arcX - Archive, Rm & Clean toolboX

ArcX is a versatile toolbox designed to streamline data management by automating key tasks. Whether you're handling experiment outputs or maintaining clean directories, ArcX simplifies the process with a range of powerful tools.

• Archiving: Seamlessly archive Oceanic experiment outputs using a flexible YAML configuration file. ArcX takes care of organizing and storing your data without manual intervention.
• Safe File Removal: Efficiently remove files that already have a local copy. ArcX ensures that only unnecessary files are deleted, safeguarding important data.
• Comprehensive Directory Cleaning: Clean multiple directories with a single command. ArcX offers various cleaning options and leverages a YAML configuration file to specify exactly what to delete and how to do it.

Installation

Via pip

pip install arcx

Via conda/mamba

mamba install arcx

Usage

Usage: arcx [OPTIONS] COMMAND [ARGS]...

Options:
  --version  Show the version and exit.
  --help     Show this message and exit.

Commands:
  rolling
  saferm
  update-md5

Rolling

To keep a directory under rolling, use the command:

Usage: cli.py rolling [OPTIONS]

Options:
  -c, --config PATH  [required]
  --dry-run          Dry run mode.
  -d, --debug        Enable debug mode.
  --help             Show this message and exit.

Rolling Configuration File

Here a template of a rolling configuration file:

- !CleanPath
    path: $PATH_UNDER_ROLLING
    fmt: ????????   # YYYYMMDD
    safe:
        to_keep: X
        reference_paths:
            - <REF1>
            - <REF2>
    conditional:
        to_keep: Y
        expected_files:
            - file.exe
            - tmp.nc
    force:
        to_keep: Z

let's comment each section.

CleanPath object

With this, we start to declare a list of !CleanPath objects with two mandatory attributes:

- !CleanPath
    path: $PATH_UNDER_ROLLING
    fmt: ????????   # YYYYMMDD

• path: it's the path to keep under rolling
• fmt: it's a string bash that represent the format of file/dir to delete. It can contains jolly characters ? or *

Safe

  safe:
      to_keep: X
      reference_paths:
          - <REF1>
          - <REF2>

• safe: means for safe rolling, which means delete a file only if an identical local copy already exists
• to_keep: how much dir/file to not include in the rolling
• reference_paths: a list of path where to find if a local copy already exists

To consider that the safe mode doesn't remove the dir under rolling.

Conditional

#### Conditional

  conditional:
      to_keep: Y
      expected_files:
          - file.exe
          - tmp.nc

• conditional: specify to remove a dir if some conditions are meet
• to_keep: how much dir/file to not include in the rolling
• expected_files: specify the exact list of files expected to find in rolling path to trigger the rm operations. The filename can contains jolly character ? and *

Force

    force: # optional
        to_keep: Z

• force: Enable path rm without any check
• to_keep: how much dir/file to not include in the rolling

---

Update md5 dir

A requirement to safe clean, is to compute in advance the md5 of all files in the rolling path using the command:

update-md5 [OPTIONS]

Options:
  -p, --path PATH  [required]
  -d, --debug      Enable debug mode.
  --help           Show this message and exit.

The command will create in the directory a file called .dir_md5.txt with the following structure:

md5hash filename1
md5hash filename2

---

Safe Clean

Safe Rm command is the equivalent of safe rm rolling section: it removes files from clean path only if exists an identical copy in keep path

Usage: cli.py saferm [OPTIONS]

Options:
  --keep PATH    [required]
  --clean PATH   [required]
  -f, --force    Delete files without confirm request
  -d, --dry-run  Disable file removal
  -d, --debug    Enable debug mode.
  --help         Show this message and exit.