s3-proxy-middleware

1.1.4 • Public • Published

s3-proxy

Build Status Test Coverage

S3 proxy middleware for returning S3 objects Express apps. Useful for streaming media files and data files from S3 without having to configure web hosting on the entire origin bucket. You can explicitly override the cache headers of the underlying S3 objects.

Added an option to remove bucket name from url

Usage

import express from 'express';
import s3Proxy from 's3-proxy';
 
const app = express();
app.get(
    '/media/*',
    s3Proxy({
        bucket: 'bucket_name',
        prefix: 'optional_s3_path_prefix',
        accessKeyId: 'aws_access_key_id',
        secretAccessKey: 'aws_secret_access_key',
        overrideCacheControl: 'max-age=100000'
    })
);

Options

accessKeyId

The AWS access key of the IAM user to connect to S3 with (environment variable recommended).

secretAccessKey

The AWS secret access key (environment variable recommended).

region

The AWS region of the bucket, i.e. "us-west-2".

bucket

The name of the S3 bucket.

prefix

Optional path to the root S3 folder where the files to be hosted live. If omitted, the http requests to the proxy need to mirror the full S3 path.

defaultCacheControl

Value of the Cache-Control header to use if the metadata from the S3 object does not specify it's own value.

overrideCacheControl

Value of the Cache-Control header that is applied to the response even if there there is a different value on the S3 object metadata.

removeBucketFromObjectUrl

Removes the bucket name from object url. /bucket/bucket

HTTP Cache Headers

The s3-proxy provides two different caching mechanisms. First you can specify either the defaultCacheControl or overrideCacheControl options to control the Cache-Control header that is sent in the proxied response. The most optimal policy is to specify a max-age=seconds value that informs the browser and any intermediary CDN and network proxies to cache the response for the specified number of seconds and not return to the origin server until that time has elapsed.

Secondly it supports the ETag value that S3 automatically creates whenever an object is written. The proxy forwards this header along in the http response. If the value of an incoming If-None-Match request header matches the ETag of the S3 object, the proxy returns an empty 304 Not Modified response. This is known as a "conditional get" request.

For a more in-depth description of the different caching headers and techniques, see the Google Developer HTTP caching documentation.

Example

Let's assume there is a bucket "mycompany-media-assets". Within this bucket is a folder named "website" where the images, videos, etc. for the company website reside.

mycompany-media-assets
└── website
    └── images
        ├── logo.png
        └── background.jpg

The corresponding s3-proxy route definition would look something like below. The Cache-Control response header will be set to have a max age of 30 days (2592000 seconds) no matter what metadata exists on the corresponding S3 object. This means whatever tool is being used to write the files to S3 need not worry about configuring proper cache metadata, the proxy will take care of that.

app.get(
    '/media/*',
    s3Proxy({
        bucket: 'mycompany-media-assets',
        prefix: 'website',
        accessKeyId: 'aws_access_key_id',
        secretAccessKey: 'aws_secret_access_key',
        overrideCacheControl: 'max-age=2592000'
    })
);

Now images can be declared in views like so:

<img src="/media/images/logo.png" />

Listing objects

It's also possible to return a JSON listing of all the keys by making a request ending with a trailing slash. For the sample above, issuing a request to /media/images/ will return: ['logo.png', 'background.jpg'].

License

Licensed under the Apache License, Version 2.0.

Package Sidebar

Install

npm i s3-proxy-middleware

Weekly Downloads

21

Version

1.1.4

License

Apache-2.0

Unpacked Size

12.6 kB

Total Files

4

Last publish

Collaborators

  • ionis_