Renaming Files

You may rename or move files, folders, and symbolic links on the server to a new pathname provided that you have sufficient permissions.

To change the path of a remote object, use the "–m" option as shown below:

movedat -m [user@]host:path newpath

All directories in the new path must exist, unless -B BuildPath is specified.


To change the name of file "foo" to "bar" within the folder "subdir":

movedat -m subdir/bar

To move the file "bar" from folder "subdir" to folder "otherdir":

movedat -m otherdir/bar

To move the file "bar" from the home folder "otherdir" to the root path (assuming the user has permission to do so):

movedat -m /bar

To rename the folder "subdir" to "subfolder":

movedat -m subfolder


The destination path must always include the intended name of the file.  If you only specify a destination directory, the operation may fail with an "Is directory" or similar error.

movedat -m subdir/ Failed: Remote Application: Invalid Pathname Is a directory

That behavior may vary depending on the server operating system, so specifying the full destination path is always recommended.

Case sensitivity varies by operating system and file system.  Changing only the capitalization of a file name may cause unexpected results.

Renaming a symbolic link will change the name or path of the link itself, but will not change the target path.  As always, moving a symbolic link which has a relative target path may break the link.

Object Handlers

Object handlers may implement the rename function differently, or not at all.  For example, as of this writing, the CloudDat gateway for Amazon S3 does not require that all directories in the path exist, even without -B BuildPath, because directories in S3 do not exist.