Factor out .rsyncsums logic into a separate checksum-caching library?

Matt McCutchen hashproduct+rsync at gmail.com
Sun Jun 24 17:03:03 GMT 2007


I notice that .rsyncsums is starting to look a lot like the index that
the git version control system ( http://git.or.cz/ ) uses to determine
whether a file has changed since it was last staged for committing.
The git index has been heavily used and tested, so you might find it
helpful when implementing a checksum cache for rsync.  Specifically,
it has protection against being fooled when a file's checksum is
cached and the file is modified again in the same second; .rsyncsums
could use this.

I think it would be even better to factor out logic for caching file
checksums into a separate library used by both rsync and git.  This
would have two advantages: the subtleties of implementing a 100%
correct cache only have to be addressed once, and different programs
can make use of each other's cached checksums.  GNU make and Beagle
desktop search might also use the library.


More information about the rsync mailing list