s3sync.rb (​http://www.s3sync.net/) is a cross-platform Ruby program that easily transfers directories between a local directory and an Amazon S3 bucket:prefix. I greatly recommend s3sync for any big S3 synchronization tasks.

Cyberduck (tested version 3.7 / 7380) does not yet support S3 directories created by s3sync, even though they are correctly displayed as directories in e.g. S3Fox (S3 Firefox plugin). The developer of s3sync quotes in his README:

In S3 there's no actual concept of folders, just keys and nodes. So, every tool
uses its own proprietary way of storing dir info (my scheme being the best
naturally) and in general the methods are not compatible.

s3sync seems to store directories as dummy nodes that all contain the same directory string "{E40327BF-517A-46e8-A6C3-AF51BC263F59}". See code snippets of s3sync.rb:

	$S3syncDirString = '{E40327BF-517A-46e8-A6C3-AF51BC263F59}'
	$S3syncDirTag = 'd66759af42f282e1ba19144df2d405d0'
	$S3syncDirFile = Tempfile.new("s3sync")
	$S3syncDirFile.puts $S3syncDirString
	$S3syncDirFile.close # not final; we need this file again to 'put' directory nodes

...

		def directory?()
			@tag == $S3syncDirTag and @size == $S3syncDirString.length

...

				# all s3 directories are dummy nodes contain the same directory string
				# so for easy comparison, set our size and tag thusly
				@size = $S3syncDirString.length
				@tag = $S3syncDirTag

It would be great if Cyberduck would support those directories as well in the future.

thanks! Regards, Philip