Pushing other people's patches to gerrit

[Jan Holesovsky]

Hi Moggi,

Markus Mohrhard píše v St 04. 01. 2017 v 02:47 +0100:

>         
>         And in general - what's the reason for such a restriction?
>         Why somebody
>         cannot eg. backport somebody else's patch to a stable branch,
>         even if
>         they don't have full commit access yet (?) - the patch will be
>         reviewed
>         anyway...
>
> That should be the "Forge Author Identity" part in
> https://gerrit.libreoffice.org/#/admin/projects/Dev-ACL-Template,access
> 
> At least for me the interpretation was always that it simplifies the
> lives of reviewers. If a patch has been committed by someone who is
> not the author and I can not be 100% sure that the license is correct
> I need to spend a lot of time researching the license situation
> myself.

I see, thanks for the explanation!

Is it possible to set this case-by-case?  If yes, I'd like to have this
for Heiko - he's pushed stuff for other people to gerrit in the past
with his name / mail in Author: (instead of the real author) which is
not correct; and yet his role is to help people with getting stuff
merged if they cannot do it themselves yet.

For Aron, I think best if he gets full commit access at some stage; I'll
ask for that at the ESC.

> Additionally I'm often not very thrilled when people who have not
> understood a patch are back porting them to a stable branch. IMHO
> backporting should always either be done by someone who has the
> experience in that part of the code, has written the patch or has
> reviewed the patch for master.

Sure; but backporting is just one of the use cases, and Aron has hit it
it in a distro branch - where the distro maintainers are responsible for
that & have more freedom I believe.

But now I understand the root of the concern here, and agree we
shouldn't change this setting overall :-)

Thank you,
Kendy