scripts/purge_mock_gold.mjs

#!/usr/bin/env node
// scripts/purge_mock_gold.mjs
// Remove gold entries with mock/test questions (e.g., "Q1?") from pipeline_gold.jsonl

import fs from 'fs/promises';
import path from 'path';
import { fileURLToPath } from 'url';

const __filename = fileURLToPath(import.meta.url);
const __dirname = path.dirname(__filename);
const PROJECT_ROOT = path.join(__dirname, '..');

const GOLD_PATH =
  process.env.GOLD_PATH ||
  path.join(PROJECT_ROOT, 'gold', 'pipeline_gold.jsonl');

const questionRegex = new RegExp(process.env.PURGE_QUESTION_REGEX || '^Q1\\?$');

async function main() {
  let raw;
  try {
    raw = await fs.readFile(GOLD_PATH, 'utf8');
  } catch (e) {
    if (e.code === 'ENOENT') {
      console.error(`Gold file not found: ${GOLD_PATH}`);
      process.exit(1);
    }
    throw e;
  }

  const lines = raw
    .split('\n')
    .map((l) => l.trim())
    .filter(Boolean);

  const kept = [];
  const removed = [];

  for (const line of lines) {
    try {
      const obj = JSON.parse(line);
      const q = obj.question || '';
      if (questionRegex.test(q)) {
        removed.push(line);
      } else {
        kept.push(line);
      }
    } catch {
      // if invalid JSON, keep it to avoid accidental loss
      kept.push(line);
    }
  }

  if (removed.length === 0) {
    console.log('No matching entries to purge.');
    return;
  }

  await fs.writeFile(GOLD_PATH, kept.join('\n') + '\n', 'utf8');
  console.log(`Purged ${removed.length} entries matching ${questionRegex}. Kept ${kept.length}.`);
}

main().catch((err) => {
  console.error('Purge error:', err);
  process.exit(1);
});