const fs = require("fs");
 const path = require("path");
-const { PDF2Pic } = require("pdf2pic");
-const { PDFDocument } = require("pdf-lib"); // untuk hitung jumlah halaman
-const gm = require("gm").subClass({ imageMagick: true });
+const { fromPath } = require("pdf2pic");
+// gunakan ImageMagick mode agar gm tidak cari binary gm sendiri
+const gm = require('gm').subClass({ imageMagick: true });
 
 const app = express();
 app.use(express.json({ limit: "50mb" }));
@@ -22,46 +22,39 @@ app.post("/ocr", async (req, res) => {
     const pdfPath = path.join(uploadDir, `file_${Date.now()}.pdf`);
     fs.writeFileSync(pdfPath, Buffer.from(file, "base64"));
 
-    // Hitung jumlah halaman PDF
-    const pdfBuffer = fs.readFileSync(pdfPath);
-    const pdfDoc = await PDFDocument.load(pdfBuffer);
-    const totalPages = pdfDoc.getPageCount();
-
-    const { fromPath } = require("pdf2pic");
-
-    const { PDF2Pic } = require("pdf2pic");
-
-const pdf2pic = new PDF2Pic({
-  density: 150,
-  saveFilename: "page",
-  savePath: uploadDir,
-  format: "png",
-  width: 1024,
-  height: 1024
-});
-
-const result = await pdf2pic.convert(pdfPath, 1);
-
-    // 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}`;
-
-      // Hapus file gambar sementara
-      if (fs.existsSync(result.path)) fs.unlinkSync(result.path);
+    // Konfigurasi pdf2pic
+    const pdf2pic = new PDF2Pic({
+      density: 150,
+      saveFilename: "page",
+      savePath: uploadDir,
+      format: "png",
+      width: 1024,
+      height: 1024
+    });
+
+    // Ambil jumlah halaman dari pdf-poppler (atau simple trick dengan pdf2pic)
+    // pdf2pic bisa langsung convert seluruh halaman
+    const convertAllPages = await pdf2pic.convertBulk(pdfPath, -1);
+
+    let finalText = "";
+    for (const page of convertAllPages) {
+      const { data: { text } } = await Tesseract.recognize(page.path, "ind+eng");
+      finalText += `\n--- Page ${page.page} ---\n${text}`;
+      // hapus PNG tiap halaman setelah OCR
+      fs.unlinkSync(page.path);
     }
 
-    // Hapus file PDF sementara
+    // Hapus file PDF
     fs.unlinkSync(pdfPath);
 
-    res.json({ success: true, text: allText });
+    res.json({ success: true, text: finalText.trim() });
   } catch (error) {
+    console.error("OCR Error:", 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}`);
 });