On Sun, Dec 21, 2008 at 01:22:59AM +0100, Kevin Kofler wrote:
Kevin Fenzi wrote:
> I would normally agree, but I have seen a number of cases where people
> got Obsoletes/Provides wrong and caused a mess. ;(
>
> I would like to see them get another pair of eyes on their package
> before pushing the renamed version out. Thats all.
Then maybe the guideline should be that the reviewer has to check for valid
Obsoletes/Provides, but any other checks are optional (as in: if the
reviewer notices something obviously wrong, he/she should report it and
request it fixed before approving the rename, but he/she shouldn't be
required to go through the whole checklist again)?
I think it would be right like that. A normal review is unneeded in my
opinion, and not checking obsoletes... is wrong too.
--
Pat