// tests/kb-update/test-audit-corpus-headers.test.mjs
// Spor 1 Step 1 — the read-only corpus-header audit that pins the migration's
// empirical ground truth: which files carry a stale (non-date) **Verified:**,
// which have a body-duplicate or pipe-delimited shape, which lack the English
// **Last updated:** / **Status:** base fields, and which header dialect each uses.
//
// auditHeaders is a pure function with an injectable reader (validate-kb-file.mjs
// pattern), so these tests never touch disk — synthetic fixtures only.

import { test } from 'node:test';
import assert from 'node:assert/strict';
import { auditHeaders } from '../../scripts/kb-update/audit-corpus-headers.mjs';

// Stale verified: value is "MCP <date>", not a clean date. Norwegian dialect,
// no English **Last updated:** (the 45-file shape).
const V_NONDATE =
  '# T\n\n**Kategori:** governance\n**Status:** GA\n**Sist oppdatert:** 2026-06\n' +
  '**Verified:** MCP 2026-06\n\n---\n\n## A\n\ntekst\n';

// Clean date verified: must NOT be flagged stale.
const V_DATE =
  '# T\n\n**Category:** governance\n**Status:** GA\n**Last updated:** 2026-06\n' +
  '**Verified:** 2026-06\n\n---\n\n## A\n\ntekst\n';

// Body-duplicate: a second **Verified:** below the first `---` but still inside
// the 500-byte header window (the mlops-fundamentals-overview.md:10 shape).
const V_BODYDUP =
  '# T\n\n**Status:** GA\n**Last updated:** 2026-06\n**Verified:** MCP 2026-06\n\n---\n\n' +
  '## A\n\n**Verified:** MCP 2026-06\n\ntekst\n';

// Pipe-delimited meta row (the ai-center-of-excellence-setup.md:4 shape).
const V_PIPE =
  '# T\n\n**Sist oppdatert:** 2026-06-17 | **Verified:** MCP 2026-06-17\n\n---\n\n' +
  '## A\n\ntekst\n';

// PLAIN (non-bold) Verified pipe-tail on the **Last updated:** line — the 87-file RX-KB1
// shape. Invisible to the bold-only RE_VERIFIED (verified stays null), which is exactly the
// M4 blindness; the audit must surface it via a SEPARATE plainVerified signal.
const V_PLAIN_PIPE =
  '# T\n\n**Last updated:** 2026-06-19 | Verified: MCP 2026-06-19\n\n---\n\n' +
  '## A\n\ntekst\n';

// Base-field gap: no **Status:** at all.
const NO_STATUS =
  '# T\n\n**Category:** x\n**Last updated:** 2026-06\n\n---\n\n## A\n\ntekst\n';

// Fully conformant, no Verified — the quiet majority.
const CLEAN =
  '# T\n\n**Category:** x\n**Status:** GA\n**Last updated:** 2026-06\n\n---\n\n## A\n\ntekst\n';

function reader(map) {
  return (p) => {
    if (!(p in map)) throw new Error(`unexpected read: ${p}`);
    return map[p];
  };
}

test('auditHeaders — non-date **Verified:** MCP is flagged stale (isDate:false)', () => {
  const r = auditHeaders(['a.md'], reader({ 'a.md': V_NONDATE }));
  const f = r.files['a.md'];
  assert.ok(f.verified, 'verified header detected');
  assert.equal(f.verified.isDate, false);
  assert.equal(f.verified.value, 'MCP 2026-06');
  assert.equal(r.aggregate.verifiedNonDate, 1);
  assert.equal(r.aggregate.verifiedDate, 0);
});

test('auditHeaders — clean date **Verified:** is NOT flagged stale', () => {
  const r = auditHeaders(['b.md'], reader({ 'b.md': V_DATE }));
  const f = r.files['b.md'];
  assert.ok(f.verified);
  assert.equal(f.verified.isDate, true);
  assert.equal(r.aggregate.verifiedNonDate, 0);
  assert.equal(r.aggregate.verifiedDate, 1);
});

test('auditHeaders — body-duplicate within 500B is flagged separately from the header hit', () => {
  const r = auditHeaders(['c.md'], reader({ 'c.md': V_BODYDUP }));
  const f = r.files['c.md'];
  assert.ok(f.verified);
  assert.equal(f.verified.duplicateWithinHeaderRegion, true);
  assert.equal(r.aggregate.verifiedDuplicate, 1);
  // the single-occurrence fixtures must not trip the duplicate flag
  const r2 = auditHeaders(['a.md'], reader({ 'a.md': V_NONDATE }));
  assert.equal(r2.files['a.md'].verified.duplicateWithinHeaderRegion, false);
});

test('auditHeaders — pipe-delimited **Verified:** row is flagged pipe, value stops at the separator', () => {
  const r = auditHeaders(['d.md'], reader({ 'd.md': V_PIPE }));
  const f = r.files['d.md'];
  assert.ok(f.verified);
  assert.equal(f.verified.pipe, true);
  assert.equal(f.verified.value, 'MCP 2026-06-17');
  assert.equal(f.verified.isDate, false);
  assert.equal(r.aggregate.verifiedPipe, 1);
  assert.equal(f.dialect, 'pipe');
});

test('auditHeaders — Norwegian **Sist oppdatert:** does not satisfy the English Last-updated check', () => {
  const r = auditHeaders(['a.md'], reader({ 'a.md': V_NONDATE }));
  const f = r.files['a.md'];
  assert.equal(f.lastUpdatedEnglish, false);
  assert.equal(r.aggregate.missingEnglishLastUpdated, 1);
  assert.equal(f.dialect, 'kategori');
});

test('auditHeaders — missing **Status:** is flagged', () => {
  const r = auditHeaders(['e.md'], reader({ 'e.md': NO_STATUS }));
  const f = r.files['e.md'];
  assert.equal(f.status, false);
  assert.equal(r.aggregate.missingStatus, 1);
  assert.equal(f.dialect, 'category');
});

test('auditHeaders — conformant file raises no flags; aggregate counts add up over a batch', () => {
  const map = {
    'a.md': V_NONDATE, 'b.md': V_DATE, 'c.md': V_BODYDUP,
    'd.md': V_PIPE, 'e.md': NO_STATUS, 'f.md': CLEAN,
  };
  const r = auditHeaders(Object.keys(map), reader(map));
  assert.equal(r.aggregate.files, 6);
  const clean = r.files['f.md'];
  assert.equal(clean.verified, null);
  assert.equal(clean.title, true);
  assert.equal(clean.status, true);
  assert.equal(clean.lastUpdatedEnglish, true);
  // a/c/d carry a non-date verified; b is the only clean date
  assert.equal(r.aggregate.verifiedPresent, 4);
  assert.equal(r.aggregate.verifiedNonDate, 3);
  assert.equal(r.aggregate.verifiedDate, 1);
  // e has no Status; d's pipe row carries no Status either
  assert.equal(r.aggregate.missingStatus, 2);
  // a (Sist oppdatert) and d (pipe row, no English Last updated) lack the English field
  assert.equal(r.aggregate.missingEnglishLastUpdated, 2);
  assert.equal(r.aggregate.missingTitle, 0);
});

test('auditHeaders — plain (non-bold) Verified pipe-tail is surfaced, though bold RE_VERIFIED is blind to it', () => {
  const r = auditHeaders(['p.md'], reader({ 'p.md': V_PLAIN_PIPE }));
  const f = r.files['p.md'];
  assert.equal(f.verified, null, 'bold-only RE_VERIFIED does not see the plain tail (M4)');
  assert.equal(f.plainVerified, true, 'the plain tail IS surfaced by the new signal');
  assert.equal(r.aggregate.plainVerifiedPipe, 1);
});

test('auditHeaders — a bold **Verified:** is NOT counted as plainVerified', () => {
  // both the header-bold and pipe-bold fixtures must leave plainVerified false
  const r = auditHeaders(['a.md', 'd.md'], reader({ 'a.md': V_NONDATE, 'd.md': V_PIPE }));
  assert.equal(r.files['a.md'].plainVerified, false);
  assert.equal(r.files['d.md'].plainVerified, false);
  assert.equal(r.aggregate.plainVerifiedPipe, 0);
});

test('auditHeaders — a clean file with no Verified at all has plainVerified false', () => {
  const r = auditHeaders(['f.md'], reader({ 'f.md': CLEAN }));
  assert.equal(r.files['f.md'].plainVerified, false);
  assert.equal(r.aggregate.plainVerifiedPipe, 0);
});

test('auditHeaders — an unreadable file is reported, not thrown', () => {
  const r = auditHeaders(['x.md'], () => { throw new Error('ENOENT'); });
  assert.equal(r.files['x.md'].error !== undefined, true);
  assert.equal(r.aggregate.files, 1);
  assert.equal(r.aggregate.readErrors, 1);
});

// ── RX-KB1b: footer date-conflict + label-whitelist ────────────────────────
// A file that claims one **Last updated:** date in the header and a DIFFERENT date
// under a norsk "Sist oppdatert" footer contradicts itself. Only the last-updated-
// EQUIVALENT label ("Sist oppdatert") is conflict-bearing; "Verifisert"/"Sist
// verifisert" record an independent axis (when content was checked vs source) and are
// whitelisted. Detection is read-only and flag-to-human — the audit never mutates.

// header 2026-06-24, italic footer "Sist oppdatert: 2026-02" — the
// stakeholder-communication-ai-decisions.md shape (genuine self-contradiction).
const FOOTER_CONFLICT =
  '# T\n\n**Category:** x\n**Status:** GA\n**Last updated:** 2026-06-24\n\n---\n\n' +
  '## A\n\ntekst\n\n*Sist oppdatert: 2026-02 av Cosmo Skyberg (AI Architect Plugin)*\n';

// header and footer agree on the month — no conflict.
const FOOTER_MATCH =
  '# T\n\n**Category:** x\n**Status:** GA\n**Last updated:** 2026-02-03\n\n---\n\n' +
  '## A\n\ntekst\n\n**Sist oppdatert:** 2026-02-03\n';

// footer carries a DIFFERENT date but under the whitelisted "Sist verifisert" label —
// an independent provenance axis, NOT a conflict.
const FOOTER_VERIFISERT_INDEP =
  '# T\n\n**Category:** x\n**Status:** GA\n**Last updated:** 2026-06-24\n\n---\n\n' +
  '## A\n\ntekst\n\n**Sist verifisert:** 2026-02-04\n';

// placeholder footer ([Dato]) is template scaffolding, not a real date — never a conflict.
const FOOTER_PLACEHOLDER =
  '# T\n\n**Category:** x\n**Status:** GA\n**Last updated:** 2026-06-24\n\n---\n\n' +
  '## A\n\n## Versjon og godkjenning\n- **Sist oppdatert:** [Dato]\n';

test('auditHeaders — footer "Sist oppdatert" date ≠ header Last updated is flagged as a conflict', () => {
  const r = auditHeaders(['a.md'], reader({ 'a.md': FOOTER_CONFLICT }));
  const f = r.files['a.md'];
  assert.ok(f.footerConflict, 'conflict object present');
  assert.equal(f.footerConflict.headerLastUpdated, '2026-06-24');
  assert.deepEqual(f.footerConflict.footerDates, ['2026-02']);
  assert.equal(r.aggregate.footerConflicts, 1);
});

test('auditHeaders — footer "Sist oppdatert" matching the header month raises no conflict', () => {
  const r = auditHeaders(['b.md'], reader({ 'b.md': FOOTER_MATCH }));
  assert.equal(r.files['b.md'].footerConflict, null);
  assert.equal(r.aggregate.footerConflicts, 0);
});

test('auditHeaders — a differing "Sist verifisert" footer is whitelisted, not a conflict', () => {
  const r = auditHeaders(['c.md'], reader({ 'c.md': FOOTER_VERIFISERT_INDEP }));
  assert.equal(r.files['c.md'].footerConflict, null);
  assert.equal(r.aggregate.footerConflicts, 0);
});

test('auditHeaders — a placeholder "[Dato]" footer never counts as a conflict', () => {
  const r = auditHeaders(['d.md'], reader({ 'd.md': FOOTER_PLACEHOLDER }));
  assert.equal(r.files['d.md'].footerConflict, null);
  assert.equal(r.aggregate.footerConflicts, 0);
});

test('auditHeaders — footer conflict needs BOTH a header date and a norsk footer date', () => {
  // V_NONDATE has a norsk "Sist oppdatert" but no English header Last updated → no conflict.
  // V_DATE has an English header date but no norsk footer → no conflict.
  const r = auditHeaders(['a.md', 'b.md'], reader({ 'a.md': V_NONDATE, 'b.md': V_DATE }));
  assert.equal(r.files['a.md'].footerConflict, null);
  assert.equal(r.files['b.md'].footerConflict, null);
  assert.equal(r.aggregate.footerConflicts, 0);
});

// ── RX-HDR (dissolved → permanent audit signal): R7-unstampable detection ───
// RX-HDR was replanned as a 10-file header-slanking unit. Premiss-verification against
// ground truth (simulating the real R7 stamper insertVerifiedFields over all 327 non-
// advisor files, 2026-07-17) collapsed the population to 1: only headers so byte-dense
// that BOTH stamps (**Verified:** + **Verified by:**) cannot fit inside the 500-byte
// window are truly "passing-men-ustempelbar". Field-reordering alone cannot fix such a
// file (its meta lines carry fat prose tails — a VALUE-slim is required, out of RX-HDR's
// scope), so the unit dissolved into R7's existing "header-slanking kreves" flag path.
// This signal makes the population permanent so it can never silently grow: the audit
// runs the exact stamper R7 uses (representative stamp date + judge-v3.1) and flags any
// file it would reject. [[gap-discipline-must-close]] — the lint signal IS the mechanism.

// A header so dense that inserting Verified + Verified by lands past 500B — the
// gpt5-gpt41-pricing-models.md shape (a ~175B **Status:** prose tail). `---` sits at
// byte 500, so the two stamps (~53B) cannot fit however the meta lines are ordered.
const UNSTAMPABLE =
  '# T\n\n**Last updated:** 2026-06\n**Status:** ' + 'GA '.repeat(140).trim() +
  '\n**Category:** x\n**Type:** reference\n\n---\n\n## A\n\ntekst\n';

test('auditHeaders — a header too dense for R7 to stamp within 500B is flagged unstampable', () => {
  const r = auditHeaders(['u.md'], reader({ 'u.md': UNSTAMPABLE }));
  assert.equal(r.files['u.md'].stampable, false, 'dense header is not R7-stampable');
  assert.equal(r.aggregate.unstampable, 1);
});

test('auditHeaders — a conformant header is stampable and not counted unstampable', () => {
  const r = auditHeaders(['f.md'], reader({ 'f.md': CLEAN }));
  assert.equal(r.files['f.md'].stampable, true, 'clean header is R7-stampable');
  assert.equal(r.aggregate.unstampable, 0);
});