server.js

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 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"));

    // Convert PDF → PNG (halaman pertama)
    const options = {
      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);

    // OCR dengan Tesseract
    const { data: { text } } = await Tesseract.recognize(result.path, "ind+eng");

    // Hapus file sementara
    fs.unlinkSync(pdfPath);
    fs.unlinkSync(result.path);

    res.json({ success: true, text });
  } catch (error) {
    res.status(500).json({ success: false, error: error.message });
  }
});

const PORT = process.env.PORT || 5000;
app.listen(PORT, "0.0.0.0", () => {
  console.log(`🚀 OCR server running on http://0.0.0.0:${PORT}`);
});