Amazon S3 is a web service that can be used to store and retrieve any amount of data which gives developers access to the same highly scalable, reliable, fast, inexpensive data storage infrastructure that Amazon uses to run its own global network of web sites.
Interarchy provides direct access to this service, allowing you to upload or download files just as if you were accessing an FTP or SFTP site. You simply select the Amazon S3 protocol and use your AWS Access Key ID and AWS Secret Access Key just as you would your FTP username and password.
You must create an Amazon Web Services account, and Amazon S3 bandwidth and storage costs will apply.
Interarchy allows you to create buckets and directories, upload or download files, and set permissions (Private, Public Read, Public Read/Write, and Authenticated Read). You can even use a Mirror or Net Disk directly to Amazon S3.
The following are some technical notes on Interarchy’s implementation of Amazon S3 access. For normal use, you probably never need to know any of this, Interarchy makes your Amazon S3 storage look just like nay other file system.
Amazon S3 file paths can be anything at all, there is no requirement for any kind of directory structure. For example, a valid Amazon S3 file might be found at http://s3.amazonaws.com/nolobe//./..//. The Key would be /./..//. However, Interarchy simulates a directory structure based on slash (/) delimited paths. Interarchy does not support such unusual keys.
The top level (root) directory in Amazon S3 is a global name space of “buckets” which is shared by all Amazon S3 users, so you cannot create a top level folder that is not unique to you, nor any files in the root directory. Also bucket names must be 3 to 255 characters long and can contain only ASCII letters, numbers, underscores and dashes. Further, you are limited to 100 buckets. So you should generally create a top level directory named after your name or domain name (eg “nolobe.com”) or containing your Amazon Access Key ID and then do your work in there.
Amazon S3 has no concept of directories, so Interarchy simulates them. If a file like http://s3.amazonaws.com/nolobe/folder/files/file.txt exists, then Interarchy simulates directories for /nolobe, /nolobe/folder and /nolobe/folder/files. If you create a directory in Interarchy, it actually creates a zero length file with a name ending in a slash. For example, if you create directory http://s3.amazonaws.com/nolobe/folder, Interarchy actually creates a zero length file with bucket nolobe and key folder/.
If you create a directory in the root, Interarchy creates a bucket for you, if available. Buckets are world-global, so they disappear almost as fast as .com domain names.
Amazon S3 does not support renaming. Because of this, if you duplicate a folder, Interarchy will prompt you for the new name.
Interarchy supports Amazon S3’s “Canned Access Policies” (Private, Public Read, Public Read/Write, and Authenticated Read). You can configure the permissions for uploaded files using the Permissions Preference Pane, or set them afterwards using the Get Info Window.
