On Wed, Mar 05, 2008 at 04:54:48PM +0100, Hans de Goede wrote:
Jesse Keating wrote:
>On Wed, 2008-03-05 at 09:54 -0600, Matt Domsch wrote:
>>On Wed, Mar 05, 2008 at 10:49:15AM -0500, Brian Pepple wrote:
>>>Hi,
>>>
>>>Please find below the list of topics that are likely to come up in the
>>>next FESCo meeting that is scheduled for tomorrow, Thursday at 18:00 UTC
>>>in #fedora-meeting on
irc.freenode.org:
>>In the buildsys call last week, Jesse said he would bring a proposal
>>to make tags in CVS "immutable", which the
'correspondingsource'
>>project on fedorahosted will require.
>
>
>OOps, yes, thanks for bringing this up again. Please add it to the
>schedule.
>
Ugh,
Can we _not_ do this please, sometimes one forgets to "cvs add
foo-bar.patch", being able to then just add it once the srpm build has
failed in koji and do a make force-tag is _very_ convenient. Is there
actual proof that people are using force-tag for other reasons then to fix
build errors. If there is no proof for this then why immediately use the
immutable big hammer, for a theoretical problem?
I didn't want the big hammer, but the buildsys folks don't want Koji
to be able to write into CVS at all, such as to create a new tag, and
there's no way to make a koji-created tag immutable, while letting the
other tags be mutable, AFAIK.
Yes, it's a theoretical problem, but if we hit it, it becomes a legal
problem.
--
Matt Domsch
Linux Technology Strategist, Dell Office of the CTO
linux.dell.com &
www.dell.com/linux