You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@bigtop.apache.org by Mark Grover <ma...@apache.org> on 2013/09/04 23:58:00 UTC
[DISCUSS] Using --signoff option with git am when committing patches
Hi fellow developers,
It's great that we encourage contributors to submit patches using git
format-patch. It allows us to attribute credit properly to the contributor
in the git along with other nice things.
For committing such patches, the command we typically use git am. However,
I would like to propose that committers use "git am --signoff" instead when
committing the patches.
The only difference with --signoff flag is that the name of the committer
who committed the code also shows up in the comment of the commit.
Here is an example,
When using "git am", the commit log looks like:
commit 00aa88222bbdddl
Author: Mr. Contributor <co...@contributor.com>
Date: Fri Jul 26 18:00:00 2013 -0400
BIGTOP-X: Test commit
When using "git am --signoff", the commit log looks like:
commit 00aa88222bbdddl
Author: Mr. Contributor <co...@contributor.com>
Date: Fri Jul 26 18:00:00 2013 -0400
BIGTOP-X: Test commit
Signed-off-by: Ms. Committer <co...@apache.org>
The commit is still attributed to the contributor but I think it would be
nice for logistical reasons to have the committer name in the log messages
of the commit.
Anyways, I have started documenting this at
https://cwiki.apache.org/confluence/display/BIGTOP/How+to+Contribute
Please let me know what you think!
Mark
Re: [DISCUSS] Using --signoff option with git am when committing
patches
Posted by Konstantin Boudnik <co...@apache.org>.
While nice to have, it seems like a complication that might or might not be
beneficial. I would abstain from making a policy.
Cos
On Sun, Sep 15, 2013 at 03:20PM, Mark Grover wrote:
> That wasn't my intent. When I commit my own changes, I simply push the
> commit that I generated the patch from instead of git applying the patch
> that I uploaded to JIRA.
>
> I am personally in favor of keeping the (unwritten?) policy the way it is -
> that is, it's ok to commit to your patches, once they have gotten a +1 from
> another committer.
>
>
> On Sun, Sep 15, 2013 at 1:28 PM, Andrew Purtell <ap...@apache.org> wrote:
>
> > Would a consequence of this be no committing of your own changes? Signing
> > off on your own commit wouldn't add any value.
> >
> >
> >
> > On Sun, Sep 8, 2013 at 6:03 PM, Mark Grover <ma...@apache.org> wrote:
> >
> > > And, about your former point:
> > >
> > > Our git format-patch policy for patches ensures that the name of the
> > actual
> > > contributor who committed the code shows up on the commit message when
> > > doing git log. However, I think it will be useful to see which committer
> > > committed that patch on behalf of the contributor. That's exactly the
> > > problem "--sign-off" solves.
> > >
> > > Linux kernel<
> > >
> > http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/Documentation/SubmittingPatches?id=HEAD
> > > >,
> > > for example, has a policy to use --sign-off flag and the same applies to
> > > contributing to the git
> > > project<
> > >
> > http://git.kernel.org/cgit/git/git.git/tree/Documentation/SubmittingPatches?id=HEAD
> > > >
> > > .
> > > And, there are some other benefits too, listed here:
> > >
> > >
> > http://stackoverflow.com/questions/1962094/what-is-the-sign-off-feature-in-git-for
> > >
> > > Mark
> > >
> >
> >
> >
> > --
> > Best regards,
> >
> > - Andy
> >
> > Problems worthy of attack prove their worth by hitting back. - Piet Hein
> > (via Tom White)
> >
Re: [DISCUSS] Using --signoff option with git am when committing patches
Posted by Mark Grover <gr...@gmail.com>.
That wasn't my intent. When I commit my own changes, I simply push the
commit that I generated the patch from instead of git applying the patch
that I uploaded to JIRA.
I am personally in favor of keeping the (unwritten?) policy the way it is -
that is, it's ok to commit to your patches, once they have gotten a +1 from
another committer.
On Sun, Sep 15, 2013 at 1:28 PM, Andrew Purtell <ap...@apache.org> wrote:
> Would a consequence of this be no committing of your own changes? Signing
> off on your own commit wouldn't add any value.
>
>
>
> On Sun, Sep 8, 2013 at 6:03 PM, Mark Grover <ma...@apache.org> wrote:
>
> > And, about your former point:
> >
> > Our git format-patch policy for patches ensures that the name of the
> actual
> > contributor who committed the code shows up on the commit message when
> > doing git log. However, I think it will be useful to see which committer
> > committed that patch on behalf of the contributor. That's exactly the
> > problem "--sign-off" solves.
> >
> > Linux kernel<
> >
> http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/Documentation/SubmittingPatches?id=HEAD
> > >,
> > for example, has a policy to use --sign-off flag and the same applies to
> > contributing to the git
> > project<
> >
> http://git.kernel.org/cgit/git/git.git/tree/Documentation/SubmittingPatches?id=HEAD
> > >
> > .
> > And, there are some other benefits too, listed here:
> >
> >
> http://stackoverflow.com/questions/1962094/what-is-the-sign-off-feature-in-git-for
> >
> > Mark
> >
>
>
>
> --
> Best regards,
>
> - Andy
>
> Problems worthy of attack prove their worth by hitting back. - Piet Hein
> (via Tom White)
>
Re: [DISCUSS] Using --signoff option with git am when committing patches
Posted by Andrew Purtell <ap...@apache.org>.
Would a consequence of this be no committing of your own changes? Signing
off on your own commit wouldn't add any value.
On Sun, Sep 8, 2013 at 6:03 PM, Mark Grover <ma...@apache.org> wrote:
> And, about your former point:
>
> Our git format-patch policy for patches ensures that the name of the actual
> contributor who committed the code shows up on the commit message when
> doing git log. However, I think it will be useful to see which committer
> committed that patch on behalf of the contributor. That's exactly the
> problem "--sign-off" solves.
>
> Linux kernel<
> http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/Documentation/SubmittingPatches?id=HEAD
> >,
> for example, has a policy to use --sign-off flag and the same applies to
> contributing to the git
> project<
> http://git.kernel.org/cgit/git/git.git/tree/Documentation/SubmittingPatches?id=HEAD
> >
> .
> And, there are some other benefits too, listed here:
>
> http://stackoverflow.com/questions/1962094/what-is-the-sign-off-feature-in-git-for
>
> Mark
>
--
Best regards,
- Andy
Problems worthy of attack prove their worth by hitting back. - Piet Hein
(via Tom White)
Re: [DISCUSS] Using --signoff option with git am when committing patches
Posted by Mark Grover <ma...@apache.org>.
And, about your former point:
Our git format-patch policy for patches ensures that the name of the actual
contributor who committed the code shows up on the commit message when
doing git log. However, I think it will be useful to see which committer
committed that patch on behalf of the contributor. That's exactly the
problem "--sign-off" solves.
Linux kernel<http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/Documentation/SubmittingPatches?id=HEAD>,
for example, has a policy to use --sign-off flag and the same applies to
contributing to the git
project<http://git.kernel.org/cgit/git/git.git/tree/Documentation/SubmittingPatches?id=HEAD>
.
And, there are some other benefits too, listed here:
http://stackoverflow.com/questions/1962094/what-is-the-sign-off-feature-in-git-for
Mark
Re: [DISCUSS] Using --signoff option with git am when committing patches
Posted by Mark Grover <gr...@gmail.com>.
Will do!
On Sat, Sep 7, 2013 at 10:56 PM, Roman Shaposhnik <rv...@apache.org> wrote:
> On Wed, Sep 4, 2013 at 2:58 PM, Mark Grover <ma...@apache.org> wrote:
> > The commit is still attributed to the contributor but I think it would be
> > nice for logistical reasons to have the committer name in the log
> messages
> > of the commit.
>
> I would be curious to hear more about why you think this would
> be helpful. The reason I'm asking is that Git is different from SVN
> in this respect -- with SVN comments are the only place for
> this info while Git tracks committers separately anyway.
>
> > Anyways, I have started documenting this at
> > https://cwiki.apache.org/confluence/display/BIGTOP/How+to+Contribute
> >
> > Please let me know what you think!
>
> Great start! Perhaps add a link to it onto our website (patch, please ;-))?
>
> Thanks,
> Roman.
>
Re: [DISCUSS] Using --signoff option with git am when committing patches
Posted by Roman Shaposhnik <rv...@apache.org>.
On Wed, Sep 4, 2013 at 2:58 PM, Mark Grover <ma...@apache.org> wrote:
> The commit is still attributed to the contributor but I think it would be
> nice for logistical reasons to have the committer name in the log messages
> of the commit.
I would be curious to hear more about why you think this would
be helpful. The reason I'm asking is that Git is different from SVN
in this respect -- with SVN comments are the only place for
this info while Git tracks committers separately anyway.
> Anyways, I have started documenting this at
> https://cwiki.apache.org/confluence/display/BIGTOP/How+to+Contribute
>
> Please let me know what you think!
Great start! Perhaps add a link to it onto our website (patch, please ;-))?
Thanks,
Roman.
Re: [DISCUSS] Using --signoff option with git am when committing patches
Posted by Mark Grover <gr...@gmail.com>.
Indeed, we are. This would be in addition to that, for added convenience:-)
On Wed, Sep 4, 2013 at 3:03 PM, Konstantin Boudnik <co...@apache.org> wrote:
> Aren't we tracking how commited what through JIRA tickets as well?
>
> --signoff seems reasonable though.
>
> Cos
>
> On Wed, Sep 04, 2013 at 02:58PM, Mark Grover wrote:
> > Hi fellow developers,
> > It's great that we encourage contributors to submit patches using git
> > format-patch. It allows us to attribute credit properly to the
> contributor
> > in the git along with other nice things.
> >
> > For committing such patches, the command we typically use git am.
> However,
> > I would like to propose that committers use "git am --signoff" instead
> when
> > committing the patches.
> >
> > The only difference with --signoff flag is that the name of the committer
> > who committed the code also shows up in the comment of the commit.
> >
> > Here is an example,
> >
> > When using "git am", the commit log looks like:
> > commit 00aa88222bbdddl
> > Author: Mr. Contributor <co...@contributor.com>
> > Date: Fri Jul 26 18:00:00 2013 -0400
> >
> > BIGTOP-X: Test commit
> >
> >
> >
> > When using "git am --signoff", the commit log looks like:
> > commit 00aa88222bbdddl
> > Author: Mr. Contributor <co...@contributor.com>
> > Date: Fri Jul 26 18:00:00 2013 -0400
> >
> > BIGTOP-X: Test commit
> >
> > Signed-off-by: Ms. Committer <co...@apache.org>
> >
> >
> > The commit is still attributed to the contributor but I think it would be
> > nice for logistical reasons to have the committer name in the log
> messages
> > of the commit.
> >
> > Anyways, I have started documenting this at
> > https://cwiki.apache.org/confluence/display/BIGTOP/How+to+Contribute
> >
> > Please let me know what you think!
> >
> > Mark
>
Re: [DISCUSS] Using --signoff option with git am when committing
patches
Posted by Konstantin Boudnik <co...@apache.org>.
Aren't we tracking how commited what through JIRA tickets as well?
--signoff seems reasonable though.
Cos
On Wed, Sep 04, 2013 at 02:58PM, Mark Grover wrote:
> Hi fellow developers,
> It's great that we encourage contributors to submit patches using git
> format-patch. It allows us to attribute credit properly to the contributor
> in the git along with other nice things.
>
> For committing such patches, the command we typically use git am. However,
> I would like to propose that committers use "git am --signoff" instead when
> committing the patches.
>
> The only difference with --signoff flag is that the name of the committer
> who committed the code also shows up in the comment of the commit.
>
> Here is an example,
>
> When using "git am", the commit log looks like:
> commit 00aa88222bbdddl
> Author: Mr. Contributor <co...@contributor.com>
> Date: Fri Jul 26 18:00:00 2013 -0400
>
> BIGTOP-X: Test commit
>
>
>
> When using "git am --signoff", the commit log looks like:
> commit 00aa88222bbdddl
> Author: Mr. Contributor <co...@contributor.com>
> Date: Fri Jul 26 18:00:00 2013 -0400
>
> BIGTOP-X: Test commit
>
> Signed-off-by: Ms. Committer <co...@apache.org>
>
>
> The commit is still attributed to the contributor but I think it would be
> nice for logistical reasons to have the committer name in the log messages
> of the commit.
>
> Anyways, I have started documenting this at
> https://cwiki.apache.org/confluence/display/BIGTOP/How+to+Contribute
>
> Please let me know what you think!
>
> Mark