Deploying in 2013, from Grunt to Amazon S3

What is your deploying chain right now? How has it changed over time?

Deploying in 2013

The actual deploy of a single page app I’m involved is based on a single Grunt script, executing on a Node.js environment.

The Grunt chain takes all the files (Backbone.js app, JS libs, HTML templates, CSS and images), checks, copies, concatenates and minificates them.

The resulting dir is versioned and uploaded to an Amazon S3 bucket and to a redundant static cache. We start with 259 files (images excluded) to end up with 13 files, serving 4 different environments. Every final file is available in debug and minified version, requests can be made with http and https protocols.

Previously, all this was made manually, for 4 different servers. Now, only an error free, time saving, automated single task. Thanks, Grunt!

p.s. If only Amazon S3 would serve gzipped files easily …

1 thought on “Deploying in 2013, from Grunt to Amazon S3

  1. avatarBetsy F. Livingston

    I was looking at Amazon.com and I noticed that their static files are versioned at the file level (example: s9-multipack-min.V171170235.js ) instead of the directory level.

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *