0d3cb498a3
Deploy local.promptfoo.app / Deploy to Cloudflare Pages (push) Waiting to run
Test and Publish Multi-arch Docker Image / test (push) Waiting to run
Test and Publish Multi-arch Docker Image / build-docker-and-push-digests (map[digest-suffix:linux-amd64 platform:linux/amd64 runner:ubuntu-latest]) (push) Blocked by required conditions
Test and Publish Multi-arch Docker Image / build-docker-and-push-digests (map[digest-suffix:linux-arm64 platform:linux/arm64 runner:ubuntu-24.04-arm]) (push) Blocked by required conditions
Test and Publish Multi-arch Docker Image / merge-docker-digests (push) Blocked by required conditions
Test and Publish Multi-arch Docker Image / Attest Multi-arch Image (push) Blocked by required conditions
Validate Renovate Config / Validate Renovate Configuration (push) Waiting to run
CI / Shell Format Check (push) Has been cancelled
CI / Check Ruby (3.4) (push) Has been cancelled
CI / CI Config (push) Has been cancelled
CI / Test on Node ${{ matrix.node }} and ${{ matrix.os }}${{ matrix.shard && format(' (shard {0}/3)', matrix.shard) || '' }} (push) Has been cancelled
CI / Build on Node ${{ matrix.node }} (push) Has been cancelled
CI / Style Check (push) Has been cancelled
CI / Generate Assets (push) Has been cancelled
CI / Check Python (3.14) (push) Has been cancelled
CI / Check Python (3.9) (push) Has been cancelled
CI / Build Docs (push) Has been cancelled
CI / Code Scan Action (push) Has been cancelled
CI / Site tests (push) Has been cancelled
CI / webui tests (push) Has been cancelled
CI / Run Integration Tests (push) Has been cancelled
CI / Run Smoke Tests (push) Has been cancelled
CI / Go Tests (push) Has been cancelled
CI / Share Test (push) Has been cancelled
CI / Redteam (Production API) (push) Has been cancelled
CI / Redteam (Staging API) (push) Has been cancelled
CI / GitHub Actions Lint (push) Has been cancelled
CI / Check Ruby (3.0) (push) Has been cancelled
release-please / release-please (push) Has been cancelled
release-please / build (push) Has been cancelled
release-please / publish-npm (push) Has been cancelled
release-please / publish-npm-backfill (push) Has been cancelled
release-please / docker (push) Has been cancelled
release-please / publish-code-scan-action (push) Has been cancelled
release-please / attest-code-scan-action (push) Has been cancelled
106 lines
4.7 KiB
TypeScript
106 lines
4.7 KiB
TypeScript
import { afterEach, describe, expect, it, vi } from 'vitest';
|
|
import Eval from '../../src/models/eval';
|
|
import EvalResult from '../../src/models/evalResult';
|
|
import { EvalEvaluationStore } from '../../src/node/evaluationStore';
|
|
import { createCompletedPrompt, createEvaluateResult } from '../factories/eval';
|
|
|
|
describe('EvalEvaluationStore', () => {
|
|
afterEach(() => {
|
|
vi.restoreAllMocks();
|
|
});
|
|
|
|
it('exposes the underlying evaluation state', () => {
|
|
const evaluation = new Eval({ description: 'adapter test' });
|
|
const store = new EvalEvaluationStore(evaluation);
|
|
|
|
expect(store.evaluation).toBe(evaluation);
|
|
expect(store.id).toBe(evaluation.id);
|
|
expect(store.config).toBe(evaluation.config);
|
|
expect(store.persisted).toBe(false);
|
|
expect(store.prompts).toBe(evaluation.prompts);
|
|
expect(store.results).toBe(evaluation.results);
|
|
expect(store.resultPersistenceFailed).toBe(false);
|
|
});
|
|
|
|
it('delegates result, prompt, resume, and read operations', async () => {
|
|
const evaluation = new Eval({});
|
|
const store = new EvalEvaluationStore(evaluation);
|
|
const result = createEvaluateResult();
|
|
const prompts = [createCompletedPrompt('Prompt')];
|
|
const modelResult = { testIdx: 0 } as EvalResult;
|
|
const completed = new Set(['0:0']);
|
|
|
|
const addResult = vi.spyOn(evaluation, 'addResult').mockResolvedValue(undefined);
|
|
const addPrompts = vi.spyOn(evaluation, 'addPrompts').mockResolvedValue(undefined);
|
|
const getCompletedIndexPairs = vi
|
|
.spyOn(EvalResult, 'getCompletedIndexPairs')
|
|
.mockResolvedValue(completed);
|
|
const getFailedResultsByTestIdx = vi
|
|
.spyOn(evaluation, 'getFailedResultsByTestIdx')
|
|
.mockResolvedValue([modelResult]);
|
|
const getResults = vi.spyOn(evaluation, 'getResults').mockResolvedValue([modelResult]);
|
|
const fetchResultsByTestIdx = vi
|
|
.spyOn(evaluation, 'fetchResultsByTestIdx')
|
|
.mockResolvedValue([modelResult]);
|
|
|
|
await store.appendResult(result);
|
|
await store.appendPrompts(prompts);
|
|
|
|
expect(await store.readCompletedIndexPairs({ excludeErrors: true })).toBe(completed);
|
|
expect(await store.readFailedResultsByTestIdx(0)).toEqual([modelResult]);
|
|
expect(await store.readResults()).toEqual([modelResult]);
|
|
expect(await store.readResultsByTestIdx(0)).toEqual([modelResult]);
|
|
expect(addResult).toHaveBeenCalledWith(result);
|
|
expect(addPrompts).toHaveBeenCalledWith(prompts);
|
|
expect(getCompletedIndexPairs).toHaveBeenCalledWith(evaluation.id, { excludeErrors: true });
|
|
expect(getFailedResultsByTestIdx).toHaveBeenCalledWith(0);
|
|
expect(getResults).toHaveBeenCalledOnce();
|
|
expect(fetchResultsByTestIdx).toHaveBeenCalledWith(0);
|
|
});
|
|
|
|
it('delegates persistence failure, final result, and save operations', async () => {
|
|
const evaluation = new Eval({});
|
|
const store = new EvalEvaluationStore(evaluation);
|
|
const result = createEvaluateResult();
|
|
const modelResult = { save: vi.fn().mockResolvedValue(undefined) } as unknown as EvalResult;
|
|
|
|
const recordFinalJsonlResult = vi.spyOn(evaluation, 'recordFinalJsonlResult');
|
|
const recordResultPersistenceFailure = vi.spyOn(evaluation, 'recordResultPersistenceFailure');
|
|
const hasResultPersistenceFailure = vi
|
|
.spyOn(evaluation, 'hasResultPersistenceFailure')
|
|
.mockReturnValue(true);
|
|
const save = vi.spyOn(evaluation, 'save').mockResolvedValue(undefined);
|
|
const setDurationMs = vi.spyOn(evaluation, 'setDurationMs');
|
|
const setVars = vi.spyOn(evaluation, 'setVars');
|
|
|
|
store.recordFinalResult(result);
|
|
store.recordResultPersistenceFailure(result);
|
|
expect(store.hasResultPersistenceFailure(result)).toBe(true);
|
|
await store.saveResult(modelResult);
|
|
store.setDurationMs(100);
|
|
store.setVars(['topic']);
|
|
await store.save();
|
|
|
|
expect(recordFinalJsonlResult).toHaveBeenCalledWith(result);
|
|
expect(recordResultPersistenceFailure).toHaveBeenCalledWith(result);
|
|
expect(hasResultPersistenceFailure).toHaveBeenCalledWith(result);
|
|
expect(modelResult.save).toHaveBeenCalledOnce();
|
|
expect(setDurationMs).toHaveBeenCalledWith(100);
|
|
expect(setVars).toHaveBeenCalledWith(['topic']);
|
|
expect(save).toHaveBeenCalledOnce();
|
|
});
|
|
|
|
it('normalizes model results while preserving plain EvaluateResult values', () => {
|
|
const store = new EvalEvaluationStore(new Eval({}));
|
|
const plainResult = createEvaluateResult();
|
|
const normalizedResult = createEvaluateResult({ score: 0.5 });
|
|
const modelResult = {
|
|
toEvaluateResult: vi.fn().mockReturnValue(normalizedResult),
|
|
} as unknown as EvalResult;
|
|
|
|
expect(store.toEvaluateResult(plainResult)).toBe(plainResult);
|
|
expect(store.toEvaluateResult(modelResult)).toBe(normalizedResult);
|
|
expect(modelResult.toEvaluateResult).toHaveBeenCalledOnce();
|
|
});
|
|
});
|