What is it?
===========

dedup is a simple data deduplication program.

Getting started
===============

To use dedup you have to first initialize the repository.

    dedup -r ~/repo -i

dedup only handles a single file at a time, so using tar is advised.
For example, to dedup a directory tree you can invoke dedup as
follows:

    tar -c ~/dir | dedup -r ~/repo -m "$(date)"

This will create .{snapshots,store} files in the ~/repo directory.
The store file contains all the unique blocks.  The snapshots file
contains all the revisions of files that have been deduplicated.  The
-m flag is used to attach an arbitrary message to the snapshot.

To list all known revisions run:

    dedup -r ~/repo -l

You will get a list of hashes.  Each hash corresponds to a single file
(in this case, a tar archive).

To extract a file from the deduplicated store run:

    dedup -r ~/repo -e <hash> > dir.tar

Portability
===========

dedup works on Linux, *BSD, macOS and possibly other UNIX-like systems.

Dependencies
============

  - liblz4
  - snappy
  - libomp (optional, see config.mk)

Contact
=======

You can reach us through irc.2f30.org/6697 at #2f30.
