Files
OmniRoute/tests/integration/chatcore-context-window-boundary.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

107 lines
3.6 KiB
TypeScript

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-ctx-boundary-"));
process.env.DATA_DIR = TEST_DATA_DIR;
process.env.REQUIRE_API_KEY = "false";
process.env.API_KEY_SECRET = process.env.API_KEY_SECRET || "test-ctx-boundary-secret";
const core = await import("../../src/lib/db/core.ts");
const providersDb = await import("../../src/lib/db/providers.ts");
const compressionDb = await import("../../src/lib/db/compression.ts");
const { handleChatCore } = await import("../../open-sse/handlers/chatCore.ts");
const { estimateTokens, getTokenLimit } = await import("../../open-sse/services/contextManager.ts");
const originalFetch = globalThis.fetch;
test.after(async () => {
globalThis.fetch = originalFetch;
core.closeDbInstance();
try {
fs.rmSync(TEST_DATA_DIR, { recursive: true, force: true, maxRetries: 5, retryDelay: 100 });
} catch {}
});
// Integration lock for the pre-dispatch context-window boundary (#7379):
// enforceOutputTokenBudget in chatCore must reject a request whose input alone
// exceeds the target's context window — before any upstream fetch — when
// compression is disabled and therefore cannot reduce it.
test("chatCore integration: over-window request is rejected before dispatch when compression cannot reduce it", async () => {
const provider = "openai";
const model = "gpt-4";
const originalContextLength = process.env.CONTEXT_LENGTH_OPENAI;
process.env.CONTEXT_LENGTH_OPENAI = "8192";
await compressionDb.updateCompressionSettings({
enabled: false,
defaultMode: "off",
autoTriggerTokens: 0,
});
const connection = await providersDb.createProviderConnection({
provider,
apiKey: "test-key",
isActive: true,
});
const body = {
model,
stream: false,
messages: [
{ role: "system", content: "You are helpful." },
{ role: "user", content: `${"Keep spacing.\n\n\n".repeat(2000)}Way over the window.` },
],
};
assert.ok(
estimateTokens(JSON.stringify(body.messages)) > getTokenLimit(provider, model),
"Test body should exceed the full context window"
);
let fetchCalls = 0;
globalThis.fetch = async () => {
fetchCalls += 1;
return new Response(
JSON.stringify({
choices: [{ message: { role: "assistant", content: "test" } }],
usage: { prompt_tokens: 10, completion_tokens: 5, total_tokens: 15 },
}),
{
status: 200,
headers: { "content-type": "application/json" },
}
);
};
try {
const result = await handleChatCore({
body,
modelInfo: { provider, model },
credentials: { apiKey: "test-key" },
log: { debug: () => {}, info: () => {}, warn: () => {}, error: () => {} },
clientRawRequest: { endpoint: "/v1/chat/completions", headers: new Map() },
connectionId: connection.id,
onCredentialsRefreshed: () => {},
onRequestSuccess: () => {},
onStreamFailure: () => {},
onDisconnect: () => {},
userAgent: "test-agent",
comboName: null,
});
assert.equal(result.success, false, "Over-window request should be rejected");
assert.equal(result.status, 400);
assert.equal(result.errorCode, "context_length_exceeded");
assert.equal(fetchCalls, 0, "Rejected request must never reach the upstream");
} finally {
globalThis.fetch = originalFetch;
if (originalContextLength === undefined) {
delete process.env.CONTEXT_LENGTH_OPENAI;
} else {
process.env.CONTEXT_LENGTH_OPENAI = originalContextLength;
}
}
});