mirror of
https://github.com/Abdulazizzn/n8n-enterprise-unlocked.git
synced 2025-12-16 09:36:44 +00:00
feat(OpenAI Node): Support custom headers for model requests (#17835)
Co-authored-by: Idir Ouhab <ouhab.idir@gmail.com> Co-authored-by: Oleg Ivaniv <me@olegivaniv.com>
This commit is contained in:
@@ -358,6 +358,16 @@ export class LmChatOpenAi implements INodeType {
|
||||
dispatcher: getProxyAgent(configuration.baseURL ?? 'https://api.openai.com/v1'),
|
||||
};
|
||||
}
|
||||
if (
|
||||
credentials.header &&
|
||||
typeof credentials.headerName === 'string' &&
|
||||
credentials.headerName &&
|
||||
typeof credentials.headerValue === 'string'
|
||||
) {
|
||||
configuration.defaultHeaders = {
|
||||
[credentials.headerName]: credentials.headerValue,
|
||||
};
|
||||
}
|
||||
|
||||
// Extra options to send to OpenAI, that are not directly supported by LangChain
|
||||
const modelKwargs: {
|
||||
|
||||
@@ -0,0 +1,488 @@
|
||||
/* eslint-disable n8n-nodes-base/node-filename-against-convention */
|
||||
/* eslint-disable @typescript-eslint/unbound-method */
|
||||
import { ChatOpenAI } from '@langchain/openai';
|
||||
import { createMockExecuteFunction } from 'n8n-nodes-base/test/nodes/Helpers';
|
||||
import type { INode, ISupplyDataFunctions } from 'n8n-workflow';
|
||||
|
||||
import { LmChatOpenAi } from '../LMChatOpenAi/LmChatOpenAi.node';
|
||||
import { N8nLlmTracing } from '../N8nLlmTracing';
|
||||
import { makeN8nLlmFailedAttemptHandler } from '../n8nLlmFailedAttemptHandler';
|
||||
|
||||
jest.mock('@langchain/openai');
|
||||
jest.mock('../N8nLlmTracing');
|
||||
jest.mock('../n8nLlmFailedAttemptHandler');
|
||||
jest.mock('@utils/httpProxyAgent', () => ({
|
||||
getProxyAgent: jest.fn().mockReturnValue({}),
|
||||
}));
|
||||
|
||||
const MockedChatOpenAI = jest.mocked(ChatOpenAI);
|
||||
const MockedN8nLlmTracing = jest.mocked(N8nLlmTracing);
|
||||
const mockedMakeN8nLlmFailedAttemptHandler = jest.mocked(makeN8nLlmFailedAttemptHandler);
|
||||
|
||||
describe('LmChatOpenAi', () => {
|
||||
let lmChatOpenAi: LmChatOpenAi;
|
||||
let mockContext: jest.Mocked<ISupplyDataFunctions>;
|
||||
|
||||
const mockNode: INode = {
|
||||
id: '1',
|
||||
name: 'OpenAI Chat Model',
|
||||
typeVersion: 1.2,
|
||||
type: 'n8n-nodes-langchain.lmChatOpenAi',
|
||||
position: [0, 0],
|
||||
parameters: {},
|
||||
};
|
||||
|
||||
const setupMockContext = (nodeOverrides: Partial<INode> = {}) => {
|
||||
const node = { ...mockNode, ...nodeOverrides };
|
||||
mockContext = createMockExecuteFunction<ISupplyDataFunctions>(
|
||||
{},
|
||||
node,
|
||||
) as jest.Mocked<ISupplyDataFunctions>;
|
||||
|
||||
// Setup default mocks
|
||||
mockContext.getCredentials = jest.fn().mockResolvedValue({
|
||||
apiKey: 'test-api-key',
|
||||
});
|
||||
mockContext.getNode = jest.fn().mockReturnValue(node);
|
||||
mockContext.getNodeParameter = jest.fn();
|
||||
|
||||
// Mock the constructors/functions properly
|
||||
MockedN8nLlmTracing.mockImplementation(() => ({}) as any);
|
||||
mockedMakeN8nLlmFailedAttemptHandler.mockReturnValue(jest.fn());
|
||||
|
||||
return mockContext;
|
||||
};
|
||||
|
||||
beforeEach(() => {
|
||||
lmChatOpenAi = new LmChatOpenAi();
|
||||
jest.clearAllMocks();
|
||||
});
|
||||
|
||||
afterEach(() => {
|
||||
jest.clearAllMocks();
|
||||
});
|
||||
|
||||
describe('node description', () => {
|
||||
it('should have correct node properties', () => {
|
||||
expect(lmChatOpenAi.description).toMatchObject({
|
||||
displayName: 'OpenAI Chat Model',
|
||||
name: 'lmChatOpenAi',
|
||||
group: ['transform'],
|
||||
version: [1, 1.1, 1.2],
|
||||
description: 'For advanced usage with an AI chain',
|
||||
});
|
||||
});
|
||||
|
||||
it('should have correct credentials configuration', () => {
|
||||
expect(lmChatOpenAi.description.credentials).toEqual([
|
||||
{
|
||||
name: 'openAiApi',
|
||||
required: true,
|
||||
},
|
||||
]);
|
||||
});
|
||||
|
||||
it('should have correct output configuration', () => {
|
||||
expect(lmChatOpenAi.description.outputs).toEqual(['ai_languageModel']);
|
||||
expect(lmChatOpenAi.description.outputNames).toEqual(['Model']);
|
||||
});
|
||||
});
|
||||
|
||||
describe('supplyData', () => {
|
||||
it('should create ChatOpenAI instance with basic configuration (version >= 1.2)', async () => {
|
||||
const mockContext = setupMockContext({ typeVersion: 1.2 });
|
||||
|
||||
// Mock getNodeParameter to handle the proper parameter names for v1.2
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options') return {};
|
||||
return undefined;
|
||||
});
|
||||
|
||||
const result = await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(mockContext.getCredentials).toHaveBeenCalledWith('openAiApi');
|
||||
expect(mockContext.getNodeParameter).toHaveBeenCalledWith('model.value', 0);
|
||||
expect(mockContext.getNodeParameter).toHaveBeenCalledWith('options', 0, {});
|
||||
expect(MockedChatOpenAI).toHaveBeenCalledWith(
|
||||
expect.objectContaining({
|
||||
apiKey: 'test-api-key',
|
||||
model: 'gpt-4o-mini',
|
||||
timeout: 60000,
|
||||
maxRetries: 2,
|
||||
configuration: {},
|
||||
callbacks: expect.arrayContaining([expect.any(Object)]),
|
||||
modelKwargs: {},
|
||||
onFailedAttempt: expect.any(Function),
|
||||
}),
|
||||
);
|
||||
|
||||
expect(result).toEqual({
|
||||
response: expect.any(Object),
|
||||
});
|
||||
});
|
||||
|
||||
it('should create ChatOpenAI instance with basic configuration (version < 1.2)', async () => {
|
||||
const mockContext = setupMockContext({ typeVersion: 1.1 });
|
||||
|
||||
// Mock getNodeParameter to handle the proper parameter names for v1.1
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model') return 'gpt-4o-mini';
|
||||
if (paramName === 'options') return {};
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(mockContext.getNodeParameter).toHaveBeenCalledWith('model', 0);
|
||||
expect(mockContext.getNodeParameter).toHaveBeenCalledWith('options', 0, {});
|
||||
expect(MockedChatOpenAI).toHaveBeenCalledWith(
|
||||
expect.objectContaining({
|
||||
apiKey: 'test-api-key',
|
||||
model: 'gpt-4o-mini',
|
||||
timeout: 60000,
|
||||
maxRetries: 2,
|
||||
configuration: {},
|
||||
callbacks: expect.arrayContaining([expect.any(Object)]),
|
||||
modelKwargs: {},
|
||||
onFailedAttempt: expect.any(Function),
|
||||
}),
|
||||
);
|
||||
});
|
||||
|
||||
it('should handle custom baseURL from options', async () => {
|
||||
const customBaseURL = 'https://custom-api.example.com/v1';
|
||||
const mockContext = setupMockContext();
|
||||
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options')
|
||||
return {
|
||||
baseURL: customBaseURL,
|
||||
timeout: 30000,
|
||||
maxRetries: 5,
|
||||
};
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(MockedChatOpenAI).toHaveBeenCalledWith(
|
||||
expect.objectContaining({
|
||||
apiKey: 'test-api-key',
|
||||
model: 'gpt-4o-mini',
|
||||
baseURL: customBaseURL,
|
||||
timeout: 30000,
|
||||
maxRetries: 5,
|
||||
configuration: {
|
||||
baseURL: customBaseURL,
|
||||
fetchOptions: {
|
||||
dispatcher: {},
|
||||
},
|
||||
},
|
||||
callbacks: expect.arrayContaining([expect.any(Object)]),
|
||||
modelKwargs: {},
|
||||
onFailedAttempt: expect.any(Function),
|
||||
}),
|
||||
);
|
||||
});
|
||||
|
||||
it('should handle custom baseURL from credentials', async () => {
|
||||
const customURL = 'https://custom-openai.example.com/v1';
|
||||
const mockContext = setupMockContext();
|
||||
|
||||
mockContext.getCredentials.mockResolvedValue({
|
||||
apiKey: 'test-api-key',
|
||||
url: customURL,
|
||||
});
|
||||
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options') return {};
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(MockedChatOpenAI).toHaveBeenCalledWith(
|
||||
expect.objectContaining({
|
||||
apiKey: 'test-api-key',
|
||||
model: 'gpt-4o-mini',
|
||||
timeout: 60000,
|
||||
maxRetries: 2,
|
||||
configuration: {
|
||||
baseURL: customURL,
|
||||
fetchOptions: {
|
||||
dispatcher: {},
|
||||
},
|
||||
},
|
||||
callbacks: expect.arrayContaining([expect.any(Object)]),
|
||||
modelKwargs: {},
|
||||
onFailedAttempt: expect.any(Function),
|
||||
}),
|
||||
);
|
||||
});
|
||||
|
||||
it('should handle custom headers from credentials', async () => {
|
||||
const mockContext = setupMockContext();
|
||||
|
||||
mockContext.getCredentials.mockResolvedValue({
|
||||
apiKey: 'test-api-key',
|
||||
header: true,
|
||||
headerName: 'X-Custom-Header',
|
||||
headerValue: 'custom-value',
|
||||
});
|
||||
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options') return {};
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(MockedChatOpenAI).toHaveBeenCalledWith(
|
||||
expect.objectContaining({
|
||||
apiKey: 'test-api-key',
|
||||
model: 'gpt-4o-mini',
|
||||
timeout: 60000,
|
||||
maxRetries: 2,
|
||||
configuration: {
|
||||
defaultHeaders: {
|
||||
'X-Custom-Header': 'custom-value',
|
||||
},
|
||||
},
|
||||
callbacks: expect.arrayContaining([expect.any(Object)]),
|
||||
modelKwargs: {},
|
||||
onFailedAttempt: expect.any(Function),
|
||||
}),
|
||||
);
|
||||
});
|
||||
|
||||
it('should handle all available options', async () => {
|
||||
const mockContext = setupMockContext();
|
||||
const options = {
|
||||
frequencyPenalty: 0.5,
|
||||
maxTokens: 1000,
|
||||
presencePenalty: 0.3,
|
||||
temperature: 0.8,
|
||||
topP: 0.9,
|
||||
timeout: 45000,
|
||||
maxRetries: 3,
|
||||
responseFormat: 'json_object' as const,
|
||||
reasoningEffort: 'high' as const,
|
||||
};
|
||||
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options') return options;
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(MockedChatOpenAI).toHaveBeenCalledWith(
|
||||
expect.objectContaining({
|
||||
apiKey: 'test-api-key',
|
||||
model: 'gpt-4o-mini',
|
||||
frequencyPenalty: 0.5,
|
||||
maxTokens: 1000,
|
||||
presencePenalty: 0.3,
|
||||
temperature: 0.8,
|
||||
topP: 0.9,
|
||||
timeout: 45000,
|
||||
maxRetries: 3,
|
||||
responseFormat: 'json_object',
|
||||
reasoningEffort: 'high',
|
||||
configuration: {},
|
||||
callbacks: expect.arrayContaining([expect.any(Object)]),
|
||||
modelKwargs: {
|
||||
response_format: { type: 'json_object' },
|
||||
reasoning_effort: 'high',
|
||||
},
|
||||
onFailedAttempt: expect.any(Function),
|
||||
}),
|
||||
);
|
||||
});
|
||||
|
||||
it('should only add valid reasoning effort to modelKwargs', async () => {
|
||||
const mockContext = setupMockContext();
|
||||
const options = {
|
||||
reasoningEffort: 'invalid' as 'low' | 'medium' | 'high',
|
||||
};
|
||||
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options') return options;
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(MockedChatOpenAI).toHaveBeenCalledWith(
|
||||
expect.objectContaining({
|
||||
apiKey: 'test-api-key',
|
||||
model: 'gpt-4o-mini',
|
||||
reasoningEffort: 'invalid',
|
||||
timeout: 60000,
|
||||
maxRetries: 2,
|
||||
configuration: {},
|
||||
callbacks: expect.arrayContaining([expect.any(Object)]),
|
||||
modelKwargs: {}, // Should not include invalid reasoning_effort
|
||||
onFailedAttempt: expect.any(Function),
|
||||
}),
|
||||
);
|
||||
});
|
||||
|
||||
it('should create N8nLlmTracing callback', async () => {
|
||||
const mockContext = setupMockContext();
|
||||
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options') return {};
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(MockedN8nLlmTracing).toHaveBeenCalledWith(mockContext);
|
||||
});
|
||||
|
||||
it('should create failed attempt handler', async () => {
|
||||
const mockContext = setupMockContext();
|
||||
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options') return {};
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(mockedMakeN8nLlmFailedAttemptHandler).toHaveBeenCalledWith(
|
||||
mockContext,
|
||||
expect.any(Function), // openAiFailedAttemptHandler
|
||||
);
|
||||
});
|
||||
|
||||
it('should use default values for timeout and maxRetries when not provided', async () => {
|
||||
const mockContext = setupMockContext();
|
||||
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options') return {};
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(MockedChatOpenAI).toHaveBeenCalledWith(
|
||||
expect.objectContaining({
|
||||
timeout: 60000,
|
||||
maxRetries: 2,
|
||||
}),
|
||||
);
|
||||
});
|
||||
|
||||
it('should prioritize options.baseURL over credentials.url', async () => {
|
||||
const optionsBaseURL = 'https://options-api.example.com/v1';
|
||||
const credentialsURL = 'https://credentials-api.example.com/v1';
|
||||
const mockContext = setupMockContext();
|
||||
|
||||
mockContext.getCredentials.mockResolvedValue({
|
||||
apiKey: 'test-api-key',
|
||||
url: credentialsURL,
|
||||
});
|
||||
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options')
|
||||
return {
|
||||
baseURL: optionsBaseURL,
|
||||
};
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(MockedChatOpenAI).toHaveBeenCalledWith(
|
||||
expect.objectContaining({
|
||||
configuration: {
|
||||
baseURL: optionsBaseURL,
|
||||
fetchOptions: {
|
||||
dispatcher: {},
|
||||
},
|
||||
},
|
||||
}),
|
||||
);
|
||||
});
|
||||
|
||||
it('should handle text response format correctly', async () => {
|
||||
const mockContext = setupMockContext();
|
||||
const options = {
|
||||
responseFormat: 'text' as const,
|
||||
};
|
||||
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options') return options;
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(MockedChatOpenAI).toHaveBeenCalledWith(
|
||||
expect.objectContaining({
|
||||
responseFormat: 'text',
|
||||
modelKwargs: {
|
||||
response_format: { type: 'text' },
|
||||
},
|
||||
}),
|
||||
);
|
||||
});
|
||||
|
||||
it('should handle all reasoning effort values correctly', async () => {
|
||||
const reasoningEffortValues = ['low', 'medium', 'high'] as const;
|
||||
|
||||
for (const effort of reasoningEffortValues) {
|
||||
const mockContext = setupMockContext();
|
||||
|
||||
mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
|
||||
if (paramName === 'model.value') return 'gpt-4o-mini';
|
||||
if (paramName === 'options')
|
||||
return {
|
||||
reasoningEffort: effort,
|
||||
};
|
||||
return undefined;
|
||||
});
|
||||
|
||||
await lmChatOpenAi.supplyData.call(mockContext, 0);
|
||||
|
||||
expect(MockedChatOpenAI).toHaveBeenCalledWith(
|
||||
expect.objectContaining({
|
||||
reasoningEffort: effort,
|
||||
modelKwargs: {
|
||||
reasoning_effort: effort,
|
||||
},
|
||||
}),
|
||||
);
|
||||
|
||||
jest.clearAllMocks();
|
||||
}
|
||||
});
|
||||
});
|
||||
|
||||
describe('methods', () => {
|
||||
beforeEach(() => {
|
||||
setupMockContext();
|
||||
});
|
||||
|
||||
it('should have searchModels method', () => {
|
||||
expect(lmChatOpenAi.methods).toEqual({
|
||||
listSearch: {
|
||||
searchModels: expect.any(Function),
|
||||
},
|
||||
});
|
||||
});
|
||||
});
|
||||
});
|
||||
@@ -18,16 +18,28 @@ export async function apiRequest(
|
||||
endpoint: string,
|
||||
parameters?: RequestParameters,
|
||||
) {
|
||||
const { body, qs, option, headers } = parameters ?? {};
|
||||
const { body, qs, option } = parameters ?? {};
|
||||
|
||||
const credentials = await this.getCredentials('openAiApi');
|
||||
|
||||
let uri = `https://api.openai.com/v1${endpoint}`;
|
||||
|
||||
let headers = parameters?.headers ?? {};
|
||||
if (credentials.url) {
|
||||
uri = `${credentials?.url}${endpoint}`;
|
||||
}
|
||||
|
||||
if (
|
||||
credentials.header &&
|
||||
typeof credentials.headerName === 'string' &&
|
||||
credentials.headerName &&
|
||||
typeof credentials.headerValue === 'string'
|
||||
) {
|
||||
headers = {
|
||||
...headers,
|
||||
[credentials.headerName]: credentials.headerValue,
|
||||
};
|
||||
}
|
||||
|
||||
const options = {
|
||||
headers,
|
||||
method,
|
||||
|
||||
Reference in New Issue
Block a user