Hi Ian, long time no hear,
> Obviously it has the disadvantage of using twice the disk space.
Plus the fact that with the XOR format, common parts of old and new
become zeroes which compress really well.
> We might want to only generate
> offset == 0
> and
> (oldlen == newlen) || (oldlen == filelen - offset)
> for now at least.
Well, it'd be easy enough to detect common prefixes and suffixes of
the files.
Juliusz
_______________________________________________
darcs-devel mailing list
[email protected]
http://www.abridgegame.org/cgi-bin/mailman/listinfo/darcs-devel
