Easily deploy to S3 with multiple environment support. Version 2 now supports multiprocessing for faster uploads.

Installation & Usage

To install, run pip install d3ploy. To use, run d3ploy. Additional arguments may be specified. Run d3ploy --help for more information.


Your AWS credentials can be set in a number of ways:

  1. In a ”.boto” file in your home folder. See Boto’s documentation for how to create this file.
  2. In a ”.aws” file in the folder you’re running d3ploy in. Follows the same format as ”.boto”.
  3. In the environment variables “AWS_ACCESS_KEY_ID” and “AWS_SECRET_ACCESS_KEY”.
  4. Passed in as arguments. -a or --access-key for the Access Key ID and -s or --access-secret for the Secret Access Key.
  5. In the per-enviroment configuration outlined below.

Configuration options

When you run d3ploy, it will look in the current directory for a “deploy.json” file that defines the different deploy enviroments and their options. At a minimum, a “default” environment is required and is the environment used if you pass no arguments to d3ploy. Additionally, you may pass in a different path for you config file with the -c or --config options.

To supress all output, pass -q or --quiet to the command. Note that there is not a way to set the quiet option in the config file(s).

To set the number of separate processes to use, pass -p 10 or --processess 10 where ‘10’ is the number to use. If you do not want to use multiple processes, set this to ‘0’.

You can add as many environments as needed. Deploy to an environment by passing in its key like d3ploy staging. Environments besides “default” will inherit any settings not explicitly set from the default configuration.

The only required option for any environment is “bucket” for the S3 bucket to upload to. Additionally, you may define:

  • “local_path” to upload only the contents of a directory under the current one; defaults to ”.” (current directory)
  • “bucket_path” to upload to a subfolder in the bucket; defaults to “/” (root)
  • “aws_key” to specify the AWS Access Key ID to use for uploading
  • “aws_secret” to specify the AWS Secret Access Key to use for uploading
  • “exclude” to specify patterns to not upload
  • “gzip” to automatically gzip files before uploading to S3
  • “gzip_skip” to specify mimetypes to not gzip when gzip is set to true
  • “delete” to remove files on S3 that are not present in the local directory
  • “charset” to set the charset flag on ‘Content-Type’ headers of text files
  • “cache” to set the Cache-Control header for various mimetypes. See below for more.
  • “gitignore” to add all entries in a .gitignore file to the exclude patterns
  • “cloudfront” to invalidate all paths in the given CloudFront distribution ID

Cache-Control Headers

If you want to set Cache-Control headers on various files, add a cache object to your config file like:

"cache": {
  "text/css": 2592000,
  "application/javascript": 2592000,
  "image/png": 22896000,
  "image/jpeg": 22896000,
  "image/webp": 22896000,
  "image/gif": 22896000

Each key is the mimetype of the kind of file you want to have cached, with a value that is the seconds the max-age flag set to. In the above example, CSS and JavaScript files will be cached for 30 days while images will be cached for 1 year. For more about Cache-Control, read Leverage Browser Caching.

OS X Notification Center

d3ploy will attempt to alert you via Notification Center when it is completed. To enable this feature, you need pyobjc; run pip install pyobjc to install.

Progress Bar

d3ploy will use the progressbar module if it’s available to display output. This includes a percentage completed and an ETA. To enable, run pip install progressbar.

Caution About Using the gzip Option

Almost all modern browsers will support files that are served with gzip compression. The notable exception is non-smartphone mobile browsers. If you have significant traffic over those browsers, it is advisable to avoid the gzip option.