TypeScript
Learn how to use the TypeScript SDK for Node and Browsers.
The SDK is mostly generated from the OpenAPI specification. You can read the API documentation here: https://cloud.squidex.io/api/docs.
This means that the API is fully covered, including all endpoints used by the frontend. You can also use the SDK to create and configure Apps, schemas and rules.
Use this SDK in your TypeScript or JavaScript application to integrate with Squidex CMS.
npm install @squidex/squidex --save
or
yarn add @squidex/squidex
The SDK has no peer dependencies and will add/install all required packages.
The
SquidexClient is the main entry point for the SDK. It provides the properties for all endpoints.You can instantiate the client using the following code snippet:
import { SquidexClient } from "@squidex/squidex";
const sdk = new SquidexClient({
appName: "your-app",
clientId: "your-app:default",
clientSecret: "xxx",
environment: "https://cloud.squidex.io"
});
The
environment parameter is optional if you are using Squidex Cloud.In this article, we will cover some of the important configuration values only. Have a look at the source code for all available options: https://github.com/Squidex/sdk-node/blob/main/src/wrapper/SquidexClient.ts#L8
Configure a timeout in milliseconds to cancel unresponsive requests.
const client = new SquidexClient({
...
timeoutInMs: 5000
});
The SDK uses the client ID and the client secret to acquire a bearer token, and handles invalidation and expiration automatically. By default the token is stored inside the client and therefore a new token is acquired when the client is cleaned by the garbage collector. However, one can define where the token is stored.
You can store the token in the local store.
const client = new SquidexClient({
...
tokenStore: new SquidexClient.StorageTokenStore()
});
One can also store the token in the session store using the following configuration:
const client = new SquidexClient({
...
tokenStore: new SquidexClient.StorageTokenStore(sessionStore, 'MyTokenKey')
});
You can control the HTTP request flow with interceptors. They work similar to a middleware in NodeJS. Because the SDK has separate pipelines for normal requests and streaming requests (when downloading a file), you should ideally implement both interceptors.
For example you can add custom headers using the following code snippet:
const client = new SquidexClient({
...
fetcherInterceptor: next => {
return args => {
args.headers ??= {};
args.headers['Key'] = 'Value';
return next(args);
}
},
streamingFetcherInterceptor: next => {
return args => {
args.headers ??= {};
args.headers['Key'] = 'Value';
return next(args);
}
},
});
Use the following code snippet example to implement retry requests:
const client = new SquidexClient({
...
fetcherInterceptor: next => {
return args => {
const NUM_ATTEMPTS = 5;
for (let attempt = 1; attempt <= NUM_ATTEMPTS; attempt++) {
try {
return next(args);
} catch (ex) {
if (attempt === NUM_ATTEMPTS) {
throw ex;
}
}
}
throw new Error('Max attempts reached without error.');
}
}
});
To use the client you have to use the correct property.
For example, to create a schema use the
schemas property.const createdSchema =
await client.schemas.postSchema({
name: "my-schema",
fields: [{
name: "field1",
properties: {
fieldType: "String",
},
}],
isPublished: true,
});
In order to query contents, use the
contents property.const createdContent =
await client.contents.postContent("my-schema", {
body: {
field1: {
iv: "My-Value"
}
}
});
The SDK uses exceptions for error handling.
Refer to the code snippet below for an example:
import { SquidexClient, SquidexError } from "@squidex/squidex";
const client = new SquidexClient({
...
});
try {
await client.contents.postContent("my-schema", { ... });
} catch (ex) {
if (ex instanceof SquidexError && ex.statusCode === 400) {
console.log("Validation Error");
}
}
As of this writing, the SDK has the following limitations:
- 1.Endpoints to download files (e.g. assets) return a stream only and not metadata information like the content type or the content length.
- 2.Endpoints to download files work for Node only and are not available in the browser. This is generally not an issue as assets are handled via direct links in the browser.
- 3.When logging an instance of the
SquidexErrorclass, only the message is written. But the instance also contains the error object from the API that has detailed information about the problem, especially for validation errors. Therefore, it is recommended to convert the error to a JSON string before logging it. - 4.Deprecated methods and properties are not annotated yet.
Last modified 7d ago