Le 20/05/2025 à 12:07, Jonathan Dowland a écrit :
I'm deeply concerned that the "surface area" of Debian packaging tools
is too large already for new folk, and adding something like gwh into
the archive makes that problem worse.
Fair enough, although I'm hoping to make gwh good enough to encapsulate
part of that surface area in one tool.
I would really like to see an effort to improve gbp itself directly at
least first. Have you tried to do that?
I got in touch with Guido Guenther before sending this ITP, and he
seemed to agree that gwh can only be implemented as a wrapper around gbp
(mainly due to adherence to pbuilder hooks while gbp can be used with
other builders).
From the README, I see
I'm still in the process of turning these from a bunch of files in my
~/bin and ~/.config and so on into something that can be of use to
more people. This warning will disappear once I'm confident most of
the rough edges have been smoothed out.
IMHO, putting this in the archive is premature.
I wouldn't upload to the archive before that warning is gone :-)
(Besides, it's almost gone, I just need to pick up the nerves to commit
its removal). I'm also stabilizing the option names before pushing further.
I think most feedback on ITPs (like this) just goes into the void. Can
I be explicit in asking you Roland, to withdraw this ITP, at least
until you have exhausted options for improving gbp itself?
I've read all the feedback on this ITP, and will take it into account,
including the renaming part. I'm uncomfortable with withdrawing it
altogether, but I propose to sit on it for a few weeks, at least long
enough to make the tool good enough to add real value to the archive.
Then I'll restart a discussion on -devel, and if the consensus then is
that gwh has no place in Debian I'll withdraw the ITP.
How does that sound?
Roland.
