mirror of
https://github.com/diegosouzapw/OmniRoute.git
synced 2026-09-13 18:32:12 +03:00
65e81158ab
Landed with the design call resolved per the owner's pick — **option 1**: the synced store is now endpoint-agnostic (persistDiscoveredModels and managedModelImport no longer drop non-chat models at write time), and chat selectability moved to read time (auto-pool expansion in autoStrategy applies filterChatSelectableModels; the models-route projection already had its chatOnly filter). Your discovery test now passes end-to-end (3/3): /api/show capabilities persist per connection and image/embedding requests route through the advertising host. Reconciliation notes: conflicted areas merged onto the current tip (adobe discovery import, requestedModel preflight signature, resolvedProvider fast-path coexists with the synced-route override — explicit resolution wins); carried base-red drains (#10055 memoization, #11071 test variants) dropped as already-landed; the managed-model-import exclusion test was propagated to the new contract (image/video models persist; the read filter still hides them from chat pickers — pinned by a new assertion). Full battery: 205/206 focused (the one red is a confirmed periodic-timer timing flake on the loaded devbox — 20/20 isolated), autoCombo vitest 30/30, combo suites 46/46, gates + typecheck clean. Thank you @yourspraveen — the capability probe + routing design was right; it just needed the store contract opened up. Fixes #11087.
53 lines
1.8 KiB
TypeScript
53 lines
1.8 KiB
TypeScript
import { describe, it } from "node:test";
|
|
import assert from "node:assert/strict";
|
|
|
|
import { getPromptCache } from "../../src/lib/cacheLayer.ts";
|
|
import {
|
|
clearMemoryCache,
|
|
getMemoryCacheStats,
|
|
setCachedResponse,
|
|
} from "../../src/lib/semanticCache.ts";
|
|
|
|
// Regression guard for /api/cache/stats.
|
|
//
|
|
// The route used to read getPromptCache() — an LRU that no request path writes
|
|
// to. It answered "0 hit / 0 miss, size 0" no matter how much traffic the
|
|
// semantic cache served, and two dashboard pages rendered that as fact.
|
|
//
|
|
// The first assertion fails against the old wiring: caching a response fills the
|
|
// semantic cache and leaves the prompt cache empty.
|
|
describe("cache stats report the cache that requests actually use", () => {
|
|
it("counts an entry written through the semantic cache", () => {
|
|
clearMemoryCache();
|
|
getPromptCache().clear();
|
|
|
|
const before = getMemoryCacheStats();
|
|
setCachedResponse("sig-cache-stats-guard", "gpt-4.1", { choices: [] }, 42);
|
|
const after = getMemoryCacheStats();
|
|
|
|
assert.equal(after.size, before.size + 1);
|
|
assert.equal(getPromptCache().getStats().size, 0);
|
|
});
|
|
|
|
it("keeps the shape the dashboards read, with a numeric hit rate", () => {
|
|
const stats = getMemoryCacheStats();
|
|
|
|
for (const key of ["size", "maxSize", "hits", "misses", "hitRate"]) {
|
|
assert.ok(key in stats, `missing ${key}`);
|
|
}
|
|
// Both dashboard pages call hitRate.toFixed(1); a string would throw there.
|
|
assert.equal(typeof stats.hitRate, "number");
|
|
assert.equal(typeof stats.size, "number");
|
|
assert.equal(typeof stats.maxSize, "number");
|
|
});
|
|
|
|
it("clears the in-memory entries", () => {
|
|
setCachedResponse("sig-cache-stats-clear", "gpt-4.1", { choices: [] }, 1);
|
|
assert.ok(getMemoryCacheStats().size > 0);
|
|
|
|
clearMemoryCache();
|
|
|
|
assert.equal(getMemoryCacheStats().size, 0);
|
|
});
|
|
});
|