[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Ddiff version 0.2 has been released.



This is just a note to announce that ddiff version 0.2 has been
released and is available to download from 

	 http://onegeek.org/~tom/software/ddiff/

This new version includes several new features. In semi-random order,
they are:

- Gzip headers are stored. This usually leads to output files that are
  completely md5sum identical to the package files used to produce
  them. This assumes that, for any given input stream and compression
  option, gzip produces an output stream that's identical except for
  headers. I'd appreciate it if people could test this assumption
  out on different platforms. (ie. run ddiff and dpatch on machines
  with different endiannesses and see if the results are identical to
  the source packages.)

- In the case where a diff is being made against a file in a source
  package, a md5 of the member of the source package is stored in the
  debdiff. This provides an additional check to ensure package
  integrity.

- Ddiff now supports the -z option, which compresses the body of the
  produced debdiffs. (The dpatch program has been modified to handle
  these compressed ddiffs.)

Dpatch, especially, is fast. It's runtime tends to be dominated by the
gzip compression process that it invokes.

This version contains the full set of features required for ddiff to
hopefully be useful in the real world. The only feature that's been
planned but not implemented is the ability to do a dpatch against the
files installed on this system.

At this time I'd like to ask if anyone would be willing to sponsor
this package for inclusion into Debian?

I'm also wondering if anyone has any ideas on how debdiff support
could be integrated with apt or other parts of the package system. My
first idea involved an apt method, but that seems to me to be less
than ideal. 

Any comments or suggestions?

-- 
Tom Rothamel --------- http://onegeek.org/~tom/ ---------- Using GNU/Linux
		The Moon is Waning Crescent (16% of Full)


Reply to: