Grabbags is an enhanced implementation of the Library of Congress's BagIt Library. Grabbags allows users to do bulk creation and validation of bags. Grabbags can also eliminate system files before bagging. Even better, it can delete system files automatically in existing bags if they haven't been written to the bag manifest.

For installation, see getting_started.rst

To run grabbags, use the command: $ grabbags (optional flags) (target directory path)

By default, grabbags will do bulk creation of bags. It assumes that a target directory contains many other subdirectories inside of it that will be turned into bags. So set up your directories accordingly. You can also give the command multiple target directories $ grabbags (optional flags) (target directory path 1) (target directory path 2)

Since grabbags uses the bagit Python library, all the functionality of bagit (including adding metadata fields and choosing checksum algorithms) should be available for bag creation.

Using the --no-system-files flag when creating bags will find and remove any system files before bagging. The current system files that the script can find are:

• .DS_Store
• Thumbs.db
• ._ files (AppleDoubles)
• Icon files (Apple custom icons)

Please send a pull request or issue if you have additional information about new system files that users would want to delete.

Validation of bags has two possible options:

The default behavior of $ grabbags --validate is to validate the bag by comparing the checksums of all files with the checksums contained in the manifest. Users can optionally use the flags --validate --no-checksums. This only validates the Oxsum of the bag, the number of files, and the proper files according to the bagit specification. Using the --no-checksums flag is equivalent to running --validate --completeness-only

Grabbags can delete system files within existing bags if they haven't already been written to the bag manifest. To use this feature, run the following:

$ grabbags --clean (target directory path)

Remember, that all of your bags should be in subdirectories inside of the target directory.

Just as in bagit python, users can use the --log (path to place log file) flag to create a log when creating or validating bags. At the end of the output grabbags will display summary data about the numbers of bags created or validated (number of successes, number of failures and path to all failures).

Grabbags was originally produced as part of AMIA/DLF Hack Day 2019

The initial project team was:

• Henry Borchers
• Helyx Chase Scearce Horwitz
• Jonathan Farbowitz
• Nıck Krabbenhöft

As part of AMIA/DLF Hack Day 2021, Grabbags was given a little more love.

The 2021 Project Team was:

• Henry Borchers
• Jonathan Farbowitz
• Bryn Knowles
• Milo Thiesen