* test(#3145): bound the installer/runtime cluster onto the process seam Migrates 156 unbounded sync spawn sites across 47 files. Allowlist 120 to 73. Timeouts are sized from evidence already in the tree rather than a house default, because this wave spawns installers rather than git plumbing and an undersized bound does not catch a hang -- it manufactures CI flake, which is worse, since a flake gets re-run instead of investigated. install.test.cjs records a real spawnSync ETIMEDOUT at a 60000ms cap on a loaded bench while another lane passed the same commit in 12.7s, so full installs are bound at 120000ms against that recorded incident. Also adds an auditable escape to the guard's timeout ceiling. The 600000ms cap was set in #3143 from partial evidence, but fragment-single-edit- propagation carries a documented, load-tested 900000ms bound on a run that chains a full build plus eight generators -- the guard would have rejected a correct timeout the moment that file left the allowlist. A value above the ceiling is now permitted only with an inline allow-spawn-timeout-ceiling marker carrying a non-empty reason. It raises the ceiling; it never waives the requirement for a bound, which is asserted directly. install-shared.cjs keeps its hand-rolled assert rather than routing through throwIfFailed: its message embeds both streams, and throwIfFailed carries only a trimmed stderr. The message now also names the outcome, so a bounded timeout reads as such across its 38 importers instead of as expected null to equal 0. * test(#3145): extract class-norm timeouts and correct the build-hooks sizing A pre-PR review found 52 copies of four class-norm timeout constants across this wave. These are not per-suite fixture bindings -- they are shared facts about how long a class of subprocess takes, derived from a recorded bench incident. That norm already moved once (60000 to 120000 after a real ETIMEDOUT), and 52 copies would have drifted the next time it moved. Extracts tests/helpers/timeouts.cjs, where each norm is justified once, and converts the copies. A site that genuinely differs -- a real tsc compile, or regen:derived -- keeps its own local constant with its own justification. Also corrects a misclassification: scripts/build-hooks.js was sized as a build at 120000 in twelve places and 60000 in another, but it compiles and bundles nothing. Its own header says no bundling needed; it copies pre-built files and syntax-checks them with vm. Three different values bounded one script; now there is one. * test(#3145): fix red CI — lint self-match and a Windows chunk overrun Two failures on PR 3176. lint-allow-test-rule-refs read a RuleTester fixture as a real exemption. The fixture exists to prove an unrelated marker does NOT suppress the rule, so it carries that marker's literal text as test data. Split via concatenation, the same idiom no-unbounded-spawn-allowlist.test.cjs already uses for its own self-match problem. The explanatory comment needed the same treatment. The Windows shard 3/3 chunk was killed at its 600000ms budget. Output stopped seven minutes before the kill, so this was an overrun rather than a slow chunk: regenDerivedPropagatesSingleFragmentEditWithNoSecondSourceSurface runs regen:derived bounded at 900000ms, which is larger than the whole chunk budget, so the chunk killer always fires first and it can never complete there. Both the test and that bound predate this change; modifying the file pulled it into the Windows targeted set and exposed it. Skipped on Windows with the reason recorded; the Linux lanes cover it. The 900000 bound and its ceiling marker are unchanged -- they are correct. * test(#3145): refresh the stale test-timings cost table The Windows shard was killed at its 600000ms per-chunk budget. run-tests.cjs packs chunks by measured duration from tests/test-timings.json, and an unknown file falls back to the table's median weight -- advisory by design, but it silently underweights exactly the files that matter. Four of the failing chunk's 22 files were absent from the table, including the two heaviest: fragment-single-edit-propagation.install.test.cjs at 230s (it runs regen:derived) and agent-fragments-emission.install.test.cjs at 79s. Both were weighted as average, so the chunk's total weight read 53.68 against a budget of 60 and the packer produced a single chunk. Regenerated from a passing full-suite run, per the remedy the script itself documents. 700 to 770 entries, 70 added, 0 dropped -- verified, since gen-test-timings.cjs replaces the table wholesale rather than merging. Proven against the real packer: the same 22 files now weigh 103.91 and split into two chunks. No logic, budget, or timeout was changed; raising a budget to make a red gate pass is not a fix. --------- Co-authored-by: sim <sim@local>
184 lines
8.5 KiB
JavaScript
184 lines
8.5 KiB
JavaScript
'use strict';
|
|
/**
|
|
* Tests for scripts/ci-rebase-check.cjs — Codex round 4 P1 regression.
|
|
*
|
|
* The root bug: run() used execFileSync with stdio:'inherit', which returns null
|
|
* on success. The caller checked `result !== null` to detect success, so the
|
|
* condition was ALWAYS false (null !== null === false) and every successful fetch
|
|
* fell through to the "failed after 3 attempts" exit-1 path.
|
|
*
|
|
* Fix: run() now returns true on success, false on failure, making the boolean
|
|
* check unambiguous regardless of the stdio mode.
|
|
*
|
|
* These tests exercise the run() logic in isolation via subprocess execution,
|
|
* verifying the sentinel behaviour rather than internal module state.
|
|
*/
|
|
|
|
const { describe, test } = require('node:test');
|
|
const assert = require('node:assert/strict');
|
|
const { spawnSync } = require('node:child_process');
|
|
const path = require('node:path');
|
|
const fs = require('node:fs');
|
|
const os = require('node:os');
|
|
|
|
const ROOT = path.resolve(__dirname, '..');
|
|
const SCRIPT = path.join(ROOT, 'scripts', 'ci-rebase-check.cjs');
|
|
const NODE = process.execPath;
|
|
const { cleanup } = require('./helpers.cjs');
|
|
const { gitOrThrow } = require('./helpers/git-fixture.cjs');
|
|
// #3145: class-norm timeout, not a per-suite value — see helpers/timeouts.cjs.
|
|
const { GIT_TIMEOUT_MS } = require('./helpers/timeouts.cjs');
|
|
|
|
// ---------------------------------------------------------------------------
|
|
// Helper: run a small inline Node snippet that requires the run() helper
|
|
// directly from the source (extracted via a thin wrapper).
|
|
//
|
|
// Because the script has top-level side-effects (git config calls), we cannot
|
|
// require() it. Instead we test the run() sentinel by embedding the function
|
|
// body verbatim in a one-shot subprocess.
|
|
// ---------------------------------------------------------------------------
|
|
|
|
function evalRunHelper(stmts) {
|
|
// Inline the exact fixed run() body so the test is tightly coupled to the
|
|
// contract, not some mock.
|
|
const code = `
|
|
'use strict';
|
|
const { execFileSync } = require('child_process');
|
|
function run(cmd, args, opts) {
|
|
try {
|
|
execFileSync(cmd, args, { stdio: 'inherit', ...opts });
|
|
return true;
|
|
} catch (e) {
|
|
return false;
|
|
}
|
|
}
|
|
${stmts}
|
|
`;
|
|
const r = spawnSync(NODE, ['-e', code], { encoding: 'utf8', timeout: 10_000 });
|
|
return { status: r.status ?? 1, stdout: r.stdout ?? '', stderr: r.stderr ?? '' };
|
|
}
|
|
|
|
// ---------------------------------------------------------------------------
|
|
// Test group 1 — run() sentinel correctness
|
|
// ---------------------------------------------------------------------------
|
|
|
|
describe('ci-rebase-check: run() helper — success sentinel (Codex round 4 P1)', () => {
|
|
|
|
test('run() returns true when the command succeeds', () => {
|
|
// Use `node -e ""` (no-op) as a guaranteed-success command that produces no output,
|
|
// avoiding stdout contamination when stdio:'inherit' writes to the same stream.
|
|
const { status, stdout, stderr } = evalRunHelper(`
|
|
const result = run(process.execPath, ['-e', '']);
|
|
process.stdout.write(String(result));
|
|
`);
|
|
assert.strictEqual(status, 0, `subprocess should exit 0; stderr: ${stderr}`);
|
|
assert.strictEqual(stdout, 'true', `run() must return true on success; got: ${stdout}`);
|
|
});
|
|
|
|
test('run() returns false when the command fails', () => {
|
|
// An invalid binary name causes execFileSync to throw ENOENT.
|
|
const { status, stdout, stderr } = evalRunHelper(`
|
|
const result = run('__nonexistent_binary_that_cannot_exist__', []);
|
|
process.stdout.write(String(result));
|
|
`);
|
|
assert.strictEqual(status, 0, `subprocess should exit 0; stderr: ${stderr}`);
|
|
assert.strictEqual(stdout, 'false', `run() must return false on failure; got: ${stdout}`);
|
|
});
|
|
|
|
test('run() returns true (not null) — counter-test for pre-fix null behaviour', () => {
|
|
// Pre-fix: execFileSync with stdio:'inherit' returns null on success.
|
|
// The old check was `result !== null`, which would be `null !== null === false`.
|
|
// Post-fix: result must be strictly true, making `if (result)` correct.
|
|
// Use `node -e ""` (no-op) so stdio:'inherit' does not pollute our stdout capture.
|
|
const { status, stdout } = evalRunHelper(`
|
|
const result = run(process.execPath, ['-e', '']);
|
|
process.stdout.write(JSON.stringify({ isTrue: result === true, isNull: result === null }));
|
|
`);
|
|
assert.strictEqual(status, 0);
|
|
const { isTrue, isNull } = JSON.parse(stdout);
|
|
assert.strictEqual(isNull, false, 'run() must NOT return null on success (pre-fix bug)');
|
|
assert.strictEqual(isTrue, true, 'run() must return exactly true on success');
|
|
});
|
|
|
|
test('run() returns false (not null) — failure path also returns boolean', () => {
|
|
const { status, stdout } = evalRunHelper(`
|
|
const result = run('__nonexistent__', []);
|
|
process.stdout.write(JSON.stringify({ isFalse: result === false, isNull: result === null }));
|
|
`);
|
|
assert.strictEqual(status, 0);
|
|
const { isFalse, isNull } = JSON.parse(stdout);
|
|
assert.strictEqual(isNull, false, 'run() must NOT return null on failure');
|
|
assert.strictEqual(isFalse, true, 'run() must return exactly false on failure');
|
|
});
|
|
|
|
});
|
|
|
|
// ---------------------------------------------------------------------------
|
|
// Test group 2 — fetch-retry loop uses the boolean correctly
|
|
//
|
|
// We cannot run the actual git fetch against GitHub in unit tests, but we can
|
|
// verify the fetch-loop logic by running the full script against a local git
|
|
// repo where GITHUB_TOKEN and GITHUB_REPOSITORY are absent (so remote set-url
|
|
// is skipped) and GITHUB_BASE_REF points to a branch that exists locally.
|
|
// ---------------------------------------------------------------------------
|
|
|
|
describe('ci-rebase-check: fetch-retry loop resolves when git fetch succeeds', () => {
|
|
|
|
test('script exits 0 when fetch succeeds (local bare remote, clean merge)', () => {
|
|
// Set up: create a temp dir with a local git repo that has a `main` branch.
|
|
// The script will fetch `origin main` from this local "remote".
|
|
const tmpDir = fs.mkdtempSync(path.join(os.tmpdir(), 'gsd-431-ci-rebase-'));
|
|
const remoteDir = path.join(tmpDir, 'remote.git');
|
|
const workDir = path.join(tmpDir, 'work');
|
|
|
|
try {
|
|
// Build a bare remote with a `main` branch containing one commit. Each
|
|
// setup step uses gitOrThrow (not the bare seam) so a failure here aborts
|
|
// loudly at the point of failure instead of surfacing as a baffling
|
|
// assertion mismatch against the script's exit code further down —
|
|
// matching the throw-on-failure behavior the pre-migration `spawnSync`
|
|
// calls this replaced never had either.
|
|
fs.mkdirSync(remoteDir, { recursive: true });
|
|
gitOrThrow(['init', '--bare', remoteDir], { timeoutMs: GIT_TIMEOUT_MS });
|
|
|
|
// Create a working clone to push an initial commit.
|
|
gitOrThrow(['clone', remoteDir, workDir], { timeoutMs: GIT_TIMEOUT_MS });
|
|
fs.writeFileSync(path.join(workDir, 'seed.txt'), 'init\n');
|
|
gitOrThrow(['-C', workDir, 'config', 'user.email', 'ci@test'], { timeoutMs: GIT_TIMEOUT_MS });
|
|
gitOrThrow(['-C', workDir, 'config', 'user.name', 'CI Test'], { timeoutMs: GIT_TIMEOUT_MS });
|
|
gitOrThrow(['-C', workDir, 'checkout', '-b', 'main'], { timeoutMs: GIT_TIMEOUT_MS });
|
|
gitOrThrow(['-C', workDir, 'add', 'seed.txt'], { timeoutMs: GIT_TIMEOUT_MS });
|
|
gitOrThrow(['-C', workDir, 'commit', '-m', 'init'], { timeoutMs: GIT_TIMEOUT_MS });
|
|
gitOrThrow(['-C', workDir, 'push', 'origin', 'main'], { timeoutMs: GIT_TIMEOUT_MS });
|
|
|
|
// Run the script from `workDir` with origin pointing at our bare remote.
|
|
// GITHUB_BASE_REF=main so it fetches `origin main`.
|
|
// No GITHUB_TOKEN so remote set-url is skipped.
|
|
const r = spawnSync(NODE, [SCRIPT], {
|
|
cwd: workDir,
|
|
encoding: 'utf8',
|
|
timeout: 20_000,
|
|
env: {
|
|
...process.env,
|
|
GITHUB_BASE_REF: 'main',
|
|
GITHUB_TOKEN: '',
|
|
GITHUB_REPOSITORY: '',
|
|
},
|
|
});
|
|
|
|
assert.strictEqual(
|
|
r.status, 0,
|
|
`Script should exit 0 when fetch+merge succeed.\nstdout: ${r.stdout}\nstderr: ${r.stderr}`
|
|
);
|
|
// Must NOT emit the "failed after 3 attempts" error message.
|
|
assert.ok(
|
|
!(r.stderr || '').includes('failed after 3 attempts'),
|
|
`Script must not emit "failed after 3 attempts" when fetch succeeded.\nstderr: ${r.stderr}`
|
|
);
|
|
} finally {
|
|
cleanup(tmpDir);
|
|
}
|
|
});
|
|
|
|
});
|