Spaces:
Runtime error
Runtime error
File size: 6,300 Bytes
077865a | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 | import { afterEach, beforeAll, beforeEach, describe, expect, it, vi } from 'vitest';
import type { Express } from 'express';
import { createApp } from '../../app.js';
import { getDb, initDb } from '../../db/index.js';
import { mintDashboardToken, isGatedApiPath } from '../helpers/auth.js';
let dashToken = '';
async function request(app: Express, path: string) {
const server = app.listen(0);
const addr = server.address() as any;
const url = `http://127.0.0.1:${addr.port}${path}`;
const res = await fetch(url, {
headers: isGatedApiPath(path) ? { Authorization: `Bearer ${dashToken}` } : {},
});
const data = await res.json().catch(() => null);
server.close();
return { status: res.status, body: data };
}
function insertRequest(createdAt: string) {
const db = getDb();
db.prepare(`
INSERT INTO requests (platform, model_id, status, input_tokens, output_tokens, latency_ms, error, created_at)
VALUES ('test', 'test-model', 'success', 1, 2, 3, NULL, ?)
`).run(createdAt);
}
function insertTokensRequest(
platform: string,
modelId: string,
status: 'success' | 'error',
inputTokens: number,
outputTokens: number,
createdAt: string,
) {
const db = getDb();
db.prepare(`
INSERT INTO requests (platform, model_id, status, input_tokens, output_tokens, latency_ms, error, created_at)
VALUES (?, ?, ?, ?, ?, 3, NULL, ?)
`).run(platform, modelId, status, inputTokens, outputTokens, createdAt);
}
describe('Analytics API', () => {
let app: Express;
beforeAll(() => {
process.env.ENCRYPTION_KEY = '0'.repeat(64);
initDb(':memory:');
app = createApp();
dashToken = mintDashboardToken();
});
beforeEach(() => {
getDb().prepare('DELETE FROM requests').run();
vi.useFakeTimers();
vi.setSystemTime(new Date('2026-05-29T12:00:00.000Z'));
});
afterEach(() => {
vi.useRealTimers();
});
it('uses a rolling 24-hour window for summary analytics', async () => {
insertRequest('2026-05-28 11:59:59');
insertRequest('2026-05-28 12:00:00');
insertRequest('2026-05-29 11:59:59');
const { status, body } = await request(app, '/api/analytics/summary?range=24h');
expect(status).toBe(200);
expect(body.totalRequests).toBe(2);
expect(body.totalInputTokens).toBe(2);
expect(body.totalOutputTokens).toBe(4);
});
it.each([
['7d', '2026-05-22 11:59:59', '2026-05-22 12:00:00'],
['30d', '2026-04-29 11:59:59', '2026-04-29 12:00:00'],
])('uses a rolling %s window for summary analytics', async (range, outside, boundary) => {
insertRequest(outside);
insertRequest(boundary);
insertRequest('2026-05-29 11:59:59');
const { status, body } = await request(app, `/api/analytics/summary?range=${range}`);
expect(status).toBe(200);
expect(body.totalRequests).toBe(2);
});
it('prices savings at the served model paid-equivalent rate', async () => {
// groq/llama-3.3-70b-versatile is mapped at $0.10/M in, $0.32/M out
// (db/model-pricing.ts): 10M in + 5M out → 1.00 + 1.60 = $2.60
insertTokensRequest('groq', 'llama-3.3-70b-versatile', 'success', 10_000_000, 5_000_000, '2026-05-29 11:00:00');
const { status, body } = await request(app, '/api/analytics/summary?range=24h');
expect(status).toBe(200);
expect(body.estimatedCostSavings).toBe(2.6);
// Drives the client's span-based 30-day projection
expect(body.firstRequestAt).toBe('2026-05-29 11:00:00');
});
it('falls back to modest default pricing for unmapped models', async () => {
// Unknown model → $0.20/M in, $0.80/M out: 10M in + 5M out → 2.00 + 4.00 = $6.00
insertTokensRequest('custom', 'mystery-model', 'success', 10_000_000, 5_000_000, '2026-05-29 11:00:00');
const { status, body } = await request(app, '/api/analytics/summary?range=24h');
expect(status).toBe(200);
expect(body.estimatedCostSavings).toBe(6);
});
it('excludes failed requests from savings', async () => {
insertTokensRequest('groq', 'llama-3.3-70b-versatile', 'error', 10_000_000, 0, '2026-05-29 11:00:00');
const { status, body } = await request(app, '/api/analytics/summary?range=24h');
expect(status).toBe(200);
expect(body.estimatedCostSavings).toBe(0);
});
it('returns per-model estimated cost in the by-model breakdown', async () => {
insertTokensRequest('groq', 'llama-3.3-70b-versatile', 'success', 10_000_000, 5_000_000, '2026-05-29 11:00:00');
const { status, body } = await request(app, '/api/analytics/by-model?range=24h');
expect(status).toBe(200);
expect(body[0].estimatedCost).toBe(2.6);
});
describe('pinned vs auto tracking', () => {
function insertPinnedRequest(modelId: string, requestedModel: string | null, createdAt: string) {
getDb().prepare(`
INSERT INTO requests (platform, model_id, requested_model, status, input_tokens, output_tokens, latency_ms, error, created_at)
VALUES ('test', ?, ?, 'success', 1, 2, 3, NULL, ?)
`).run(modelId, requestedModel, createdAt);
}
it('summary splits pinned, honored, and auto requests', async () => {
insertPinnedRequest('model-a', 'model-a', '2026-05-29 11:00:00'); // pin honored
insertPinnedRequest('model-b', 'model-a', '2026-05-29 11:01:00'); // pin overridden by failover
insertPinnedRequest('model-b', null, '2026-05-29 11:02:00'); // auto-routed
const { status, body } = await request(app, '/api/analytics/summary?range=24h');
expect(status).toBe(200);
expect(body.totalRequests).toBe(3);
expect(body.pinnedRequests).toBe(2);
expect(body.pinHonoredRequests).toBe(1);
});
it('by-model counts only requests the model served because it was pinned', async () => {
insertPinnedRequest('model-a', 'model-a', '2026-05-29 11:00:00'); // pinned + served
insertPinnedRequest('model-a', null, '2026-05-29 11:01:00'); // auto, same model
insertPinnedRequest('model-a', 'model-x', '2026-05-29 11:02:00'); // failover landed here
const { status, body } = await request(app, '/api/analytics/by-model?range=24h');
expect(status).toBe(200);
const row = body.find((r: any) => r.modelId === 'model-a');
expect(row.requests).toBe(3);
expect(row.pinnedRequests).toBe(1);
});
});
});
|