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
160 lines
5.5 KiB
TypeScript
160 lines
5.5 KiB
TypeScript
import { describe, expect, it } from 'vitest';
|
|
import { handleFinishReason } from '../../src/assertions/finishReason';
|
|
|
|
import type { AssertionParams } from '../../src/types/index';
|
|
|
|
describe('finishReason assertion', () => {
|
|
it('should pass when finish reason matches', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'finish-reason', value: 'stop' },
|
|
providerResponse: { finishReason: 'stop' },
|
|
};
|
|
|
|
const result = handleFinishReason(params as AssertionParams);
|
|
expect(result.pass).toBe(true);
|
|
expect(result.score).toBe(1);
|
|
});
|
|
|
|
it('should fail when stop reason does not match', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'finish-reason', value: 'stop' },
|
|
providerResponse: { finishReason: 'length' },
|
|
};
|
|
|
|
const result = handleFinishReason(params as AssertionParams);
|
|
expect(result.pass).toBe(false);
|
|
expect(result.score).toBe(0);
|
|
});
|
|
|
|
it('should fail when provider does not return stop reason', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'finish-reason', value: 'stop' },
|
|
providerResponse: {},
|
|
};
|
|
|
|
const result = handleFinishReason(params as AssertionParams);
|
|
expect(result.pass).toBe(false);
|
|
expect(result.score).toBe(0);
|
|
expect(result.reason).toContain('did not supply');
|
|
});
|
|
|
|
it('should handle renderedValue override', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'finish-reason', value: 'stop' },
|
|
renderedValue: 'length',
|
|
providerResponse: { finishReason: 'length' },
|
|
};
|
|
|
|
const result = handleFinishReason(params as AssertionParams);
|
|
expect(result.pass).toBe(true);
|
|
expect(result.score).toBe(1);
|
|
});
|
|
|
|
it('should prioritize renderedValue over assertion.value', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'finish-reason', value: 'stop' },
|
|
renderedValue: 'length',
|
|
providerResponse: { finishReason: 'stop' },
|
|
};
|
|
|
|
const result = handleFinishReason(params as AssertionParams);
|
|
expect(result.pass).toBe(false);
|
|
expect(result.score).toBe(0);
|
|
expect(result.reason).toContain('Expected finish reason "length" but got "stop"');
|
|
});
|
|
|
|
it('should throw for non-string assertion value', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'finish-reason', value: 123 as any },
|
|
providerResponse: { finishReason: 'stop' },
|
|
};
|
|
|
|
expect(() => handleFinishReason(params as AssertionParams)).toThrow(
|
|
'"finish-reason" assertion type must have a string value',
|
|
);
|
|
});
|
|
|
|
it('should throw for non-string renderedValue', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'finish-reason', value: 'stop' },
|
|
renderedValue: 123 as any,
|
|
providerResponse: { finishReason: 'stop' },
|
|
};
|
|
|
|
expect(() => handleFinishReason(params as AssertionParams)).toThrow(
|
|
'"finish-reason" assertion type must have a string value',
|
|
);
|
|
});
|
|
|
|
it('should handle undefined finishReason gracefully', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'finish-reason', value: 'stop' },
|
|
providerResponse: { finishReason: undefined },
|
|
};
|
|
|
|
const result = handleFinishReason(params as AssertionParams);
|
|
expect(result.pass).toBe(false);
|
|
expect(result.score).toBe(0);
|
|
expect(result.reason).toContain('did not supply');
|
|
});
|
|
|
|
it('should perform case-insensitive matching', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'finish-reason', value: 'stop' },
|
|
providerResponse: { finishReason: 'STOP' }, // Different case
|
|
};
|
|
|
|
const result = handleFinishReason(params as AssertionParams);
|
|
expect(result.pass).toBe(true); // Now passes with case-insensitive comparison
|
|
expect(result.score).toBe(1);
|
|
expect(result.reason).toBe('Assertion passed');
|
|
});
|
|
|
|
it('should handle mixed case in both assertion and response', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'finish-reason', value: 'LENGTH' },
|
|
providerResponse: { finishReason: 'length' },
|
|
};
|
|
|
|
const result = handleFinishReason(params as AssertionParams);
|
|
expect(result.pass).toBe(true);
|
|
expect(result.score).toBe(1);
|
|
});
|
|
|
|
it('should invert result when inverse is true (not-finish-reason)', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'not-finish-reason' as any, value: 'stop' },
|
|
inverse: true,
|
|
providerResponse: { finishReason: 'stop' },
|
|
};
|
|
|
|
const result = handleFinishReason(params as AssertionParams);
|
|
expect(result.pass).toBe(false);
|
|
expect(result.score).toBe(0);
|
|
});
|
|
|
|
it('should pass when inverse is true and finish reason does not match', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'not-finish-reason' as any, value: 'stop' },
|
|
inverse: true,
|
|
providerResponse: { finishReason: 'length' },
|
|
};
|
|
|
|
const result = handleFinishReason(params as AssertionParams);
|
|
expect(result.pass).toBe(true);
|
|
expect(result.score).toBe(1);
|
|
});
|
|
|
|
it('should pass when inverse is true and provider has no finish reason', () => {
|
|
const params: Partial<AssertionParams> = {
|
|
assertion: { type: 'not-finish-reason' as any, value: 'stop' },
|
|
inverse: true,
|
|
providerResponse: {},
|
|
};
|
|
|
|
const result = handleFinishReason(params as AssertionParams);
|
|
expect(result.pass).toBe(true);
|
|
expect(result.score).toBe(1);
|
|
});
|
|
});
|