chore: import upstream snapshot with attribution
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
Deploy local.promptfoo.app / Deploy to Cloudflare Pages (push) Has been cancelled
Test and Publish Multi-arch Docker Image / test (push) Has been cancelled
Test and Publish Multi-arch Docker Image / build-docker-and-push-digests (map[digest-suffix:linux-amd64 platform:linux/amd64 runner:ubuntu-latest]) (push) Has been cancelled
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) Has been cancelled
Test and Publish Multi-arch Docker Image / merge-docker-digests (push) Has been cancelled
Test and Publish Multi-arch Docker Image / Attest Multi-arch Image (push) Has been cancelled
Validate Renovate Config / Validate Renovate Configuration (push) Has been cancelled
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
Deploy local.promptfoo.app / Deploy to Cloudflare Pages (push) Has been cancelled
Test and Publish Multi-arch Docker Image / test (push) Has been cancelled
Test and Publish Multi-arch Docker Image / build-docker-and-push-digests (map[digest-suffix:linux-amd64 platform:linux/amd64 runner:ubuntu-latest]) (push) Has been cancelled
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) Has been cancelled
Test and Publish Multi-arch Docker Image / merge-docker-digests (push) Has been cancelled
Test and Publish Multi-arch Docker Image / Attest Multi-arch Image (push) Has been cancelled
Validate Renovate Config / Validate Renovate Configuration (push) Has been cancelled
This commit is contained in:
@@ -0,0 +1,233 @@
|
||||
import fs from 'fs';
|
||||
|
||||
import * as yaml from 'js-yaml';
|
||||
import { afterAll, afterEach, beforeAll, beforeEach, describe, expect, it, vi } from 'vitest';
|
||||
import { synthesizeFromTestSuite } from '../../../src/assertions/synthesis';
|
||||
import { disableCache } from '../../../src/cache';
|
||||
import { doGenerateAssertions } from '../../../src/commands/generate/assertions';
|
||||
import telemetry from '../../../src/telemetry';
|
||||
import { resolveConfigs } from '../../../src/util/config/load';
|
||||
import { loadYaml } from '../../../src/util/yamlLoad';
|
||||
|
||||
import type { Assertion, TestSuite } from '../../../src/types/index';
|
||||
|
||||
const fsMocks = vi.hoisted(() => ({
|
||||
readFileSync: vi.fn(),
|
||||
writeFileSync: vi.fn(),
|
||||
existsSync: vi.fn(),
|
||||
}));
|
||||
|
||||
vi.mock('fs', async (importOriginal) => {
|
||||
const actual = await importOriginal<typeof import('fs')>();
|
||||
return {
|
||||
...actual,
|
||||
default: {
|
||||
...actual,
|
||||
...fsMocks,
|
||||
},
|
||||
...fsMocks,
|
||||
};
|
||||
});
|
||||
|
||||
vi.mock('fs/promises', () => ({
|
||||
default: {
|
||||
readFile: fsMocks.readFileSync,
|
||||
writeFile: fsMocks.writeFileSync,
|
||||
},
|
||||
readFile: fsMocks.readFileSync,
|
||||
writeFile: fsMocks.writeFileSync,
|
||||
}));
|
||||
vi.mock('js-yaml');
|
||||
vi.mock('../../../src/util/yamlLoad');
|
||||
vi.mock('../../../src/assertions/synthesis');
|
||||
vi.mock('../../../src/util/config/load', () => ({
|
||||
resolveConfigs: vi.fn(),
|
||||
}));
|
||||
vi.mock('../../../src/cache', () => ({
|
||||
disableCache: vi.fn(),
|
||||
}));
|
||||
vi.mock('../../../src/logger', () => ({
|
||||
default: {
|
||||
info: vi.fn(),
|
||||
warn: vi.fn(),
|
||||
error: vi.fn(),
|
||||
debug: vi.fn(),
|
||||
child: vi.fn().mockReturnValue({}),
|
||||
},
|
||||
}));
|
||||
vi.mock('../../../src/telemetry', () => ({
|
||||
default: {
|
||||
record: vi.fn(),
|
||||
send: vi.fn().mockResolvedValue(undefined),
|
||||
},
|
||||
}));
|
||||
vi.mock('../../../src/util', () => ({
|
||||
printBorder: vi.fn(),
|
||||
setupEnv: vi.fn(),
|
||||
}));
|
||||
|
||||
vi.mock('../../../src/util/promptfooCommand', () => ({
|
||||
promptfooCommand: vi.fn().mockReturnValue('promptfoo eval'),
|
||||
detectInstaller: vi.fn().mockReturnValue('unknown'),
|
||||
isRunningUnderNpx: vi.fn().mockReturnValue(false),
|
||||
}));
|
||||
|
||||
describe('assertion generation', () => {
|
||||
beforeAll(() => {
|
||||
vi.useFakeTimers();
|
||||
});
|
||||
|
||||
beforeEach(() => {
|
||||
vi.resetAllMocks();
|
||||
});
|
||||
|
||||
afterEach(() => {
|
||||
vi.resetAllMocks();
|
||||
});
|
||||
|
||||
afterAll(() => {
|
||||
vi.useRealTimers();
|
||||
});
|
||||
|
||||
describe('doGenerateAssertions', () => {
|
||||
const mockTestSuite: TestSuite = {
|
||||
prompts: [{ raw: 'test prompt', label: 'test' }],
|
||||
tests: [
|
||||
{
|
||||
assert: [
|
||||
{
|
||||
type: 'llm-rubric',
|
||||
value: 'test question',
|
||||
},
|
||||
],
|
||||
},
|
||||
],
|
||||
providers: [
|
||||
{
|
||||
id: () => 'test-provider',
|
||||
callApi: vi.fn() as any,
|
||||
},
|
||||
],
|
||||
};
|
||||
|
||||
const mockResults: Assertion[] = [
|
||||
{ type: 'pi', value: 'additional assertion' },
|
||||
{ type: 'pi', value: 'additional assertion 2' },
|
||||
];
|
||||
|
||||
beforeEach(() => {
|
||||
vi.mocked(synthesizeFromTestSuite).mockResolvedValue(mockResults);
|
||||
vi.mocked(yaml.dump).mockReturnValue('yaml content');
|
||||
vi.mocked(loadYaml).mockReturnValue(mockTestSuite);
|
||||
vi.mocked(fs.readFileSync).mockReturnValue('mock config content');
|
||||
vi.mocked(fs.existsSync).mockReturnValue(true);
|
||||
vi.mocked(fs.writeFileSync).mockImplementation(() => undefined);
|
||||
vi.mocked(disableCache).mockImplementation(() => undefined);
|
||||
vi.mocked(telemetry.record).mockImplementation(() => undefined);
|
||||
|
||||
vi.mocked(resolveConfigs).mockResolvedValue({
|
||||
testSuite: mockTestSuite,
|
||||
config: {},
|
||||
basePath: '',
|
||||
});
|
||||
});
|
||||
|
||||
it('should write YAML output', async () => {
|
||||
const configPath = 'config.yaml';
|
||||
|
||||
await doGenerateAssertions({
|
||||
cache: true,
|
||||
config: configPath,
|
||||
numAssertions: '2',
|
||||
type: 'pi',
|
||||
output: 'output.yaml',
|
||||
write: false,
|
||||
defaultConfig: {},
|
||||
defaultConfigPath: configPath,
|
||||
});
|
||||
|
||||
expect(fs.writeFileSync).toHaveBeenCalledWith('output.yaml', 'yaml content');
|
||||
});
|
||||
|
||||
it('should throw error for unsupported file type', async () => {
|
||||
const configPath = 'config.yaml';
|
||||
|
||||
await expect(
|
||||
doGenerateAssertions({
|
||||
cache: true,
|
||||
config: configPath,
|
||||
numAssertions: '2',
|
||||
type: 'pi',
|
||||
output: 'output.txt',
|
||||
write: false,
|
||||
defaultConfig: {},
|
||||
defaultConfigPath: configPath,
|
||||
}),
|
||||
).rejects.toThrow('Unsupported output file type: output.txt');
|
||||
});
|
||||
|
||||
it('should write to config file when write option is true', async () => {
|
||||
const configPath = 'config.yaml';
|
||||
|
||||
await doGenerateAssertions({
|
||||
cache: true,
|
||||
config: configPath,
|
||||
numAssertions: '2',
|
||||
type: 'pi',
|
||||
write: true,
|
||||
defaultConfig: {},
|
||||
defaultConfigPath: configPath,
|
||||
});
|
||||
|
||||
expect(fs.writeFileSync).toHaveBeenCalledWith(configPath, expect.any(String));
|
||||
});
|
||||
|
||||
it('should throw error when no config file found', async () => {
|
||||
await expect(
|
||||
doGenerateAssertions({
|
||||
cache: true,
|
||||
numAssertions: '2',
|
||||
type: 'pi',
|
||||
write: false,
|
||||
defaultConfig: {},
|
||||
defaultConfigPath: undefined,
|
||||
}),
|
||||
).rejects.toThrow('Could not find a config file');
|
||||
});
|
||||
|
||||
it('should handle output without file extension', async () => {
|
||||
const configPath = 'config.yaml';
|
||||
|
||||
await expect(
|
||||
doGenerateAssertions({
|
||||
cache: true,
|
||||
config: configPath,
|
||||
numAssertions: '2',
|
||||
type: 'pi',
|
||||
output: 'output',
|
||||
write: false,
|
||||
defaultConfig: {},
|
||||
defaultConfigPath: configPath,
|
||||
}),
|
||||
).rejects.toThrow('Unsupported output file type: output');
|
||||
});
|
||||
|
||||
it('should handle synthesis errors', async () => {
|
||||
vi.mocked(synthesizeFromTestSuite).mockRejectedValue(new Error('Synthesis failed'));
|
||||
const configPath = 'config.yaml';
|
||||
|
||||
await expect(
|
||||
doGenerateAssertions({
|
||||
cache: true,
|
||||
config: configPath,
|
||||
numAssertions: '2',
|
||||
type: 'pi',
|
||||
output: 'output.yaml',
|
||||
write: false,
|
||||
defaultConfig: {},
|
||||
defaultConfigPath: configPath,
|
||||
}),
|
||||
).rejects.toThrow('Synthesis failed');
|
||||
});
|
||||
});
|
||||
});
|
||||
Reference in New Issue
Block a user