Prune

KO | EN

gitlite — search

#python #java #python3 #arduino #golang #machine-learning #rust #html #flask #javascript #seismology #nodejs

★ 11 Open GitHub ↗

Prune is a simple tool that lets you remove any files not matching the specified retention options.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Prune

![a person in a purple shirt pulling old folders from a tree](https://raw.githubusercontent.com/binarypatrick/Prune/main/docs/pruning-tree.png)

[![Build & Test](https://github.com/BinaryPatrick/Prune/actions/workflows/build-test.yml/badge.svg?branch=main)](https://github.com/BinaryPatrick/Prune/actions/workflows/build-test.yml)

Prune is a simple tool that lets you ~remove~ _prune_ files in a folder, deleting any files not matching the specified retention options.

Any file type can be an archive and prune allows you to specify which files are in scope to be pruned. Prune is developed using .NET 8.0 and can be compiled for x64 (Windows, Linux, MacOS) and arm/arm64 (Raspberry Pi). So far it is running in the wild on a Debian 12 VM, a Raspberry Pi 4B 2GB and 4GB, and a Raspberry Pi 3B.

## Disclaimer

The purpose of this tool is to **DELETE** files. Please take care when using it and ensure you do not delete the wrong files. I highly recommend using the `--verbose` and `--dry-run` flags until your are sure what will be pruned. This tool is provided without warranty of any kind.

## Usage

```bash
prune --help
```

```bash
prune --path /backups/images --verbose --dry-run /
   --ext img /
   --keep-last 5 /
   --keep-daily 3 /
   --keep-weekly 2 /
   --keep-monthly 4 /
```

```bash
prune --path /backups/images -e img -l 5 -d 3 -w 2 -m 4
```

> **Warning** Use the `--verbose` and `--dry-run` flags until you confirm what will be pruned

| Flag               | Description                                                                              |
| ------------------ | ---------------------------------------------------------------------------------------- |
| --path             | Required. Path to files location                                                         |
| --dry-run          | (Default: false) Do not make any changes and display simulated output                    |
| -v, --verbose      | (Default: false) Enable verbose logging                                                  |
| -s, --silent       | (Default: false) Disable all logging                                                     |
| -p, --prefix       | File name prefix to use when matching archives                                           |
| -e, --ext          | File extension to use when matching archives, i.e. img, txt, tar.gz (do not include dot) |
| -l, --keep-last    | (Default: 5) Number of archives to keep at a minimum                                     |
| -h, --keep-hourly  | (Default: 0) Number of hourly archives to keep                                           |
| -d, --keep-daily   | (Default: 0) Number of daily archives to keep                                            |
| -w, --keep-weekly  | (Default: 0) Number of weekly archives to keep                                           |
| -m, --keep-monthly | (Default: 0) Number of monthly archives to keep                                          |
| -y, --keep-yearly  | (Default: 0) Number of yearly archives to keep                                           |
| --help             | Display this information                                                                 |
| --version          | Display version information                                                              |

> If there is more than one backup for a time interval, only the latest is kept

The time intervals are processed in the same order as above. Each interval only covers backups within its segment. A single backup will not be matched by multiple time intervals, and if a time interval spans over a previous interval, it will prune until it finds a matching unique file. For example, if the last daily time interval backup is on a Thursday, the weekly time interval will skip to the preceding week to find a file, pruning any other files it iterates through.

The backups are designed to follow the same logic as Proxmox Backup Server pruning. [Use their very useful simulator](https://pbs.proxmox.com/docs/prune-simulator/) to test different configurations.

## Build

To build from source, you will need to have:

- [DotNet SDK for .Net 8.0](https://dotnet.microsoft.com/en-us/download)
- Git

Pull down this repo and you can compile the project from the `src/` folder.

```bash
# linux-x64 environment
dotnet publish --runtime linux-x64 --self-contained -p:PublishSingleFile=true -c Release -o ./out
```

```bash
# linux-arm64 environment
dotnet publish --runtime linux-arm64 --self-contained -p:PublishSingleFile=true -c Release -o ./out
```

```bash
# win-x64 environment
dotnet publish --runtime win-x64 --self-contained -p:PublishSingleFile=true -c Release -o ./out
```

You should see the binary in the `/src/out` folder. You can discard the `.pdb` file.

Prune

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README