diff --git a/scripts/kb-update/backfill-category.mjs b/scripts/kb-update/backfill-category.mjs index d55094c..cd87e64 100644 --- a/scripts/kb-update/backfill-category.mjs +++ b/scripts/kb-update/backfill-category.mjs @@ -8,19 +8,26 @@ // change by more than that single line, the run aborts and writes nothing. Idempotent: // files that already carry **Category:**/**Kategori:** are skipped, so a re-run is a no-op. // +// The planning is factored into pure/importable pieces (categoryForFile + planCategoryBackfill) +// behind an isMain guard, so the frozen taxonomy and the plan can be tested without executing +// the corpus scan as an import side effect. +// +// Recovery contract: writes are crash-safe (atomicWriteSync tmp+rename — a reader sees the old +// file or the new one, never a partial); an interrupted run is recovered by re-running (idempotent). +// // Usage: node scripts/kb-update/backfill-category.mjs [--dry] -import { readFileSync, writeFileSync } from 'node:fs'; +import { readFileSync, realpathSync } from 'node:fs'; import { execSync } from 'node:child_process'; import { join, dirname } from 'node:path'; import { fileURLToPath } from 'node:url'; import { insertMetaField } from './lib/transform.mjs'; +import { atomicWriteSync } from './lib/atomic-write.mjs'; const __dirname = dirname(fileURLToPath(import.meta.url)); const PLUGIN_ROOT = join(__dirname, '..', '..'); -const dry = process.argv.includes('--dry'); -// Immediate-parent-directory basename → Category. Operator-approved 2026-07-06. -const FOLDER_CATEGORY = { +// Immediate-parent-directory basename → Category. Operator-approved 2026-07-06 (frozen). +export const FOLDER_CATEGORY = { architecture: 'Solution Architecture & Advisory', platforms: 'Microsoft AI Platforms', development: 'Microsoft AI Platforms', @@ -30,53 +37,88 @@ const FOLDER_CATEGORY = { const RE_CAT = /^\s*\*\*(Category|Kategori):\*\*/im; -const files = execSync("find skills/*/references -name '*.md'", { cwd: PLUGIN_ROOT, encoding: 'utf8' }) - .trim() - .split('\n'); - -const targets = files.filter((rel) => !RE_CAT.test(readFileSync(join(PLUGIN_ROOT, rel), 'utf8'))); - -const planned = []; -const unmatched = []; -for (const rel of targets) { - const folder = rel.split('/').slice(-2, -1)[0]; - const cat = FOLDER_CATEGORY[folder]; - if (!cat) { - unmatched.push(`${rel} (folder=${folder})`); - continue; - } - const old = readFileSync(join(PLUGIN_ROOT, rel), 'utf8'); - const out = insertMetaField(old, 'Category', cat); - // Hard invariant: exactly one line inserted (the Category line), body byte-identical. - const o = old.split('\n'); - const n = out.split('\n'); - if (n.length !== o.length + 1) throw new Error(`ABORT ${rel}: not exactly one line inserted`); - let d = 0; - while (d < o.length && o[d] === n[d]) d++; - if (n[d] !== `**Category:** ${cat}`) { - throw new Error(`ABORT ${rel}: inserted line mismatch → ${JSON.stringify(n[d])}`); - } - if (o.slice(d).join('\n') !== n.slice(d + 1).join('\n')) { - throw new Error(`ABORT ${rel}: body not byte-identical below the insertion`); - } - planned.push({ rel, cat, out }); +/** + * The Category for a reference file, keyed on its immediate-parent folder. Pure. + * @param {string} rel — plugin-relative path + * @returns {string|null} the approved Category, or null when the folder is outside the rule + */ +export function categoryForFile(rel) { + const folder = String(rel ?? '').split('/').slice(-2, -1)[0]; + return FOLDER_CATEGORY[folder] ?? null; } -if (unmatched.length) { - console.error(`ABORT — ${unmatched.length} target(s) outside the approved folder rule:`); - unmatched.forEach((u) => console.error(' ' + u)); - process.exit(1); +/** + * Plan the Category backfill under `root` (reads files; performs NO writes). Discovers every + * reference file, keeps only those missing a Category, and for each in an approved folder builds + * the insert-only mutation behind the one-line/byte-identical invariant. Files outside the + * approved folder rule are collected as `unmatched` (never planned for a write). + * @param {string} root — corpus root (contains skills/…) + * @returns {{files: string[], targets: string[], planned: {rel:string,cat:string,out:string}[], unmatched: string[]}} + */ +export function planCategoryBackfill(root) { + const files = execSync("find skills/*/references -name '*.md'", { cwd: root, encoding: 'utf8' }) + .trim() + .split('\n') + .filter(Boolean); + + const targets = files.filter((rel) => !RE_CAT.test(readFileSync(join(root, rel), 'utf8'))); + + const planned = []; + const unmatched = []; + for (const rel of targets) { + const cat = categoryForFile(rel); + if (!cat) { + const folder = String(rel).split('/').slice(-2, -1)[0]; + unmatched.push(`${rel} (folder=${folder})`); + continue; + } + const old = readFileSync(join(root, rel), 'utf8'); + const out = insertMetaField(old, 'Category', cat); + // Hard invariant: exactly one line inserted (the Category line), body byte-identical. + const o = old.split('\n'); + const n = out.split('\n'); + if (n.length !== o.length + 1) throw new Error(`ABORT ${rel}: not exactly one line inserted`); + let d = 0; + while (d < o.length && o[d] === n[d]) d++; + if (n[d] !== `**Category:** ${cat}`) { + throw new Error(`ABORT ${rel}: inserted line mismatch → ${JSON.stringify(n[d])}`); + } + if (o.slice(d).join('\n') !== n.slice(d + 1).join('\n')) { + throw new Error(`ABORT ${rel}: body not byte-identical below the insertion`); + } + planned.push({ rel, cat, out }); + } + return { files, targets, planned, unmatched }; } -const byCat = {}; -for (const p of planned) byCat[p.cat] = (byCat[p.cat] || 0) + 1; -console.log(`Reference files: ${files.length} | without a category: ${targets.length} | planned: ${planned.length}`); -console.log('By category:'); -for (const [c, n] of Object.entries(byCat).sort()) console.log(` ${n} ${c}`); +function run({ root = PLUGIN_ROOT, dry }) { + const { files, targets, planned, unmatched } = planCategoryBackfill(root); -if (dry) { - console.log('\n(dry run — no writes)'); - process.exit(0); + if (unmatched.length) { + console.error(`ABORT — ${unmatched.length} target(s) outside the approved folder rule:`); + unmatched.forEach((u) => console.error(' ' + u)); + process.exit(1); + } + + const byCat = {}; + for (const p of planned) byCat[p.cat] = (byCat[p.cat] || 0) + 1; + console.log(`Reference files: ${files.length} | without a category: ${targets.length} | planned: ${planned.length}`); + console.log('By category:'); + for (const [c, n] of Object.entries(byCat).sort()) console.log(` ${n} ${c}`); + + if (dry) { + console.log('\n(dry run — no writes)'); + return; + } + for (const p of planned) atomicWriteSync(join(root, p.rel), p.out); + console.log(`\nWrote ${planned.length} files.`); } -for (const p of planned) writeFileSync(join(PLUGIN_ROOT, p.rel), p.out); -console.log(`\nWrote ${planned.length} files.`); + +const isMain = (() => { + try { + return realpathSync(process.argv[1]) === realpathSync(fileURLToPath(import.meta.url)); + } catch { + return false; + } +})(); +if (isMain) run({ dry: process.argv.includes('--dry') }); diff --git a/scripts/kb-update/backfill-status.mjs b/scripts/kb-update/backfill-status.mjs index 53e00c3..64a757a 100644 --- a/scripts/kb-update/backfill-status.mjs +++ b/scripts/kb-update/backfill-status.mjs @@ -22,11 +22,15 @@ // **Status:** in its header window is skipped, so a re-run — including the 14 R21 files already // applied — is a no-op. Aborts and writes nothing on any drift or invariant breach. // +// Recovery contract: writes are crash-safe (atomicWriteSync tmp+rename — a reader sees the old +// file or the new one, never a partial); an interrupted run is recovered by re-running (idempotent). +// // Usage: node scripts/kb-update/backfill-status.mjs [--dry] -import { readFileSync, writeFileSync, existsSync, realpathSync } from 'node:fs'; +import { readFileSync, existsSync, realpathSync } from 'node:fs'; import { join, dirname, basename } from 'node:path'; import { fileURLToPath } from 'node:url'; import { insertMetaField } from './lib/transform.mjs'; +import { atomicWriteSync } from './lib/atomic-write.mjs'; const __dirname = dirname(fileURLToPath(import.meta.url)); const PLUGIN_ROOT = join(__dirname, '..', '..'); @@ -143,7 +147,7 @@ function run({ dry }) { for (const p of planned) console.log(` + ${p.rel} → **Status:** ${p.status}`); return; } - for (const p of planned) writeFileSync(join(PLUGIN_ROOT, p.rel), p.out); + for (const p of planned) atomicWriteSync(join(PLUGIN_ROOT, p.rel), p.out); console.log(`\nWrote ${planned.length} files.`); } diff --git a/scripts/kb-update/dedup-plain-header.mjs b/scripts/kb-update/dedup-plain-header.mjs index a384b15..553dd5e 100644 --- a/scripts/kb-update/dedup-plain-header.mjs +++ b/scripts/kb-update/dedup-plain-header.mjs @@ -21,11 +21,15 @@ // // Closes the last 4 of the corpus's Missing-Status and all 4 Missing-English-Last-updated. // +// Recovery contract: writes are crash-safe (atomicWriteSync tmp+rename — a reader sees the old +// file or the new one, never a partial); an interrupted run is recovered by re-running (idempotent). +// // Usage: node scripts/kb-update/dedup-plain-header.mjs [--dry] -import { readFileSync, writeFileSync, existsSync, realpathSync } from 'node:fs'; +import { readFileSync, existsSync, realpathSync } from 'node:fs'; import { join, dirname } from 'node:path'; import { fileURLToPath } from 'node:url'; import { boldifyPlainField, dropRedundantPlainField } from './lib/transform.mjs'; +import { atomicWriteSync } from './lib/atomic-write.mjs'; const __dirname = dirname(fileURLToPath(import.meta.url)); const PLUGIN_ROOT = join(__dirname, '..', '..'); @@ -125,7 +129,7 @@ function run({ dry }) { for (const p of planned) console.log(` ~ ${p.rel}`); return; } - for (const p of planned) writeFileSync(join(PLUGIN_ROOT, p.rel), p.out); + for (const p of planned) atomicWriteSync(join(PLUGIN_ROOT, p.rel), p.out); console.log(`\nWrote ${planned.length} files.`); } diff --git a/scripts/kb-update/lib/backup.mjs b/scripts/kb-update/lib/backup.mjs index 7a55eee..44fa2b9 100644 --- a/scripts/kb-update/lib/backup.mjs +++ b/scripts/kb-update/lib/backup.mjs @@ -1,9 +1,12 @@ -// backup.mjs — Backup + sentinel-guarded rollback for skills/-tree. +// backup.mjs — Backup + sentinel-guarded, SCOPED rollback for skills/-tree. // Zero dependencies. Uses fs.cpSync (recursive + preserveTimestamps) without // dereference (Node 22.17.x regression) and without filter (Windows symlink- -// type bug). Rollback writes a .rollback-in-progress sentinel at backupRoot -// BEFORE destructive operations and removes it on success — a crash mid- -// restore leaves the sentinel behind so detectStaleRollback() can flag it. +// type bug). restore(relPaths) rolls back ONLY the relative paths the calling +// run wrote — never the whole tree — so a parallel session's writes to sibling +// files survive and there is no destructive rm→cp window. It writes a +// .rollback-in-progress sentinel at backupRoot BEFORE the first per-file restore +// and removes it on success — a crash mid-restore leaves the sentinel behind so +// detectStaleRollback() can flag it. import { cpSync, @@ -15,8 +18,8 @@ import { unlinkSync, mkdirSync, } from 'node:fs'; -import { join, basename } from 'node:path'; -import { atomicWriteJson } from './atomic-write.mjs'; +import { join, basename, dirname } from 'node:path'; +import { atomicWriteJson, atomicWriteSync } from './atomic-write.mjs'; const META_FILENAME = '.backup-meta.json'; const SENTINEL_FILENAME = '.rollback-in-progress'; @@ -84,39 +87,41 @@ export function backupDir(srcDir, backupRoot, opts = {}) { schema_version: 1, }); - const restore = () => { + // Scoped rollback: restore ONLY the relative paths (relative to srcDir) that the calling run + // wrote this session. A file present in the backup is restored to its prior bytes atomically; a + // file ABSENT from the backup was created this run, so rolling back means deleting it. Sibling + // files a parallel session touched after this backup are never read or written, and there is no + // window where srcDir is deleted. The sentinel is written before the first mutation and cleared + // only on the success path — a throw mid-restore leaves it for detectStaleRollback(). + const restore = (relPaths) => { + if (!Array.isArray(relPaths)) { + throw new Error( + "backupDir.restore: relPaths must be an array — restore is scoped to the run's own writes", + ); + } const sentinelPath = join(backupRoot, SENTINEL_FILENAME); atomicWriteJson(sentinelPath, { backup_path: backupPath, src_dir: srcDir, + rel_paths: relPaths, started_at: new Date().toISOString(), - schema_version: 1, + schema_version: 2, }); - try { - rmSync(srcDir, { - recursive: true, - force: true, - maxRetries: 3, - retryDelay: 200, - }); - cpSync(backupPath, srcDir, { - recursive: true, - force: true, - preserveTimestamps: true, - }); - // Remove the meta file we copied back into srcDir so srcDir is clean. - const restoredMeta = join(srcDir, META_FILENAME); - if (existsSync(restoredMeta)) { + for (const rel of relPaths) { + const from = join(backupPath, rel); + const to = join(srcDir, rel); + if (existsSync(from)) { + mkdirSync(dirname(to), { recursive: true }); + atomicWriteSync(to, readFileSync(from)); // prior bytes, crash-safe (Buffer → byte-exact) + } else if (existsSync(to)) { try { - unlinkSync(restoredMeta); + unlinkSync(to); // created this run — must not exist in the rolled-back state } catch { // best-effort } } - } finally { - // Sentinel removed only on success path — leave it on throw so the - // post-mortem detector can see the orphan. } + // Success — clear the sentinel. On a throw above it is left behind for detectStaleRollback(). try { unlinkSync(sentinelPath); } catch { diff --git a/scripts/kb-update/migrate-corpus.mjs b/scripts/kb-update/migrate-corpus.mjs index 5190ab1..1a7ee32 100644 --- a/scripts/kb-update/migrate-corpus.mjs +++ b/scripts/kb-update/migrate-corpus.mjs @@ -18,7 +18,10 @@ // - advisor allowlist: only the 4 non-advisor skills are walked; a target path that // matches /ms-ai-advisor/ is a hard throw (belt-and-suspenders on the pure fn). // - atomic: atomicWriteSync only, never the raw fs writer. -// - backup: backupDir(root, backupRoot) before the first write; restore() retained. +// - backup: backupDir(root, backupRoot) before the first write. Recovery contract — restore() +// is SCOPED: it rolls back ONLY the files this run wrote, so a parallel session's writes to +// sibling files survive; a stale rollback sentinel aborts the run; old backups are pruned +// past retention after a successful batch. // - per-field idempotent: delegated to insertHeaderFields (skips set Type/Source) and // insertToc (skips a file that already hasToc / is small). // - post-write assertion: re-read each written file, assert parseTypeHeader (and, when a @@ -41,7 +44,7 @@ import { } from './lib/transform.mjs'; import { parseTypeHeader, parseSourceHeader } from './lib/kb-headers.mjs'; import { atomicWriteSync } from './lib/atomic-write.mjs'; -import { backupDir } from './lib/backup.mjs'; +import { backupDir, detectStaleRollback, cleanupOldBackups } from './lib/backup.mjs'; // The 4 non-advisor skills, enumerated explicitly (NO shell brace-expansion). Advisor // (ms-ai-advisor) is deliberately absent — it is classified in the manifest but never @@ -170,16 +173,30 @@ export function migrateCorpus({ root, backupRoot, manifest, write = false }) { return { root, write, backupPath: null, restore: null, results, counts }; } - // ONE backup before the first write — restore() undoes the whole batch. - const { backupPath, restore } = backupDir(skillsDir, backupRoot); + // A stale rollback sentinel means a prior run crashed mid-restore and the tree may be + // inconsistent — refuse to write over it; the operator must recover and clear it first. + if (detectStaleRollback(backupRoot)) { + throw new Error( + `migrateCorpus: a stale rollback sentinel is present under ${backupRoot} — a prior run ` + + `crashed mid-restore. Recover the corpus and clear .rollback-in-progress before rerunning.`, + ); + } + + // ONE backup before the first write. restore() rolls back ONLY the files THIS run wrote + // (scoped) — a parallel session's writes to sibling files are never touched. + const { backupPath, retentionDays, restore: restoreFiles } = backupDir(skillsDir, backupRoot); + const writtenRel = []; + const restore = () => restoreFiles(writtenRel); for (const r of changed) { + const rel = relative(skillsDir, r.full).split(sep).join('/'); atomicWriteSync(r.full, r.content); + writtenRel.push(rel); const written = readFileSync(r.full, 'utf8'); const typeOk = !r.expectType || parseTypeHeader(written) !== null; const sourceOk = !r.expectSource || parseSourceHeader(written) !== null; if (!typeOk || !sourceOk) { - restore(); // roll the whole batch back — never leave a partial corpus + restore(); // roll back only this run's own writes — never leave a partial corpus throw new Error( `migrateCorpus: post-write header truncated for ${r.relpath} ` + `(inserted field fell past the ${500}-byte scan window; corpus restored from backup)`, @@ -189,6 +206,9 @@ export function migrateCorpus({ root, backupRoot, manifest, write = false }) { counts.written += 1; } + // Successful batch — prune backups past retention (best-effort hygiene). + cleanupOldBackups(backupRoot, retentionDays); + return { root, write, backupPath, restore, results, counts }; } diff --git a/scripts/kb-update/relabel-dato.mjs b/scripts/kb-update/relabel-dato.mjs index dd50bfa..ccce2bb 100644 --- a/scripts/kb-update/relabel-dato.mjs +++ b/scripts/kb-update/relabel-dato.mjs @@ -19,11 +19,15 @@ // creation-vs-modified date pair) → carved out to a later dedup unit, and the collision-guard aborts on // them regardless. 8 further occurrences are body/list/placeholder and are never in scope. // +// Recovery contract: writes are crash-safe (atomicWriteSync tmp+rename — a reader sees the old +// file or the new one, never a partial); an interrupted run is recovered by re-running (idempotent). +// // Usage: node scripts/kb-update/relabel-dato.mjs [--dry] -import { readFileSync, writeFileSync, existsSync, realpathSync } from 'node:fs'; +import { readFileSync, existsSync, realpathSync } from 'node:fs'; import { join, dirname } from 'node:path'; import { fileURLToPath } from 'node:url'; import { relabelHeaderDialect } from './relabel-dialect.mjs'; +import { atomicWriteSync } from './lib/atomic-write.mjs'; const __dirname = dirname(fileURLToPath(import.meta.url)); const PLUGIN_ROOT = join(__dirname, '..', '..'); @@ -125,7 +129,7 @@ function run({ dry }) { } return; } - for (const p of planned) writeFileSync(join(PLUGIN_ROOT, p.rel), p.out); + for (const p of planned) atomicWriteSync(join(PLUGIN_ROOT, p.rel), p.out); console.log(`\nWrote ${planned.length} files.`); } diff --git a/scripts/kb-update/relabel-dialect.mjs b/scripts/kb-update/relabel-dialect.mjs index ca8f8ba..33e99e5 100644 --- a/scripts/kb-update/relabel-dialect.mjs +++ b/scripts/kb-update/relabel-dialect.mjs @@ -18,10 +18,14 @@ // label token swapped, everything else byte-identical). Idempotent: a file with no norsk header // label is skipped, so a re-run is a no-op. Aborts and writes nothing on any drift or invariant breach. // +// Recovery contract: writes are crash-safe (atomicWriteSync tmp+rename — a reader sees the old +// file or the new one, never a partial); an interrupted run is recovered by re-running (idempotent). +// // Usage: node scripts/kb-update/relabel-dialect.mjs [--dry] -import { readFileSync, writeFileSync, existsSync, realpathSync } from 'node:fs'; +import { readFileSync, existsSync, realpathSync } from 'node:fs'; import { join, dirname } from 'node:path'; import { fileURLToPath } from 'node:url'; +import { atomicWriteSync } from './lib/atomic-write.mjs'; const __dirname = dirname(fileURLToPath(import.meta.url)); const PLUGIN_ROOT = join(__dirname, '..', '..'); @@ -180,7 +184,7 @@ function run({ dry }) { } return; } - for (const p of planned) writeFileSync(join(PLUGIN_ROOT, p.rel), p.out); + for (const p of planned) atomicWriteSync(join(PLUGIN_ROOT, p.rel), p.out); console.log(`\nWrote ${planned.length} files.`); } diff --git a/tests/kb-update/test-backfill-category.test.mjs b/tests/kb-update/test-backfill-category.test.mjs new file mode 100644 index 0000000..b8097bb --- /dev/null +++ b/tests/kb-update/test-backfill-category.test.mjs @@ -0,0 +1,93 @@ +// tests/kb-update/test-backfill-category.test.mjs +// TDD for the Category-backfill driver (R20). RX-OPS2 makes it importable (isMain +// guard + extracted run()/planCategoryBackfill) so its deterministic contract can be +// pinned without executing the corpus scan as an import side effect, and routes its +// writes through the crash-safe atomic writer. The pure, frozen contract: +// - FOLDER_CATEGORY: the 5 operator-approved folder→category entries (2026-07-06). +// - categoryForFile: immediate-parent-folder lookup; unknown folder → null. +// - planCategoryBackfill: reads a corpus root, plans insert-only Category lines behind +// the same one-line/byte-identical invariant as the sibling drivers; files that +// already carry Category are idempotently skipped; unknown folders are unmatched +// (never written). + +import { test } from 'node:test'; +import assert from 'node:assert/strict'; +import { mkdtempSync, mkdirSync, rmSync, writeFileSync } from 'node:fs'; +import { tmpdir } from 'node:os'; +import { join, dirname } from 'node:path'; +import { + FOLDER_CATEGORY, + categoryForFile, + planCategoryBackfill, +} from '../../scripts/kb-update/backfill-category.mjs'; + +function withTmp(fn) { + const dir = mkdtempSync(join(tmpdir(), 'bcat-test-')); + try { + return fn(dir); + } finally { + rmSync(dir, { recursive: true, force: true }); + } +} + +function writeFile(root, rel, content) { + const path = join(root, rel); + mkdirSync(dirname(path), { recursive: true }); + writeFileSync(path, content, 'utf8'); +} + +test('FOLDER_CATEGORY — frozen operator-approved taxonomy (5 folders, 2026-07-06)', () => { + assert.deepEqual(FOLDER_CATEGORY, { + architecture: 'Solution Architecture & Advisory', + platforms: 'Microsoft AI Platforms', + development: 'Microsoft AI Platforms', + 'responsible-ai': 'Responsible AI & Governance', + 'mlops-genaiops': 'MLOps & GenAIOps', + }); +}); + +test('categoryForFile — maps immediate-parent folder; unknown folder → null', () => { + assert.equal(categoryForFile('skills/x/references/architecture/foo.md'), 'Solution Architecture & Advisory'); + assert.equal(categoryForFile('skills/x/references/mlops-genaiops/bar.md'), 'MLOps & GenAIOps'); + assert.equal(categoryForFile('skills/x/references/responsible-ai/baz.md'), 'Responsible AI & Governance'); + assert.equal(categoryForFile('skills/x/references/unknown-folder/qux.md'), null); +}); + +test('planCategoryBackfill — plans a Category-less file in a known folder (one line, right value)', () => { + withTmp((root) => { + writeFile( + root, + 'skills/ms-ai-engineering/references/mlops-genaiops/x.md', + '# X\n\n**Status:** Reference\n\n---\n\nBody.\n', + ); + const { planned, unmatched } = planCategoryBackfill(root); + assert.equal(unmatched.length, 0); + assert.equal(planned.length, 1); + assert.equal(planned[0].cat, 'MLOps & GenAIOps'); + // Insert-only invariant: exactly one extra line, the Category line. + assert.match(planned[0].out, /^\*\*Category:\*\* MLOps & GenAIOps$/m); + }); +}); + +test('planCategoryBackfill — file already carrying Category is skipped (idempotent)', () => { + withTmp((root) => { + writeFile( + root, + 'skills/ms-ai-engineering/references/mlops-genaiops/x.md', + '# X\n\n**Category:** MLOps & GenAIOps\n**Status:** Reference\n\n---\n\nBody.\n', + ); + const { planned, unmatched } = planCategoryBackfill(root); + assert.equal(planned.length, 0); + assert.equal(unmatched.length, 0); + }); +}); + +test('planCategoryBackfill — file in an unknown folder is unmatched, never planned for a write', () => { + withTmp((root) => { + writeFile(root, 'skills/ms-ai-x/references/mystery/x.md', '# X\n\n**Status:** Reference\n\n---\n\nBody.\n'); + const { planned, unmatched } = planCategoryBackfill(root); + assert.equal(planned.length, 0); + assert.equal(unmatched.length, 1); + assert.match(unmatched[0], /mystery/); + }); +}); diff --git a/tests/kb-update/test-backup-restore.test.mjs b/tests/kb-update/test-backup-restore.test.mjs index b5fa89a..247cd6d 100644 --- a/tests/kb-update/test-backup-restore.test.mjs +++ b/tests/kb-update/test-backup-restore.test.mjs @@ -141,35 +141,95 @@ test('backupDir — writes .backup-meta.json sentinel inside backup', () => { }); }); -test('restore — round-trips content after src is mutated', () => { +// --- restore(relPaths): SCOPED per-file rollback (RX-OPS2) --- +// The old restore() rm+cp'd the WHOLE srcDir, which deletes a parallel session's +// writes made after the backup and opens a destructive rm→cp window. restore() now +// takes the explicit list of relative paths THIS run wrote and rolls back only those. + +test('restore(relPaths) — scoped: rolls back only listed files, preserves parallel writes', () => { withTmp((tmp) => { const src = join(tmp, 'skills'); const root = join(tmp, '.kb-backup'); - makeSrc(src, { 'a.md': 'original', 'sub/b.md': 'original-b' }); - const original = readAll(src); + makeSrc(src, { 'a.md': 'orig-a', 'b.md': 'orig-b' }); const handle = backupDir(src, root); - // Mutate src. - writeFileSync(join(src, 'a.md'), 'mutated', 'utf8'); - writeFileSync(join(src, 'new.md'), 'extra', 'utf8'); - rmSync(join(src, 'sub'), { recursive: true, force: true }); + // This run mutates a.md; a PARALLEL session writes b.md AFTER the backup. + writeFileSync(join(src, 'a.md'), 'mutated-a', 'utf8'); + writeFileSync(join(src, 'b.md'), 'parallel-b', 'utf8'); - handle.restore(); + handle.restore(['a.md']); // roll back ONLY this run's own write - const restored = readAll(src); - assert.deepEqual(restored, original); + assert.equal(readFileSync(join(src, 'a.md'), 'utf8'), 'orig-a', 'listed file rolled back'); + assert.equal(readFileSync(join(src, 'b.md'), 'utf8'), 'parallel-b', "parallel session's write preserved"); }); }); -test('restore — sentinel is removed after successful restore', () => { +test('restore(relPaths) — scoped: deletes a file newly created this run', () => { withTmp((tmp) => { const src = join(tmp, 'skills'); const root = join(tmp, '.kb-backup'); - makeSrc(src, { 'foo.md': 'A' }); + makeSrc(src, { 'a.md': 'orig-a' }); + const handle = backupDir(src, root); - handle.restore(); - assert.equal(detectStaleRollback(root), false); + writeFileSync(join(src, 'new.md'), 'created', 'utf8'); // absent at backup time + + handle.restore(['new.md']); + + assert.equal(existsSync(join(src, 'new.md')), false, 'file created this run removed on rollback'); + assert.equal(readFileSync(join(src, 'a.md'), 'utf8'), 'orig-a', 'sibling untouched'); + }); +}); + +test('restore(relPaths) — scoped: restores a nested file to its pre-write bytes', () => { + withTmp((tmp) => { + const src = join(tmp, 'skills'); + const root = join(tmp, '.kb-backup'); + makeSrc(src, { 'sub/deep/c.md': 'orig-c', 'sub/other.md': 'orig-other' }); + + const handle = backupDir(src, root); + writeFileSync(join(src, 'sub/deep/c.md'), 'mutated-c', 'utf8'); + writeFileSync(join(src, 'sub/other.md'), 'parallel-other', 'utf8'); + + handle.restore(['sub/deep/c.md']); + assert.equal(readFileSync(join(src, 'sub/deep/c.md'), 'utf8'), 'orig-c', 'nested file rolled back'); + assert.equal(readFileSync(join(src, 'sub/other.md'), 'utf8'), 'parallel-other', 'unlisted nested sibling preserved'); + }); +}); + +test('restore(relPaths) — leaves no atomic-write temp file behind', () => { + withTmp((tmp) => { + const src = join(tmp, 'skills'); + const root = join(tmp, '.kb-backup'); + makeSrc(src, { 'a.md': 'orig-a' }); + const handle = backupDir(src, root); + writeFileSync(join(src, 'a.md'), 'mutated-a', 'utf8'); + handle.restore(['a.md']); + const names = Object.keys(readAll(src)); + assert.equal(names.some((n) => /\.tmp\./.test(n)), false, 'no restore temp file survived'); + }); +}); + +test('restore — throws when relPaths is not an array (rollback must be scoped)', () => { + withTmp((tmp) => { + const src = join(tmp, 'skills'); + const root = join(tmp, '.kb-backup'); + makeSrc(src, { 'a.md': 'A' }); + const handle = backupDir(src, root); + assert.throws(() => handle.restore(), /relPaths|scoped/i); + assert.throws(() => handle.restore('a.md'), /relPaths|scoped/i); + }); +}); + +test('restore — sentinel is written then removed after a successful scoped restore', () => { + withTmp((tmp) => { + const src = join(tmp, 'skills'); + const root = join(tmp, '.kb-backup'); + makeSrc(src, { 'a.md': 'orig-a' }); + const handle = backupDir(src, root); + writeFileSync(join(src, 'a.md'), 'mut', 'utf8'); + handle.restore(['a.md']); + assert.equal(detectStaleRollback(root), false, 'sentinel cleared on success'); }); }); diff --git a/tests/kb-update/test-driver-atomic-writes.test.mjs b/tests/kb-update/test-driver-atomic-writes.test.mjs new file mode 100644 index 0000000..3500e39 --- /dev/null +++ b/tests/kb-update/test-driver-atomic-writes.test.mjs @@ -0,0 +1,29 @@ +// tests/kb-update/test-driver-atomic-writes.test.mjs +// RX-OPS2 acceptance: every corpus-writing driver routes its writes through the +// crash-safe atomicWriteSync (tmp+rename), never a bare writeFileSync that can leave a +// half-written reference file. Mirrors the existing migrate-corpus source guard +// (test-migrate-apply.test.mjs) for the five standalone drivers. + +import { test } from 'node:test'; +import assert from 'node:assert/strict'; +import { readFileSync } from 'node:fs'; +import { join, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; + +const KB = join(dirname(fileURLToPath(import.meta.url)), '..', '..', 'scripts', 'kb-update'); + +const DRIVERS = [ + 'backfill-status.mjs', + 'backfill-category.mjs', + 'dedup-plain-header.mjs', + 'relabel-dato.mjs', + 'relabel-dialect.mjs', +]; + +for (const driver of DRIVERS) { + test(`${driver} — writes corpus files via atomicWriteSync, never bare writeFileSync`, () => { + const src = readFileSync(join(KB, driver), 'utf8'); + assert.match(src, /atomicWriteSync/, `${driver} must import/use the atomic writer`); + assert.doesNotMatch(src, /\bwriteFileSync\b/, `${driver} must not call writeFileSync directly`); + }); +} diff --git a/tests/kb-update/test-migrate-apply.test.mjs b/tests/kb-update/test-migrate-apply.test.mjs index 0147258..6aff493 100644 --- a/tests/kb-update/test-migrate-apply.test.mjs +++ b/tests/kb-update/test-migrate-apply.test.mjs @@ -279,3 +279,44 @@ test('migrateCorpus — post-write parse assertion catches a deep-preamble (>500 assert.deepEqual(readAll(root), original); }); }); + +// --- RX-OPS2: backup hygiene wired into the applier --- + +test('migrateCorpus — aborts when a stale rollback sentinel is present (prior crash)', () => { + withTmp((tmp) => { + const root = join(tmp, 'skills'); + const backupRoot = join(tmp, '.kb-backup'); + makeCorpus(root, { 'ms-ai-engineering/references/a.md': REF_SMALL }); + mkdirSync(backupRoot, { recursive: true }); + // A previous run crashed mid-restore and left its sentinel behind. + writeFileSync(join(backupRoot, '.rollback-in-progress'), JSON.stringify({ started_at: '2026-07-01T00:00:00Z' }), 'utf8'); + const manifest = { [key('ms-ai-engineering/references/a.md')]: { type: 'reference', source: 'https://learn.microsoft.com/x' } }; + assert.throws( + () => migrateCorpus({ root, backupRoot, manifest, write: true }), + /stale|rollback|recover/i, + ); + // Aborted before any write — corpus untouched. + assert.equal(readFileSync(join(root, 'ms-ai-engineering/references/a.md'), 'utf8'), REF_SMALL); + }); +}); + +test('migrateCorpus — prunes backups older than retention after a successful write', () => { + withTmp((tmp) => { + const root = join(tmp, 'skills'); + const backupRoot = join(tmp, '.kb-backup'); + makeCorpus(root, { 'ms-ai-engineering/references/a.md': REF_SMALL }); + // Seed an old backup dir (>7 days) that cleanup must remove after a good run. + const oldDir = join(backupRoot, '2026-01-01T00-00-00'); + mkdirSync(oldDir, { recursive: true }); + writeFileSync( + join(oldDir, '.backup-meta.json'), + JSON.stringify({ created_at: '2026-01-01T00:00:00.000Z', schema_version: 1 }), + 'utf8', + ); + const manifest = { [key('ms-ai-engineering/references/a.md')]: { type: 'reference', source: 'https://learn.microsoft.com/x' } }; + const report = migrateCorpus({ root, backupRoot, manifest, write: true }); + assert.equal(report.counts.written, 1); + assert.equal(existsSync(oldDir), false, 'stale-age backup pruned after successful write'); + assert.ok(existsSync(report.backupPath), "this run's fresh backup retained"); + }); +});