Files
OmniRoute/tests/unit/compression/preview-outer-engine-token-reconcile-6488.test.ts
Diego Rodrigues de Sa e Souza 3d4f3e4960 test(infra): retry recursive temp-dir removal instead of failing a shard on ENOTEMPTY (#11966) (#11968)
* test(infra): retry recursive temp-dir removal instead of failing a shard on ENOTEMPTY (#11966)

Two shards on release/v3.8.51 went red in one day with the same signature —
"ENOTEMPTY, Directory not empty: /tmp/omniroute-<test>-XXXXXX" — from
combo-same-provider-cascade (Unit Tests fast-path 4/4, on a PR that touches only
.github/) and auth-policy-embeddings-webfetch-7785 (the 20k-test TIA step). Both pass
alone and on re-run: the cleanup races something still writing into the directory
(SQLite WAL/-shm checkpoint, a worker, the backup) and under a loaded hosted runner
the window opens. 1154 test files do their own cleanup with
fs.rmSync(dir, { recursive: true, force: true }); 57 already asked for retries.

One-shot codemod (scripts/ad-hoc/codemod-rm-maxretries.mjs, kept for the record):
every rm / rmSync / rmdirSync option object with `recursive: true` and no
`maxRetries` gains `maxRetries: 5, retryDelay: 100` — Node itself then retries
ENOTEMPTY/EBUSY/EPERM for up to ~0.5 s before giving up. 2243 call sites in 1292
files under tests/, the shared tests/_setup/isolateDataDir.ts exit hook included.
Only the option object changes: no call site, assertion or import is touched.

Validation: prettier and ESLint (with the frozen suppressions) clean on all 1292
files; a random 20-file sample runs green (quota-redis-store hangs identically on
the untouched tree — it needs a Redis on localhost, an environment matter). The
four unit shards on this PR are the full run.

* fix(quality): let check-forgotten-sibling-tests read a 1,000-file diff

The gate shells out to `git diff` through execFileSync with Node's default 1 MB
maxBuffer; the 1,292-file codemod in this PR is the first diff large enough to
overflow it, and the gate died with `spawnSync git ENOBUFS` before comparing
anything. 64 MB is far above any real PR and costs nothing when unused.
2026-08-29 01:17:40 -03:00

80 lines
3.2 KiB
TypeScript

import test from "node:test";
import assert from "node:assert/strict";
import { join } from "node:path";
import { mkdtempSync, rmSync } from "node:fs";
import { tmpdir } from "node:os";
const TEST_DATA_DIR = mkdtempSync(join(tmpdir(), "preview-reconcile-6488-"));
process.env.DATA_DIR = TEST_DATA_DIR;
process.env.API_KEY_SECRET = process.env.API_KEY_SECRET ?? "test-secret-32-chars-min-aaaaaaaa";
delete process.env.INITIAL_PASSWORD;
const core = await import("../../../src/lib/db/core.ts");
const route = await import("../../../src/app/api/compression/preview/route.ts");
function makeReq(body: unknown) {
return new Request("http://localhost/api/compression/preview", {
method: "POST",
headers: { "content-type": "application/json" },
body: JSON.stringify(body),
});
}
test.beforeEach(() => core.resetDbInstance());
test.after(() => {
core.resetDbInstance();
rmSync(TEST_DATA_DIR, { recursive: true, force: true, maxRetries: 5, retryDelay: 100 });
});
// Regression for #6488: outer originalTokens/compressedTokens (real tiktoken counter over
// extracted message text) and engineBreakdown[].originalTokens/compressedTokens (internal
// JSON.stringify(body).length/4 estimate) used to diverge on small/degenerate input because
// they measured different things. For a single-engine breakdown, the entry represents the
// exact same before/after transformation as the overall response, so it must be reconciled
// to match the outer counts exactly.
test("degenerate input with pipeline=['lite']: engineBreakdown[0] matches outer token counts", async () => {
const res = await route.POST(
makeReq({
messages: [{ role: "user", content: "user: " }],
pipeline: ["lite"],
})
);
const body = await res.json();
assert.equal(res.status, 200, `expected 200, got ${res.status}: ${JSON.stringify(body)}`);
const engines = (body.engineBreakdown ?? []).map((e: { engine: string }) => e.engine);
assert.ok(
engines.every((e: string) => e === "lite"),
`expected engineBreakdown to only contain 'lite', got ${JSON.stringify(engines)}`
);
assert.equal(body.engineBreakdown.length, 1);
const [step] = body.engineBreakdown;
assert.equal(
step.originalTokens,
body.originalTokens,
`outer originalTokens=${body.originalTokens} vs engine ${step.engine} originalTokens=${step.originalTokens}`
);
assert.equal(
step.compressedTokens,
body.compressedTokens,
`outer compressedTokens=${body.compressedTokens} vs engine ${step.engine} compressedTokens=${step.compressedTokens}`
);
});
// Same reconciliation must hold for the single-engine (non-pipeline) dispatch path, where
// engineBreakdown is synthesized by ensureEngineBreakdown from the overall stats.
test("single-engine dispatch (engineId='rtk'): engineBreakdown[0] matches outer token counts", async () => {
const res = await route.POST(
makeReq({
messages: [{ role: "user", content: "a" }],
engineId: "rtk",
})
);
const body = await res.json();
assert.equal(res.status, 200, `expected 200, got ${res.status}: ${JSON.stringify(body)}`);
assert.equal(body.engineBreakdown.length, 1);
const [step] = body.engineBreakdown;
assert.equal(step.originalTokens, body.originalTokens);
assert.equal(step.compressedTokens, body.compressedTokens);
});