On Thu, May 3, 2012 at 12:27 AM, Marc Delisle <marc(a)infomarc.info> wrote:
Le 2012-05-02 14:54, Atul Pratap Singh a écrit :
On Thu, May 3, 2012 at 12:03 AM, Marc Delisle
<marc(a)infomarc.info>
wrote:
> Le 2012-05-02 14:25, Marc Delisle a écrit :
>> Le 2012-05-02 14:13, Atul Pratap Singh a écrit :
>>
>>> But what if the local changes are not desired to be published to the
> origin
>>> right now. Marc, should we always experiment on a separate branch or
> repo
>>> or is there any other way to make selective push to origin like
HEAD~1
or
>> something?
>
> Atul,
> you should always push your changes to the origin (on github), this is
> how your mentor (me) will look at your work. You can experiment on a
> separate branch if you want, but this will be part of the same
repository.
>
About experimenting, of course you should test a chunk of code that
relates to one feature or one refactoring, before committing it to some
branch (master or testing branch).
Thanks Marc,
I would always push properly tested commits to github origin. And I
would
use separate branch or repo clones for
experimenting.
Thilina, About pushing as in your case, if you don't wanna push your
current local changes to github origin, you can use:-
git checkout master
git push origin master~x:master (x being integer = number of
commits to
leave from HEAD)
After you have properly merged upstream to master.
Atul,
a local change is not a commit. I suspect Thilina has not committed his
changes.
Marc, yeah maybe he has not committed, so pushing won't get affected. I