[IPython-dev] pull requests

MinRK benjaminrk at gmail.com
Thu Sep 19 12:03:15 EDT 2013


Yes, open a PR against master.  I don't know the provenance of those
gitwash docs, but they say several things that are not true. We should
figure out if we should update them to be accurate, or just remove them
altogether (I would be in favor of the latter).

-MinRK


On Thu, Sep 19, 2013 at 5:21 AM, Cyrille Rossant
<cyrille.rossant at gmail.com>wrote:

> 2013/9/19 Matthias BUSSONNIER <bussonniermatthias at gmail.com>:
> >
> > Le 19 sept. 2013 à 12:14, Cyrille Rossant a écrit :
> >
> >>> I'm not sure I see where this state that PR should not be done gains
> master.
> >>> Anyway.
> >>
> >> Ah, I misunderstood then, PR should be made against master on ipython,
> >> but not necessarily on master on one's fork, right?
> > Right
> >
> >>
> >>> Usually you will push a named branched on your repo. and you ask to
> merge this branch into ipython/master
> >>> So github will send a mail
> >>>
> >>> Cyrille want to merge crossant/my-awesome-feature into ipython/master
> >>
> >> So we should always use a new branch in our fork, and never make edits
> >> in master?
> >
> > Technically you can. It is just a pragmatic reasons.
> >         - Having people writing on their master increase chance of weird
> history is one submit many PR and then pull an update version.
> >         - when you speak of "master" it is not ambiguous.
> >         - it's easier for a dev to $ git checkout
> crossant/vizpy-integration as you know what you are expecting and it's
> easier to search
> >         - some commit are more descriptive "merge branch
> fperez/fix-bug-6374 into master" is better than "merge branch
> ellisonbg/master into master"
> >         - …
> >
> >
> > Also technically you can cross-push branches, ($ git push github
> local-name:remote-name)
> > so it happen to me to work locally on master and to push on gihub with
> another name,
> > but those are details that we shouldn't expose to newcomers.
> >
> > This in one of the reason our doc says (or should say at least) that it
> is preferable to make the changes
> > in named branches which names are as explicit as possible to help future
> workflow. It is not strictly enforced, but it's a good habit.
>
> Great, thanks very much for clarifying.
>
> Cyrille
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ipython-dev/attachments/20130919/f04704fa/attachment.html>


More information about the IPython-dev mailing list