Files
OmniRoute/tests/integration/chatcore-context-window-boundary.test.ts
Diego Rodrigues de Sa e Souza 93265eede3 test(infra): retry recursive temp-dir removal on main (main twin of #11968) (#12246)
* test(infra): retry recursive temp-dir removal on main (main twin of #11968)

`main` has been red since b342c1a361 on the vitest and integration gates:

  ✖ tests/unit/autoCombo/provider-family-combos.test.ts > auto/<family>
  ✖ chat pipeline applies Codex OAuth fingerprint and priority tier inside combos

Both call resetStorage() from beforeEach, which does an fs.rmSync(TEST_DATA_DIR,
{recursive: true, force: true}) with no retry, and intermittently loses the race
with a not-yet-released SQLite handle (ENOTEMPTY).

release/v3.8.51 fixed this in #11968 with a mechanical codemod adding
maxRetries/retryDelay to every recursive rm/rmSync/rmdirSync under tests/, but
that PR landed only on the release branch. Because main only receives work at
the release squash, it stayed broken for the whole cycle — and repo-wide gates
then turn every open PR into main red on checks unrelated to their diff.

This is the --base main twin: re-runs the same codemod that already shipped on
the release branch (scripts/ad-hoc/codemod-rm-maxretries.mjs), so the two
branches converge on identical test-teardown semantics. Test-only; no product
logic is touched.

The remaining three failures reported on #12133 (unit full suite exceeding its
4800s ceiling, package-artifact exceeding 1200s, and the boot-smoke that is
skipped as a consequence) are runner-contention timeouts, not code defects —
validate-release-green.mjs runs those heavy gates concurrently on one shared
hosted runner. There is no fix to port for those.

* chore(scripts): carry the rm-maxretries codemod onto main alongside its output

The codemod that generated the previous commit lives in the repo on
release/v3.8.51 (added by #11968) but was never on main. Bringing it over keeps
the tool next to the change it produced, so the transformation stays
reproducible and auditable from either branch.
2026-09-01 01:48:00 -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;
}
}
});