A synchronisation describes a task: what is reconciled from where to where. It is created with File → New synchronisation… and changed with Edit → Edit synchronisation… or a double click on the entry.
The name identifies the synchronisation and must be unique. It appears in the overview, in the log and – if switched on – as the file name of the backup proof.
| Type | What is reconciled |
|---|---|
| File | Exactly one file. Source and target are complete file names. |
| Directory | All files in one directory, without its subdirectories. |
| Directory tree | A directory with all its subdirectories. Missing target directories are created. |
Both fields take a path. The button next to each opens a chooser; for a computer at the other end of an SSH connection there is a button of its own, see Synchronising over SSH.
A target that does not exist yet is created on the first run. If the source is on a removable drive or a network drive that is not connected at the moment, AutoFileSync skips the synchronisation and leaves the target alone – this matters, so that an unplugged drive is never mistaken for “everything was deleted”.
For the types Directory and Directory tree, the Wildcard field limits which files are looked at at all, for example *.txt for text files or *.* for all of them. Files that do not match are invisible to this synchronisation: they are neither copied nor deleted, not even when deleted files are also deleted in the target.
The Exclusions tab holds one pattern per line, for example:
*.tmp
node_modules
Thumbs.db
Exclusions apply to copying as well as to deleting. In a directory tree, an excluded directory is not even entered – this saves time and spares checking the files inside it one by one.
A synchronisation without the Synchronisation active tick stays in the list but is not run. This way a task can be put aside for a while without losing it.
On to the options.