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

Re: Uninstallable Packages



     [My response to this message, and my question are more suited to
-mentors than -devel, so I have set Reply-to to -mentors.]

Anthony Towns <aj@azure.humbug.org.au> writes:

> In particular, a number of binary packages could be produced from
> either of two source packages:
> 
> 	dict-elements and dict-misc both include source for:
> 		dict-elements.deb
> 			(presumably dict-misc needs to be changed)

     This is also true of dict-foldoc and dict-misc, except that I
failed to comment out the lines pertaining to elements in
Makefile.in. 

     The upstream tarball for dict-misc includes the sources for the
elements, foldoc, jargon, easton, and hitchcock databases.  Since
foldoc is updated very frequently, I forked it into a separate package
in the initial release.  Subsequently, I received an updated version of
elements, so I forked this as well.  This made it unnecessary to build
five packages every time one of the actively maintained dictionaries
was updated.

     Because of this, the sources for these two dictionaries appear in
the archives twice.  There would be a saving in archive and mirror
disk space and in bandwidth if I removed these databases from my
source tree and made a new orig.tar.gz as non-pristine source.  All of
the dict-* packages include formatting software produced by the
upstream author and a database taken from other sources and dropped 
into the ../data directory of the source tree.  IMHO, the structure of
these packages is such that "pristine" sources is less meaningful than
it is for more typical packages.  The forked dictionaries are no
longer "pristine source".

     Would it be desirable for me to create a new, non-pristine
orig.tar.gz to be included with the forthcoming upload to adopt the
FHS, and thereby reduce the size of the archive slightly?

Bob     
-- 
   _
  |_)  _  |_       Robert D. Hilliard    <hilliard@flinet.com>
  |_) (_) |_)      Palm City, FL  USA    PGP Key ID: A8E40EB9


Reply to: