server.js

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 app = express();
app.use(express.json({ limit: "50mb" }));

// Buat folder uploads (dalam /tmp agar bisa ditulis di HuggingFace)
const uploadDir = path.join("/tmp", "uploads");
if (!fs.existsSync(uploadDir)) fs.mkdirSync(uploadDir, { recursive: true });

app.post("/ocr", async (req, res) => {
  try {
    const { file } = req.body;
    if (!file) return res.status(400).json({ success: false, error: "Base64 file string not provided" });

    // Simpan base64 → PDF
    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();

    // Konfigurasi pdf2pic
    const pdf2pic = new PDF2Pic({
      density: 150,
      saveFilename: "page",
      savePath: uploadDir,
      format: "png",
      width: 1024,
      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}`;

      // Hapus file gambar sementara
      if (fs.existsSync(result.path)) fs.unlinkSync(result.path);
    }

    // Hapus file PDF sementara
    fs.unlinkSync(pdfPath);

    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", () => {
  console.log(`🚀 OCR server running on http://0.0.0.0:${PORT}`);
});