The windows-test-parity ratchet greps test source for fs.rmSync-without-
maxRetries (and six other Windows-portability anti-patterns), failing when an
integer offender COUNT exceeds a frozen baseline (rmSync: 95). A count ratchet
is a Goodhart metric: fixing one offender and adding another keeps the count
constant, so a new defect slips through green. Replace it — and every other
count ratchet in the repo — with a layered, masking-proof design.
Behavioral seam test
- tests/helpers-cleanup.test.cjs proves helpers.cleanup() carries the Windows
EBUSY retry budget. cleanup() delegates retries to Node's fs.rmSync via
maxRetries (it owns no loop), so the test asserts the option contract
(recursive/force/maxRetries>0/retryDelay>0) + real-FS removal + the cwd-guard,
rather than a loop that does not exist. The EBUSY risk is now tested ONCE at
the helper, not approximated textually at every call site.
Write-time ESLint rule (AST-accurate, replaces the grep)
- eslint-rules/no-raw-rmsync-in-tests.cjs (error in tests/**/*.test.cjs) bans
raw fs.rmSync, steering to cleanup(). Catches member, computed (fs['rmSync']),
destructured and aliased forms; escape hatch is inline
`// eslint-disable-next-line local/no-raw-rmsync-in-tests -- <reason>` only.
- Migrated 336 raw fs.rmSync teardown calls across ~116 test files to cleanup().
~18 genuinely load-bearing sites (mid-test SUT/fault-injection removals,
error-swallowing or name-colliding local teardown helpers) keep the raw call
with an inline eslint-disable + reason.
Shared anti-ratchet primitive
- scripts/lib/allowlist-ratchet.cjs:
- assertWithinAllowlist: fails on NOVEL ids (new offender introduced) AND on
STALE ids (a known offender was fixed but not pruned) — identity, not count,
and a ratchet DOWN toward zero.
- assertTightCeiling: a size/length budget whose ceiling must stay within a
grace band of the high-water mark, so budgets may only tighten, never creep.
Ratchets converted onto the primitive
- windows-test-parity-guard.test.cjs: rmSync rule deleted (now ESLint-enforced);
the remaining six patterns moved from integer baselines to named-set
allowlists with ratchet-down.
- scripts/lint-test-file-count.{cjs,allowlist.json}: per-module integer counts →
named filename sets (closes the swap-a-file-keep-the-count blind spot); a
module dropping under cap now FAILS to force pruning its allowlist entry.
- enh-2790 skill-count `<= 63` → named skill allowlist (ratchets toward ~58).
Size budgets hardened (tighten-only)
- agent-size / workflow-size / feat-3039 help-tiered: ceilings lowered to the
current high-water mark and an assertTightCeiling anti-creep check added per
tier. Fixed external-contract limits (description ≤100 chars, agent ≤100 KB)
are intentionally left as-is — they are not grandfathered creeping budgets.
No user-facing behavior change (tests + tooling only); no USER_FACING_PREFIXES
touched, so no changeset fragment is required.
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
246 lines
10 KiB
JavaScript
246 lines
10 KiB
JavaScript
'use strict';
|
|
|
|
process.env.GSD_TEST_MODE = '1';
|
|
|
|
/**
|
|
* Bug #2969: /gsd-reapply-patches Step 5 hunk verification gate reports
|
|
* success on lost content because the LLM-driven workflow fills in
|
|
* "verified: yes" without actually checking content presence.
|
|
*
|
|
* Fix: deterministic verifier script (scripts/verify-reapply-patches.cjs)
|
|
* that the workflow calls.
|
|
*
|
|
* Per the repo's no-source-grep testing standard (CONTRIBUTING.md):
|
|
* tests must assert on TYPED structured fields — not regex/substring
|
|
* matching against script output, formatter prose, or file content.
|
|
*
|
|
* The script's --json mode emits a structured report whose `reason`
|
|
* field is a stable enum (exposed as REASON), and whose `missing` field
|
|
* is an array of typed strings (exact set membership, not substring).
|
|
* Every assertion below is a deepEqual / equal / Array.includes against
|
|
* those typed fields. Zero regex, zero String#includes on text.
|
|
*/
|
|
|
|
const { test, describe, before, after } = require('node:test');
|
|
const assert = require('node:assert/strict');
|
|
const fs = require('node:fs');
|
|
const os = require('node:os');
|
|
const path = require('node:path');
|
|
const cp = require('node:child_process');
|
|
const { cleanup } = require('./helpers.cjs');
|
|
|
|
const ROOT = path.join(__dirname, '..');
|
|
// Script lives at get-shit-done/bin/ so the installer ships it under
|
|
// `${GSD_HOME}/get-shit-done/bin/` (issue #2994). The top-level scripts/
|
|
// directory is not copied to user installs.
|
|
const SCRIPT = path.join(ROOT, 'get-shit-done', 'bin', 'verify-reapply-patches.cjs');
|
|
const { REASON } = require(SCRIPT);
|
|
|
|
let tmpRoot;
|
|
let patchesDir;
|
|
let configDir;
|
|
let pristineDir;
|
|
|
|
function writeFile(absPath, content) {
|
|
fs.mkdirSync(path.dirname(absPath), { recursive: true });
|
|
fs.writeFileSync(absPath, content);
|
|
}
|
|
|
|
function resetFixture({ withPristine = true } = {}) {
|
|
for (const dir of [patchesDir, configDir, pristineDir]) {
|
|
cleanup(dir);
|
|
}
|
|
fs.mkdirSync(patchesDir);
|
|
fs.mkdirSync(configDir);
|
|
if (withPristine) fs.mkdirSync(pristineDir);
|
|
}
|
|
|
|
/** Runs the verifier with --json. Returns parsed structured report. */
|
|
function runVerifier({ includePristine = true } = {}) {
|
|
const args = [
|
|
SCRIPT,
|
|
'--patches-dir', patchesDir,
|
|
'--config-dir', configDir,
|
|
...(includePristine ? ['--pristine-dir', pristineDir] : []),
|
|
'--json',
|
|
];
|
|
const r = cp.spawnSync(process.execPath, args, { encoding: 'utf8' });
|
|
return {
|
|
status: r.status,
|
|
report: r.stdout && r.stdout.length ? JSON.parse(r.stdout) : null,
|
|
};
|
|
}
|
|
|
|
before(() => {
|
|
tmpRoot = fs.mkdtempSync(path.join(os.tmpdir(), 'gsd-2969-'));
|
|
patchesDir = path.join(tmpRoot, 'patches');
|
|
configDir = path.join(tmpRoot, 'installed');
|
|
pristineDir = path.join(tmpRoot, 'pristine');
|
|
resetFixture();
|
|
});
|
|
|
|
after(() => {
|
|
cleanup(tmpRoot);
|
|
});
|
|
|
|
describe('Bug #2969: deterministic Step 5 verification gate', () => {
|
|
test('REASON enum exposes the documented set of stable codes', () => {
|
|
// Locks the public diagnostic surface — adding a code requires updating
|
|
// this assertion, removing one breaks consumers that switch on the enum.
|
|
// Bug #3657 added OK_PRISTINE_DRIFT_DETECTED.
|
|
assert.deepEqual(
|
|
Object.keys(REASON).sort(),
|
|
[
|
|
'FAIL_INSTALLED_MISSING',
|
|
'FAIL_INSTALLED_NOT_REGULAR_FILE',
|
|
'FAIL_READ_ERROR',
|
|
'FAIL_USER_LINES_MISSING',
|
|
'OK_NO_SIGNIFICANT_BACKUP_LINES',
|
|
'OK_NO_USER_LINES_VS_PRISTINE',
|
|
'OK_PRISTINE_DRIFT_DETECTED',
|
|
],
|
|
);
|
|
});
|
|
|
|
test('exits 0 with status=ok when every user-added line is present in the merged file', () => {
|
|
resetFixture();
|
|
const pristine = 'line one of stock content here\nline two of stock content here\nline three of stock content here\n';
|
|
const userAdded = 'a custom line the user added for behavior X\nanother substantial line that the user inserted\n';
|
|
|
|
writeFile(path.join(pristineDir, 'skills', 'foo', 'SKILL.md'), pristine);
|
|
writeFile(path.join(patchesDir, 'skills', 'foo', 'SKILL.md'), pristine + userAdded);
|
|
writeFile(path.join(configDir, 'skills', 'foo', 'SKILL.md'), pristine + userAdded);
|
|
|
|
const { status, report } = runVerifier();
|
|
assert.equal(status, 0);
|
|
assert.equal(report.failures, 0);
|
|
assert.equal(report.checked, 1);
|
|
assert.equal(report.results[0].status, 'ok');
|
|
assert.deepEqual(report.results[0].missing, []);
|
|
});
|
|
|
|
test('reason=FAIL_USER_LINES_MISSING with the exact dropped line in .missing[]', () => {
|
|
resetFixture();
|
|
const pristine = 'first stock line in the original file here\nsecond stock line in the original file here\n';
|
|
const lostLine = 'this is the visual companion block that must survive';
|
|
writeFile(path.join(pristineDir, 'skills', 'discuss-phase', 'SKILL.md'), pristine);
|
|
writeFile(path.join(patchesDir, 'skills', 'discuss-phase', 'SKILL.md'), `${pristine}${lostLine}\n`);
|
|
writeFile(path.join(configDir, 'skills', 'discuss-phase', 'SKILL.md'), pristine);
|
|
|
|
const { status, report } = runVerifier();
|
|
assert.equal(status, 1);
|
|
assert.equal(report.failures, 1);
|
|
const r0 = report.results[0];
|
|
// Normalize separators: on Windows the SUT emits 'skills\discuss-phase\SKILL.md'.
|
|
assert.equal(r0.file.replace(/\\/g, '/'), 'skills/discuss-phase/SKILL.md');
|
|
assert.equal(r0.status, 'fail');
|
|
assert.equal(r0.reason, REASON.FAIL_USER_LINES_MISSING);
|
|
assert.ok(
|
|
r0.missing.includes(lostLine),
|
|
`dropped line should be in .missing[]; got ${JSON.stringify(r0.missing)}`,
|
|
);
|
|
});
|
|
|
|
test('reason=FAIL_INSTALLED_NOT_REGULAR_FILE when installed path is a directory', () => {
|
|
resetFixture();
|
|
writeFile(path.join(pristineDir, 'a.md'), 'pristine line of substantial content here\n');
|
|
writeFile(path.join(patchesDir, 'a.md'), 'pristine line of substantial content here\nuser added line that is substantial\n');
|
|
fs.mkdirSync(path.join(configDir, 'a.md')); // EISDIR trap
|
|
|
|
const { status, report } = runVerifier();
|
|
assert.equal(status, 1);
|
|
assert.equal(report.results[0].status, 'fail');
|
|
assert.equal(report.results[0].reason, REASON.FAIL_INSTALLED_NOT_REGULAR_FILE);
|
|
});
|
|
|
|
test('reason=FAIL_INSTALLED_MISSING when the merged file has been deleted', () => {
|
|
resetFixture();
|
|
const pristine = 'stock line one with substantial content for the test\n';
|
|
writeFile(path.join(pristineDir, 'workflow.md'), pristine);
|
|
writeFile(path.join(patchesDir, 'workflow.md'), `${pristine}user line that should survive but does not\n`);
|
|
// configDir intentionally missing the file.
|
|
|
|
const { status, report } = runVerifier();
|
|
assert.equal(status, 1);
|
|
assert.equal(report.results[0].status, 'fail');
|
|
assert.equal(report.results[0].reason, REASON.FAIL_INSTALLED_MISSING);
|
|
});
|
|
|
|
test('--json report has the documented shape: { checked, failures, results: [{ file, status, missing, reason }] }', () => {
|
|
resetFixture();
|
|
const pristine = 'pristine line that is sufficiently long to be significant\n';
|
|
const userAdded = 'extra line the user wrote for their workflow customisation';
|
|
writeFile(path.join(pristineDir, 'a.md'), pristine);
|
|
writeFile(path.join(patchesDir, 'a.md'), `${pristine}${userAdded}\n`);
|
|
writeFile(path.join(configDir, 'a.md'), pristine);
|
|
|
|
const { status, report } = runVerifier();
|
|
assert.equal(status, 1);
|
|
// Bug #3657 (Finding 1): drifted + drifted_files are additive fields added to surface
|
|
// pristine-drift skips distinctly from failures. Shape-lock updated to include them.
|
|
assert.deepEqual(Object.keys(report).sort(), ['checked', 'drifted', 'drifted_files', 'failures', 'results']);
|
|
const r0 = report.results[0];
|
|
assert.deepEqual(Object.keys(r0).sort(), ['file', 'missing', 'reason', 'status']);
|
|
assert.equal(typeof r0.file, 'string');
|
|
assert.equal(typeof r0.status, 'string');
|
|
assert.equal(typeof r0.reason, 'string');
|
|
assert.ok(Array.isArray(r0.missing));
|
|
});
|
|
|
|
test('ignores backup-meta.json — it is metadata, not a patched file', () => {
|
|
resetFixture();
|
|
writeFile(path.join(patchesDir, 'backup-meta.json'), JSON.stringify({ files: [] }));
|
|
|
|
const { status, report } = runVerifier();
|
|
assert.equal(status, 0);
|
|
assert.equal(report.checked, 0);
|
|
assert.equal(report.failures, 0);
|
|
assert.deepEqual(report.results, []);
|
|
});
|
|
|
|
test('without --pristine-dir, treats every significant backup line as required (safe over-broad fallback)', () => {
|
|
resetFixture({ withPristine: false });
|
|
const presentLine = 'this is a substantial line of user content here';
|
|
const droppedLine = 'another substantial line that should survive';
|
|
writeFile(path.join(patchesDir, 'b.md'), `${presentLine}\n${droppedLine}\n`);
|
|
writeFile(path.join(configDir, 'b.md'), `${presentLine}\n`);
|
|
|
|
const { status, report } = runVerifier({ includePristine: false });
|
|
assert.equal(status, 1);
|
|
assert.equal(report.results[0].reason, REASON.FAIL_USER_LINES_MISSING);
|
|
assert.ok(report.results[0].missing.includes(droppedLine));
|
|
assert.ok(!report.results[0].missing.includes(presentLine));
|
|
});
|
|
|
|
test('treats gsd-hook-version install-time substitution as upstream-owned, not missing user content (#229)', () => {
|
|
resetFixture();
|
|
const rel = path.join('hooks', 'gsd-statusline.js');
|
|
const pristine = [
|
|
'// gsd-hook-version: {{GSD_VERSION}}',
|
|
'console.log("statusline hook");',
|
|
'',
|
|
].join('\n');
|
|
const backup = [
|
|
'// gsd-hook-version: 1.41.0',
|
|
'console.log("statusline hook");',
|
|
'',
|
|
].join('\n');
|
|
const installed = [
|
|
'// gsd-hook-version: 1.42.3',
|
|
'console.log("statusline hook");',
|
|
'',
|
|
].join('\n');
|
|
|
|
writeFile(path.join(pristineDir, rel), pristine);
|
|
writeFile(path.join(patchesDir, rel), backup);
|
|
writeFile(path.join(configDir, rel), installed);
|
|
|
|
const { status, report } = runVerifier();
|
|
assert.equal(status, 0, `expected pass for upstream-owned version substitution; report=${JSON.stringify(report)}`);
|
|
assert.equal(report.failures, 0);
|
|
assert.equal(report.checked, 1);
|
|
assert.equal(report.results[0].status, 'ok');
|
|
assert.deepStrictEqual(report.results[0].missing, []);
|
|
});
|
|
});
|