kaboot/server/sandbox/convert.js

import { createServer } from 'http';
import { exec } from 'child_process';
import { writeFile, readFile, unlink, mkdir, readdir } from 'fs/promises';
import { existsSync } from 'fs';
import { join } from 'path';
import { randomUUID } from 'crypto';

const PORT = process.env.PORT || 3002;
const TEMP_DIR = '/tmp/convert';
const TIMEOUT_MS = 30000;
const MAX_FILE_SIZE = 50 * 1024 * 1024;
const MAX_CONCURRENT = parseInt(process.env.MAX_CONCURRENT || '3', 10);

const EXTENSION_TO_OUTPUT = {
  '.ppt': 'pptx',
  '.doc': 'docx',
  '.xls': 'xlsx',
};

let activeJobs = 0;
let queuedJobs = 0;

async function cleanup(dir) {
  if (!existsSync(dir)) return;
  try {
    for (const file of await readdir(dir)) {
      await unlink(join(dir, file)).catch(() => {});
    }
    await unlink(dir).catch(() => {});
  } catch {}
}

function runLibreOffice(inputPath, outputDir, outputExt) {
  return new Promise((resolve, reject) => {
    const timer = setTimeout(() => {
      reject(new Error('timeout'));
    }, TIMEOUT_MS);

    exec(
      `libreoffice --headless --convert-to ${outputExt} --outdir "${outputDir}" "${inputPath}"`,
      { maxBuffer: 10 * 1024 * 1024 },
      (error) => {
        clearTimeout(timer);
        if (error) reject(error);
        else resolve();
      }
    );
  });
}

async function processConversion(buffer, ext) {
  const jobId = randomUUID();
  const jobDir = join(TEMP_DIR, jobId);
  const inputPath = join(jobDir, `input${ext}`);
  const outputExt = EXTENSION_TO_OUTPUT[ext];
  const outputPath = join(jobDir, `input.${outputExt}`);

  try {
    await mkdir(jobDir, { mode: 0o700 });
    await writeFile(inputPath, buffer, { mode: 0o600 });

    await runLibreOffice(inputPath, jobDir, outputExt);

    if (!existsSync(outputPath)) {
      throw new Error('no_output');
    }

    return await readFile(outputPath);
  } finally {
    await cleanup(jobDir);
  }
}

async function handleConvert(req, res) {
  if (activeJobs >= MAX_CONCURRENT) {
    queuedJobs++;
    if (queuedJobs > MAX_CONCURRENT * 2) {
      queuedJobs--;
      res.writeHead(503, { 'Content-Type': 'application/json' });
      res.end(JSON.stringify({ error: 'Server busy, try again later' }));
      return;
    }

    await new Promise((resolve) => {
      const check = setInterval(() => {
        if (activeJobs < MAX_CONCURRENT) {
          clearInterval(check);
          queuedJobs--;
          resolve();
        }
      }, 100);
    });
  }

  activeJobs++;

  const chunks = [];
  let size = 0;

  req.on('data', (chunk) => {
    size += chunk.length;
    if (size <= MAX_FILE_SIZE) {
      chunks.push(chunk);
    }
  });

  req.on('end', async () => {
    if (size > MAX_FILE_SIZE) {
      activeJobs--;
      res.writeHead(413, { 'Content-Type': 'application/json' });
      res.end(JSON.stringify({ error: 'File too large' }));
      return;
    }

    const url = new URL(req.url, `http://localhost:${PORT}`);
    const ext = url.searchParams.get('ext');

    if (!ext || !EXTENSION_TO_OUTPUT[ext]) {
      activeJobs--;
      res.writeHead(400, { 'Content-Type': 'application/json' });
      res.end(JSON.stringify({ error: 'Invalid or missing extension parameter' }));
      return;
    }

    try {
      const converted = await processConversion(Buffer.concat(chunks), ext);
      res.writeHead(200, {
        'Content-Type': 'application/octet-stream',
        'Content-Length': converted.length,
      });
      res.end(converted);
    } catch (err) {
      const message = err.message === 'timeout'
        ? 'Conversion timed out'
        : err.message === 'no_output'
        ? 'Conversion produced no output'
        : 'Conversion failed';
      res.writeHead(500, { 'Content-Type': 'application/json' });
      res.end(JSON.stringify({ error: message }));
    } finally {
      activeJobs--;
    }
  });
}

const server = createServer((req, res) => {
  if (req.method === 'GET' && req.url === '/health') {
    res.writeHead(200, { 'Content-Type': 'application/json' });
    res.end(JSON.stringify({ status: 'ok', active: activeJobs, queued: queuedJobs }));
    return;
  }

  if (req.method === 'POST' && req.url?.startsWith('/convert')) {
    handleConvert(req, res);
    return;
  }

  res.writeHead(404, { 'Content-Type': 'application/json' });
  res.end(JSON.stringify({ error: 'Not found' }));
});

if (!existsSync(TEMP_DIR)) {
  await mkdir(TEMP_DIR, { recursive: true, mode: 0o700 });
}

server.listen(PORT, '0.0.0.0', () => {
  console.log(`Sandbox running on port ${PORT} (max ${MAX_CONCURRENT} concurrent)`);
});