ultralight-s3

🪽 ultralight-s3

~15KB lightweight S3 client with zero dependencies, designed for Node.js, edge computing like Cloudflare workers, AWS Lambda (and browsers - not implemented yet).

Dev:

Release:

Features

• 🚀 Lightweight: Only ~16KB minified
• 🔧 Zero dependencies
• 💻 Works on NodeJS, Cloudflare workers, ideal for edge computing (browser support - not implemented yet)
• 🔑 Supports essential S3 APIs (list, put, get, delete and a few more)
• 🔁 Streaming support & multipart uploads for large files
• 📦 Bring your own S3 bucket

Table of Contents

• Features
• Usage & examples
• Installation
• Configuration
  • Minio ✅
  • Cloudflare R2 ✅
  • AWS Lambda
  • Google Cloud
  • Azure
  • Ceph
  • Others
• Public functions
• API
• Community
• License

Simple usage

import { S3 } from 'ultralight-s3';

// ... your configuration
const s3 = new S3({
  accessKeyId: 'your-access-key-id',
  secretAccessKey: 'your-secret-access-key',
  region: 'auto',
  bucket: 'your-bucket-name',
});

// List objects
const objects = await s3.list();
// or with prefix
// const specificObjectsUnderPrefix = await s3.list('/', 'prefix');
console.log(objects);

// Check if a file exists
const exists = await s3.fileExists('path/to/file.txt');

// Get a life
const data = await s3.get('path/to/life.txt');
console.log(data);

// get a stream of a large file (first chunk)
const firstChunk = await s3.getResponse('path/to/large-file.mp4', false, 0).body;

// get a stream of a large file (all chunks)
const allChunks = await s3.getResponse('path/to/large-file.mp4', true);
for await (const chunk of allChunks.body) {
  console.log(chunk);
}

// Upload a large file
// by default is 5MB per request (minimum for AWS S3 is 5MB)
const chunkSize = s3.getMaxRequestSizeInBytes();
// Initiate multipart upload
const uploadId = await s3.getMultipartUploadId('randomFileName.fastaq', 'text/plain');
const buffer = randomBytes(chunkSize * 2 + 1024); // Just over 2 parts
const upload = await s3.uploadPart('randomFileName.fastaq', buffer, uploadId, 1);
const upload2 = await s3.uploadPart('randomFileName.fastaq', buffer, uploadId, 2);
const upload3 = await s3.uploadPart('randomFileName.fastaq', buffer, uploadId, 3);

// Complete multipart upload
const result = await s3.completeMultipartUpload('randomFileName.fastaq', uploadId, [
  { partNumber: 1, ETag: upload.ETag },
  { partNumber: 2, ETag: upload2.ETag },
  { partNumber: 3, ETag: upload3.ETag },
]);
console.log(result);

// Get file size
const size = await s3.getContentLength('path/to/file.txt');
console.log(size);

// Put a file
await s3.put('path/to/file.txt', Buffer.from('Hello, World!'));

// Delete a file
await s3.delete('path/to/file.txt');

For some examples, check the dev directory and try to use it with Hono or Cloudflare Workers.

Installation

npm install ultralight-s3

# or

yarn add ultralight-s3

# or

pnpm add ultralight-s3

# or
# Not yet implemented
# <script src="https://unpkg.com/ultralight-s3/dist/ultralight-s3.min.js" defer></script>

Configuration

Minio (✅ tested)

import { S3 } from 'ultralight-s3';

const s3 = new S3({
  accessKeyId: 'your-access-key-id',
  secretAccessKey: 'your-secret-access-key',
  endpoint: 'https://your-s3-endpoint.com' || 'http://127.0.0.1:9000',
  bucketName: 'your-bucket-name',
  region: 'auto', //optional -  by default is auto
  maxRequestSizeInBytes: 5242880, // optional - by default is 5MB
  requestAbortTimeout: undefined, // optional - for aborting requests
  logger: console, // optional - for debugging
});

Cloudflare R2 (✅ tested)

import { S3 } from 'ultralight-s3';

const s3 = new S3({
  accessKeyId: 'your-access-key-id',
  secretAccessKey: 'your-secret-access-key',
  endpoint: 'https://your-clouflare-id.r2.cloudflarestorage.com/your-bucket-name',
  bucketName: 'your-bucket-name',
  region: 'auto', //optional -  by default is auto
  maxRequestSizeInBytes: 5242880, // optional - by default is 5MB
  requestAbortTimeout: undefined, // optional - for aborting requests
  logger: console, // optional - for debugging
});

Others

(AWS Lambda, Azure, Google Cloud, Ceph, etc)

Not tested, but should work with other S3 compatible services. Full list - soon to come. PRs are welcome.

Public functions

Bucket Operations

• bucketExists(): Check if a bucket exists.

Object Operations

• get(key, opts): Get an object from the bucket
• getObjectWithETag(key, opts): Get an object and its ETag from the bucket
• getEtag(key, opts): Get the ETag of an object
• getResponse(key, wholeFile, rangeFrom, rangeTo, opts): Get a response of an object from the bucket
• put(key, data): Put an object into the bucket
• delete(key): Delete an object from the bucket
• getContentLength(key): Get the content length of an object
• fileExists(key): Check if an object exists in the bucket

Listing Operations

• list(delimiter, prefix, maxKeys, method, opts): List objects in the bucket
• listMultiPartUploads(delimiter, prefix, method, opts): List multipart uploads in the bucket

Multipart Upload Operations

• getMultipartUploadId(key, fileType): Initiate a multipart upload
• uploadPart(key, data, uploadId, partNumber, opts): Upload a part in a multipart upload
• completeMultipartUpload(key, uploadId, parts): Complete a multipart upload
• abortMultipartUpload(key, uploadId): Abort a multipart upload

Configuration and Utility Methods

• getBucketName(): Get the current bucket name
• setBucketName(bucketName): Set the bucket name
• getRegion(): Get the current region
• setRegion(region): Set the region
• getEndpoint(): Get the current endpoint
• setEndpoint(endpoint): Set the endpoint
• getMaxRequestSizeInBytes(): Get the maximum request size in bytes
• setMaxRequestSizeInBytes(maxRequestSizeInBytes): Set the maximum request size in bytes
• sanitizeETag(etag): Sanitize an ETag string
• getProps(): Get all configuration properties
• setProps(props): Set all configuration properties

API

new S3(config: Object)

• Input: A configuration object with the following properties:
  • accessKeyId: string: The access key ID for authentication.
  • secretAccessKey: string: The secret access key for authentication.
  • endpoint: string: The endpoint URL of the S3-compatible service.
  • bucketName: string: The name of the bucket to operate on.
  • region?: string (optional): The region of the S3 service (default: 'auto').
  • maxRequestSizeInBytes?: number (optional): The maximum size of a single request in bytes (minimum 5MB).
  • requestAbortTimeout?: number (optional): The timeout in milliseconds after which a request should be aborted.
  • logger?: Object (optional): A logger object with methods like info, warn, error.
• Behavior: Creates a new instance of the S3 class with the provided configuration.
• Returns: S3: An instance of the S3 class.

list(delimiter?: string, prefix?: string, maxKeys?: number, method?: string, opts?: Object): Promise<Array<Object>

• Input:
  • delimiter?: string (optional): The delimiter to use for grouping objects in specific path (default: '/').
  • prefix?: string (optional): The prefix to filter objects in specific path (default: '').
  • maxKeys?: number (optional): The maximum number of keys to return (default: 1000).
  • method?: string (optional): The HTTP method to use (default: 'GET').
  • opts?: Object (optional): Additional options for the list operation.
• Behavior: Lists objects in the bucket, supporting pagination and filtering.
• Returns: Promise<Array<Object>: A promise that resolves to an array of objects or object metadata.

put(key: string, data: Buffer | string): Promise<Object>

• Input:
  • key: string: The key of the object to put.
  • data: Buffer | string: The content of the object.
• Behavior: Uploads an object to the bucket.
• Returns: Promise<Object>: A promise that resolves to the response from the put operation.

get(key: string, opts?: Object): Promise

• Input:
  • key: string: The key of the object to get.
  • opts?: Object (optional): Additional options for the get operation.
• Behavior: Retrieves an object from the bucket.
• Returns: Promise<string | null>: A promise that resolves to the content of the object or null if the object doesn't exist or there's an ETag mismatch.

getObjectWithETag(key: string, opts?: Object): Promise<{ etag: string | null; data: string | null }>

• Input:
  • key: string: The key of the object to get.
  • opts?: Object (optional): Additional options for the get operation.
• Behavior: Retrieves an object and its ETag from the bucket.
• Returns: Promise<{ etag: string | null; data: string | null }>: A promise that resolves to an object containing the ETag and content of the object, or null values if the object doesn't exist or there's an ETag mismatch.

getResponse(key: string, wholeFile?: boolean, rangeFrom?: number, rangeTo?: number, opts?: Object): Promise<Response>

• Input:
  • key: string: The key of the object to get.
  • wholeFile?: boolean (optional): Whether to get the whole file or a part (default: true).
  • rangeFrom?: number (optional): The byte range from to get if not getting the whole file (default: 0).
  • rangeTo?: number (optional): The byte range to to get if not getting the whole file (default: maxRequestSizeInBytes). Note: rangeTo is inclusive.
  • opts?: Object (optional): Additional options for the get operation.
• Behavior: Retrieves a response of an object from the bucket.
• Returns: Promise<Response>: A promise that resolves to a Response of the object content. Use readableStream() to get the stream from .body.

getEtag(key: string, opts?: Object): Promise<string | null>

• Input:
  • key: string: The key of the object to get the ETag for.
  • opts?: Object (optional): Additional options for the operation.
• Behavior: Retrieves the ETag of an object from the bucket.
• Returns: Promise<string | null>: A promise that resolves to the ETag of the object or null if the ETag doesn't match the provided conditions.

delete(key: string): Promise

• Input: key: string: The key of the object to delete.
• Behavior: Deletes an object from the bucket.
• Returns: Promise<bollean>: A promise that resolves to true if the delete operation was successful, false otherwise. ⚠️ Note: Delete will NOT return false if the object does not exist. Use fileExists() to check if an object exists before/after deleting it.

fileExists(key: string): Promise

• Input: key: string: The key of the object to check.
• Behavior: Checks if an object exists in the bucket.
• Returns: Promise: A promise that resolves to a value indicating the existence status: false (not found), true (found), or null (ETag mismatch).

getContentLength(key: string): Promise<number>

• Input: key: string: The key of the object.
• Behavior: Gets the content length of an object.
• Returns: Promise<number>: A promise that resolves to the content length of the object in bytes.

listMultiPartUploads(delimiter?: string, prefix?: string, method?: string, opts?: Object): Promise<Array<Object>

• Input:
  • delimiter?: string (optional): The delimiter to use for grouping objects in specific path (default: '/').
  • prefix?: string (optional): The prefix to filter objects in specific path (default: '').
  • method?: string (optional): The HTTP method to use (default: 'GET').
  • opts?: Object (optional): Additional options for the list operation.
• Behavior: Lists multipart uploads in the bucket.
• Returns: Promise<Array<Object>: A promise that resolves to an array of multipart uploads or multipart upload metadata.

getMultipartUploadId(key: string, fileType?: string): Promise

• Input:
  • key: string: The key of the object to upload.
  • fileType?: string (optional): The MIME type of the file (default: 'application/octet-stream').
• Behavior: Initiates a multipart upload.
• Returns: Promise<string>: A promise that resolves to the upload ID for the multipart upload.

uploadPart(key: string, data: Buffer | string, uploadId: string, partNumber: number, opts?: Object): Promise<{ partNumber: number, ETag: string }>

• Input:
  • key: string: The key of the object being uploaded.
  • data: Buffer | string: The content of the part.
  • uploadId: string: The upload ID of the multipart upload.
  • partNumber: number: The part number.
  • opts?: Object (optional): Additional options for the upload.
• Behavior: Uploads a part in a multipart upload.
• Returns: Promise<{ partNumber: number, ETag: string }>: A promise that resolves to an object containing the ETag and part number of the uploaded part.

completeMultipartUpload(key: string, uploadId: string, parts: Array<{ partNumber: number, ETag: string }>): Promise<Object>

• Input:
  • key: string: The key of the object being uploaded.
  • uploadId: string: The upload ID of the multipart upload.
  • parts: Array<{ partNumber: number, ETag: string }>: An array of objects containing PartNumber and ETag for each part.
• Behavior: Completes a multipart upload.
• Returns: Promise<Object>: A promise that resolves to the result of the complete multipart upload operation.

abortMultipartUpload(key: string, uploadId: string): Promise<Object>

• Input:
  • key: string: The key of the object being uploaded.
  • uploadId: string: The ID of the multipart upload to abort.
• Behavior: Aborts a multipart upload.
• Returns: Promise<Object>: A promise that resolves to the abort response.

bucketExists(): Promise<boolean>

• Behavior: Checks if the configured bucket exists.
• Returns: Promise: A promise that resolves to a boolean indicating whether the bucket exists.

sanitizeETag(etag: string): string

• Input: etag: string: The ETag to sanitize.
• Behavior: Removes surrounding quotes and HTML entities from the ETag.
• Returns: string: The sanitized ETag.

Also all essential getters and setters for the config object.

Community

Stay connected with the community and get support.

• Issues: Report bugs or request features
• GH Discussions: Ask questions and share ideas
• X/Twitter: @SentienHQ
• Webbsite: sentienhq.com

Contributing

Contributions are welcome! Please open an issue or submit a pull request.

License

This project is licensed under the MIT License - see the LICENSE.md file for details.