Deprecating the wiki-based Packaging Guide
Scott Kitterman
ubuntu at kitterman.com
Wed Dec 19 02:56:41 UTC 2012
Steve Langasek <steve.langasek at ubuntu.com> wrote:
>On Tue, Dec 18, 2012 at 09:33:18PM -0500, Scott Kitterman wrote:
>
>> >One concrete example where UDD shines and the non-UDD workflow is
>> >inadequate is for sponsoring of package merges. If someone hands me
>a
>> >branch that properly merges the new Debian version into the Ubuntu
>> >branch, I can review that with the standard bzr diff tools and
>ascertain
>> >that the sponsoree has done the merge correctly. If someone hands
>me a
>> >debdiff for a Debian merge, that's useless; I effectively have to do
>the
>> >merge myself as part of the review, and no time is saved.
>
>I should clarify here that I meant a merge of a new upstream version
>packaged in Debian. For packaging-only merges, debdiffs work fine.
>
>> It only works better if you are using UDD. I agree that if your
>primary
>> workflow is UDD based, then UDD branches are better. If I get a
>branch
>> it's as useless for me as a debdiff is for you. When asked to
>sponsor
>> things that have a branch, I generally decline or ask for a debdiff.
>
>Your decision to boycott UDD doesn't make a UDD branch "useless". A
>debdiff
>for a merge of a new upstream package version actually *is* useless and
>is a
>waste of the sponsoree's time, for the stated reason that the "review"
>of
>such a debdiff involves re-doing the merge myself.
Right. What I really want (and what our docs asked for at one point) is a packaging diff (debian dir) and a pointer to the upstream tarball. I said the branch would be useless to me, because I'd have to extract out the packaging diff and redo the merge to make sense of it.
BTW, I didn't come to my perspective on UDD without trying it.I gave it a real go, but it just didn't work for me.
Scott K
More information about the ubuntu-devel
mailing list