mirror of
https://github.com/langgenius/dify.git
synced 2026-04-28 06:28:05 +08:00
78ab09b521
Fix critical bugs in response body parsing: 1. Error responses (non-OK status): - Response body can only be read once - Previous code: response.json() fails → response.text() throws 'Body already read' - Fix: Read as text first, then parse JSON with fallback to raw text 2. Success responses (OK status): - Malformed JSON was not caught, causing NetworkError misclassification - Fix: Wrap JSON.parse in try-catch, fallback to raw text These fixes prevent unhandled promise rejections and provide better error messages when servers return malformed JSON responses. Reported-by: gemini-code-assist Severity: Critical (issue 1), High (issue 2)
Dify Node.js SDK
This is the Node.js SDK for the Dify API, which allows you to easily integrate Dify into your Node.js applications.
Install
npm install dify-client
Usage
After installing the SDK, you can use it in your project like this:
import {
DifyClient,
ChatClient,
CompletionClient,
WorkflowClient,
KnowledgeBaseClient,
WorkspaceClient
} from 'dify-client'
const API_KEY = 'your-app-api-key'
const DATASET_API_KEY = 'your-dataset-api-key'
const user = 'random-user-id'
const query = 'Please tell me a short story in 10 words or less.'
const chatClient = new ChatClient(API_KEY)
const completionClient = new CompletionClient(API_KEY)
const workflowClient = new WorkflowClient(API_KEY)
const kbClient = new KnowledgeBaseClient(DATASET_API_KEY)
const workspaceClient = new WorkspaceClient(DATASET_API_KEY)
const client = new DifyClient(API_KEY)
// App core
await client.getApplicationParameters(user)
await client.messageFeedback('message-id', 'like', user)
// Completion (blocking)
await completionClient.createCompletionMessage({
inputs: { query },
user,
response_mode: 'blocking'
})
// Chat (streaming)
const stream = await chatClient.createChatMessage({
inputs: {},
query,
user,
response_mode: 'streaming'
})
for await (const event of stream) {
console.log(event.event, event.data)
}
// Chatflow (advanced chat via workflow_id)
await chatClient.createChatMessage({
inputs: {},
query,
user,
workflow_id: 'workflow-id',
response_mode: 'blocking'
})
// Workflow run (blocking or streaming)
await workflowClient.run({
inputs: { query },
user,
response_mode: 'blocking'
})
// Knowledge base (dataset token required)
await kbClient.listDatasets({ page: 1, limit: 20 })
await kbClient.createDataset({ name: 'KB', indexing_technique: 'economy' })
// RAG pipeline (may require service API route registration)
const pipelineStream = await kbClient.runPipeline('dataset-id', {
inputs: {},
datasource_type: 'online_document',
datasource_info_list: [],
start_node_id: 'start-node-id',
is_published: true,
response_mode: 'streaming'
})
for await (const event of pipelineStream) {
console.log(event.data)
}
// Workspace models (dataset token required)
await workspaceClient.getModelsByType('text-embedding')
Notes:
- App endpoints use an app API token; knowledge base and workspace endpoints use a dataset API token.
- Chat/completion require a stable
useridentifier in the request payload. - For streaming responses, iterate the returned AsyncIterable. Use
stream.toText()to collect text.
License
This SDK is released under the MIT License.