ktg-plugin-marketplace/scripts/check-okf-parity.mjs

#!/usr/bin/env node
// check-okf-parity.mjs — the running, n-way per-file OKF §3 parity gate.
//
// WHY THIS EXISTS (docs/okf-second-brain/log.md, STEG 3): the convention's §3
// minimal contract is enforced by several INDEPENDENT checkers, and two of them
// have already diverged on the same input — catalog's frozen okf-check.mjs walks
// everything and does not normalize bytes; okr's hardened copy skips innboks/+dot-dirs
// and strips BOM/CRLF. A one-shot lift-time parity check cannot catch that drift.
// This gate makes parity a RUNNING red/green signal over an adversarial corpus, and
// it CARRIES A PROOF THAT IT CAN GO RED (the corpus's `expected: "diverge"` fixtures).
//
// SCOPE (honest): only two implementations expose a runnable per-file conformance
// surface today — catalog + okr. The gate is architected n-way; llm-ingestion-okf's
// Python checker (fase 2) and its future Node port drop into IMPLS behind an
// available() guard when they legitimately exist. It does NOT fake a third impl.
//
// CONTRACT: comparison unit is the PER-FILE outcome, not the bundle verdict — two
// checkers can both say "OK" while disagreeing on what IS a concept file. Each impl's
// existing checkBundle(root) return is normalized to { conceptCount, untyped[], okfVersion };
// the gate diverges iff those normalized outcomes are not identical across the
// available impls. Defined over DEFAULT read-mode (strictIngest OFF, not files-scoped).
//
// The runner COMPARES; it does not author the fasit. Expected agree/diverge per fixture
// lives in the corpus manifest (catalog owns the axes; propagation is spec -> fixtures ->
// implementations, never the reverse). Zero npm dependencies (node: builtins).

import { readdirSync, existsSync, readFileSync, statSync } from 'node:fs';
import { join } from 'node:path';
import { fileURLToPath } from 'node:url';

// --- Implementation registry (n-way; guarded by available()) ---
// modulePath is resolved relative to THIS file. Sibling-repo impls (okr) resolve via
// the polyrepo layout (catalog + okr siblings). available() lets a catalog-only checkout
// (e.g. CI without the sibling) degrade gracefully instead of crashing.
export const IMPLS = [
  { name: 'catalog', modulePath: new URL('./okf-check.mjs', import.meta.url) },
  { name: 'okr', modulePath: new URL('../../okr/scripts/okf-check.mjs', import.meta.url) },
];

export function implAvailable(impl) {
  try {
    return existsSync(fileURLToPath(impl.modulePath));
  } catch {
    return false;
  }
}

// Normalize one impl's checkBundle(root) return to the per-file parity contract.
// Returns null if the impl module is unavailable (guarded skip, reported by caller).
export async function normalizedOutcome(impl, root) {
  if (!implAvailable(impl)) return null;
  const mod = await import(impl.modulePath);
  if (typeof mod.checkBundle !== 'function') {
    throw new Error(`impl "${impl.name}" exports no checkBundle()`);
  }
  const r = mod.checkBundle(root); // default read-mode: no options (strictIngest off, no files)
  return {
    name: impl.name,
    conceptCount: r.scanned,
    untyped: [...(r.missingType ?? [])].sort(),
    okfVersion: r.okfVersion ?? null,
  };
}

// Signature over the fields that define per-file parity. Distinct signatures => divergence.
function signature(o) {
  return `${o.conceptCount}|${o.untyped.join(',')}|${o.okfVersion}`;
}

// Run all impls on one bundle root; decide agree vs diverge among the comparable ones.
export async function evaluateBundle(root, impls = IMPLS) {
  const outcomes = [];
  const skipped = [];
  for (const impl of impls) {
    const o = await normalizedOutcome(impl, root);
    if (o === null) skipped.push(impl.name);
    else outcomes.push(o);
  }
  const signatures = new Set(outcomes.map(signature));
  return {
    outcomes,
    skipped,
    comparableCount: outcomes.length,
    diverges: outcomes.length >= 2 && signatures.size > 1,
  };
}

// Run the whole corpus against its manifest. Pure over (corpusRoot, impls); no process.exit.
export async function runCorpus(corpusRoot, impls = IMPLS) {
  const manifestPath = join(corpusRoot, 'manifest.json');
  if (!existsSync(manifestPath)) {
    throw new Error(`corpus manifest not found: ${manifestPath}`);
  }
  const manifest = JSON.parse(readFileSync(manifestPath, 'utf8'));
  const fixtures = manifest.fixtures ?? {};
  const results = [];

  for (const [name, spec] of Object.entries(fixtures)) {
    const dir = join(corpusRoot, name);
    if (!existsSync(dir) || !statSync(dir).isDirectory()) {
      results.push({ fixture: name, axis: spec.axis, expected: spec.expected, status: 'MISSING' });
      continue;
    }
    const ev = await evaluateBundle(dir, impls);
    if (ev.comparableCount < 2) {
      results.push({ fixture: name, axis: spec.axis, expected: spec.expected, status: 'SKIPPED', ...ev });
      continue;
    }
    const actual = ev.diverges ? 'diverge' : 'agree';
    const status = actual === spec.expected ? 'PASS' : 'FAIL';
    results.push({ fixture: name, axis: spec.axis, expected: spec.expected, actual, status, ...ev });
  }

  // Orphan directories present on disk but absent from the manifest (no silent coverage gaps).
  const onDisk = readdirSync(corpusRoot, { withFileTypes: true })
    .filter((e) => e.isDirectory())
    .map((e) => e.name);
  const orphans = onDisk.filter((d) => !(d in fixtures));

  const failed = results.filter((r) => r.status === 'FAIL' || r.status === 'MISSING');
  const availableImpls = impls.filter(implAvailable).map((i) => i.name);
  return { results, orphans, failed, availableImpls, exitCode: failed.length === 0 ? 0 : 1 };
}

// --- CLI ---
const isMain = process.argv[1] && fileURLToPath(import.meta.url) === process.argv[1];
if (isMain) {
  const arg = process.argv[2];
  const corpusRoot = arg
    ? arg
    : fileURLToPath(new URL('../test/okf-parity-corpus/', import.meta.url));
  const { results, orphans, failed, availableImpls, exitCode } = await runCorpus(corpusRoot);

  const out = [];
  out.push(`OKF parity gate over: ${corpusRoot}`);
  out.push(`Implementations available (${availableImpls.length}): ${availableImpls.join(', ') || 'NONE'}`);
  if (availableImpls.length < 2) {
    out.push('  ! Fewer than 2 impls available — parity cannot be established; every fixture skipped.');
    out.push('    (This is an environment limitation, not a divergence. In a polyrepo checkout,');
    out.push('     ensure the sibling checker repos are present.)');
  }
  out.push('');
  for (const r of results) {
    const tag =
      r.status === 'PASS' && r.expected === 'diverge' ? 'PASS (red-proof)' : r.status;
    out.push(`${r.fixture.padEnd(22)} [${(r.axis || '').padEnd(16)}] expect=${(r.expected || '').padEnd(8)} ${tag}`);
    for (const o of r.outcomes ?? []) {
      out.push(`    ${o.name.padEnd(10)} count=${o.conceptCount}  untyped=[${o.untyped.join(', ')}]  okf=${o.okfVersion}`);
    }
    if (r.skipped?.length) out.push(`    (skipped, unavailable: ${r.skipped.join(', ')})`);
  }
  out.push('');
  const pass = results.filter((r) => r.status === 'PASS').length;
  const skip = results.filter((r) => r.status === 'SKIPPED').length;
  out.push(`Summary: ${results.length} fixtures — ${pass} pass, ${failed.length} fail, ${skip} skipped.`);
  if (orphans.length) out.push(`  ! Unmanifested corpus directories (add to manifest): ${orphans.join(', ')}`);
  out.push(exitCode === 0 ? 'OK: parity holds (red fixtures went red, green fixtures agreed).' : `FAIL: ${failed.length} fixture(s) off expectation.`);

  process.stdout.write(`${out.join('\n')}\n`);
  process.exit(exitCode);
}