apex-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chetan Narsude <che...@datatorrent.com>
Subject Re: Assigning pull requests
Date Tue, 08 Sep 2015 23:59:39 GMT
Thanks for the detailed answer, Chris. It does help to make sure that we
are not reinventing the wheel.

We are relying heavily on Jira to have the open technical communication
before the development happens. Good point on closing it with the github
comments as well when necessary.


On Tue, Sep 8, 2015 at 4:23 PM, Chris Nauroth <cnauroth@hortonworks.com>

> Spark is the only other Apache project I'm aware of that makes heavy use
> of GitHub pull requests.  I did a quick scan of their pull requests, and
> just as you noted, it appears the assignee field is unusable.
> https://github.com/apache/spark/pulls
> Their contribution guide also advises using "@username" notation in
> comments to request reviews from engineers who have worked on the same
> code in the past.
> https://cwiki.apache.org/confluence/display/SPARK/Contributing+to+Spark#Con
> tributingtoSpark-PullRequest
> I'd say it's safe to assume this is an acceptable norm and proceed with
> this as the process for Apex.  Of course, you're also empowered to try
> other things too if any Apex community members have alternative proposals.
>  Apache does relatively little to dictate specific technical choices, and
> instead defers to the individual communities to make the best choices for
> themselves.
> One thing I will advise though is that we seek to use JIRA as the primary
> means of communication.  Please make sure each pull request has a
> corresponding JIRA issue, and use the JIRA issue for significant technical
> discussions.  If discussion starting on a pull request results in a
> significant technical decision for the project, then please at least
> summarize that decision back on the corresponding JIRA issue.  (I don't
> think it's necessary to echo every pull request comment back to JIRA
> though, especially if it's minor spot comments on individual lines of
> code.)
> If your JIRA issues contain detailed discussion between community members,
> then this will be more visible to IPMC members.  This will make it easier
> for them to see that Apex strives to satisfy the openness criteria to
> qualify for eventual graduation.  Of course, all of this requires that the
> JIRA setup be completed in Apache infrastructure first.  :-)  I plan to
> follow up again if we don't hear anything by tomorrow.
> I hope this helps.
> --Chris Nauroth
> On 9/8/15, 3:51 PM, "Chetan Narsude" <chetan@datatorrent.com> wrote:
> >Sounds like a good idea.
> >
> >Is where we could follow the established norm though? I checked httpd,
> >hadoop but could not really find an example on outstanding pull request
> >where such a notation is followed.
> >
> >If no such norm, wondering if it's intentional. We cannot be the first
> >project to have this issue.
> >
> >--
> >Chetan
> >
> >
> >
> >
> >On Tue, Sep 8, 2015 at 3:15 PM, Thomas Weise <thomas@datatorrent.com>
> >wrote:
> >
> >> You may have noticed that after switch to the ASF repositories, there
> >>is no
> >> option to assign pull requests for review (tied to write access to the
> >> repository).
> >>
> >> As workaround, I thought that a comment on the pull request like
> >>"@username
> >> please review" could be used.
> >>
> >> Any other ideas?
> >>

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message