-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 8/25/10 3:45 AM, Stanislav Ochotnicky wrote:
Excerpts from Daniel P. Berrange's message of Wed Aug 25 11:08:16
+0200 2010:
> On Tue, Aug 24, 2010 at 11:34:36AM -0700, Jesse Keating wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> On 8/24/10 6:43 AM, Richard W.M. Jones wrote:
>>> Is it OK to use 'git rebase -i' to compress my mistakes together
into
>>> a single working Fedora git commit? (Provided I don't push things in
>>> between or otherwise try to rewrite public history)
>>>
>>> I'm a bit confused by whether 'fedpkg commit', 'fedpkg
build', 'fedpkg
>>> push' etc are doing magic that will be broken by this.
>>>
>>> Rich.
>>>
>>
>> You are free to do any sort of history altering actions you want prior
>> to a push. fedpkg will prevent you from trying to build something that
>> hasn't been pushed unless you're doing a scratch build. commit and push
>> are very thin wrappers over the git equivs.
>
> Is there a server side hook/check to prevent accidentally pushing
> non-fastforward commits ?
Standard git repositories deny pushing non fast-forward commits unless
you --force the push. So yes I guess there is protection, I haven't
tried if --force is allowed or not (I guess it should not be on allowed
on standard branches)
Right now the ACL system prevents non-fast forward pushes, even with
- --force, for everything. I have on my "white board" to look into
allowing non-fastforward changes within the user/* branch space, as well
as completely open ACLs within that branch space. It's not the top of
my whiteboard yet though.
- --
Jesse Keating
Fedora -- Freedom² is a feature!
identi.ca:
http://identi.ca/jkeating
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Darwin)
Comment: Using GnuPG with Mozilla -
http://enigmail.mozdev.org/
iEYEARECAAYFAkx1fDIACgkQ4v2HLvE71NXX7ACfRVnwQ8xhGqlJgrPijmR101FB
PoAAmgJ3I4TCQ0GK3B1WWIbrE5IlqSd3
=tLfM
-----END PGP SIGNATURE-----