2016-2021, Tobias Küster

This is a very simple tool for creating backups. I made this primarily for my personal use and still use it regularly. At first it was just a simple script and a file with the directories I want to back up, but eventually it got a proper UI and some more convenience features, but it is still very basic. For instance, detects whether a backup of a directory is needed at all based on the changed-dates of the files, and can also do simple icremental backups.

• select directories to be backed up
• select zip or tar archive (for files that can not be compressed)
• detect whether a directory needs to be backed up again
• create and collect backup archives with current date

The program is started with python3 backup/main.py. On the first start, a configuration file is created in ~/.config/t-kuester/backup.json holding the different directories to be backed up as well as some more information. The file is automatically read when the program is started and updated when it is closed. The content of the file is explained below.

    {
        "target_pattern": "~/BACKUP/{parent}/{dirname} {date}{inc}",
        "directories": [
            {
                "path": "/home/user/.config",
                "archive_type": "zip",
                "last_backup": "2020-11-07 19:58:41",
                "include": true,
                "incremental": false
            },
            ...
        ]
    }

The target_pattern indicates the directories where new backups should be stored. It allows for different placeholders:

• {date}: the current date, e.g. 2020-11-07
• {datetime}: the current date and time, e.g. 2020-11-07 19:58:41
• {parent}: the full absolute path of the parent of the directory to be backed up
• {dirname}: the name of the actual directory to be backed up
• {inc}: can be used to add a suffic _inc to incremental backup archives

The directories list shows the individual directories to be backed up. Their defining feature, of course, is the path. Besides that, you can chose whether to use zip or tar for each directory. The last_backup field indicates exactly that, and is set automatically. The include field shows whether the directory should be included in the next backup and can either be set manually or derived from the dates of the last backup and change. The incremental field means that only those files will be included into the archive that have been changed since the last backup (i.e. the backup will be faster, but previous backups have to be keps in order to restore all files)

As seen in the following figure, the user interface mainly consists of a large table of all the directories to be backed up and a few buttons and text inputs. For a description of the individual attributes and placeholders, please refer to the description of the configuration above.

• use the Target Pattern entry to define where and how to store the created archive files (see above for valid placeholders)
• use the Add and Remove buttons to select new directories to add to the list or to remove entries from the list
• the Refresh button is used to check whether any files were changed since the last backup and to set the include attribute accordingly
• the Backup button is used to create the actual backup
• the table can be sorted by each of the columns, but only the Type, Incl. and Incr. columns are editable; valid archive types are zip and tar
• when the backup has been triggered, the bottom of the UI shows the progress
• the configuration is automatically saved when the UI is closed

To create a backup, add and select the directories to be backed up and hit the Backup button. For each directory, an archive of the selected type is created and moved to the target directory. If the file or the parent directories start with a . (i.e. indicating that they are hidden) the . is replaced with a _ in the backup so that no backups are missed when moving them to storage.

Depending on the kind of files to be backed up, you may chose to use either zip (i.e. with compression) or tar (without compression), and whether to create an incremental backup or a full backup. For example:

• for a collection of already compressed files, where new files are added but files are rarely removed or changed, you should use tar and incremental
• for regular, non-compressed files that often change, like a documents folder or your mail box, use zip and a non-incremental backup

Note: Any existing files with the same name in those directories will be overwritten without further warning!

Afterwards, the collected backups can be moved to the target drive, e.g. a CD, removeable USB drive, betwork share, or cloud storage.