Get an s3 object and stream it to disk
Switch branches/tags
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
test
.gitignore
History.md
LICENSE
README.md
gulpfile.js
index.js
package.json

README.md

lambduh-s3-get-object

Download an object from S3.

Install

npm i --save lambduh-get-s3-object

Usage

var Q = require('q');
var download = require('lambduh-get-s3-object');

//your lambda function
exports.handler = function(event, context) {

  download(null, {
    srcBucket: "source-bucket"
    srcKey: "path/to/s3/key.txt"
    downloadFilepath: "/tmp/path/to/local/file.txt"
  })
  .then(function(result) {
  	context.done();
  })
  .fail(function(err) {
  	context.done(err);
  });

}

This module expects three fields on the passed options object: .srcBucket, .srcKey, and .downloadFilepath

By default this will use the default region the lambda operates in. If you need to operate on an S3 bucket in another region you can set the region field on the options object: .region.

It will download an object to the specified filepath from the specifed S3 bucket, key and region (if specified).

Note: for whatever reason, this func is resolving before the stream is completely finished. In practice, I'm solving this with a 500ms timeout. Which sucks, but has been good enough for now.

Full disclosure

This module's tests don't yet cover the aws-sdk implementation - only the validation and other basic things.

I'm hoping to get back to this soon.......PRs welcome....