mirror of
https://github.com/diegosouzapw/OmniRoute.git
synced 2026-09-13 18:32:12 +03:00
* 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.
91 lines
3.7 KiB
TypeScript
91 lines
3.7 KiB
TypeScript
// tests/unit/combo-context-prefix-resolution.test.ts
|
|
// Regression guard for computeComboContextLength()'s provider-prefix bug.
|
|
//
|
|
// resolveNestedComboTargets() returns target.modelStr in "provider/model" form
|
|
// (e.g. "glm/glm-5.2"), but computeComboContextLength() used to pass that
|
|
// qualified string straight into getCanonicalModelMetadata() without stripping
|
|
// the prefix first — unlike the catalog's own getComboTargetCatalogMetadata(),
|
|
// which strips it via getComboTargetModelId()/getProviderPrefixes() before the
|
|
// lookup. The alias-resolution chain (getResolvedModelCapabilities ->
|
|
// resolveCanonicalProviderModel -> resolveProviderModelAlias) does an exact-match
|
|
// lookup keyed by the BARE registry id, so a "provider/model" string only
|
|
// resolved for the handful of models with a curated MODEL_SPECS alias in that
|
|
// exact qualified form — every other registry-only model (the vast majority)
|
|
// silently fell out of the min() computation, and computed_context_length was
|
|
// dropped from the /api/combos response entirely.
|
|
//
|
|
// This test uses glm-5.2 (open-sse/config/providers/registry/glm — real
|
|
// 1,000,000-token context, no "glm/glm-5.2"-form curated alias) — the exact
|
|
// class of model the bug affected. Confirmed empirically before this fix
|
|
// landed: computeComboContextLength() returned `undefined` for a combo whose
|
|
// only member was "glm/glm-5.2", even though the bare "glm-5.2" resolves fine
|
|
// via getCanonicalModelMetadata() directly.
|
|
import test from "node:test";
|
|
import assert from "node:assert/strict";
|
|
import fs from "node:fs";
|
|
import os from "node:os";
|
|
import path from "node:path";
|
|
|
|
const TEST_DATA_DIR = fs.mkdtempSync(
|
|
path.join(os.tmpdir(), "omniroute-combo-context-prefix-resolution-")
|
|
);
|
|
process.env.DATA_DIR = TEST_DATA_DIR;
|
|
|
|
const core = await import("../../src/lib/db/core.ts");
|
|
const { computeComboContextLength } = await import("../../src/lib/combos/comboContext.ts");
|
|
const { setModelContextOverride, removeModelContextOverride } =
|
|
await import("../../src/lib/db/modelContextOverrides.ts");
|
|
|
|
test.after(() => {
|
|
core.resetDbInstance();
|
|
fs.rmSync(TEST_DATA_DIR, { recursive: true, force: true, maxRetries: 5, retryDelay: 100 });
|
|
});
|
|
|
|
test("computeComboContextLength resolves a registry-known, prefixed member (glm/glm-5.2) to its real context window", () => {
|
|
const combo = {
|
|
name: "prefix-resolution-probe-single",
|
|
models: ["glm/glm-5.2"],
|
|
};
|
|
|
|
const result = computeComboContextLength(combo, []);
|
|
|
|
assert.equal(
|
|
result,
|
|
1000000,
|
|
"glm/glm-5.2 is a real registry model with a 1,000,000-token context window " +
|
|
"(open-sse/config/providers/registry/glm) — the prefix must be stripped " +
|
|
"before the canonical-model lookup so it is not silently excluded"
|
|
);
|
|
});
|
|
|
|
test("computeComboContextLength takes the minimum across multiple prefixed, registry-known members", () => {
|
|
const combo = {
|
|
name: "prefix-resolution-probe-multi",
|
|
// glm-4.5 (128,000) is the smaller of the two known windows.
|
|
models: ["glm/glm-5.2", "glm/glm-4.5"],
|
|
};
|
|
|
|
const result = computeComboContextLength(combo, []);
|
|
|
|
assert.equal(
|
|
result,
|
|
128000,
|
|
"the minimum across all resolved (prefix-stripped) members should win, " +
|
|
"matching the catalog's minKnownNumber semantics"
|
|
);
|
|
});
|
|
|
|
test("computeComboContextLength honors a larger persisted Codex GPT-5.6 window", () => {
|
|
const modelId = "gpt-5.6-terra";
|
|
assert.equal(setModelContextOverride("codex", modelId, 500000, "manual"), true);
|
|
try {
|
|
assert.equal(
|
|
computeComboContextLength({ models: [`codex/${modelId}`] }, []),
|
|
500000,
|
|
"the combo aggregate must use the effective override instead of the Codex registry default"
|
|
);
|
|
} finally {
|
|
removeModelContextOverride("codex", modelId);
|
|
}
|
|
});
|