Process for handlng fakesyncs

[Martin Pitt]

Stefan Potyra [2010-02-13 23:27 +0100]:
> erm, doesn't dpkg (or another tool which I forgot) disallow using an Ubuntu 
> address in changelog (which would be the case for the fakesync) and have a 
> maintainer with a non-Ubuntu address?

No, the implementation actually behaves according to Emmet's/my gut
feeling: It only enforces an ubuntu-ish Maintainer field if the
version number has "ubuntu" in it.

> Anyways, imho this thread tries to overregulate things while completely 
> ignoring the real problem: Why is a fake sync necessary in the first place. 
> Imho we should find rules for the real problem, e.g. if you plan to upload a 
> new upstream version before debian does so, please don't fuss with 
> the .orig.tar.gz

In most of these cases it happens when upstream only has a tar.bz2, or
has non-redistributable contents and thus need a repack. I guess the
cases so far were pretty justifyable.

> and more importantly please coordinate with the debian 
> maintainer. 

That's indeed something that can't be stressed enough. Fakesyncs are
quite inconvenient, compared to a real sync, and we should avoid them
as much as possible. (Ever tried to fakesync a java-something package?
You need to download 200 MB of build dependencies first just to build
a _source_ package..)

Martin

-- 
Martin Pitt                        | http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
Url : https://lists.ubuntu.com/archives/ubuntu-devel/attachments/20100214/82644b9d/attachment.pgp