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

Bug#407654: O: slidentd -- minimal ident (RFC 1413) daemon



Daniel Baumann <daniel@debian.org> writes:

> David Smith wrote:
>> Heh, let's try that one more time. I've reuploaded to mentors
>> and I've verified the same package from my personal archive at
>> http://bosabosa.org/~dds/debian/ . If mentors is still fubared
>> please try the bosabosa archive.
>
> still wrong on both sites:
>
> the one in debian:
> 59b89db95d447262894c5a56bd24ec04  slidentd_1.0.0.orig.tar.gz
>
> the one from you:
> 16e443115ee48aadfd75deef9ebac624  slidentd_1.0.0.orig.tar.gz

I see. The md5sum of the file is different because the
timestamps inside the tarball are different, and the timestamps
are different because I'm using darcs-buildpackage which when
you tell it to make an upstream tarball, checks it out of the
darcs repository it created for upstream sources. So while
diff won't show you any difference, the md5sum will. I hadn't
thought of that.

And I guess darcs-buildpackage was just not expecting to have
to handle this situation of needing to publish a original
source tarball without actually having any changes inside it,
as would be the case with an upload to mentors.debian.net.

Since this seems to be a special case, I've rebuilt the package
by hand with the current source tarball and my debian
directory and the next time a package needs to be made
using darcs-buildpackage should not produce any problems. I
will check with John Goerzen if my understanding of the problem
is accurate and if there is any solution.

Sorry for going back and forth but please take another look
again,
-- 
=========================================================
  David D. Smith <davidsmith at acm.org>
Jabber/GoogleTalk: dds at jabber.org ; GPG: 0xE6511C7E
IRC: dds on irc.freenode.net ; MSN: dds4418 at hotmail.com
==========================================================

Attachment: pgpfAOnSnRyCb.pgp
Description: PGP signature


Reply to: