cve-dashboard/backend/helpers/excelProcessor.js

const { spawn } = require('child_process');
const path = require('path');
const fs = require('fs');

/**
 * Process vulnerability report Excel file by splitting CVE IDs into separate rows
 * @param {string} inputPath - Path to original Excel file
 * @param {string} outputPath - Path for processed Excel file
 * @returns {Promise<{original_rows: number, processed_rows: number, output_path: string}>}
 */
function processVulnerabilityReport(inputPath, outputPath) {
  return new Promise((resolve, reject) => {
    const scriptPath = path.join(__dirname, '..', 'scripts', 'split_cve_report.py');

    // Verify script exists
    if (!fs.existsSync(scriptPath)) {
      return reject(new Error(`Python script not found: ${scriptPath}`));
    }

    // Verify input file exists
    if (!fs.existsSync(inputPath)) {
      return reject(new Error(`Input file not found: ${inputPath}`));
    }

    const python = spawn('python3', [scriptPath, inputPath, outputPath]);

    let stdout = '';
    let stderr = '';
    let timedOut = false;

    // 30 second timeout
    const timeout = setTimeout(() => {
      timedOut = true;
      python.kill();
      reject(new Error('Processing timed out. File may be too large or corrupted.'));
    }, 30000);

    python.stdout.on('data', (data) => {
      stdout += data.toString();
    });

    python.stderr.on('data', (data) => {
      stderr += data.toString();
    });

    python.on('close', (code) => {
      clearTimeout(timeout);

      if (timedOut) return;

      if (code !== 0) {
        // Parse Python error messages
        if (stderr.includes('Sheet') && stderr.includes('not found')) {
          return reject(new Error('Invalid Excel file. Expected "Vulnerabilities" sheet with "CVE ID" column.'));
        }
        if (stderr.includes('pandas') || stderr.includes('openpyxl')) {
          return reject(new Error('Python dependencies missing. Run: pip3 install pandas openpyxl'));
        }
        return reject(new Error(`Python script failed: ${stderr || 'Unknown error'}`));
      }

      // Parse output for row counts
      const originalMatch = stdout.match(/Original rows:\s*(\d+)/);
      const newMatch = stdout.match(/New rows:\s*(\d+)/);

      if (!originalMatch || !newMatch) {
        return reject(new Error('Failed to parse row counts from Python output'));
      }

      // Verify output file was created
      if (!fs.existsSync(outputPath)) {
        return reject(new Error('Processed file was not created'));
      }

      resolve({
        original_rows: parseInt(originalMatch[1]),
        processed_rows: parseInt(newMatch[1]),
        output_path: outputPath
      });
    });

    python.on('error', (err) => {
      clearTimeout(timeout);
      if (err.code === 'ENOENT') {
        reject(new Error('Python 3 is required but not found. Please install Python.'));
      } else {
        reject(err);
      }
    });
  });
}

module.exports = { processVulnerabilityReport };