mirror of
https://github.com/diegosouzapw/OmniRoute.git
synced 2026-09-19 13:23:50 +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.
84 lines
2.8 KiB
TypeScript
84 lines
2.8 KiB
TypeScript
import { test } from "node:test";
|
||
import assert from "node:assert/strict";
|
||
|
||
import { estimateChatTokenCost, estimateStringTokens } from "../../src/lib/quota/tokenEstimator";
|
||
|
||
test("estimateStringTokens: chars/4 heuristic", () => {
|
||
assert.equal(estimateStringTokens(""), 0);
|
||
assert.equal(estimateStringTokens("abcd"), 1);
|
||
assert.equal(estimateStringTokens("abcdefgh"), 2);
|
||
});
|
||
|
||
test("estimateChatTokenCost: sums message content strings", () => {
|
||
const cost = estimateChatTokenCost({
|
||
messages: [
|
||
{ role: "user", content: "Hello world, this is a test message" },
|
||
{ role: "assistant", content: "A shorter reply" },
|
||
],
|
||
});
|
||
assert.ok(cost.inputTokens > 0);
|
||
// total = input × 1.1 (over-provision) + output budget
|
||
assert.equal(cost.totalTokens, Math.ceil(cost.inputTokens * 1.1) + cost.outputTokens);
|
||
});
|
||
|
||
test("estimateChatTokenCost: includes system prompt", () => {
|
||
const withoutSystem = estimateChatTokenCost({
|
||
messages: [{ role: "user", content: "hi there" }],
|
||
});
|
||
const withSystem = estimateChatTokenCost({
|
||
system: "You are a helpful assistant with a fairly long system prompt to count",
|
||
messages: [{ role: "user", content: "hi there" }],
|
||
});
|
||
assert.ok(withSystem.inputTokens > withoutSystem.inputTokens);
|
||
});
|
||
|
||
test("estimateChatTokenCost: honors max_tokens as output budget", () => {
|
||
const cost = estimateChatTokenCost({
|
||
messages: [{ role: "user", content: "hi" }],
|
||
max_tokens: 2000,
|
||
});
|
||
assert.equal(cost.outputTokens, 2000);
|
||
});
|
||
|
||
test("estimateChatTokenCost: honors max_completion_tokens (Responses API)", () => {
|
||
const cost = estimateChatTokenCost({
|
||
messages: [{ role: "user", content: "hi" }],
|
||
max_completion_tokens: 500,
|
||
});
|
||
assert.equal(cost.outputTokens, 500);
|
||
});
|
||
|
||
test("estimateChatTokenCost: defaults output allowance when unset", () => {
|
||
const cost = estimateChatTokenCost({ messages: [{ role: "user", content: "hi" }] });
|
||
assert.equal(cost.outputTokens, 1024);
|
||
});
|
||
|
||
test("estimateChatTokenCost: handles multimodal content arrays", () => {
|
||
const cost = estimateChatTokenCost({
|
||
messages: [
|
||
{
|
||
role: "user",
|
||
content: [
|
||
{ type: "text", text: "Describe this image" },
|
||
{ type: "image_url", image_url: { url: "data:image/png;base64,xxx" } },
|
||
],
|
||
},
|
||
],
|
||
});
|
||
assert.ok(cost.inputTokens > 0);
|
||
});
|
||
|
||
test("estimateChatTokenCost: handles Responses API input array", () => {
|
||
const cost = estimateChatTokenCost({
|
||
input: [{ role: "user", text: "What is the capital of France" }],
|
||
});
|
||
assert.ok(cost.inputTokens > 0);
|
||
});
|
||
|
||
test("estimateChatTokenCost: never throws on malformed bodies", () => {
|
||
for (const bad of [null, undefined, {}, { messages: "nope" }, { messages: [null, 42] }]) {
|
||
const cost = estimateChatTokenCost(bad as Record<string, unknown>);
|
||
assert.ok(cost.totalTokens >= 0);
|
||
}
|
||
});
|