Files
OmniRoute/tests/unit/guardrails/vision-bridge-callmodel.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

122 lines
4.7 KiB
TypeScript

/**
* callVisionModel fallback behavior — Integration test (PR #3377, Rule #18)
*
* Verifies that when the primary vision model fails, callVisionModel falls
* through to the next model in the fallback list, and that when ALL models
* fail it throws the last error (not a silent empty result).
*
* Run: node --import tsx/esm --test tests/unit/guardrails/vision-bridge-callmodel.test.ts
*/
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-vision-bridge-"));
process.env.DATA_DIR = TEST_DATA_DIR;
// Prevent vision bridge from routing through a real API
process.env.VISION_BRIDGE_ENABLED = "false";
const { callVisionModel } = await import("../../../src/lib/guardrails/visionBridgeHelpers.ts");
const { createProviderConnection } = await import("../../../src/lib/db/providers.ts");
// PR #8433 taught getFallbackModels() to exclude any candidate without a
// usable active connection (see visionBridgeRouter.ts::getVisionCapableModels).
// This test's isolated DATA_DIR starts with zero provider connections, so
// without a seeded connection every fallback candidate is confirmed
// unusable and callVisionModel has nothing left to retry — seed one
// credentialed connection so the fallback-retry mechanics under test here
// stay independent of that (unrelated) credential-filtering behavior.
await createProviderConnection({
provider: "anthropic",
authType: "apikey",
name: "vision-bridge-callmodel-test-fallback",
apiKey: "sk-test-anthropic-fallback",
isActive: true,
});
const originalFetch = globalThis.fetch;
test.after(() => {
globalThis.fetch = originalFetch;
fs.rmSync(TEST_DATA_DIR, { recursive: true, force: true, maxRetries: 5, retryDelay: 100 });
});
test.afterEach(() => {
globalThis.fetch = originalFetch;
});
// Helper: build a minimal OpenAI-compat image data URI
const TINY_PNG =
"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==";
test("callVisionModel falls through to next model when primary fails", async () => {
let fetchCallCount = 0;
// The fallback candidate can legitimately resolve to either an OpenAI-compatible
// model (POST .../chat/completions, { choices: [{ message: { content } }] }) or an
// Anthropic model (POST .../v1/messages, { content: [{ type: "text", text }] }) —
// vision-bridge router priority (#7204) now ranks credentialed providers (openai/
// anthropic) ahead of opencode-*, so the mock must match whichever shape the
// fallback attempt actually requests instead of assuming OpenAI's shape.
const FALLBACK_TEXT = "fallback model description";
globalThis.fetch = async (url: RequestInfo | URL, _init?: RequestInit) => {
fetchCallCount++;
if (fetchCallCount === 1) {
// First call (primary model) — simulate API error
throw new Error("mock: primary model unavailable");
}
// Second call (fallback model) — return a valid response shaped for whichever
// API the fallback model actually calls.
const urlStr = typeof url === "string" ? url : url.toString();
const isAnthropicCall = urlStr.includes("/v1/messages");
const body = isAnthropicCall
? JSON.stringify({ content: [{ type: "text", text: FALLBACK_TEXT }] })
: JSON.stringify({ choices: [{ message: { content: FALLBACK_TEXT } }] });
return new Response(body, {
status: 200,
headers: { "Content-Type": "application/json" },
});
};
const result = await callVisionModel(
TINY_PNG,
{ model: "openai/gpt-4o-mini", prompt: "Describe this image." },
"sk-test-key",
{ fixedModel: "openai/gpt-4o-mini", maxFallbackAttempts: 2 }
);
assert.equal(fetchCallCount, 2, "must have attempted exactly 2 models (primary + 1 fallback)");
assert.equal(result, FALLBACK_TEXT, "must return the fallback model's response");
});
test("callVisionModel throws when ALL models fail", async () => {
let fetchCallCount = 0;
globalThis.fetch = async () => {
fetchCallCount++;
throw new Error(`mock: model-${fetchCallCount} unavailable`);
};
await assert.rejects(
() =>
callVisionModel(
TINY_PNG,
{ model: "openai/gpt-4o-mini", prompt: "Describe this image." },
"sk-test-key",
{ fixedModel: "openai/gpt-4o-mini", maxFallbackAttempts: 2 }
),
(err: Error) => {
assert.ok(
err.message.includes("unavailable") || err.message.includes("All vision models failed"),
`error should indicate failure, got: ${err.message}`
);
return true;
}
);
assert.ok(fetchCallCount >= 1, "must have attempted at least 1 model");
});