Using Multiple Buckets with the AWS Amplify Storage Module

I’m currently working with React and AWS Amplify to create a front end for a web application. The app allows users to upload images, which are subsequently resized with a lambda function (post to follow) and placed into a new S3 bucket. This means the app needs to be able to upload to one bucket and get images from another. Weirdly the AWS Amplify documentation covers setup for a single bucket using global configuration. There is an open Github Issue on the Amplify project, but I had a hunt around and found a quick solution already in the code.

AWS Amplify is a Javascript library that provides easier ways to interact with AWS services often used by app developers, such as API Gateway and S3.

The documentation for the Storage module, that interacts with S3, suggests the following code for configuring a bucket to use.

Amplify.configure({
    Auth: {
        identityPoolId: 'XX-XXXX-X:XXXXXXXX-XXXX-1234-abcd-1234567890ab', //REQUIRED - Amazon Cognito Identity Pool ID
        region: 'XX-XXXX-X', // REQUIRED - Amazon Cognito Region
        userPoolId: 'XX-XXXX-X_abcd1234', //OPTIONAL - Amazon Cognito User Pool ID
        userPoolWebClientId: 'XX-XXXX-X_abcd1234', //OPTIONAL - Amazon Cognito Web Client ID
    },
    Storage: {
        bucket: '', //REQUIRED -  Amazon S3 bucket
        region: 'XX-XXXX-X', //OPTIONAL -  Amazon service region
    }
});

As you can see bucket is mandatory. When carrying out some of the operations detailed in the documentation such as puts and gets, this is the bucket that gets used.

However, there are a few more options available for each operation than the documentation outlines.

For example under the Get section of the documentation, the expires option is explained.

Storage.get('test.txt', {expires: 60})
    .then(result => console.log(result))
    .catch(err => console.log(err));

The expires keywork is contained in the Get call. Other than fooling around with React in an attempt to throw together a front end, I am not a Javascript developer. But the expires keyword here seems to work similarly to kwargs in Python.

Diving into the code and looking at the Get and Put operations I found this snippet in the Storage code.

public async get(key: string, options?): Promise<Object> {
    ...
    const opt = Object.assign({}, this._options, options);
    const { bucket, region, credentials, level, download, track, expires } = opt;
    ...
    const params: any = {
            Bucket: bucket,
            Key: final_key
    };

The options here expose an optional bucket value. So I had a go at setting that in my code and it works!

Storage.vault.get(this.props.image_key, 
    {
        bucket: 'alternative-bucket-name'
    }
);

So as it happens it is possible to use multiple S3 buckets with AWS Amplify. Hopefully the documentation will be updated soon to reflect this. If I get time I may even make a pull request myself.