[Duplicity-talk] file format

[Rob Braun]

I came across the duplicity project and found the new archive format
described on the web page.  I couldn't help but notice that this is
surprisingly similar to the xar project I and a few others are working
on.  http://www.opendarwin.org/projects/xar/
This project is to build an archiver around an archive format that uses
an xml based table of contents at the beginning of the file, with a
seekable file "heap" at the end.  
We currently have a library which contains all the archive manipulation
code, and a command line tool to expose the library's functionality.
The project and the TOC are changing frequently, but we are currently
archiving and extracting files, hardlinks, directories, symlinks,
device files, and more.  We have been actively developing on OS X and
Linux.  OS X provides an opportunity to exercise some of xar's unique
features by archiving filesystem metadata and resource forks and 
addressing the cross-platform handling of these pieces of information.

Feel free to check out the code from our cvs repository, make a few
archives, and examine the file format for yourself.  Since the table
of contents is in XML, you can easily view it and see the kinds of
things we are doing in this new archive format.

If this looks compatible with duplicity's needs, we'd be happy to 
work with the duplicity project to improve both projects.

Rob