// test-cosmo-labels.test.mjs — R13b: neutralise the MECHANICAL persona occurrences that
// sit OUTSIDE headings in the reference corpus — bold labels and table cells.
//
// THE SCOPE IS A MEASUREMENT, NOT THE ORDER'S PREMISE. The dispatching order carried
// "MEKANISK 85 / REDAKSJONELT 47", itself a correction of an earlier "87/45". Re-measured
// 2026-09-12 over skills/*/references (389 files), that split is wrong too — and wrong in
// the CLASSIFIER, not in the count. Three findings, all pinned by tests below:
//
//   1. FOURTEEN bold labels are DIALOGUE SPEAKER ATTRIBUTIONS, not labels. The dialogue
//      bucket only ever matched the italic form `*Cosmo:*`; the bold form sits in front of
//      quoted speech —  **Naar kunden sier:** "..." / **Cosmo svarer:** "..."  — and
//      deleting it leaves the quote with no speaker. Editorial, not mechanical.
//   2. TEN table cells are PROVENANCE CLAIMS in the source-quality column ("Moenstre er
//      Cosmo-design", "Raadgivende innhold basert paa Cosmo-persona"). Cell position splits
//      the 26 table occurrences into three classes, not one: 12 row labels (column 0),
//      4 column headers, 10 provenance. Rewriting a provenance claim asserts something new
//      about where the content came from; one of the ten has no deletion target at all.
//      Six of the order's nine compound forms live entirely inside this class.
//   3. Leaking the OTHER way, +1 mechanical: reasoning-models-o1-o3-optimization.md:549
//      `- **For arkitekten (Cosmo):** ...` is exactly the target class, but a line-start
//      anchored regex cannot see it behind the `- ` list marker, so it was booked as prose.
//
// MEASURED SPLIT: MEKANISK 62 (46 labels + 16 table cells) / REDAKSJONELT 70. Sum 132
// unchanged. Operator-ratified 2026-09-15: run the 62; dialogue and provenance go to R14's
// editorial pass, because both hang on the never-answered #R14-persona-ramme (neutral
// frame / new named persona / none) while the 62 do not — R13 already executed
// `For Cosmo` -> `For arkitekten` across 401 headings, so this only makes labels and table
// rows consistent with a decision that has ALREADY run.
import { test } from 'node:test';
import assert from 'node:assert/strict';
import { readFileSync, globSync } from 'node:fs';
import { join, dirname } from 'node:path';
import { fileURLToPath } from 'node:url';
import {
  LABEL_MAP,
  EDITORIAL_LABELS,
  TABLE_CELL_MAP,
  EDITORIAL_CELLS,
  classifyPersonaSites,
  neutralizeLabels,
} from '../../scripts/kb-update/lib/cosmo-persona.mjs';

const PLUGIN_ROOT = join(dirname(fileURLToPath(import.meta.url)), '..', '..');
const refFiles = () => globSync('skills/*/references/**/*.md', { cwd: PLUGIN_ROOT }).sort();

// ------------------------------------------------------------------ the ratified tables

test('LABEL_MAP is the 16 ratified bold-label variants', () => {
  assert.equal(Object.keys(LABEL_MAP).length, 16);
  assert.equal(LABEL_MAP['For Cosmo:'], 'For arkitekten:');
  assert.equal(LABEL_MAP['For Cosmo Skyberg:'], 'For arkitekten:');
  assert.equal(LABEL_MAP['For arkitekten (Cosmo):'], 'For arkitekten:');
  assert.equal(LABEL_MAP['Cosmo-anbefaling:'], 'Anbefaling:');
  assert.equal(LABEL_MAP['Owner (Cosmo):'], 'Owner:');
  assert.equal(LABEL_MAP['Til Cosmo:'], 'Til arkitekten:');
  assert.equal(LABEL_MAP["Cosmo's rule of thumb:"], 'Rule of thumb:');
});

test('TABLE_CELL_MAP is the 4 ratified cell variants, bold form kept bold', () => {
  assert.equal(Object.keys(TABLE_CELL_MAP).length, 4);
  assert.equal(TABLE_CELL_MAP['For arkitekten (Cosmo)'], 'For arkitekten');
  assert.equal(TABLE_CELL_MAP['**For arkitekten (Cosmo)**'], '**For arkitekten**');
  assert.equal(TABLE_CELL_MAP['Cosmo-anbefaling'], 'Anbefaling');
  assert.equal(TABLE_CELL_MAP['Cosmos raad'.replace('raad', 'råd')], 'Råd');
});

test('no target invents a word absent from its source (R13 invariant, reused)', () => {
  // Identical rule to HEADING_MAP's: dropping words is allowed, inventing one is not,
  // except the two substitutions the operator sanctioned for the persona token.
  const SANCTIONED = new Set(['arkitekten', 'å']);
  const words = (s) => s.toLowerCase().split(/[^\p{L}\p{N}]+/u).filter(Boolean);
  for (const [from, to] of Object.entries({ ...LABEL_MAP, ...TABLE_CELL_MAP })) {
    const src = new Set(words(from));
    for (const w of words(to)) {
      assert.ok(src.has(w) || SANCTIONED.has(w), `target invents "${w}": "${from}" -> "${to}"`);
    }
  }
});

test('every mapped target is itself persona-free, and the two sets are disjoint', () => {
  for (const [from, to] of Object.entries({ ...LABEL_MAP, ...TABLE_CELL_MAP })) {
    assert.ok(!/cosmo/i.test(to), `target keeps persona: "${from}" -> "${to}"`);
  }
  for (const k of EDITORIAL_LABELS) assert.equal(LABEL_MAP[k], undefined, `"${k}" in both sets`);
  for (const k of EDITORIAL_CELLS) assert.equal(TABLE_CELL_MAP[k], undefined, `"${k}" in both sets`);
});

test('the editorial sets are the measured dialogue and provenance classes', () => {
  assert.equal(EDITORIAL_LABELS.size, 8);
  assert.ok(EDITORIAL_LABELS.has('Cosmo svarer:'));
  assert.ok(EDITORIAL_LABELS.has('Cosmo:'));
  assert.equal(EDITORIAL_CELLS.size, 10);
  assert.ok(EDITORIAL_CELLS.has('Rådgivende innhold basert på Cosmo-persona'));
});

// ------------------------------------------------------------------ KLAUSUL 2: reach

test('the corpus carries ZERO mechanical sites, and the editorial ones are all still there', () => {
  // Asserts the state AFTER the transform. The pre-transform reference point — 62 mechanical
  // in 50 files — lives in cosmo-labels-baseline.json and is read from there rather than
  // restated here: a number asserted in two places drifts, and the baseline is the artefact
  // the gate compares against. The test below pins that the two agree.
  const t = { mechLabel: 0, mechCell: 0, edLabel: 0, edCell: 0, unmapped: [] };
  for (const rel of refFiles()) {
    const s = classifyPersonaSites(readFileSync(join(PLUGIN_ROOT, rel), 'utf8'));
    t.mechLabel += s.label.mechanical;
    t.mechCell += s.table.mechanical;
    t.edLabel += s.label.editorial;
    t.edCell += s.table.editorial;
    for (const u of s.unmapped) t.unmapped.push(`${rel}: ${u}`);
  }
  assert.deepEqual(t.unmapped, [], 'an unmeasured persona site exists — the net has a hole');
  assert.equal(t.mechLabel + t.mechCell, 0, 'R13b is executed: no mechanical site remains');
  assert.equal(t.edLabel, 16, 'dialogue (14) + the order\u2019s own 2 sentence forms, held for R14');
  assert.equal(t.edCell, 10, 'provenance cells held back for R14');
});

test('the ratified tables and the gate baseline describe the SAME 62 occurrences', () => {
  // The baseline is a reference-point artefact: it records what the corpus looked like BEFORE
  // the transform, and must never be silently re-emitted. This pins its arithmetic against the
  // tables that produced it, so a regenerated-over baseline shows up as a failing test.
  const b = JSON.parse(readFileSync(join(PLUGIN_ROOT, 'scripts/kb-update/data/cosmo-labels-baseline.json'), 'utf8'));
  assert.equal(b.mechanicalLabels, 46);
  assert.equal(b.mechanicalCells, 16);
  assert.equal(b.mechanicalTotal, 62);
  assert.equal(b.mechanicalFiles, 50);
  // The 70 editorial occurrences split 16 labels + 10 cells + 44 that sit at no label or
  // cell site at all (plain prose, italic dialogue, provenance footers) — R14's, every one.
  assert.equal(b.editorialLabels + b.editorialCells, 26);
  assert.equal(b.editorialTotal - b.editorialLabels - b.editorialCells, 44);
  assert.equal(b.personaProseBefore - b.mechanicalTotal, b.personaProseAfter);
  assert.equal(b.personaProseBefore, 132, 'R13 left 132 persona occurrences outside headings');
  assert.equal(b.personaProseAfter, 70, 'R13b leaves 70 for R14');
  assert.equal(b.productOccurrences, 451, 'the Azure Cosmos DB mentions R1/R3 protect');
});

test('reach: the list-item label behind a "- " marker is INSIDE scope', () => {
  // Finding 3. A line-start-anchored net books this as prose and silently misses it.
  const src = '- **For arkitekten (Cosmo):** Baseline (erfaring + Cosmo-persona)\n';
  const s = classifyPersonaSites(src);
  assert.equal(s.label.mechanical, 1);
  const { content } = neutralizeLabels(src);
  assert.equal(content, '- **For arkitekten:** Baseline (erfaring + Cosmo-persona)\n');
});

test('reach: a label form the tables do not know is REPORTED, never silently skipped', () => {
  // KLAUSUL 2's known-positive: construct an in-class occurrence the transform cannot
  // handle and assert it surfaces as unmapped rather than passing through untouched.
  const s = classifyPersonaSites('**Cosmo-hjørnet:** noe tekst\n');
  assert.deepEqual(s.unmapped, ['label: "Cosmo-hjørnet:"']);
  assert.throws(() => neutralizeLabels('**Cosmo-hjørnet:** noe tekst\n'), /ukjent persona-etikett/);
});

// ------------------------------------------------------------------ KLAUSUL 1: referent

test('known-negative: a product label and a product table row must NOT be touched', () => {
  // The second is the dangerous one: it LOOKS like a persona label and is product.
  const src = '**Cosmos DB-anbefaling:** bruk autoscale\n\n| Azure Cosmos DB | 8000 NOK |\n';
  const s = classifyPersonaSites(src);
  assert.equal(s.label.mechanical + s.table.mechanical, 0);
  assert.deepEqual(s.unmapped, []);
  assert.equal(neutralizeLabels(src).content, src, 'product content was rewritten');
});

test('known-positive: both injected forms trip the net, genitive included', () => {
  const s = classifyPersonaSites('**For Cosmo:** x\n\n| Trade-off | Cosmos råd |\n');
  assert.equal(s.label.mechanical, 1);
  assert.equal(s.table.mechanical, 1);
});

// ------------------------------------------------------------------ KLAUSUL 3: what stays

test('neutralizeLabels leaves dialogue and provenance byte-identical', () => {
  const src = [
    '**Når kunden sier:** "Vi må teste sikkerheten"',
    '**Cosmo svarer:**',
    '1. Start med PyRIT',
    '',
    '| Beslutningsveiledning | **Baseline** | Cosmo-syntese av verified sources |',
    '',
  ].join('\n');
  assert.equal(neutralizeLabels(src).content, src);
});

test('neutralizeLabels is idempotent and never changes the line count', () => {
  const src = '**For Cosmo:** x\n| For arkitekten (Cosmo) | Baseline | y |\n';
  const once = neutralizeLabels(src).content;
  assert.equal(neutralizeLabels(once).content, once);
  assert.equal(once.split('\n').length, src.split('\n').length);
  assert.equal(once, '**For arkitekten:** x\n| For arkitekten | Baseline | y |\n');
});

test('a table row rewrites only the mapped cell, leaving its siblings byte-identical', () => {
  const src = '| For arkitekten (Cosmo) | **Baseline** | Best practices + erfaring |\n';
  assert.equal(
    neutralizeLabels(src).content,
    '| For arkitekten | **Baseline** | Best practices + erfaring |\n',
  );
});