Download or getObject feature

[levithomason]

We're storing css regression reference images (phantomjs screenshots) on S3. I'm using gulp-awspublish to publish them. When running tests, we need to download the reference images first, to compare our test images against the references.

We are currently using vinyl-s3 to download from S3. This allows you to use S3 as a source or destination of vinyl files, replacing gulp.src and gulp.dest in your task. This is really powerful, but does not take advantage of the other great things about gulp-awspublish, like caching, etag comparison, and downstream plugin capabilities.

What are your thoughts on adding some kind of download, getObject, etc. method to accommodate this?